Split (1)

train · ~481k rows (showing the first 208k)

text stringlengths 1.36k 1.27M
Performance Models for Data Transfers: A Case Study with Molecular Chemistry Kernels Suraj Kumar\textsuperscript{1}, Lionel Eyraud-Dubois\textsuperscript{2}, and Sriram Krishnamoorthy\textsuperscript{1} \textsuperscript{1} Pacific Northwest National Laboratory, Richland, Washington, USA \{suraj.kumar, sriram\}@pnnl.gov \textsuperscript{2} Inria Bordeaux – Sud-Ouest, Université de Bordeaux, France email@example.com Abstract. With increasing complexity of hardwares, systems with different memory nodes are ubiquitous in High Performance Computing (HPC). It is paramount to develop strategies to overlap the data transfers between memory nodes with computations in order to exploit the full potential of these systems. In this article, we consider the problem of deciding the order of data transfers between two memory nodes for a set of independent tasks with the objective to minimize the makespan. We prove that with limited memory capacity, obtaining the optimal order of data transfers is a NP-complete problem. We propose several heuristics for this problem and provide details about their favorable situations. We present an analysis of our heuristics on traces, obtained by running 2 molecular chemistry kernels, namely, Hartree–Fock (HF) and Coupled Cluster Single Double (CCSD) on 10 nodes of an HPC system. Our results show that some of our heuristics achieve significant overlap for moderate memory capacities and are very close to the lower bound of makespan. Keywords: Communication Scheduling · Memory Nodes · Runtime Systems · Communication-Computation Overlap · Molecular Chemistry. 1 Introduction With the advent of multicore, and the use of accelerators, it is notoriously cumbersome to exploit the full capability of a machine. Indeed, there are several challenges that come into picture. First, every architecture provides its own efficacy and interface. Therefore, a steep learning curve is required for programmers to take good utilization of all resources. Second, scheduling is a well known NP-Complete optimization problem, and hybrid and distributed resources make this problem harder (we refer [1] for a survey on the complexity of scheduling problems and [2] for a recent survey in the case of hybrid nodes). Third, due to shared buses and parallel resources, it is challenging to obtain a precise model based on prediction of computation and communication times. Fourth, the number of architectures has increased drastically in recent years, therefore it is almost impossible to develop hand tuned optimized code for all these architectures. All these observations led to the development of different task based runtime systems. Among several runtimes, we may cite QUARK [3] and PaRSEC [4] from ICL, Univ. of Tennessee Knoxville (USA), StarPU [5] form Inria Bordeaux (France), Legion [6] from Stanford Univ. (USA), StarSs [7] from Barcelona Supercomputing Center (Spain), KAAPI [8] from Inria Grenoble (France). All these runtime systems allow programmers to express their algorithms at the abstract level in the form of direct acyclic graphs (DAG), where vertices represent computations and edges represent dependencies among them. Sometimes some static information is also provided along with the DAG, such as distance to exit (last) node as a priority or affinity of computation towards resources. The runtime is then responsible for managing scheduling of computations and communications, data transfers among different memories, computation-communication overlap, and load balance. In the last few decades, we have witnessed a drastic improvement in the hardware to provide higher rate of computation, but comparatively smaller improvement has been achieved for the rate of data movement. With extreme scale computing, supercomputers face bottlenecks due to the need of large amount of data [9,10]. Therefore, the HPC community is now focusing on avoiding, hiding and minimizing communication costs. Certain applications such as dense linear algebra kernels have regular structure. Therefore, it is possible to associate priorities to computations, based on the task graph structure, and to use them at runtime to make the execution efficient. In irregular applications, programmers do not know the precise structure of the task graphs in advance: tasks are added recursively based on certain sentinel constraints. For such applications, the runtime system sees a set of independent tasks and schedules them on different processing units. It is extremely important for runtimes to decide the order of data transfers for these scheduled computations so as to maximize the overlap between computations and communications. This is the main topic of this article. We prove that the order of communications on two memory nodes with the objective of minimizing the makespan is a NP-Complete problem if the memory of the target node is limited. Our proof is inspired from work by [11], which applies a similar technique for 2-machine flowshop problem with bounded capacity. The main difference between both approaches is that they consider all tasks have the same occupation on the second machine and the memory occupation starts when the processing finishes on the first machine. On the contrary, our approach is designed for tasks appearing in scientific workloads whose memory requirements are highly irregular and we consider that memory is acquired before starting the data transfer on the communication resource. We propose different runtime strategies in order to maximize the overlap of computations and communications. We evaluate our strategies on the context of a cluster of homogeneous nodes. However, our approach is generic and easily adaptable to any system which operates on different memory spaces. Here are the important contributions of this article: - NP-Completeness proof for the general data-transfer problem • Proposed different scheduling strategies with the objective to minimize the makespan • Linear programming formulation of the problem • Numerous experiments to assess the effectiveness of our strategies on molecular chemistry kernels The outline of the article is the following. Section 2 describes past work on the computations with limited memory and similar problems in the literature. In section 3, we present an algorithm to obtain the order of data transfers when there is not any memory capacity restriction. Then, we also prove that in general data transfer problem is NP-complete. In Section 4, we propose several heuristics and describe their favorable scenarios. We mainly consider three categories of heuristics: static heuristics, dynamic heuristics and static heuristics with dynamic corrections. Sections 5 describes our experimental setup and we evaluate our proposed strategies on two molecular chemistry kernels in Section 6. Our results show that static heuristics with dynamic corrections achieve good performance in most cases. We finally propose conclusions and perspectives in Section 7. 2 Related Work Historically there has been a great emphasis on the development of parallel algorithms and minimizing the complexity of computations. As the number of computation cores has increased drastically in recent years, supercomputers face bottleneck due to communication required by an application. Hence, in recent years the focus has changed towards developing communication avoiding algorithms, strategies to hiding communications and minimizing the data accessed by applications [10]. The problem of scheduling tasks has been highly studied in the literature and many formulations are known to be NP-Complete [12]. Many static and dynamic strategies have been proposed and analyzed for scheduling tasks on heterogeneous resources [13,14,15]. There is also a number of studies in the direction of task scheduling with the emphasis on improving locality and minimizing the communication cost [5,13]. Stanisic et. al [16] proposed a heuristic to schedule tasks on a computational resource where most of its data is available. A similar approach has been adopted by Agullo et. al for the scheduling of sparse linear algebra kernels [17]. Predari et. al proposed heuristics to partition the task graph across a number of processors such that inter-processor communication can be minimized [18]. The problem considered in this article also can be viewed as a flow shop problem: the communication link can be seen as a processing resource, and each task needs to first be handled by the communication link and then by the computational resource. Communication and computation times of a task can thus be considered as processing times on different machines. Johnson has provided scheduling strategies for 2 and 3-machine flow shop problems with infinite memory capacity [19]. 2-machine flow shop problem with finite buffer has been proven NP-Complete by Papadimitriou et. al [11], in which a constraint is imposed on the number of tasks that can await execution on the second machine. A number of other studies have focused on scheduling with limited memory and storage, starting with the work of register allocation for arithmetic expressions by Sethi and Ulman [20]. Sarkar et. al worked on the scheduling of graphs of smaller-grain tasks with limited memory, where each task requires homogeneous data size [21]. The same work has been extended by Marchal et. al for task graphs where memory requirement of each task is highly irregular [22]. 3 Data Transfer Problem Formulation To exploit the full potential of a system it may be necessary to execute tasks on processing units where all of their data does not reside. A task may require all of its input data in local memory before starting the computation. There may be multiple tasks scheduled on a processing unit, which require to transfer data from the same memory node. Ordering data transfers for such tasks is very crucial for the communication-computation overlap, thus for the overall performance. In general, order of task execution with input and output data transfers can be viewed as a 3-machine flowshop problem, where processing time on the first machine is input data transfer time, processing time on the second machine is task computation time, and processing time on the third machine is output data transfer time; and the objective is to minimize the total makespan. This is a well known NP-complete problem [23]. In many cases, output data that needs to be retrieved after task execution is much smaller than the input data. It is often the case that future tasks running on the same memory node require output data of the past tasks. Therefore, most runtime systems transfer data to other memory nodes based on the demand – not immediately after they were produced. It is also possible that all output data can be stored in preallocated separate buffer on a memory node. Hence, we do not consider output data explicitly in our analysis and assume that output data is negligible or stored in a separate buffer for each task. Thus problem considered here is more similar to a 2-machine flowshop problem. We prove that ordering the execution of such tasks with finite memory capacity is a NP-complete problem: Problem DT: A set of tasks $ST = \{T_1, \cdots, T_n\}$ is scheduled on a processing unit $P$ with memory unit $M$ of capacity $C$. Input data for tasks of $ST$ reside on another memory unit $M'$. $CM_i$ is the communication time to transfer input data from $M'$ to $M$ for task $i$ and $CP_i$ is the computation time of task $i$ on $P$. We assume that these tasks do not produce any output data. There can be only one communication at a time, and $P$ can only process one task at a time. A task uses an amount of memory in $M$ from the start of its communication to the end of its computation. Given $L$, is there a feasible schedule $S$ for $ST$ such that makespan of $S$, $\mu(S) \leq L$? Given a schedule, $S_{COM}(i)$ and $S_{COMP}(i)$ represent the start times of task $i$ on communication and computation resources. A schedule is feasible if for every time $t$, the amount of memory required by all tasks such that $ S_{\text{COMM}}(i) \leq t \leq S_{\text{COMP}}(i) + CP_i $ is not more than the memory capacity $ C $. For simplicity, we assume throughout the article that tasks require memory only to store their input data, and thus that the amount of memory required by a task is proportional to its communication time. Without loss of generality, we consider in all examples of Sections 3 and 4 that the memory requirement of a task is equal to its communication time. We call a task $ i $ compute intensive if $ CP_i \geq CM_i $, and communication intensive otherwise. ### 3.1 Special case: Infinite Memory When the computational resource has a very large memory, our problem becomes a classic 2-machine flowshop problem: communication time is the processing time on the first machine and computation time is the processing time on the second machine. Johnson’s algorithm [19] is known to provide an ordering for the tasks which results in an optimal makespan. This algorithm is rewritten in Algorithm 1. Algorithm 1: Johnson’s [19] algorithm (infinite memory case). ``` 1: Divide ready tasks in two sets $ S_1 $ and $ S_2 $. If computation time of a task $ T $ is not less than its communication time, then $ T $ is in $ S_1 $ otherwise in $ S_2 $. 2: Sort $ S_1 $ in queue $ Q $ by non-decreasing communication times 3: Sort $ S_2 $ in queue $ Q' $ by non-increasing computation times 4: Append $ Q' $ to $ Q $ 5: $ \tau_{\text{COMM}} \leftarrow 0 $ \hspace{1cm} \{Available time of communication resource\} 6: $ \tau_{\text{COMP}} \leftarrow 0 $ \hspace{1cm} \{Available time of computation resource\} 7: while $ Q \neq \emptyset $ do 8: Remove a task $ T $ from beginning of $ Q $ for processing 9: $ S_{\text{COMM}}(T) \leftarrow \tau_{\text{COMM}} $ 10: $ S_{\text{COMP}}(T) \leftarrow \max(S_{\text{COMM}}(T) + CM_T, \tau_{\text{COMP}}) $ 11: $ \tau_{\text{COMM}} \leftarrow S_{\text{COMM}}(T) + CM_T $ 12: $ \tau_{\text{COMP}} \leftarrow S_{\text{COMP}}(T) + CP_T $ 13: end while ``` We prove optimality of Algorithm 1 differently. Our proof rely on the following lemma. Lemma 1. Swapping two contiguous tasks $ A $ and $ B $ in a schedule does not improve the makespan if one of the conditions is true. - $ i) $ $ CP_A \geq CM_A, CP_B \geq CM_B, CM_A \leq CM_B $ - $ ii) $ $ CP_A < CM_A, CP_B < CM_B, CP_A \geq CP_B $ - $ iii) $ $ CP_A \geq CM_A, CP_B < CM_B $ Proof. Let $t_1$ and $t_2$ be the early start time on communication and computation resources just before the task $A$ starts. We write the following constraints based on the two schedules of the Fig 1. \[ S_{\text{COMP}}(A) = \max(t_1 + CM_A, t_2) \] \[ S_{\text{COMP}}(B) = \max(S_{\text{COMP}}(A) + CP_A, t_1 + CM_A + CM_B) \] Completion time of $B$ in original Schedule = $S_{\text{COMM}}(B) + CP_B$ Completion time of $B$ in swapped Schedule, $S'_{\text{COMP}}(B) = \max(t_1 + CM_B, t_2)$ Completion time of $A$ in swapped Schedule, $S'_{\text{COMP}}(A)$ \[ = \max(S'_{\text{COMP}}(B) + CP_B, t_1 + CM_B + CM_A) \] In both schedules, early available time on communication resource after scheduling $A$ and $B$ is same. If we show that early available time on computation resource in swapped schedule after scheduling both tasks is not less than the time of original schedule, then the proof is complete. Hence our goal is to prove that, \[ S_{\text{COMP}}(B) + CP_B \leq S'_{\text{COMP}}(A) + CP_A \] \[ S_{\text{COMP}}(B) + CP_B = \max(S_{\text{COMP}}(A) + CP_A + CP_B, t_1 + CM_A + CM_B + CP_B) \] \[ = \max(t_1 + CM_A + CP_A + CP_B, t_2 + CP_A + CP_B, t_1 + CM_A + CM_B + CP_B) \] Case I: $ CP_A \geq CM_A, CP_B \geq CM_B, CM_A \leq CM_B $ \[ S_{\text{COMP}}(B) + CP_B = \max(t_1 + CM_A + CP_A + CP_B, t_2 + CP_A + CP_B, t_1 + CM_A + CM_B + CP_B) \] \[ \leq \max(t_1 + CM_A + CP_B, t_2 + CP_B, t_1 + CM_B + CP_B) + CP_A \] \[ = \max(t_1 + CM_B + CP_B, t_2 + CP_B) + CP_A \] \[ = \max(\max(t_1 + CM_B, t_2) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = \max(S_{\text{COMP}}(B) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = S'_{\text{COMP}}(A) + CP_A \] Case II: $ CP_A < CM_A, CP_B < CM_B, CP_A \geq CP_B $ \[ S_{\text{COMP}}(B) + CP_B = \max(t_1 + CM_A + CP_A + CP_B, t_2 + CP_A + CP_B, t_1 + CM_A + CM_B + CP_B) \] \[ \leq \max(t_1 + CM_A + CP_B, t_2 + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = \max(t_1 + CM_A + CM_B, t_2 + CP_B) + CP_A \] \[ = \max(t_1 + CM_B + CP_B, t_2 + CP_B, t_1 + CM(A) + CM_B) + CP_A \] \[ = \max(\max(t_1 + CM_B, t_2) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = \max(S_{\text{COMP}}(B) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = S'_{\text{COMP}}(A) + CP_A \] Case III: $ CP_A \geq CM_A, CP_B < CM_B $ \[ S_{\text{COMP}}(B) + CP_B = \max(t_1 + CM_A + CP_A + CP_B, t_2 + CP_A + CP_B, t_1 + CM_A + CM_B + CP_B) \] \[ \leq \max(t_1 + CM_A + CP_B, t_2 + CP_B, t_1 + CM_B + CP_B) + CP_A \] \[ \leq \max(t_1 + CM_A + CM_B, t_2 + CP_B, t_1 + CM_B + CP_B) + CP_A \] \[ = \max(\max(t_1 + CM_B, t_2) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = \max(S_{\text{COMP}}(B) + CP_B, t_1 + CM_A + CM_B) + CP_A \] \[ = S'_{\text{COMP}}(A) + CP_A \] Theorem 1. Scheduled constructed by Algorithm 1 achieves optimal makespan. Proof. Let $ O $ be an optimal schedule. We assume that $ O $ is a permutation schedule. If it is not the case, we make the order of computations same as the order of communications without increasing the makespan. Suppose two tasks have opposite order on both resources then we position the second task just before the first task on the computation resource. It is evident that this change does not alter the optimal makespan, we repeat this procedure until order of communications and computations is same in $O$. Let $S$ be the schedule obtained from Algorithm 1. We prove the theorem by converting $S$ to $O$ and showing that at each step makespan of intermediate schedule is not less than the original makespan. We rely on Lemma 1 to convert $S$ to $O$. We traverse schedule $O$ from left to right and for each $i$th task in sequence, we apply Lemma 1 repetitively until order of task in the swapped schedule is same. It is obvious that after moving the $i$th task at the beginning remaining schedule (schedule after $i$th task) still satisfies one of the conditions of Lemma 1. Let the final swapped schedule is $S_{final}$. From Lemma 1, $\text{makespan}(S_{final}) \geq \text{makespan}(S)$. From the construction, $\text{makespan}(S_{final}) \leq \text{makespan}(O)$. As $O$ is an optimal schedule, hence $\text{makespan}(S) = \text{makespan}(O)$. This completes our proof. ### 3.2 Finite Memory We now consider the general case, in which the memory limit is a constraint for the schedule. This is related to previous work by Papadimitriou et. al [11], in which the second machine can only handle a bounded number of tasks. Our problem generalizes this work to heterogeneous memory consumption among tasks, with an additional difference: memory usage starts at the beginning of the first part of a task (instead of at the end of the first part). This requires to provide a slightly different NP-completeness proof, as given below. Theorem 2. Problem $DT$ is NP-complete. Proof. It is easy to see that the $DT$ belongs in NP: given a schedule, one can check in linear time that at each start of a communication, the memory constraint is satisfied, and that task starts computation only after its input data is transferred to $M$. In order to prove NP-hardness, we use a reduction from the well-known NP-complete problem 3 Partition [12]. Three Partition Problem (3PAR): Given a set of $3m$ integers $A = \{a_1, \cdots, a_{3m}\}$, is there a partition of $A$ into $m$ triplets $TR_i = \{a_{i_1}, a_{i_2}, a_{i_3}\}$, such that $\forall i, a_{i_1} + a_{i_2} + a_{i_3} = b$, where $b = (1/m) \sum a_i$? Let us first show that 3PAR problem reduces in polynomial time to problem $DT$. Suppose that we are given an instance $A = \{a_1, \cdots, a_{3m}\}$ of 3PAR. It is immediately obvious that $a_i > 1$, since we can always add sufficiently large integers to the $a_i$ values and scale the problem accordingly. This scaling will not affect in any way the existence of a solution for the instance of 3PAR problem. From such an instance, we define $x = \max\{a_i : 1 \leq i \leq 3m\}$, and we construct an instance $I$ of the problem $DT$ with $4m + 1$ tasks, whose characteristics are given in Table 1. As mentioned previously, we consider memory requirement of a task is equal to its communication time. If it is not the case, we can adjust $C$ such that at any point in a schedule at max one $K_i$ and three $A_i$ tasks can be active. \| Task \| Communication time \| Computation time \| \|------------\|--------------------\|------------------\| \| $K_0$ \| 0 \| 3 \| \| $K_1, \ldots, K_{m-1}$ \| $b' = b + 6x$ \| 3 \| \| $K_m$ \| $b' = b + 6x$ \| 0 \| \| $1 \leq i \leq 3m, A_i$ \| 1 \| $a'_i = a_i + 2x$\| Memory capacity: $C = b' + 3$ Target makespan: $L = m(b' + 3)$ Table 1: Definition of tasks in the reduction from 3Par. We show that $I$ has a schedule $S$ with makespan at most $L$ if and only if the original 3Par instance has a solution. Notice that the sum of communication times and the sum of computation times are both equal to $L$, therefore a valid schedule of makespan at most $L$ has makespan exactly $L$, with no idle time on both resources. It indicates that the first task is $K_0$ and the last task is $K_m$. ![Pattern of feasible schedule $S$.](image) Fig. 2: Pattern of feasible schedule $S$. If the 3Par instance has a solution, $A$ can be partitioned into $m$ triplets $TR_i = \{a_{i_1}, a_{i_2}, a_{i_3}\}$ such that $\forall i, a_{i_1} + a_{i_2} + a_{i_3} = b$, then we can construct a feasible schedule $S$ without idle times by the pattern depicted in Figure 2. The communications of tasks in $TR_i$ take place during the computation of task $K_{i-1}$, and the computations of tasks in $TR_i$ take place during the communication of task $K_i$. Since the memory capacity is $C = b' + 3$, all tasks from a triplet can fit in memory with a task $K_i$, and their durations are exactly equal to the communication time of $K_i$. This schedule is thus feasible, and has length exactly $L$. We now prove that any feasible schedule of $I$ corresponds to a valid decomposition of $A$ for 3Par. Indeed, we argue that every feasible schedule has to consist of $m$ segments like the one shown in Figure 2. Each segment provides a triplet $\{a_{i_1}, a_{i_2}, a_{i_3}\}$ such that $a_{i_1} + a_{i_2} + a_{i_3} = b$. Any schedule $S$ of $I$ having no idle time must start with $K_0$. We first show that no other $K_i$ task can be active with $K_0$, otherwise we would get idle time on computation resource. Indeed, the communication of such a task $K_i$ would end at time at least $b' > 3 + 6x$, but at most two $A_i$ tasks can be computed, and they end at time at most $3 + 2\max\{a'_i : 1 \leq i \leq 3m\} = 3 + 6x$. Hence three $A_i$ tasks must follow $K_0$. The memory requirement of other $K_i$ tasks is $b'$ and $2b' > C$, therefore at any point in the schedule at most one $K_i$ task can be active. Since the total duration of all $K_i$ tasks is $3 + (m - 1)(b' + 3) + b' = m(b' + 3) = L$, at each point in $S$ exactly one $K_i$ task is active. With these $K_i$ tasks in place, the schedule on the computation resource contains $m$ slots of length exactly $b'$, in which all $A_i$ tasks must fit without preemption. We can thus define triplet $TR_i$ as the set of tasks which execute during the communication phase of task $K_i$. Since at each point in $S$, exactly one $K_i$ task is active, and since $S$ has no idle time on the computation resource, the total computation time of tasks in $TR_i$ is exactly $b'$, and thus $a_{i_1} + a_{i_2} + a_{i_3} = b$. This partition is thus a valid solution for the 3Par instance $A$. This theorem shows that adding a memory constraint to our problem makes it more difficult. One additional difficulty compared to infinite memory capacity 2-machine flowshop [19] is that it may not be optimal to consider the same ordering on both machines: Proposition 1. There exists an instance of DT for which in all optimal schedules, the communication order of tasks is different from their computation order. \| Task \| Memory Req =Comm Volume \| Comm Time \| Comp Time \| \|------\|--------------------------\|-----------\|-----------\| \| A \| 0 \| 0 \| 5 \| \| B \| 4 \| 4 \| 3 \| \| C \| 1 \| 1 \| 6 \| \| D \| 3 \| 3 \| 7 \| \| E \| 6 \| 6 \| 0.5 \| \| F \| 7 \| 7 \| 0.5 \| Table 2: Example instance where ordering on both resources has to be different. Proof. Consider the instance described on Table 2, in which memory capacity is $C = 10$. Figure 3a shows the best possible schedule when tasks are scheduled in the same order on both resources (obtained by exhaustive search). On the other hand, Figure 3b shows another schedule with lower makespan, in which the order is different. In the infinite memory case, the standard proof that an optimal schedule exists with the same order on both resources claims that it is possible to swap two tasks which do not satisfy this property. On Figure 3, this would mean swapping tasks $D$ and $E$. But the communication of task $E$ can not start earlier because it would not fit in memory with tasks $B$ and $C$, and delaying the computation of task $E$ after task $D$ would delay task $F$ because $E$ and $F$ do not fit in memory together. We can see that this claim does not hold in the constrained memory case. 4 Data Transfer Order heuristics Algorithm 1 presented in Section 3 achieves an optimal makespan when there is no memory constraint. This optimal value indicates a lower bound on the makespan of the constrained case. We denote this value with optimal makespan infinite memory (OMIM). In the present Section, we propose different heuristics for the limited memory case, and we assess their efficiency with respect to this lower bound in Section 6. We classify our heuristics into mainly three categories. In the first category, the order of all computations and communications is computed in advance and the same order is followed on both resources. In the second category, the next task to schedule is dynamically chosen based on different criteria. The final category is based on combining strategies from the first two categories: a static ordering is precomputed and corrected dynamically to avoid idle time caused by memory limitations. In all of our strategies (except linear programming based strategy), communication and computations take place in the same order. 4.1 Static Ordering In this class of strategies, we compute the order of processing in advance based on criteria such as communication time and computation time. After computing the order, we follow the same sequence on computation and communication resources and make sure that the memory constraint is respected at each point in the schedule. In Algorithm 1, compute intensive tasks are sorted in increasing order of communication times. It allows tasks to utilize the computation resource maximally and make enough margin on the communication resource to accommodate more communication intensive tasks with maximum overlap. Communication intensive tasks are sorted in decreasing order of computation time, which allows tasks to utilize the margin created on communication resource. Hence, in this section, we obtain the orders by sorting tasks based on different combinations of communication and computation times. i) \textit{order of optimal strategy infinite memory (OOSIM)}: This heuristic uses the order given by Algorithm 1, but respects the memory constraint at each point in the schedule. Hence the makespan of this heuristic may be completely different from \textit{OMIM}. ii) \textit{increasing order of communication strategy (IOCMS)}: Tasks are ordered in non-decreasing order of communication time. iii) \textit{decreasing order of computation strategy (DOCPS)}: Tasks are ordered in non-increasing order of computation time. iv) \textit{increasing order of communication plus computation strategy (IOCCS)}: Tasks are ordered in non-decreasing order of the sum of their communication and computation times. v) \textit{decreasing order of communication plus computation strategy (DOCCS)}: Tasks are ordered in non-increasing order of the sum of their communication and computation times. \begin{table}[h] \centering \begin{tabular}{\|c\|c\|c\|c\|} \hline Task & Memory Req = Comm Volume & Comm Time & Comp Time \\ \hline A & 3 & 3 & 2 \\ B & 1 & 1 & 3 \\ C & 4 & 4 & 4 \\ D & 2 & 2 & 1 \\ \hline \end{tabular} \caption{A task set for static order schedules.} \end{table} In order to highlight the different behaviors of these static strategies, we propose on Table 3 an example instance, and on Figure 4 the corresponding schedules for all these heuristics. \subsection{Dynamic Selection} Dynamic strategies are based on the following principle: when the communication resource is idle, a task is chosen based on a selection criterion which differs depending on the heuristic, among those which fit in memory and induce minimum idle time on the computation resource. For example, if the selection criterion is to choose a highly compute intensive task, then we compute the ratio of computation time and communication time for all tasks, and we select a task with the maximum ratio among those which induce minimum idle time on the computation resource and fit in the currently available memory. If no task fits in memory then we leave the resource idle at that point and proceed to the next event point. We also ensure that the order on both resources is the same, by ordering tasks on the computation resource accordingly. i) \textit{largest communication task respects memory restriction (LCMR)}: A task with the largest communication time is chosen. ii) \textit{smallest communication task respects memory restriction (SCMR)}: A task with the smallest communication time is chosen. iii) \textit{maximum accelerated task respects memory restriction (MAMR)}: A task with the maximum ratio of computation time to communication time is chosen. We highlight the different dynamic heuristics with the instance described on Table 4 (these heuristics are too similar on the instance from the previous class), and Figure 5 shows the corresponding schedules. \subsection{Static Order with Dynamic Corrections} In this class of strategy, we precompute the order of tasks based on some criterion and then follow this ordering as much as possible. But when the communication \| Task \| Memory Req =Comm Volume \| Comm Time \| Comp Time \| \|------\|--------------------------\|-----------\|-----------\| \| A \| 3 \| 3 \| 2 \| \| B \| 1 \| 1 \| 6 \| \| C \| 4 \| 4 \| 6 \| \| D \| 5 \| 5 \| 1 \| Table 4: A task set for dynamic schedules. ![Different dynamic heuristic schedules](image) Fig. 5: Different dynamic heuristic schedules for a task set of Table 4 with a memory capacity of 6. resource is idle because the memory requirement of the next task is too high, then we select a task with a dynamic strategy. After a task is dynamically selected, we update the remaining order without this task. This class of strategy takes advantage of static information in the form of precomputed order and dynamic corrections to minimize the idle time due to memory constraint. In all strategies of this class, the initial order is OMIM order, obtained by Algorithm 1. We define the following heuristics based on how we select a task from the set of tasks which fit in memory and induce minimum idle time on the computation resource. If no task fits in memory then we leave the resource idle and forward to the next event point. i) optimal order infinite memory largest communication task respects memory restriction (OOLCMR): A task with the largest communication time is chosen from the set. ii) optimal order infinite memory smallest communication task respects memory restriction (OOSCMR): A task with the smallest communication time is chosen from the set. iii) optimal order infinite memory maximum accelerated task respects memory restriction (OOMAMR): A task with the highest ratio of computation time to communication time is chosen from the set. \| Task \| Memory Req =Comm Volume \| Comm Time \| Comp Time \| \|------\|--------------------------\|-----------\|-----------\| \| A \| 4 \| 4 \| 1 \| \| B \| 2 \| 2 \| 6 \| \| C \| 8 \| 8 \| 8 \| \| D \| 5 \| 5 \| 4 \| \| E \| 3 \| 3 \| 2 \| Table 5: A task set for static order dynamic corrections schedules. As previously, we propose on Table 5 an example instance for this class of strategies, and provide on Figure 6 the corresponding schedules. ### 4.4 Additional heuristics from previous work We also consider two other static heuristics for evaluation. The first heuristic is based on an algorithm, proposed by Gilmore and Gomory, to obtain the minimal cost sequence for a set of jobs [24]. This is a classical algorithm for 2-machine no-wait flow shop. In this algorithm, each job has a start and end state and a cost is associated to change the state. In our context, this cost can be seen as non-overlap time of computation for two adjacent tasks. Here is the main idea of this algorithm. Initially, a partial sequence of jobs is represented by a graph such that their overlap is maximum. Subsequently edges are greedily added to this graph to connect two components while minimizing the total non-overlap cost. When the graph is connected, an edge interchange mechanism is used to determine the sequence of jobs, which ensures that the sequence has minimal cost. More details can be found in the original paper [24] and our implementation is publicly available [25]. This algorithm does not take memory constraints into account and only provides the sequence of processing. We use this sequence with a memory capacity restriction just like for other static heuristics, and we call this heuristic Gilmore–Gomory (GG). The second heuristic is based on the First-Fit algorithm for the bin packing problem. The idea of this heuristic is to identify groups of tasks which can fit in memory together, called bins. In First-Fit, tasks are considered in an arbitrary order and added to the first bin in which they can fit. If no suitable bin is found then a new bin is created and this task is added to it. When all tasks have been assigned to bins, we consider the sequence made of all tasks from the first bin, then tasks for the second bin, and so on. We call this heuristic Bin Packing (BP). 4.5 Solving Linear Program Iteratively We use a mixed integer linear program to obtain the order of communications and computations. Recall that $CP_i$ and $CM_i$ represent computation and communication times of task $i$, and the memory capacity of the target system is $C$. In the linear program formulation, $s_i$ and $e_i$ (resp. $s'_i$ and $e'_i$) represent the start and end times of communication (resp. computation) for task $i$, and $MC(i)$ is the memory capacity requirement of task $i$. The formulation also contains for each pair of tasks $i$ and $j$ i) a boolean variable $a_{ij}$ to denote the order of $i$ and $j$ on the communication resource ii) a boolean variables $b_{ij}$ to denote the order of $i$ and $j$ on the computation resource, and iii) a boolean variables $c_{ij}$ to denote the order of $s_i$ and $e'_j$. Let $L = \sum_i (CP_i + CM_i)$. It is evident that $e_i = s_i + CM_i$ and $e'_i = s'_i + CP_i$. The linear program is given below. \[ \begin{align} \text{Minimize } l & \text{ subject to:} \\ \forall i, & \quad e'_i \leq l & \text{(task } i \text{ completes)} \\ \forall i, & \quad e_i \leq s'_i & \text{(task } i \text{ valid ordering)} \\ \forall i, \forall j \neq i, & \quad \begin{cases} e_j \leq s_i + (1 - a_{ij})L \\ e_i \leq s_j + a_{ij}L \end{cases} & \text{(exclusive use of communication link)} \\ \forall i, \forall j \neq i, & \quad \begin{cases} e'_j \leq s'_i + (1 - b_{ij})L \\ e'_i \leq s'_j + b_{ij}L \end{cases} & \text{(exclusive use of computation resource)} \\ \forall i, \forall j \neq i, & \quad \begin{cases} e'_j \leq s_i + (1 - c_{ij})L \\ s_i < e'_j + c_{ij}L \end{cases} & \text{(respect ordering of variables } c_{ij}) \\ \forall i, & \quad \sum_{r \neq i} (a_{ir} - c_{ir})MC(r) + MC(i) \leq C & \text{(memory constraint)} \end{align} \] We use GLPK solver v4.65 to solve the above formulation. We also add the following constraints to help the solver: $\forall i, \forall j \neq i$, $a_{ij} + a_{ji} = 1$, $b_{ij} + b_{ji} = 1$, The solver was unable to solve this MILP at the scale of our interest in limited time. Hence, we solve the linear program iteratively for a small subset of size $k = 3, 4, 5, 6$: at the boundary of two iterations we fix the event (communication or computation) of an unfinished task started before the boundary point and consider other events flexible. The subsets are formed in the order in which tasks are submitted, which is arbitrary. For a given size $k$, we represent the makespan calculated by this heuristic as $lp.k$. We compute various $lp.k$ values for different memory capacities and observe that most of the other heuristics perform better than this heuristic. Figure 7 shows the performance of different heuristics with MILP based heuristics for various memory capacities of a single trace file. ![Comparison of proposed heuristics heuristics with MILP solution based heuristic for different memory capacities of a single trace file](image) Fig. 7: Comparison of proposed heuristics heuristics with MILP solution based heuristic for different memory capacities of a single trace file. Here minimum memory requirement to process all tasks is $m_c = 176KB$. 4.6 Favorable Situations for Heuristics Based on the definition of proposed strategies and the optimality of Algorithm 1, we present the scenarios which should be more favorable for each heuristic in Table 6. This allows programmers to use appropriate strategies to maximize communication-computation overlap for their applications. In this table, “moderate memory capacity” refers to the case where memory is constrained, but close to the maximal memory requirement of the $OMIM$ schedule. \| Heuristic \| Favorable Situation \| \|-----------\|---------------------\| \| $OOSIM$ \| Memory capacity is not a restriction (Optimal) \| \| $IOCMS$ \| Memory capacity is not a restriction and tasks are compute intensive (Optimal) \| \| $DOCPS$ \| Memory capacity is not a restriction and tasks are communication intensive (Optimal) \| \| $IOCCS$ \| Moderate memory capacity and most tasks are highly compute intensive \| \| $DOCCS$ \| Moderate memory capacity and most tasks are highly communication intensive \| \| $LCMR$ \| Limited memory capacity and significant percentage of tasks with large communication times are compute intensive \| \| $SCMR$ \| Limited memory capacity and significant percentage of tasks with small communication times are compute intensive \| \| $MAMR$ \| Limited memory capacity and significant percentage of tasks of both types \| \| $OOLCMR$ \| Moderate memory capacity and significant percentage of tasks are communication intensive \| \| $OOSCMR$ \| Moderate memory capacity and significant percentage of tasks are compute intensive \| \| $OOMAMR$ \| Moderate memory capacity and significant percentage of highly compute and communication intensive tasks \| Table 6: Heuristics and their favorable scenarios. Some of these favorable scenarios can be clearly observed in our experimental results, on Figures 9 and 11. For example, HF compute intensive tasks have small communication times, which explains why the $SCMR$ heuristic exhibits very good performance in limited memory cases. CCSD has significant percentage of both types of tasks, and indeed the performance of $OOLCMR$ and $OOSCMR$ is very close to optimality in moderate memory cases. 5 Experimental Settings We consider a machine called Cascade [26], available at PNNL, for our experiments. We obtain traces by running two molecular chemistry applications, double precision version of HF and CCSD of NWChem [27] package on 10 nodes of this machine. Each node is composed of 16 Intel Xeon E5-2670 cores. NWChem takes advantages of a Partitioned Global Address Space Programming Model called Global Arrays (GA) [28] to use shared-memory programming APIs on distributed memory computers. GA dedicates one core of each node to handle other cores, hence we can view a node as being composed of 15 computational cores. We use 150 processes for each application and obtain 150 trace files. We run CCSD with Uracil molecules input and HF with SiOSi molecules (for Uracil molecules, HF has a much smaller workload, each processor executes only around 20 tasks, that is why we chose SiOSi input for HF execution). Each process executes around 300-800 tasks. Our data transfer model is quite simple and we consider that all data transfers between the local memory of each process and the GA memory take the same route. Modeling of different routes of data transfers for the same source-destination pair; bandwidth sharing for different source-destination pairs and network congestion is more challenging and part of our future work. This simple model is enough to provide insight to the application developers (or runtime system) about the ordering of data transfers for the same source-destination pair so as to maximize communication-computation overlap. Our model is easily adaptable to any source-destination pair when there is one fixed route between source and destination (such as between CPU and GPU, one copy engine to transfer data from CPU (resp. GPU) to GPU (resp. CPU)). Both applications mainly perform two types of computations, tensor transpose and tensor contraction. HF expects to specify a tile size and we set it to 100, so that each core can be used efficiently. CCSD automatically determines tile sizes at different program points based on the input molecules. Hence, HF operates on almost homogeneous tiles while CCSD uses more heterogeneous tiles. 5.1 Workload Characteristics To get more insights into the considered workloads, we provide information about the instances we consider in Figure 8. For each instance, we compute the sum of communication times ($sum\ comm$) and sum of computation times ($sum\ comp$), and normalize it relatively to the $OMIM$ value. Figure 8 also shows the maximum of both values, which is a lower bound on the possible makespan of a schedule for this instance, and their sum, which is an upper bound: this represents the makespan of the sequential schedule, obtained with zero overlap between computation and communication. We can see that HF is a communication intensive application and at most 20% overlap can be expected in the best scenario. On the other hand, in the CCSD workload, communications and computations are almost evenly distributed and a more significant overlap is possible. 6 Experimental Results We evaluate our scheduling heuristics for several memory capacities. From the obtained traces, we first determine the minimum requirement of the memory capacity $m_c$ to execute all tasks. Then we observe the behavior of all heuristics with memory capacity $m_c$ to $2m_c$, in increments of 0.125$m_c$. Our performance metric is the ratio to optimal $r$: if heuristic $H$ has makespan $M_H$ on an instance, and the optimal makespan for the infinite memory case is $OMIM$, then $r(H) = \frac{M_H}{OMIM}$ (lower values are better). This ratio is at least 1, and a value close to 1 indicates a well-suited heuristic which achieves maximum possible communication-computation overlap. Figures 9 and 11 show the distribution of the performance of each heuristic for different memory capacities, where plots are categorized by memory capacities. For each memory capacity and each heuristic, the box on the plot displays the median, first and last quartile, and the whiskers indicate minimum and maximum values, with outliers are shown by black dots. ### 6.1 HF Performance As indicated above, HF tasks operate on less heterogeneous tiles, this is also noticeable in Figure 9. All heuristics depict similar behavior for minimum memory capacity $m_c$ and increasing the memory capacity slightly does not change the performance of all heuristics. As memory capacity is increased further, dynamic variants of heuristics start performing better. For the moderate memory capacities (close to $2m_c$), static order with dynamic corrections variants outperform others. $GG$ heuristic does not achieve good performance, because its task sequence is obtained considering no extra memory is available, but is then applied in a different scenario where memory is limited. Surprisingly, the $BP$ heuristic which considers only memory constraint obtains good performance for a static heuristic, but is outperformed by more dynamic approaches. Figure 10 shows the performance comparison of the best variant in each category, in addition to the order of submission (OS) strategy which orders tasks in the (arbitrary) sequence in which they are given. Static strategies are expected to perform better when there is not any memory capacity restriction, and indeed this plot shows that static strategies face capacity bottleneck and underperform with limited memory. Dynamic strategies achieve slightly better performance with limited memory capacity, but when memory capacity is larger, static order with dynamic corrections strategies outperform all others. ### 6.2 CCSD Performance The CCSD application operates on tasks of different sizes, hence different heuristics exhibit distinct behaviors even at minimum memory capacity $m_c$. Heterogeneity favors dynamic strategies, as can be seen by the fact that both dynamic and static order with dynamic corrections based strategies perform better than static based strategies. Similar to HF, static order with dynamic corrections based strategies outperform others as memory capacity becomes moderate. Figure 12 shows that best variants of dynamic and static order with dynamic corrections strategies achieve similar performance at minimum memory capacity $m_c$. But as memory capacity increases, heterogeneity allows static order with dynamic corrections based strategies to take advantage of static knowledge to get maximum overlap and dynamic correction to select another task in case of memory capacity limitation. Static strategies also start performing better at the end, which indicates that this application has potential for significant communication-computation overlap and pure dynamic strategies are unable to take this information into account while making scheduling decisions. ### 6.3 Scheduling in Batches In most applications, the runtime scheduler may only observe a limited batch of independent tasks. Therefore we organize tasks of each trace file in batches of 100 (the last batch may have less than 100 tasks). We apply each heuristic on each group in succession. Figure 13 shows the performance of the best variants of each category for both applications. The plots exhibit behavior similar to Figures 10 and 12: static order with dynamic corrections variants attain maximum communication-computation overlap and outperform other heuristics. 7 Conclusion and Perspectives In this article, we consider the problem of deciding the order of data transfers between two memory nodes such that overlap of communications and computations is maximized. With Exascale computing, applications face bottlenecks due to communications. Hence, it is extremely important to achieve the maximum communication-computation overlap in order to exploit the full potential of the system. We show that determining the order of data transfers is a NP complete problem. We propose several data transfer heuristics and evaluate them on two molecular chemistry kernels, HF and CCSD. Our results show that some of our heuristics achieve significant overlap and perform very close to the lower bound. We plan to evaluate our strategies on different applications coming from multiple domains. We also plan to study the behavior of our strategies in the context of overlapping CPU-GPU communications with computations. A runtime system aiming at exposing different heuristics to maximize the communication-computation overlap at the developer level and automatically selecting the best one is currently underway. Fig. 12: Comparison of best variants of all categories for CCSD. References 1. P. Brucker and S. Knust, “Complexity results for scheduling problems,” Web document. URL: http://www2.informatik.uni-osnabrueck.de/knust/class/. 2. R. Bleuse, S. Kedad-Sidhoum, F. Monna, G. Mounié, and D. Trystram, “Scheduling independent tasks on multi-cores with gpu accelerators,” Concurrency and Computation: Practice and Experience, vol. 27, no. 6, pp. 1625–1638, 2015. 3. A. YarKhan, J. Kurzak, and J. Dongarra, QUARK Users’ Guide: QUeuing And Runtime for Kernels. UTK ICL, 2011. 4. G. Bosilca, A. Bouteiller, A. Danalis, M. Faverge, T. Héroult, and J. Dongarra, “PaRSEC: A programming paradigm exploiting heterogeneity for enhancing scalability,” Computing in Science and Engineering, 2013. 5. C. Angonnet, S. Thibault, R. Namyst, and P.-A. Wacrenier, “StarPU: A Unified Platform for Task Scheduling on Heterogeneous Multicore Architectures,” Concurrency and Computation: Practice and Experience, Special Issue: EuroPar 2009, vol. 23, pp. 187–198, Feb. 2011. [Online]. Available: http://hal.inria.fr/inria-00550877 6. M. Bauer, S. Treichler, E. Slaughter, and A. Aiken, “Legion: Expressing locality and independence with logical regions,” in Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, ser. SC ’12. Los Alamitos, CA, USA: IEEE Computer Society Press, 2012, Fig. 13: Best variants of all categories where heuristics are applied in the batches of 100 tasks. 7. A. Duran, E. Ayguadé, R. M. Badia, J. Labarta, L. Martinell, X. Martorell, and J. Planas, “Ompss: a proposal for programming heterogeneous multi-core architectures,” Parallel Processing Letters, vol. 21, no. 2, pp. 173–193, 2011. [Online]. Available: https://doi.org/10.1142/S012962641100151 8. E. Hermann, B. Raffin, F. Faure, T. Gautier, and J. Allard, “Multi-GPU and Multi-CPU Parallelization for Interactive Physics Simulations,” in Euro-Par (2), 2010, pp. 235–246. 9. “Top ten exascale research challenges,” ASCAC committee report, URL: https://science.energy.gov/~/media/ascr/ascac/pdf/meetings/20140210/Top10reportFEB14.pdf, 2014. 10. K. Yelick, “Avoiding, hiding and managing communication at the exascale,” 2016. [Online]. Available: https://people.eecs.berkeley.edu/~yelick/talks/exascale/Communication-Yelick-China16.pdf 11. C. H. Papadimitriou and P. C. Kanellakis, “Flowshop scheduling with limited temporary storage,” J. ACM, vol. 27, no. 3, pp. 533–549, Jul. 1980. [Online]. Available: http://doi.acm.org/10.1145/322203.322213 12. M. R. Garey and D. S. Johnson, Computers and Intractability, a Guide to the Theory of NP-Completeness. W.H. Freeman and Company, 1979. 13. H. Topcuoglu, S. Hariri, and M.-y. Wu, “Performance-Effective and Low-Complexity Task Scheduling for Heterogeneous Computing,” IEEE Trans. Parallel Distrib. Syst., vol. 13, no. 3, pp. 260–274, Mar. 2002. [Online]. Available: http://dx.doi.org/10.1109/71.993206 14. O. Beaumont, T. Cojean, L. Eyraud-Dubois, A. Guermouche, and S. Kumar, “Scheduling of Linear Algebra Kernels on Multiple Heterogeneous Resources,” in International Conference on High Performance Computing, Data, and Analytics (HPC 2016), Hyderabad, India, Dec. 2016. [Online]. Available: https://hal.inria.fr/hal-01361992 15. E. Agullo, O. Beaumont, L. Eyraud-Dubois, and S. Kumar, “Are Static Schedules so Bad? A Case Study on Cholesky Factorization,” in 2016 IEEE International Parallel and Distributed Processing Symposium, IPDPS 2016, Chicago, IL, USA, May 23-27, 2016, 2016, pp. 1021–1030. [Online]. Available: http://dx.doi.org/10.1109/IPDPS.2016.90 16. L. Stasić, S. Thibault, A. Legrand, B. Videau, and J.-F. Méhaut, “Modeling and simulation of a dynamic task-based runtime system for heterogeneous multicore architectures,” in Euro-Par 2014 Parallel Processing, F. Silva, I. Dutra, and V. Santos Costa, Eds. Cham: Springer International Publishing, 2014, pp. 50–62. 17. E. Agullo, B. Bramas, O. Coulaud, E. Darve, M. Messner, and T. Takahashi, “Task-based fmm for heterogeneous architectures,” Concurrency and Computation: Practice and Experience, vol. 28, no. 9, pp. 2608–2629, 2016. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/cpe.3723 18. M. Predari, “Load Balancing for Parallel Coupled Simulations,” Theses, Université de Bordeaux, LaBRI - Inria Bordeaux Sud-Ouest, Dec. 2016. [Online]. Available: https://hal.inria.fr/tel-01518956 19. S. M. Johnson, “Optimal two- and three-stage production schedules with setup times included,” Naval Research Logistics Quarterly, vol. 1, no. 1, pp. 61–68, 1954. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/nav.3800010110 20. R. Sethi and J. D. Ullman, “The generation of optimal code for arithmetic expressions,” J. ACM, vol. 17, no. 4, pp. 715–728, Oct. 1970. [Online]. Available: http://doi.acm.org/10.1145/321607.321620 21. D. Sbrilea, Z. Budimlić, and V. Sarkar, “Bounded memory scheduling of dynamic task graphs,” in 2014 23rd International Conference on Parallel Architecture and Compilation Techniques (PACT), Aug 2014, pp. 343–355. 22. L. Marchal, H. Nagy, B. Simon, and F. Vivien, “Parallel scheduling of dags under memory constraints,” in 2018 IEEE International Parallel and Distributed Processing Symposium (IPDPS), May 2018, pp. 204–213. 23. M. R. Garey, D. S. Johnson, and R. Sethi, “The complexity of flowshop and jobshop scheduling,” Math. Oper. Res., vol. 1, no. 2, pp. 117–129, May 1976. [Online]. Available: http://dx.doi.org/10.1287/moor.1.2.117 24. P. C. Gilmore and R. E. Gomory, “Sequencing a one state-variable machine: A solvable case of the traveling salesman problem,” Operations Research, vol. 12, no. 5, pp. 655–679, 1964. [Online]. Available: https://doi.org/10.1287/opre.12.5.655 25. “Communication scheduling,” https://github.com/surakuma/communication-scheduling, 2019. 26. “Computing: Cascade,” 2019. [Online]. Available: https://www.emsl.pnl.gov/emslweb/10.25582/inst.34218 27. “Nwchem: A comprehensive and scalable open-source solution for large scale molecular simulations,” Computer Physics Communications, vol. 181, no. 9, pp. 1477 – 1489, 2010. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S0010465510001438 28. J. Nieplocha, R. J. Harrison, and R. J. Littlefield, “Global arrays: A nonuniform memory access programming model for high-performance computers,” The Journal of Supercomputing, vol. 10, no. 2, pp. 169–189, Jun 1996. [Online]. Available: https://doi.org/10.1007/BF00130708
Chairman Mark Culver called the 10:00 A.M. meeting to order, established a quorum was present and welcomed everyone. Rev. Mike Pearson, Pastor, Lafayette Street United Methodist Church, gave the invocation. Commissioner Forrester led in the Pledge of Allegiance. Commissioner Cook made a motion to approve the minutes of the last meeting. Commissioner Harvey seconded; and the motion carried unanimously. Announcements and Public Comments from Chairman and Commissioners Chairman Culver reported he had already received a letter from the Salvation Army about the Christmas Kettle Community Project. He stated last year, the county challenged the city to raise money at the Kettle Drive, and there is a traveling trophy. The chairman reported the Salvation Army wants the commission to pick a date between November 18th and December 24th to adopt a kettle. He stated whichever department raises the most money can keep the plaque. Chairman Culver asked the commissioners to check their schedules and let their friends know when they would be ringing the bell and try to raise the most money. The chairman reported that he had received a letter from Paula Todd, President and CEO of the Jackson County Mississippi Chamber of Commerce, thanking the commission and everyone in our community who was involved in the Hurricane Katrina relief effort. Chairman Culver reported the public has seen press conferences on the Envision Project and he encouraged everyone to go to the website www.envisiondothan.com, and fill out the survey. He reported they wanted everyone who lives in the county to participate in the process and have input on the direction our community takes. Chairman Culver reported there will also be hard copies for anyone who wishes to fill them out. He pointed out the copies will be available in the Commission office as well as some of the surveys will be brought down to the Commission Chambers on meeting days. The chairman reported the Chattahoochee State Park has had logging activities, and he asked the public to be patient as they get the park back in the condition it needs to be and prepare it for disabled deer hunting. Chairman Culver reported they hoped to have safe conditions for the hunters within the next two weeks. Consent Agenda 1. Request to award bid on garbage truck. 2. Request from Sheriff to take bids on vehicles. Commissioner Cook made a motion to approve the consent agenda to award the bid on the garbage truck to the low bidder meeting specs, and to approve the request from the Sheriff to take bids on vehicles that are within the budget. Commissioner Harvey seconded; and the motion carried unanimously. (Please see attached bid summary sheet in Minute Book) Regular Agenda 1. Appointment – Houston County Department of Human Resources. Chairman Culver reported he had spoken with Mrs. Suzanne Geiger over the weekend and she is interested in serving and will do a great job. He stated she had a couple of questions and he had tried to call her back with the answers and she was not available. Chairman Culver reported he told her he hoped the Commission would go ahead and appoint her, and if she did not like the answers to her questions she would not have to accept the appointment. Commissioner Forrester made a motion to appoint Mrs. Suzanne Geiger to the Houston County Department of Human Resources Board. Commissioner Snellgrove seconded; and the motion carried unanimously. 2. Request to enter into NACO Prescription Drug Discount Program Contract. Chairman Culver reported at this point, the commission would be signing the contract and getting it back. He pointed out this is a discount program through NACO, National Association of Counties, and it will help some people get discounts on drugs in our community. He reported there will be no cost to the commission nor to the citizens. Commissioner Cook made a motion to approve the request to enter into the NACO Prescription Drug Discount Program Contract. Commissioner Harvey seconded; and the motion carried unanimously. 3. Request to approve lease with SE Alabama Regional Planning and Development Commission. Chairman Culver reported the SE Alabama Regional Planning and Development Commission has been a tenant for 10 years in the administrative building on the fourth floor. He stated it is time to renew the lease. Chairman Culver pointed out the commission has asked for an increase and their outgoing director thinks it is reasonable. Commissioner Snellgrove made a motion to grant the request to approve the lease with SE Alabama Regional Planning and Development Commission. Commissioner Harvey seconded; and the motion carried unanimously. 4. Request from Houston County In-Home Service to hire a senior aide on a permanent basis. Chairman Culver reported they discussed on Thursday the difficulty in providing the service in a certain area of the county. He stated they have someone who is currently doing that but the way the Senior Aide Program works this individual is supposed to be rotated off to another job after a certain period of time, and that time is up. He pointed out Mrs. Mary Lou Strickland, In-Home Service, has been unable to find someone to fill that position. Commissioner Harvey made a motion to approve the request from Houston County In-Home Service to hire a senior aide on a permanent basis. Commissioner Cook seconded the motion. Chairman Culver stated if this is done an accompanying budget amendment will be needed in the amount of $16,900. Commissioner Cook asked if anything extra was budgeted? Chairman Culver stated no. Commissioner Forrester asked if this was to replace someone who has been doing the job on a volunteer basis? Chairman Culver stated this is to hire the person who has been doing it under the Senior Aide Program. He reported it is not voluntary but it is paid through a program. He pointed out the requirements of the other program are that after a certain length of time, that person is either to be hired or he rotates into a different job. The chairman reported the purpose of the Senior Aide Program is to find employment for the seniors who are willing to work. Commissioner Cook asked if this is in addition to the one that was approved in the budget? Chairman Culver stated yes. He reported about six months ago the commission approved one aide that was in an area that they were having a hard time getting someone and this would be an additional person. Chairman Culver asked Mr. Roberts how many paid positions would this be? Mr. Roberts stated in the budget the director is full time and they have two people who appear to be included as part-time. Chairman Culver stated the amount of $16,900 seemed to be high. Mr. Roberts was informed it would not be full time and he stated it would be half the amount, $8,450.00. Mr. Roberts stated if they work no more than 39 hours then they will not have any benefits other than the taxes. He stated it would be around $6,500 - $7,000. Chairman Culver asked if they could make the budget amendment not to exceed $7,000 and find out about the hours? Mrs. Mary Lou Strickland reported the hours would be 19 hours a week. Chairman Culver reported it would not be for the whole budget year. Chairman Culver reported there is a motion to approve the request from Houston County In-Home Service to hire a senior aide on a part time permanent basis and the accompanying budget amendment not to exceed $7,000. Commissioner Snellgrove asked if the county was going to amend the budget for $7,000? Mr. Roberts reported he quoted the $16,900 because he thought it was a full time position. Mr. Snellgrove asked Mrs. Strickland how many people she has that volunteer to do the same service? Mrs. Strickland stated she did not have any at this time. He asked if this couple would cover the entire county? Mrs. Strickland stated they cover Cottonwood, Ashford, Gordon, and Pansey. Commissioner Snellgrove asked if those who cover the other areas of the county are paid or are volunteers? Mrs. Strickland stated they were paid. The Chairman called for the question; and the motion carried unanimously. (Please see attached budget amendment in Minute Book) 5. Request from Sheriff – amend budget. Chairman Culver stated this is a corrective action, and he asked Mr. Roberts if he knew the amount? Mr. Roberts did not know the exact amount but stated he thought it would be less than $.25 per hour. Commissioner Snellgrove made a motion to approve the request from the Sheriff to amend the budget. Commissioner Cook seconded; and the motion carried unanimously. (Please see attached in Minute Book) 6. Request from Landmark Park to use Omussee Creek Park as public boarding location for The Chattanooga Star. Commissioner Forrester made a motion to approve the request from Landmark Park to use Omussee Creek Park as the public boarding location for The Chattanooga Star. Commissioner Snellgrove seconded the motion. Mr. William Holman, Executive Director, Landmark Park, announced that The Chattanooga Star will be back this year providing rides on the Chattahoochee River. He stated over the last 3 years they have had over 7,000 school children to ride on the riverboat. He stated they are asking to use Omussee Creek Park as the public boarding location for the rides which will primarily be on the weekends for the public and during the week for the school children. He reported last year they had over 2,500 people ride The Chattanooga Star. Mr. Holman reported the rides will begin the first Saturday after Thanksgiving and go through the beginning of the year. He reported there will be some evening charters and they will have to be in the park after hours. Chairman Culver reported it will be basically the same as it has been over the past 3 years. Chairman Culver asked Mr. Holman to advise Mr. Pool on the dates and he would advise the person who oversees the park. Commissioner Snellgrove stated he was out at Landmark on Saturday and anyone who was unable to attend missed a fun time. Mr. Holman thanked the commission for their support during the year. The chairman then called for the question on the motion and it carried unanimously. 7. Request to install groundwater monitoring well on county property. Chairman Culver reported this procedure is requested by ADEM on the former fuel dispensing location across the street. He pointed out CDG Engineers & Associates did it previously, and they did a great job, and ADEM has asked them to put in another well. Commissioner Cook made a motion to approve the request from CDG Engineers & Associates to install a groundwater monitoring well on the county’s property. Commissioner Snellgrove seconded the motion. Chairman Culver asked Mr. Sean Curtis to coordinate the installation of the well with the representative from CDG Engineers & Associates. The chairman then called for the question on the motion and it carried unanimously. 8. Request to approve budget amendment from Gasoline Fund, Misc. Equipment for $600.00 to Gasoline Fund, Communications Equipment for the purchase of two GPS Verifying Tracking Devices @$300.00 each. Commissioner Snellgrove made a motion to approve a budget amendment in the amount of $600.00 transferring from the Gasoline Fund, Misc, Equipment for $600.00 to Gasoline Fund, Communications Equipment for the purchase of two GPS Verifying Tracking Devices @$300.00 each. Commissioner Harvey seconded the motion; and it carried unanimously. (Please see attached budget amendment in Minute Book) Reports from Staff: County Administrator Mr. Roberts reported he had attended his 40th class reunion over the weekend and he was glad to get to see all of them. County Engineer Mr. Pool reported they had paved the parking lot at Rehobeth’s City Hall, and they had completed the paving on McNeil Road. Mr. Pool stated they were working on the intersection at Beverlye and Prevatt Road and they were getting ready to start on the Rehobeth Volunteer Fire Department project. County Attorney. There was no report. Chairman Culver congratulated Ms. Dothan, Ms. Ka’lea Lisenby, who won the National Peanut Festival pageant, and thanked Ms. Houston County, Ms. Lindsay Shirley, for the outstanding job she did in representing us in the National Peanut Festival Pageant. Chairman Culver reported the following public hearings will be held for comments concerning issues with the polling places they are looking at: Thursday, November 10th, from 9:00 A.M. -10:00 A.M., in the Commission Chambers prior to the next administrative meeting. Monday, November 7th, from 5:30 P.M. – 6:30 P.M., in the small building at the Houston County Farm Center. Commissioner Snellgrove stated some of the people in the Dupree, Harmon School area, and Lovetown had contacted him over the weekend to request a public hearing in a more central location than where they are located. He asked if it would be possible to relocate from the Farm Center to the Lovetown Volunteer Fire Department building? Chairman Culver stated he thought it had gone too far at this point. He reported it was discussed on Thursday and the consensus was to have it at the Farm Center. Adjourn. Commissioner Harvey made a motion to adjourn. Commissioner Cook seconded; and the motion carried unanimously.
An Investigation into Multicasting Nikolay Manolov†, Adel Gamil†, and Stephan Wong‡ †Department of Computer Systems, Faculty of Computer Systems and Control, Technical University of Sofia, Sofia, Bulgaria {monolov,firstname.lastname@example.org http://www.tu-sofia.bg ‡Computer Engineering Laboratory, Electrical Engineering Department, Delft University of Technology, Delft, The Netherlands email@example.com http://ce.et.tudelft.nl Abstract— Nowadays, the Internet is being utilized increasingly more for communicating real-time (multimedia) data to multiple recipients, i.e., multicasting. Our goal is to investigate the network processing operations that are involved in supporting multicasting and to determine which operations are time-critical. Traditionally, investigations on such routers have been focussing on packet forwarding and functionalities of other protocols found in lower layers of the TCP/IP protocol stack. In this paper, we present our investigation of the Internet Group Management Protocol that belongs to a higher layer of the TCP/IP protocol stack. The methodology encompasses the creation of benchmarks reflecting the IGMP functionalities and subsequently the profiling of the created benchmarks in order to determine the most time-critical operations. Next to determining the time-critical operations, the mentioned simulation environment also provided results on instruction distribution, cache behavior, and branch prediction accuracy. Preliminary results suggest that the operations involved in multicasting are memory-intensive as about 50% of the operations are memory-related operations. Such results can be utilized in the design of future network processors. Keywords— Benchmarking, profiling, multicast, IGMP, videoconferencing. I. INTRODUCTION In today’s world, increasingly more people are connected to each other via computer networks and are demanding more types of services in order to communicate with each other. Depending on the type of service that is required, different requirements are placed on the utilized networks. The foremost requirement is the speed in which data is transmitted through the network. Additional requirements include reliability, security, etc. Currently, we are witnessing a new type of service that needs support, namely “one-to-many” transmission$^1$ in which multicasting plays an important role. Example applications include video conferencing, corporate communications, distance learning, and distribution of software, stock quotes, and news broadcasts. A requirement posed by the mentioned applications is to transmit data from a single source to multiple receivers. In this scenario, it is impractical for the sender to transmit the same data multiple times (i.e., as many times as there are members), because this would place a huge burden on the network utilized to relay the data. If there are thousands of receivers, even low-bandwidth applications benefit from using multicast. In these applications, the only way to send to more than one receiver simultaneously is by using IP Multicast. It is a bandwidth-conserving methodology that reduces traffic by simultaneously delivering a single stream of information to thousands of recipients. ![IP Multicast saves bandwidth by sending packets only to the group members](image) Fig. 1. IP Multicast saves bandwidth by sending packets only to the group members IP Multicast delivers source traffic to multiple receivers without adding additional burden on the source or the receivers while using the least network bandwidth of any competing technology. Multicast is based on the concept of a group. An arbitrary group of receivers expresses an interest in receiving a particular data stream (see Figure $^1$The term one-to-many is used to the transmission of data from one source to many destinations. 1). This group does not have any physical or geographical boundaries. The hosts can be located anywhere on the Internet. Multicast packets are replicated in the network by routers that employ multicast-aware routing protocols, e.g., MOSPF, PIM, and DVRMP. In this approach, routers know the exact addresses of group members allowing them to more efficiently route the multicast packets. Therefore, hosts interested in receiving data from a particular group must join the group. This is achieved by utilizing Internet Group Management Protocol (IGMP). More specifically, hosts inform the closest router they are connected to that they are interested in receiving multicast messages sent to certain multicast groups. The IGMP also provide support for the routers to periodically check whether the members of a certain group are still active. Consequently, IGMP must be available in last hop routers, i.e., the router directly connected to the group member’s network, and host operating system network stacks, and it must be used by the applications running on those hosts. The latest version of the protocol is IGMPv3. IGMPv3 supports applications that explicitly signal sources from which they want to receive traffic. The benefits from this capability are better bandwidth utilization and security. Our goal is to investigate the processing requirements for routers in order to support IGMPv3. This will be achieved by creating benchmarks based on existing source codes. Profiling of the benchmarks will help us to identify where the protocol bottlenecks are. Furthermore the simulation results from will also provide an insight to determine eventual hardware implementations of time-critical functions. This paper is organized as follows. Section II discusses in more detail what multicasting is and the associated protocol IGMPv3. Subsequently, we take a brief look at the benchmarking. In Section III, we explain the details of our benchmarking, e.g., the simulator that we use, the datasets, the exact data structures in the benchmark, etc. Section IV presents the simulation results. Section V presents the conclusion of this paper. II. BACKGROUND In this section, we provide that background on multicasting and discuss it in more detail. More specifically, we present the Internet Group Management Protocol (IGMP), multicast addressing, and the benefits of the latest version of IGMP, namely IGMPv3. Finally, we provide some background on benchmarking. The Internet Group Management Protocol (IGMP) Multicast is a point-to-multipoint routing technique that allows IP traffic to be sent from one source to multiple recipients. There are many reasons which make the multicasting capability desirable. The first reason and advantage is that multicasting decreases the network load. Assume that an application, e.g., a stock ticker, wants to transmit packets to hundreds of hosts. It is impractical for the sender to generated hundreds of the same packets and then route these packets through the network. Additionally, this approach will also generate a huge network load. In this case, multicasting reduces network load by replicating the packet(s) at the forks of the multicast delivery tree only when it is necessary. The second reason pertains resource discovery in which a router would query other routers to determine their services. Again, it is impractical for such a router to send the same query to each and every other router. Similarly, using multicast only a single query needs to be sent to the multicast group comprising the router to which a query must be sent. In a small local area network, multicasting can be implemented by introducing a central multicasting-aware router that keeps track of the number of multicast groups and which hosts belong to which group. In a wide area network, e.g., the Internet, the approach of using a single central router is impractical and not efficient. The solution is to turn some routers into multicast routers. Such routers must be aware that certain hosts that they are directly connected to are members of certain groups and they must maintain this information. The manner in which to establish multicasting on the Internet has been specified in the Internet Group Management Protocol (IGMP). Additionally, specific IP addresses were defined to specify multicast groups. Thus, by sending a packet to such a multicast address, all members belonging to that group will receive the packet. The routing is performed through standard routers and the replication of packets is performed by the multicast routers. Therefore, in joining or leaving a multicast group, the immediate multicast router must be notified. The IGMP is utilized by (multicast) routers to periodically check whether the known group members are still active. It provides the information required in the last stage of forwarding a multicast message to its destinations. This way multicast routers within networks know about the members of multicast groups on their directly attached networks and can decide whether to forward a multicast message on their network$^2$. In case there is more than one multicast router on a given subnetwork (LAN), one of the routers is elected as the “querier” and assumes the responsibility of keeping track of the membership state of the multicast groups which have active members on its subnetwork. Based on the information obtained from the IGMP $^2$The forwarding technique is much more complicated in IGMPv3 the router can decide whether to forward multicast messages it receives to its subnetwork(s) or not. After receiving a multicast packet sent to a certain multicast group, the router will check and determine whether there is at least one member of that particular group on its subnetwork. If that is the case the router will forward the message to that subnetwork. Otherwise, it will discard the multicast packet. In the development of the IGMP, three versions have been standardized over time. We present a short summary of them and focus more on version 3 (IGMPv3) since it is provides the most extensive set of operations to enable efficient multicasting. This is also the reason why we have chosen to focus on IGMPv3 in our investigation on multicasting. The three version are discussed in the following: - Version 1: specified in RFC-1112 [4], was the first widely-deployed version and the first version to become an Internet Standard. - Version 2: specified in RFC-2236 [5], added support for "low leave latency", that is, a reduction in the time it takes for a multicast router to learn that there are no longer any members of a particular group present on an attached network. - Version 3: specified in RFC-3376 [1] supports applications that explicitly signal sources from which they want to receive traffic. With IGMPv3, receivers signal membership to a multicast host group in the following two modes: - INCLUDE mode. In this mode, the receiver announces membership to a host group and provides a list of IP addresses (the INCLUDE list) from which it wants to receive traffic. - EXCLUDE mode. In this mode, the receiver announces membership to a host group and provides a list of IP addresses (the EXCLUDE list) from which it does not want to receive traffic. This indicates that the host wants to receive traffic only from other sources whose IP addresses are not listed in the EXCLUDE list. To receive traffic from all sources, like in the case of the Internet Standard Multicast (ISM) service model, a host expresses EXCLUDE mode membership with an empty EXCLUDE list. Version 3 is designed to be interoperable with versions 1 and 2. According to the protocol, in networks where present hosts running different versions of IGMP, the routers must operate in version 1 and version 2 compatible modes. The same rule holds for the hosts, when there is a router running lower version of the protocol. For delivering a multicast packet from the source to the destination nodes on other networks, multicast routers need to exchange the information they have gathered from the group membership of the hosts directly connected to them. There are many different algorithms such as "flooding", "spanning tree", "reverse path multicasting" for exchanging the routing information among the routers. Some of these algorithms have been used in dynamic multicast routing protocols such as Distance Vector Multicast Routing Protocol (DVMRP), Multicast extension to Open Shortest Path First (MOSPF), and Protocol Independent Multicast (PIM), that exists in two variations – Sparse Mode(PIM–SM) and Dense Mode(PIM–DM). Multicast Addressing A Class D IP address is assigned to a group of nodes defining a multicast group. The most significant four bits of Class D addresses are set to "1110". The 28-bit number following these four bits is called "multicast group ID". Some of the Class D addresses are registered with the Internet Assigned Numbers Authority (IANA) for special purposes. The block of multicast addresses ranging from 184.108.40.206 to 220.127.116.11 is reserved for the use of routing protocols and some other low-level topology discovery or maintenance protocols. Addresses ranging from 18.104.22.168 to 22.214.171.124 are reserved to be used for site-local "administratively scoped" applications, and not Internet-wide applications. There are some other Class D addresses already reserved for well-known groups such as "all routers on this subnet", "all DVMRP router" and "all OSPF routers". IGMPv3 Benefits In this section, we discuss several capabilities that the Source Specific Multicast (SSM) provides resulting in two main advantages. The first capability is that a host can specify to receive packets only from a specific source address. The second capability is that a source specify to receive packets from all but specific source addresses. These capabilities result in two main advantages of the IGMPv3, namely: - Optimized bandwidth utilization – The receiver may request to receive traffic only from explicitly known sources. This decreases the network workload, because unwanted traffic is not replicated by the routers anymore. - Improved security – No denial of service(DoS) attacks from unknown sources. In SSM, multicast traffic from each individual source will be transported across the network only if it was requested. In contrast, the older versions of IGMP allow forwarding traffic from any active source sending to a multicast group to all receivers requesting that multicast group. In Internet broadcast applications, this behavior is highly undesirable because it allows unwanted sources to easily disturb the actual Internet broadcast source by simply sending traffic to the same multicast group. This situation depletes bandwidth at the receiver side with unwanted traffic and thus disrupts the undisturbed reception of the Internet broadcast. In SSM, this type of denial of service attack cannot be made by simply sending traffic to a multicast group. All these advantages make the protocol very useful in networks with heavy IP traffic and frequent DoS attacks. III. IMPLEMENTATION In the previous section, we discussed the multicast environment and the Internet Group Management Protocol (IGMP), in particular, version 3 – IGMPv3. In this section, we discuss the methodology we utilized to perform profiling. We have taken application C code reflecting the functionality of IGMPv3 and turned it into a benchmark. Subsequently, we compiled the benchmark to run on a simulator. Utilizing the simulator, we were able to gather data on execution time of specific functions and instruction distribution of the benchmark. The profiling results are discussed in Section IV. The sim-outorder simulator In order to simulate the performance of our benchmark we have performed a simulation using SimpleScalar’s sim-outorder[2]. We simulated a 4-way superscalar processor with 128 KB of direct mapped level 1 (L1) data cache; 512 KB of direct mapped level 1 instruction cache; 1 MB unified level 2 (L2) data and instructions cache. The L1 and L2 cache latencies are set to default values of 1 and 6 cycles, respectively. The simulated processor uses a bimodal branch predictor with 2048 table entries. The IGMP benchmark For simulation purposes, we decided to use igmprt [3] routing daemon and compiled it with SimpleScalar’s gcc v126.96.36.199. The daemon implements the router side processing of IGMPv3, namely maintaining the IGMP table. Each entry in the table has the following structure: (Interface, (Group records), Filter mode for each group, (Source records)). The fields in this structure are discussed in the following: - Interface is the upstream interface where the daemon is running. - Group records represent the multicast groups that are reported on that interface. - Filter mode for each group is either INCLUDE or EXCLUDE. - Source records contains the reported IP addresses, used for source based filtering for each group. The records in the IGMP table are updated by means of periodic reports sent by hosts in response to General Queries, Group Specific Queries or Source and Group Specific Queries. We modified this behavior in such a manner that the program reads these reports from the file input_stream, instead of receiving them from the network. In this file, every line contains one report. The functionalities that we look at are those of receiving IGMPv3 membership reports, assembling them in the igmp_report_t structure and passing them to a function called igmp_interface_membership_report_v3. That function deals with the maintaining of IGMP table and sending queries to the query queue. The processing is done in the following steps: 1. It calls the function igmp_interface_group_add to determine whether the multicast group in the report is preset on the interface where the report is received from. The function creates a group entry if one is not found. 2. The function igmp_group_rep_add is invoked, which checks whether the sender’s IP address is in the group’s structure for that interface, and if it doesn’t it is added. 3. The record type (is_in, is_ex etc.) is determined and the corresponding function igmp_group_handle_isin or igmp_group_handle_isex are called, respectively. These two functions handle the updates of group records – add or remove source IP addresses, reset group and source timers, change the state of the router (from INCLUDE to EXCLUDE) and send “Group Specific Query” and “Group and Source Specific Query”. Following these steps, the IGMP table is updated, the group and source timers are updated also and the necessary query messages are sent to the output message queue. The actual sending of these messages is not implemented, because in this investigation we are only interested in the IGMP table updates. Data types and datasets The IGMPv3 Membership Report format [1] differs from that of IGMPv1 and IGMPv2 Membership Reports due to the SSM capabilities of IGMPv3 (discussed in Section II). In our benchmark, this report is represented in the structure igmp_report_t: ```c typedef struct igmp_report_t { u_char igmpr_type; // version and type of IGMP message u_char igmpr_code; // subtype for routing msgs u_short igmpr_cksum; // IP-style checksum u_short igmpr_rsv; // reserved u_short igmpr_numgrps; // number of groups igmp_grouprec_t* igmpr_group; // group records } igmp_report_t; ``` In the original code of igmprt, the amount of group records per report was 1 (igmp_grouprec_t igmpr_group[1]), therefore we decided to allocate memory dynamically in runtime (igmp_grouprec_t* igmpr_group) depending of the number of groups in every single report. Every group record (igmp_grouprec_t) has the following structure: ```c typedef struct igmp_grouprec_t{ u_char igmpg_type; //record type u_char igmpg_datalen; //amount of aux data u_short igmpg_numsrc; //number of sources struct in_addr igmpg_group; //the group being reported struct in_addr* igmpg_sources; //source addresses } igmp_grouprec_t; ``` In the original code of igmprt, the amount of source addresses per group was 1(igmpg_sources[1]), therefore we decided to allocate memory dynamically in runtime (in_addr* igmpg_sources) depending of the number of sources in every single group. The actual data that we process comes from file input_stream containing randomly generated reports. We performed 3 simulations with 3 different datasets containing 1000, 2000 and 5000 reports, respectively. Each report contains between 1 and 5 multicast groups with different record types (is_in, is_ex, to_in etc.), and each group has from 1 to 5 source addresses. IV. EXPERIMENTAL RESULTS In the previous section, we presented the simulator, the benchmark, and the data types and datasets used in our investigation. In this section, we present the results on instruction distribution, cache behavior, and branch prediction accuracy. Instruction Distribution The simulated functionality includes the operations performed when receiving IGMPv3 membership report. The function igmp_interface_group_lookup is called by igmp_interface_membership_report_v3 and it is responsible for determining whether the reported group persists on the IGMP table. This processing takes a major amount of processor cycles. It consumes more than 90 percents of the igmp_interface_membership_report_v3 cycles (see Figure 2). On the other hand, the functions involved in table updating (igmp_group_handle_xxxx) take no more than 8 percents from igmp_interface_membership_report_v3 cycles. When we use larger datasets, the IGMP interface table grows. Figure 2 shows an increase in the group lookups time for larger IGMP tables. The frequent loads from the memory, caused by the table lookups executed by igmp_interface_group_lookup, explain the data-intensive characteristic of our application (see Figure 3). As depicted in Figure 3 the memory operations take more than 50 percents of all instructions. The condition branch operations have relatively high distribution, which is caused by the high number of different flags and field checks when receiving packet. Cache Behavior and Branch Prediction The results for the cache behavior are summarized in Table I. \| Number of packets \| 1000 \| 2000 \| 5000 \| \|-------------------\|------\|------\|------\| \| Branch address predict. rate \| 0.9881 \| 0.9933 \| 0.9971 \| \| Branch direction predict. rate \| 0.9882 \| 0.9934 \| 0.9971 \| \| L1 data cache miss rate \| 0.1032 \| 0.1152 \| 0.1213 \| \| L1 instruction cache miss rate \| 0.0068 \| 0.0040 \| 0.0018 \| \| L2 unified cache miss rate \| 0.0045 \| 0.0388 \| 0.1897 \| TABLE I CACHE BEHAVIOR The table shows a high branch address and direction prediction values for all datasets. We can also observe the increase in data cache miss rate and decrease in instruction cache miss rate for larger datasets. In addition, the last row in the table shows fast increase in Level 2 cache miss when we process larger datasets. The cache size values are described in Section III. V. CONCLUSIONS In this paper, we presented our investigation of the Internet Group Management Protocol that belongs to a higher layer of the TCP/IP protocol stack. The methodology encompasses the creation of benchmarks reflecting the IGMP functionalities and subsequently the profiling of the created benchmarks in order to determine the most time-critical operations. Next to determining the time-critical operations, the mentioned simulation environment also provided results on instruction distribution, cache behavior, and branch prediction accuracy. Preliminary results suggest that the operations involved in multicasting are memory-intensive as about 50% of the operations are memory-related operations. Such results can be utilized in the design of future network processors. REFERENCES [1] B. Chain, S. Deering, I. Kouvelas, Cisco Systems B. Fenner, AT&T Labs - Research, A. Thyagarajan, Erricson, Internet Group Management Protocol, Version 3, RFC 3376 (www.ietf.org/rfc/rfc3376.txt?number=3376), October 2002. [2] Burger, D. and Austin, Todd M., The SimpleScalar Tool Set, Version 2, Tech. report, University of Wisconsin, June 1997. [3] Lahmadi, An igmpv3-router implementation, World Wide Web, http://www.loria.fr/lahmadi/igmpv3-router.html. [4] S. Deering, Stanford University, Host Extensions for IP Multicasting, RFC 1112 (www.ietf.org/rfc/rfc1112.txt?number=1112), August 1989. [5] W. Fenner, Xerox PARC, Internet Group Management Protocol, Version 2, RFC 2236 (www.ietf.org/rfc/rfc2236.txt?number=2236), November 1997.
E-Beam Lithography Stencil Planning and Optimization with Overlapped Characters Kun Yuan, Bei Yu, and David Z. Pan, Senior Member, IEEE Abstract—Electronic beam lithography (EBL) is one of the promising emerging technologies in the sub-22 nm regime. In EBL, the desired circuit patterns are directly shot into the wafer, which overcomes the diffraction limit of light in the current optical lithography system. However, the low throughput becomes its key technical hurdle. In the conventional EBL system, each rectangle in the layout will be projected by one electronic shot through a variable shaped beam (VSB). This could be extremely slow. Recently, the EBL system with character projection (CP) shoots complex shapes, so-called characters, in one time, by putting them into a predefined stencil. However, only a limited number of characters can be employed, due to the area constraint. Those patterns, not contained by any character, are still required to be written by VSB. A key problem is how to select an optimal set of characters and pack them on the CP stencil to minimize total processing time. In this paper, we investigate a problem of electronic beam lithography stencil design with overlapped characters. Different from previous works, besides selecting appropriate characters, their placements on the stencil are also optimized in our framework. We first propose a linear path-based iterative algorithm to handle 1-D stencil design problem, and an effective simulated annealing framework for the generalized 2-D case with an efficient look-ahead sequence pair evaluation technique. The experimental results show that, compared to conventional stencil design methodology without overlapped characters, we are able to reduce total projection time by 51%. Index Terms—Electronic beam lithography (EBL), overlapped characters, stencil design. I. INTRODUCTION As aggressive scaling continues, the conventional 193 nm optical photolithography technology is facing the great challenge of printing sub-32 nm. For the near future, double/multiple patterning lithography has been developed as a temporary solution for 32 nm, 22 nm, even 16 nm, technology [1]–[6], but the manufacturing cost will be proportionally higher. This makes multiple patterning a relatively temporary solution for further scaling. In the long term, the semiconductor industries and researchers have been actively pushing on alternative emerging nanolithography to print finer feature size below 16 nm, such as electronic beam lithography (EBL), extreme ultraviolet (EUV), and nanoimprint. EBL [7]–[9] is a maskless technology which shoots desired patterns directly into a silicon wafer, with a charged particle beam. The primary advantage is that it is one of the ways to beat the diffraction limit of light of current well-adopted optical lithography [10], which leads to 4× better resolution and lower cost compared to conventional optical lithography. However, EBL has one key limitation, low throughput. The conventional type of the EBL system is variable shaped beam (VSB). In VSB, the layout is usually decomposed into a set of rectangles, and each one would be shot into resist by dose of electron sequentially. As Fig. 1(a) shows, the pattern of “EHE” is divided into 11 rectangles and needs total 11 shots. The whole processing time of this technique increases with number of beam shots. This makes its throughput very low for the modern complicated design, which is commonly composed of a significant number of small rectangles. The character projection (CP) technology [7]–[9] has been invented for improving the throughput of VSB methods. The key idea is to print some complex shapes in one electronic beam shot, rather than writing multiple small rectangles. This reduces manufacturing time significantly. In detail, as the projection system of CP in Fig. 1(b) illustrates, a library of layout configurations, called Characters, or Templates, are prepared on a stencil first. During manufacturing, if any character exists in the targeted design, it will be chosen in the system and projected into the wafer. To print the example of Fig. 1(a), suppose two characters “E” and “H” are predesigned for the stencil. By adjusting and aligning the shaping aperture and stencil, we can print the patterns of “E,” “H,” “E” in a sequential manner, as Fig. 1(c)–(e) shows. In total, it only takes three shots. Due to less beam shots for the same layout, the CP system is much faster than VSB. However, the number of characters is limited due to the area constraint of the stencil. As in the example of Fig. 1(f), there are only maximum $\lfloor W/w \rfloor \lfloor H/h \rfloor$ characters. For modern design, it is not practical to fully make use of CP, due to numerous distinct circuit patterns. Those patterns which do not match any character are still required to be written by VSB. Standard cell and some regular routing pattern, which have high area utilization of stencil because of their frequent appearances in the design, are good candidates for implementing characters. On the other side, the custom layout from memory IP, and most irregular wires/vias would most likely be processed by VSB. Manuscript received June 13, 2011; revised August 12, 2011 and October 10, 2011; accepted November 7, 2011. Date of current version January 20, 2012. The preliminary conference version has been published at the International Symposium on Physical Design in 2011. This paper was recommended by Associate Editor J. Hu. K. Yuan is with Cadence Design Systems, San Jose, CA 95134 USA (e-mail: email@example.com). B. Yu and D. Z. Pan are with the Department of Electrical and Computer Engineering, University of Texas, Austin, TX 78731 USA (e-mail: firstname.lastname@example.org; email@example.com). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TCAD.2011.2179041 Several methodologies have been proposed to design and select group of circuit patterns as characters for minimizing total projection time of both CP and VSB. In [11], frequently-used standard cells are greedily chosen as characters, processed by CP technology. Sugihara et al. [12]–[15] employed integer linear programming to optimize the throughput, given a set of character candidates. Recently, EDA Vendor D2S, Inc. [7]–[9] proposed improving stencil design from a new point of view, but with no detailed algorithm presented. They showed that, in practice, when individual character/template is designed, blanking area is usually reserved around its boundaries. By sharing blanks between adjacent templates, more characters can be placed on the stencil than the regular design of Fig. 1(f), better improving the throughput. The work of [7]–[9] implies that, due to possible overlapping, besides selecting appropriate characters as [11]–[15], their relative locations on the stencil should also be taken into account for minimizing the total projection time of EBL. In this paper, we will investigate this new problem of electronic beam lithography stencil design with overlapped characters. 1-D/2-D problem is researched separately, depending on whether the available overlapping space of characters is nonuniform in either horizontal or both directions. Moreover, we only consider the application of one stencil for this paper. The main contributions of our work are stated as follows. 1) We co-optimize the selection process of characters and their physical placements on stencil for effective EBL throughput improvement. 2) We propose a three-phase iterative refinement process to conduct 1-D stencil design optimization. A Hamilton-path based approach has been developed to solve single-row reordering efficiently and effectively. 3) We develop a sequence pair (SP) based simulated annealing framework to optimize general 2-D stencil design. We also propose an efficient look-ahead sequence pair evaluation algorithm to reduce runtime. II. PRELIMINARIES AND PROBLEM FORMULATION A. Overlapped Character EBL is a maskless technique, which shoots desired patterns directly into a silicon wafer, and can potentially combat device parameter variations [16] and design congestion [17], [18]. Various investigations [12]–[15] have been conducted on the optimization of character selection for EPL technology, where no intersection is allowed between templates on the stencil, as shown by Fig. 1(f). Recently, the work of [7]–[9] has shown that the design of stencil can be further improved by overlapping adjacent characters, which allows more templates to be put and increases the throughput. As pointed out by [7]–[9], when an individual character is designed, blanking space is usually reserved around its enclosed rectangular circuit pattern, shown by Fig. 2(a). The reason is that, when the electron beam is scattered from the shaping aperture of Fig. 1(b), it could span larger area on the stencil than the layout to be printed. In order to avoid projecting any unwanted image, the white space should be preserved. These blanking areas offer great opportunity for character sharing. Suppose the required white space around layouts A and B are BlankA and BlankB, respectively, in Fig. 2(a). If the characters are conventionally aligned by edge as Fig. 2(b), it results in a waste of area. The space between layout A and B is actually BlankA + BlankB, which is more than required for both patterns. By contrast, we would greatly reduce the total area of character A and B by sharing an amount of min(BlankA, BlankB) space. In this case, max(BlankA, BlankB) white width is still reserved between layout A and B, which is sufficient for ensuring correct printing image. B. Stencil Design Challenge The main challenge of stencil design with overlapped characters comes from the fact that, for each character, the amount of required blanking space is not uniform, strongly depending on its enclosed layout patterns [9]. Because different shapes have different scattering and proximity effects under the projection of E-beam, to preserve the shape and fidelity of the written image, their minimum required shielding spacings vary and are not identical. The underlying reason is similar to that in the conventional optical lithography system; the spacing rules between metals are usually different, determined by geometrical dimension of associated design objects. In consequence, for different placements of characters, the area reduction from template overlapping may vary a lot. Therefore, unlike the traditional design of Fig. 1(f), the number of maximum allowable characters in the stencil is not fixed. To achieve a high quality solution, the detailed physical placement information of all the characters must be taken into account. This makes the problem of stencil design with overlapped characters not only different from but also more difficult than the conventional nonoverlapping one addressed in [12]–[15]. As the example of Fig. 3(a) illustrates, suppose there are three character candidates A–C, and we would like to pack them into a simple stencil of Fig. 3(b) for minimum projection time. As easily seen, their blanking spaces are quite different. In conventional design where overlapping is not considered, at most two of them can fit. On the other side, when the blanking space is shared by adjacent characters, the result is correlated with the detailed physical implementation of stencil, and could be different from traditional design. If these three candidates are tried out by the order of A–B–C like Fig. 3(c), only A and B can be put in. Pattern C is out of bound and has to be processed by the VSB technique. This does not lead to higher throughput than the conventional nonoverlapped methodology. In contrast, if rearranged as C–B–A as Fig. 3(d), all of these three patterns can be used as CP characters. Obviously, it is a better stencil optimization. C. Problem Formulation Similar to previous work [12]–[15], we assume a set of character candidates have already been given. To model overlapping information, as Fig. 4(a) illustrates, assume the blanking spaces of each candidate $c_i$, from left, right, top and bottom boundaries, are $l_i$, $r_i$, $t_i$, and $b_i$, respectively. The orientation of these candidates is not allowed to be flipped, since it actually becomes a different template, as explained in [13]. When two candidates $c_i$ and $c_j$ are put adjacent to each other horizontally, their maximum allowed overlap is set as $\alpha_{ij}^H$, which is $\min(r_i, l_j)$ as shown by Fig. 4(b). Similarly, Fig. 4(c) defines the maximum vertical overlapping margin $\alpha_{ij}^V$. $\alpha_{ij}^H$ and $\alpha_{ij}^V$ vary for different $i$ and $j$. Moreover, since the manufacturing time of EBL is dominantly determined by electronic beam shooting, in our work we make use of total number of shots as the measurement of projection time. Suppose each candidate $c_i$ is referred $r_i^c$ times in the chip. For each of its appearance, the candidate $c_i$ will be projected by either CP or VSB method, with a number of shots $n_i^{\text{CP}}$ and $n_i^{\text{VSB}}$. The total processing time (number of shots) of the entire circuit is computed by the following equation: $$\sum_{c_i \in C^{\text{CP}}} r_i^c n_i^{\text{CP}} + \sum_{c_i \in (C^C \setminus C^{\text{CP}})} r_i^c n_i^{\text{VSB}}.$$ (1) $C^C$ is the set of all the character candidates. $C^{\text{CP}}$ is the union of selected candidates processed by the CP method, which is a subset of $C^C$. In our work, as discussed in the introduction, we only design and optimize one single stencil for a given chip. Our optimization problem can be stated as follows. Problem formulation: given a design and its set of character candidate $C^C$, select a subset $C^{\text{CP}}$ out of $C^C$ as characters, and place them on the stencil $S$. The objective is to minimize the total projection time (number of shots) of this design expressed by (1), while the placement of $C^{\text{CP}}$ is bounded by the outline of $S$. The width and height of stencil is $W$ and $H$, respectively, and all the candidate has unique width $w$ and height $h$. The maximum overlapping margin between adjacent characters is given by $\alpha_{ij}^H$ and $\alpha_{ij}^V$. D. Stencil Design Guidance Multiple factors are required to be taken into account to select appropriate candidates onto stencil, which includes the times of references of each candidate in the design, the shot number of each candidate by VSB and CP respectively, the amount of blank margin and the size of each candidate, and the neighborhood relationships of characters. Note that the overall projection time, Objective (1), can be rewritten as $$\sum_{c_i \in C^C} r_i^c n_i^{\text{VSB}} - \sum_{c_i \in C^{\text{CP}}} r_i^c (n_i^{\text{VSB}} - n_i^{\text{CP}}).$$ (2) subject to $$\sum_{c_i \in C^D} area_i < Area(stencil)$$ \hspace{1cm} (3) where the first part is independent of stencil design. To reduce the processing time, $\sum_{C_i \in C^D} r_j^i(n_i^{VSB} - n_i^{CP})$ should be maximized under the area constraint. In other words, we need to maximize the shot number reduction per unit area of stencil, by putting characters onto the stencil. For the five factors we mentioned above, the shot number of each candidate by VSB and CP contributes to the shot number reduction. The amount of blank margin, the size of each candidate and their neighborhood relationship contribute to the optimization of area sharing from character placement. Guided by this observation, the criteria of our character selection is listed in the following paragraph. For a candidate $c_i$, suppose it has a area of $area_i^o$ shared with other surrounding candidates, we only count half of $area_i^o$ as its real usage of stencil and compute the effective area $area_i^{eff}$ of $c_i$ as $area_i - area_i^o/2$. As Fig. 5 shows, the red dash encloses $area_i$, and the light green region is $area_i^{eff}$. Therefore, to achieve maximum throughput improvement, we simply select the candidate with highest shot number reduction to effective area ratio, $(r_j^i(n_i^{VSB} - n_i^{CP}))/area_i^{eff}$, until no more character candidates can be fit in. When the blanks are uniform for all the candidates, the above methodology is easily adoptable. In such a case, the sharing space of adjacent characters is identical for any two candidates, so $area_i^{eff}$ is independent of character placement and can be precomputed. While overlapping margin varies, this problem becomes difficult, because $area_i^{eff}$ is not a deterministic but effected by the detailed placement of characters. In this paper, we will focus this challenging problem with different combinations of overlapping blanking space. We will first investigate into the special case of 1-D stencil design in Section III, where the amount of blanking spaces differs only in the horizontal direction. Vertical margins are identical for all the candidates. In our proposed optimization, we first put characters in the descending order of shot number reduction, i.e., $(r_j^i(n_i^{VSB} - n_i^{CP}))$, then perform optimization to maximize blanking sharing and area reduction. $\sum_{C_i \in C^D} area_i^o$. Next, in Section IV, the algorithm for a generalized 2-D problem will be developed. In this case, both vertical and horizontal blanking space are nonuniform for candidates. Our algorithm for the 2-D problem is a simulated annealing approach. In each iteration, we assign equal weight for an annealer to make two types of moves: the move that pushes in characters having higher shot number reduction $(r_j^i(n_i^{VSB} - n_i^{CP}))$, and the move which better increases the sharable space $area_i^o$. III. 1-D STENCIL DESIGN Normally, each template implements one standard cell. That is to say, the enclosed circuit patterns of all the characters have the same height, and their layouts near top and bottom boundary edges are mostly regular power rails. As a result, illustrated by Fig. 6(a), the required blanking spaces on the top $t$ and bottom $b$ are nearly identical for these candidates. Therefore, in such a case, characters are usually placed on the stencil in a row-based manner, shown by Fig. 6(b). All rows have a unique height $h$. The overlapped blanking margin $h_o$ between adjacent rows is also the same, which is $\min(t_i, b_i)$. In consequence, as Fig. 6(c) shows, the overlapping-aware stencil design becomes a 1-D problem. The number of character rows can be pre determined as $\lfloor (H-h_o)/(h-h_o) \rfloor$. The candidates would be packed into these rows with maximum width $W$. Motivated by this fact, in this section, we will discuss a 1-D stencil design problem, with the assumption that all the candidates have the same vertical blanking space $t_i$ and $b_i$. The overview of our algorithm for this special 1-D problem is given in Fig. 7. In the very beginning, all the rows are empty. Next, character candidates will be pushed onto the stencil in 1-D bin packing, until no more can further fit in. Then steps 2 and 3 serve for the same purpose, tuning the position and selection of characters to create more space while not degrading the throughput. This allows adding more candidates into stencil in the next iteration. After step 3, 1-D bin packing will be re-performed incrementally based on the optimization result of the previous iteration. In other words, the existing characters on the stencil of step 3 will be fixed when we revoke the first phase of bin packing. However, in each iteration, steps 2 and 3 will be executed for all the existing characters, with no exceptions or fixed objects. A. Greedy 1-D Bin Packing To construct a reasonable good starting point, we adopt a descending best-fit bin packing algorithm to push the character candidates into stencil, until there is no enough capacity. Recall that the overall projection time (number of shots) of Objective (1) can also be represented as $$\sum_{c_i \in C^S} r_i^n V_{i}^{SB} - \sum_{c_i \in C^P} r_i^n (n_i^{VSB} - n_i^{CP})$$ where the first part is independent of stencil design. To reduce the processing time, $\sum_{c_i \in C^S} r_i^n (n_i^{VSB} - n_i^{CP})$ should be made as large as possible, during greedy bin-packing. Therefore, as preprocessing, we first assign each candidate $c_i$ a profit value $p_i = r_i^n (n_i^{VSB} - n_i^{CP})$. The bigger $p_i$ is, the larger amount of projection efforts can be saved by printing $c_i$ using CP than VSB method. For getting a good greedy optimization result, the $c_i$ with larger profit should be given higher priority to be placed on the stencil. Guided by this heuristic, in the second step, the character candidates, which have not been on the stencil yet, will be sorted decreasingly based on their profits and packed in a sequential manner. Next, these sorted candidates will be pushed into stencil by a best-fit packing strategy. When $c_i$ is to be packed, the row, which has the most amount of capacities left after accommodating $c_i$, will be picked. This is to consider the possible shared space between adjacent objects, when we are computing the remaining room in each row. As Fig. 8(a) illustrates, suppose only two rows are available and candidate C is to be packed next. It appears that row R1 has more capacity left. However, as Fig. 8(b) illustrates, when we try out C in both rows, it C2 which has larger remaining room. As a result, candidate C is packed into R2, shown by Fig. 8(c). B. Single Row Reordering After greedy bin packing, there is no room left to accommodate more candidates. However, as motivated by Fig. 3, we can adjust the relative locations of already-placed characters in each row to shrink its occupied width and increase remaining capacity. This allows pushing in more candidates, which further reduces the overall projection time. Therefore, in this phase, our goal is to minimize the total width of its characters in each row for maximizing remaining capacity. Suppose row $r$ contains a set of $c_0^r, ..., c_n^r$ characters from left to right, its total occupied width can be computed as $\sum_{i=0}^{n} w - \sum_{i=0}^{n-1} o_{i,i+1}^H$. It is not difficult to see that $\sum_{i=0}^{n} w$ is a constant as long as the number of characters is not changed. Therefore, to minimize the total occupied width, the overall overlapped blanking margin $\sum_{i=0}^{n-1} o_{i,i+1}^H$ should be maximized. To compute optimal character permutation for maximum amount of shared blanking width, we formulate a minimum cost Hamiltonian path problem. First, a graph G is constructed as follows. Each $c_i^r$ is represented by a vertex $v_i^r$. For each pair of $v_i^r$ and $v_j^r$, we add two directed edges $e_{ij}$ and $e_{ji}$. The associated costs are $o_{ij}^H - o_{ij}^H$ and $o_{ji}^H - o_{ji}^H$, respectively. $o_{ij}^H/o_{ji}^H$ is the shared space when $c_i$ is put left/right adjacent to $c_j$, and $o_{ij}^H$ is a constant value, bigger than any of $o_{ij}^H$. To maximize $\sum_{i=0}^{n-1} o_{i,i+1}^H$, it suffices to find a path visiting each node of G exactly once such that the total edge weights ($\sum_{e \in \text{path}} (o_{\text{left}}^H - o_{\text{right}}^H)$) along this path are minimized. As Fig. 9(a) illustrates, a graph for three character placement (A, B, C) is given. Suppose the minimum cost Hamiltonian path is found as Fig. 9(b), Fig. 9(c) shows its corresponding character placement. Since the problem of minimum cost Hamiltonian path is NP-hard, it may be expensive to solve the whole row in one time. Our heuristic is to partition the row into multiple overlapped smaller segments, and solve each segment by the Hamiltonian path based method. C. Multiple Row Swapping After single row reordering, the character permutation within each row has been extensively optimized. However, it is still possible to increase their remaining capacities, by swapping characters from different rows. As Fig. 10 illustrates, by swapping $r_1^2$ and $r_2^2$, both characters find “better” neighbors with more overlapped blanking space. For rows R1 and R2, their remaining rooms are both increased. The algorithm is briefly explained as follows. We test every pair of characters from different rows. Only when the remaining capacities of both rows are increased after swapping, it is considered as a reasonable swap. This ensures that the modified placement is definitely better than the original one. The reason is that, after swapping, if one row gains more room but another has less, it is possible that the following optimization is hurt by the row with shrunk capacity. After the reasonable swap pairs are found, they are sorted increasingly by capacity gains, and performed one pair by one pair. One thing to emphasis here is that the algorithm could be quite slow if we really iterate through all the possible swapping, when the total number of characters is large. To encounter this issue, we only randomly select a predefined number of swapping candidates. IV. 2-D STENCIL DESIGN In this section, we investigate the general case of EBL stencil design with overlapped characters. The blanking spaces of templates are nonuniform along both horizontal and vertical directions. Due to NP-completeness of this problem, we adopt a simulated-annealing based heuristic approach. The basic flow is shown in Fig. 11. Parquet [21] is adopted as our simulated annealing framework. Given an initial solution and starting temperature $T_0$, each iteration we continuously make small permutation on sequence pair, and evaluate the resulting stencil design, as the blue steps indicated. The new SP/solution will definitely be adopted if reducing current best throughput (shot number $S_{\text{best}}$) of Objective (1). While it is actually a worse character placement, this nonimproving result is accepted with a probability, computed from a function of current temperature $T$ and $S_{\text{best}}$. The function $f(S_{\text{best}}, T)$ decreases as $S_{\text{best}}$ and $T$ decrease. At the end of each iteration, the temperature will be updated with decreasing trend. The algorithm terminates when the upper bound of iteration number or lower bound of temperature is reached. The detailed discussion for sequence pair evaluation and permutation can be found in Sections IV-B to IV-D, respectively. A. Sequence Pair Representation To represent a character placement solution, we make use of SP proposed in [19]. Given a set of character candidates $C^C$, its SP consists in two permutations $\overline{X}$ and $\overline{Y}$ of these templates $(c_0, c_1...c_n)$, which specifies their geometrical relationships as follows: $$\overline{X} : <.., c_i, c_j, ..>, \overline{Y} : <..c_i...c_j, ..> : c_i \text{ is left to } c_j \quad (4)$$ $$\overline{X} : <.., c_j, c_i, ..>, \overline{Y} : <..c_i...c_j, ..> : c_i \text{ is below } c_j. \quad (5)$$ Based on these constraints, we can map any SP into a solution of character placement as the following procedure. Procedure 1. Step 1: compute a packing solution of $C^C$, following the extensive methods of [19] and [20]. The details and speedups will be described in Sections IV-B and IV-C. Step 2: assuming the left-bottom coordinates of packed candidate and stencil are the same, the candidates, which are located completely within the outline of stencil, are considered as selected characters. Step 1 is the critical one in the above transformation. Due to specific properties of our problem, its implementation actually differs from the conventional approaches of [19] and [20], explained as follows. B. Fast Look-Ahead Sequence Pair Evaluation The key step of packing solution evaluation from SP is to determine the physical coordinates of each block. This problem has been well investigated, when overlapping is not considered between adjacent blocks. The original algorithm is proposed in [19], and improved by [20] with a new solution pruning technique. The work of [19] is easily extensible for our overlapping-enabled character placement problem, described in Section IV-B1, but it is well known that this type of method is very computationally-expensive. On the other side, in Section IV-B2 we show that the key speedup idea in [20] is not applicable directly or extensible trivially, although it is much faster. We then present an efficient look-ahead algorithm based on framework of [20] to solve this specific sequence pair computation problem. 1) Extension for Longest Path Based Algorithm: The method of [19] is based on a longest path algorithm, and starts from constraint graph construction. Given a SP, a H/V graph is built first to capture the horizontal/vertical relationship between different blocks. Assume there are totally $\|C^C\|$ candidates. The H/V graph has $\|C^C\|+2$ vertexes, one $v_i$ for each candidate $c_i$ plus a source $s$ and sink $t$. If $c_j$ is (left adjacent to)/(below) $c_k$, a directed edge $e_{jk}$ is added from $v_j$ to $v_k$. The weight of $e_{jk}$ is the minimum possible horizontal/vertical distance between the centers of $c_j$ and $c_k$. Beside these, there is a zero-weight edge from source to every $v_i$, and a zero-weight edge from every $v_i$ to sink. For the example of Fig. 12(a), Fig. 12(b) and (c) shows the resulting H and V constraint graphs, respectively. Algorithm 1 Calculate x Coordinates of Candidates Given a Sequence Pair 1: for i = 1 to n do 2: match[X[i]].x = i; match[Y[i]].y = i; L[i] = 0; 3: end for 4: for i = 1 to n do 5: b = X[i].p = match[b].y; pos[b] = L[p].t = pos[b] + width[b]; 6: for j = p to n do 7: if t > L[j] then 8: L[j] = t; 9: else 10: break; 11: end if 12: end for 13: end for After that, the x/y coordinates of these candidates can be obtained by finding its weighted longest path algorithm from source. As easy to see, this methodology is also applicable for our problem, where overlapped space is allowed between adjacent vertexes. The only difference is that, when the weights of edge are assigned, the amount of shared blanking space must be considered, as highlighted by the red cycles in Fig. 12(b) and (c). This type of longest path based algorithm is quite slow, as shown by [20]. In our work, we implement this extension as the baseline to show the effectiveness of our look-ahead pruning technique, which is proposed in Section IV-B2. 2) Extension for Longest Common Subsequence Based Algorithm: The work of [20] does not explicitly build the constraint graphs, but depends on the longest common subsequence computation. They evaluated the placement of character candidates much faster than [19]. In this subsection, we first show that the pruning algorithm in [20] is not applicable directly due to the breakdown of its assumption. Then we proposed a look-ahead technique to revive the core idea of [20] for fast sequence pair computation. a) Review of original pruning algorithm: To better present our proposed pruning technique, first, we review the key steps of [20] in Algorithm 1, and its pitfall for overlapping-induced character floorplanning. The notations are listed in Table I. Their algorithm is based on longest common subsequence. Here, we only consider the situation where the x coordinate is computed. Algorithm 2 Initial look-ahead algorithm for sequence pair evaluation 1: for i = 1 to n do 2: match[X[i]].x = i; match[Y[i]].y = i; L[i] = 0; Min[T][i] = 0 3: end for 4: for i = 1 to n do 5: b = X[i].p = match[b].y; pos[b] = L[p]; 6: t = pos[b] + width[b]; Min[T][n] = L[n] + o[b][Y[n]]; 7: for j = n-1 to p, j = j - 1 do 8: Min[T][j] = min(Min[T][j+1], L[j]+o[b][Y[j]]); 9: end for 10: for j = p to n do 11: if t < Min[T][j] then 12: break; 13: end if 14: if t > L[j]+o[b][Y[j]] then 15: L[j] = t-o[b][Y[j]]; 16: end if 17: end for 18: end for Lines 6–11 implement the essential pruning idea of [20]. In Line 8, L[j] will be updated if and only if the new required x coordinate t for candidate Y[j] is larger than current L[j]. Without considering the item of o[b][Y[j]] in the nonoverlapping case, original algorithm can do effective pruning, lying on the foundation of the following ascending property of L[j]: if i is larger than j, then L[i] is guaranteed no less than L[j]. Once t is smaller than certain L[j] in Lines 7, we can directly break out of the loop Lines 6–11. Any element in array L after index j is definitely bigger than t, and there is no need to do further evaluation using Lines 6–11. Unfortunately, while overlapping is considered, the original algorithm cannot be simply modified by updating Lines 7 and 8 using (6) and (7), respectively \[ t - o[b][Y[j]] > L[j] \] (6) \[ L[j] = t - o[b][Y[j]]. \] (7) The reason is that the array of L does not have such an ascending property any more. This is due to the variations of o[b][Y[j]]. When we update array L by (7), L[m] may be larger than L[n] even m is smaller than n, because o[b][Y[m]] may be smaller than o[b][Y[n]]. b) Initial look-ahead pruning algorithm: Motivated by the above analysis, we first present our initial look-ahead sequence pair evaluation algorithm. The main direction is to find a good threshold, e.g., MinT, which maintains an ascending property and hence can replace the usage of L[j] for correct pruning. This is actually not difficult. As we may find out, $ Min[T][j] = min(L[j]+o[b][Y[q]]), j < q < n, $ satisfies the requirement. The original algorithm can be then revised as Algorithm 2. Note that Line 11 is equivalent to checking the inequality specified by (6). Lines 7–12 are the key steps here. If the condition of Line 11 is satisfied when j = p1, the Line 15 will not have chance to be executed for following iterations (j >= p1). We save computational operation here. It works correctly, and seems helping reducing runtime with the pruning of Lines 11–12. However, if we take a closer look, Algorithm 3 Efficient look-ahead algorithm to speed up sequence pair evaluation 1: for i = 1 to n do 2: match[X[i]].x = i; match[Y[i]].y = i; L[i] = 0; MinT[i] = 0 end for 4: for i = 1 to n do 5: b = X[i].p = match[b].y; pos[b] = L[p]; t = pos[b] + width[b]; 6: index = n; 7: for j = p to n do 8: if t < MinT[j] then 9: index = j; 10: break; 11: end if 12: if t > L[j]+o[b][Y[j]] then 13: L[j]= t-o[b][Y[j]]; 14: end if 15: end for 16: for k = index-1 to p, k=k-1 do 17: MinT[k] = min(MinT[k+1], L[k]); 18: end for 19: end for the step of calculating $MinT[j] = \min(L[q] - o[b][Y[q]])$, $j < q < n$ is expensive. As in Lines 7–9 of Algorithm 2, because $o[b][Y[q]]$ varies with the values of $b$, $MinT[j]$ have to be fully updated in every iteration of loop Lines 4–17, which requires going through every element from $p$ to $n$ in Lines 7 and 8. This native look-ahead algorithm does not really benefit us in terms of runtime. Our next step is to think out a faster way to compute $MinT[j]$. c) Efficient look-ahead pruning algorithm: To resolve the issue of expensive updating in Algorithm 2, we come up with a little relaxed metric $MinT[j] = \min(L[j])$, $j < q < n$ with a much faster computation strategy. The implementation is described as Algorithm 3. By taking advantage of this new $MinT[k]$, we only need to perform partial updating shown in Lines 16–18. The correctness of this updating is easy to see. Any $MinT[k]$ can only be changed if some $L[q]$, $k < q < n$, has been updated by Line 13, and the value of index-1 is the largest entry in $L$ which has been modified during each iteration. Therefore, any $MinT[k]$, $k > index$, would keep the same and there is no need to update them in Lines 16–18. Theoretically, in the worst case, Algorithm 3 is as slow as Algorithm 2 because complete updating may still be needed every iteration in Lines 16–18, when the index is equal to $n$. However, in practice, Algorithm 3 would perform better. The significant speedup comes from the fact that in most of time, the index is much smaller than $n$. Although the relaxation of $MinT[q]$, from $\min(L[j] + o[b][Y[q]])$, $j < q < n$ to $\min(L[j])$, $j < q < n$, makes the pruning step Lines 8–10 less frequent, the benefits from efficient $MinT[q]$ updating compensate this degradation. C. Partial Packing Evaluation After evaluating packing solution, in step 2 of Procedure 1, the candidates outside the outline of stencil will not be taken as characters. This implies that the detailed locations of these candidates are not important, and do not have to be computed in step 1. Great speedup can be achieved by making use of this property. In detail, in the implementation of SP-based minimum area packing, we stop placement evaluation as soon as the contour of already-packed character candidates is completely outside the outline of stencil by at least a margin of $o_{\text{max}}$, given that $o_{\text{max}}$ is the maximum value of $o_{ij}^H$ and $o_{ij}^V$. D. Sequence Pair Permutation In this subsection, we present two effective SP perturbation methods for better local search toward shorter projection time: throughput-driven swapping and slack-based insertion. 1) Throughput-Driven Swapping: The first type of perturbation we perform is throughput-driven swapping. The basic idea is to try reducing overall projection time by swapping the positions of two candidates in the $X\&Y$ SP. This is equivalent to exchanging their relative locations in the packing solution. Fig. 13 illustrates a motivational example, which has five blocks A–E to be packed. The required number of shots, to project any of these candidates once, are assumed as 1 and 10 for CP ($n_1^{CP}$) and VSB ($n_1^{VSB}$) methods, respectively. The digit in the parentheses denotes how many times $r_i$ of each component will be used and printed in the design. Fig. 13(a) gives a SP representation and its corresponding stencil design, based on the Procedure 1 in Section IV-A. Following the definition of Objective 1, the total processing time (number of shots) is $3 + 2 + 1 + 2 + 10 \times 2 = 28$, since A–D are selected as characters while E is not. If swapping the locations of C and E in SP as Fig. 13(b), we would end up with a better stencil design with less amount processing time. It only takes a number of 19 shots, which is computed as $3 + 2 + 10 + 2 + 2 = 19$ in this case. In the detailed implementation, we enforce two heuristic swapping constraints to enable efficient and effective shot number reduction. First, given a SP, out of the pair of elements to be changed, we require that one candidate $c_x$ should have been selected as characters by its corresponding stencil packing, while the other one, $c_o$, is not. For the example of Fig. 13(a), we only allow the exchange of the positions between E and any of A–D. The swapping among any two of A–D is not enabled. The reason is that if the two candidates to be swapped are both in or out of stencil already, most likely the new SP generates a stencil solution with the same set of selected characters and just different geometrical ordering. As an example, if we swap candidates B and D which are both already in the stencil, like from Fig. 13(a) to (c), the resulting packing result also selects A–D as characters, still requiring 28 shots in total. Second, after randomly picking in-stencil candidate $c_x$ and out-of-stencil one $c_o$ for swapping, we will compute the difference of their profits $p_o - p_x$, to decide whether this swapping would be tried on. The profit $p_o/p_x$ is defined as same as $r_i(n_{\text{YSB}}^i - n_{\text{CP}}^i)$ in Section III-A, which reflects the reduction of the shoot number by printing this candidate with CP rather than VSB. If we swap the locations of $c_x$ and $c_o$, it is highly likely that $c_x$ will be pushed out of stencil but the $c_o$ would be selected as a character in turn. Assuming all the other candidates stay either in or outside the stencil, as the state before the swapping, the total shot reduction by this exchange can be approximated as $p_o - p_x$. Therefore, if the difference $p_o - p_x$ is smaller than zero, it is in high possibility that the swapping under consideration will not lead to a better packing result. For the example of Fig. 13(a), suppose $c_x$ and $c_o$ are A and E, respectively, and it turns out $p_o - p_x$ is $-9$. In this case, the corresponding stencil design indeed becomes worse, taking 35 shots as Fig. 13(d) shows. 2) Slack-Based Insertion: Given a SP and its corresponding character solution, our purpose of slack-based insertion is to add-in a new candidate, which is currently not serving as a character, into the stencil. To ensure robust throughput improvement, we would like to find a good strategy to insert such an extra candidate so that all the previously already-placed characters are still kept on the stencil in most trials. This equals to increase the number of usable templates. In this subsection, we make use of the concept of slack, applied in [21] and [22], to search such a good insertion location. Given a character $c_x$ on the stencil, its x/y slack is defined as the allowed movement range of x/y coordinates of $c_x$, under the constraint that none of all the other already-placed characters would be pushed outside the stencil after such a move. Fig. 14(a) and (b) illustrates a simple example, with four characters A–D. Their leftmost and rightmost packing solutions are shown by Fig. 14(a) and (b), respectively. Based on these two extreme cases, the x slack of C, for example, can be computed as $X_{\text{right}}^C - X_{\text{left}}^C$. Once slacks are known, we randomly pick a base character $c_b$, which has large slacks in both x and y directions, and insert a new candidate $c_{\text{new}}$ before it. The reason is that the location of such base can be moved in relatively big amount to make space for additional characters. In terms of SP operation, this can be done by simply changing the position of $c_{\text{new}}$ right before $c_b$ in $\bar{X}$ and $\bar{Y}$ permutations. As illustrated by Fig. 14(c), suppose the $c_b$ and $c_{\text{new}}$ are candidate C and E, respectively. The resulting new SP is obtained by inserting E right in front of the position of C, as shown by Fig. 14(d). V. EXPERIMENTAL RESULTS We implement our algorithm in C++ and test on an Intel Core 3.0GHz Linux machine with 32 G RAM. LKH [23] is chosen as the solver for min-cost Hamilton path, where the maximum problem size is set as 50. In other words, if the number of characters in a row exceeds 50, we will chop it into smaller segments, each less than 50, and perform LKH individually. Moreover, Parquet [21] is adopted as our simulated annealing framework. To test the efficiency of proposed methods, we randomly generate eight benchmarks. The size of stencil is set as $100 \mu m \times 100 \mu m$, and a total number of 1000 character candidates with unique size are generated. The sharable blanking area within each candidate is randomly decided, uniformly distributed between 0% and 50% character width. For the special case of 1-D problem, the blanking space along the vertical direction is set as a constant value. Moreover, for each candidate $c_i$, we randomly assign a triple of value $(r_i, n_{\text{YSB}}^i, n_{\text{CP}}^i)$ as its referred time in chip, and respective number of shots by VSB and CP. $n_{\text{YSB}}^i$ is made five to ten times larger than $n_{\text{CP}}^i$. The detailed statistical data for individual testcase is shown in Table II. The first column denotes the name of benchmarks, where “1D-x” and “2D-x” are applied for one and 2-D problem, respectively. “csize” is the size of each character candidate, formatted by “$\mu m \times \mu m$”. The units of all the other columns are “$1e^6 \mu m^2$”. “Total area” shows the total area of all the character candidates, and “total blanks” is the summation of their sharable blanking space. “Optimal area” is computed as “total area” minus “total blanks”, typically larger than the area of given stencil. This matches the fact that even under best possible case of stencil design, where all the blanking areas are indeed shared by adjacent characters, the entire set of the candidates can not be fully pushed into the stencil. For comparative reasons, we implement two different stencil design approaches. The first one, No-Overlap, is based on the ### TABLE II Statistics on Testcases \| cks \| csizes \| Total Area \| Total Blanks \| Optimal Area \| \|------\|----------\|------------\|--------------\|--------------\| \| 1D-1 \| 3.8 × 3.8\| 1.444 \| 0.416 \| 1.028 \| \| 1D-2 \| 4.0 × 4.0\| 1.6 \| 0.479 \| 1.121 \| \| 1D-3 \| 4.2 × 4.2\| 1.764 \| 0.514 \| 1.25 \| \| 1D-4 \| 4.4 × 4.4\| 1.936 \| 0.569 \| 1.367 \| \| 2D-1 \| 3.8 × 3.8\| 1.444 \| 0.414 \| 1.03 \| \| 2D-2 \| 4.0 × 4.0\| 1.6 \| 0.529 \| 1.071 \| \| 2D-3 \| 4.2 × 4.2\| 1.764 \| 0.662 \| 1.102 \| \| 2D-4 \| 4.4 × 4.4\| 1.936 \| 0.774 \| 1.162 \| work of [15], where no overlapped characters are allowed. A little difference is that, in its implementation, only one stencil with unique character size is considered. For our problem, their algorithm is somewhat degenerated into a method of selecting the most profitable candidates, which profit is judged by $ r_i(t_i^{\text{CP}} - n_i^{\text{SB}}) $. In the second comparative approach, Greedy, possible sharing is taken into account, but a greedy methodology is applied to choose character candidates. In the 1-D problem, only the first phase of heuristic descending best-fit (DBF) packing in Section III-A is performed. For 2-D problem, 2-D DBF packing is conducted. #### A. 1-D Stencil Design Table III lists the comparison of stencil design in the 1-D case. “#shot” shows the total processing time (number of shots) of the circuit by using corresponding stencil design methodologies, which is computed by the equation of Objective 1. “#char” is the number of characters that fits into stencils, and “#CPU” tells the runtime of these stencil optimization methods in terms of seconds. As we can see, compared to No-overlap, we are able to put on average 42% more characters on the stencil, and reduce the total projection time (number of shots) by 51%. With respect to Greedy algorithm, our approach still achieves on average 14% more projection time reduction, by allowing 7% more characters placed. The CPU time of our approach is relatively large but its absolute value is only around 20s. These results show the effectiveness and efficiency of our proposed three-phase iterative refinement algorithm. For this special one-dimension problem, Greedy looks also quite useful. The reason is that the vertical blanking spaces of these candidates are uniform in this case, and have been fully shared during the stencil design. Next, we investigate the benefits achieved by sharing horizontal and vertical spacing, respectively. Note that we can enable or disable vertical space margin by setting appropriate number of available rows. In Fig. 15, light blue column “No-overlap” illustrates the number of characters put on the stencil when no space overlapping is allowed. Dark blue column “H-overlap” lists the results, when horizontal blank sharing is enabled but vertical is forbidden. Green column “H+V overlap” shows the character number when both inside-row and inter-row sharing are enabled which is our default setting. The only difference between “H-overlap” and “H+V overlap” is that, in “H+V overlap,” the overlapping between rows plays roles. ![Fig. 15. Investigation of the throughput improvement from horizontal and vertical overlapping, respectively.](image) ![Fig. 16. Trend of area reduction ratio with respect to the number of iterations (for 1D-1).](image) As the result shows, the improvement of throughput is mainly from sharing horizontal spacing within each row. This alone increases the number of used characters by 32%. The vertical margin optimization contributes another 10% character number increase. This demonstrates our three-step iterative refinement method, designed for horizontal overlapping optimization, is effective and contributes major portion of throughput improvement. To illustrate how our algorithm performs over iterations, we also plot the trend of area reduction for testcase 1-D in Fig. 16. The y-axis shows the amount of reduced area after we perform Hamilton-based compaction and multirow swapping in each iteration, with normal to the stencil area. It can be concluded that the first iteration plays the key role in our optimizations. Further, we break down and study the statistics of different steps of proposed algorithm. In Fig. 17, [1]–[3] correspond to the stages of 1-D bin-packing, single row reordering, multirow, and inter-stencil tuning, respectively. Here, we only focus on one testcase 1D-1. Fig. 17(a) illustrates the CPU time distribution of these steps. On the other hand, Fig. 17(b) reflects their individual contributions for throughput improvement. In our three-step iterative refinement flow, only the first step 1-D bin-packing directly increases the number of characters. Steps 2 and 3 are all used for compacting currently-existing characters on the stencil, which creates more spacing for pushing more candidates in bin-packing the next iteration. Therefore, we only list the reduced packing area by steps 2 and 3 in Fig. 17(b). Seen from Fig. 17, the major CPU bottleneck comes from single row reordering and inter-stencil tuning. Step 2 of with-in-row optimization contributes most for packing area reduction, which is the key step for further occupation of additional characters. To evaluate how far Greedy and our proposed solution are close to the optimal one, we also implemented an ILP Fig. 17. Distribution of CPU time and area saving of distinct steps in our proposed 1-D flow (for testcase 1D-1 only). (a) Breakdown of CPU time. (b) Breakdown of area saving. TABLE III RESULT COMPARISON FOR 1-D PROBLEM \| cks \| NO-OVERLAP \| GREEDY \| Our Approach \| \|-----\|------------\|--------\|--------------\| \| \| #shot \| #char \| CPU(s) \| #shot \| #char \| CPU \| #shot \| #char \| CPU \| \| 1D-1 \| 28 \| 651 \| 1.2 \| 13 \| 530 \| 901 \| 2.2 \| 10 \| 863 \| 951 \| 22.3 \| \| 1D-2 \| 4 \| 727 \| 625 \| 1.1 \| 17 \| 929 \| 836 \| 2.1 \| 14 \| 921 \| 886 \| 17.8 \| \| 1D-3 \| 38 \| 460 \| 529 \| 0.9 \| 25 \| 155 \| 727 \| 1.9 \| 22 \| 503 \| 768 \| 20.6 \| \| 1D-4 \| 41 \| 260 \| 484 \| 0.8 \| 29 \| 462 \| 665 \| 1.8 \| 26 \| 756 \| 702 \| 20.1 \| \| Total \| 150 \| 101 \| 2314 \| 4 \| 86 \| 074 \| 3129 \| 8 \| 74 \| 263 \| 3301 \| 84.8 \| \| ratio \| 2.0 \| 1 \| 0.05 \| 1.16 \| 1.35 \| 0.10 \| 1 \| 1.42 \| 1 \| TABLE IV RESULT COMPARISON WITH RESPECT TO ILP-DERIVED OPTIMAL SOLUTION \| cks \| #char \| #shot \| CPU (s) \| \|-----\|-------\|-------\|---------\| \| NO-OVERLAP \| 16 \| 1096 \| 0.00 \| \| GREEDY \| 20 \| 846 \| 0.02 \| \| Our Approach \| 22 \| 704 \| 0.1 \| \| ILP \| 24 \| 515 \| >3600 \| formulation, and considered its solution as “optimal.” Since the complexity of ILP is forbiddingly high, we pick up one small design. The sizes of candidates and stencil are set as $4 \mu m \times 4 \mu m$ and $16 \mu m \times 16 \mu m$, respectively. The number of candidates is 30. The results are shown in Table IV. In terms of shot number, our approach is 56% and 14% better than Nonoverlap and Greedy respectively, which shows the effectiveness of proposed flow. When compared to ILP, it is 27% worse. This indicates possible improvement space of our methodology, which we are going to work on in future. On the other side, as we can see, ILP is very slow even for such a small testcase, taking more than 1 h. All the other three methods finish in seconds. It is easily seen that ILP definitely cannot handle our real case of $1e^3$ candidates. For Greedy algorithm, it is 41% worse than ILP but only 14% inferior than our approach, judged by shot number. This again supports our previous conclusion that the greedy algorithm in 1-D is not that bad as expected. B. 2-D Stencil Design In our 2-D stencil design, we reused the same parameters in Parquet for our simulated annealing framework, including initial and stopping temperature, temperature degrading factors, and so on. There are three types of different SP perturbation methods, throughput driven swapping and slack based insertion (proposed in Section IV-D), plus random permutation. Random permutation is adopted here just for avoiding getting stuck into local space. In our implementation, we rely on a pseudo number generator to determine which type of perturbation to apply in each iteration. Table V lists the comparison of stencil design in the general 2-D case. Column $CPU(eval)$ shows the runtime for SP-evaluation phase. The meaning of labels is the same as Table III. Compared to No-overlap and Greedy methods, in average, our proposed SP-based algorithm places 28% and 24% more characters on stencil, which reduces the projection time (number of shots) by 31% and 25%, respectively. The Greedy algorithm does not work that well in this 2-D problem, because the blanking area varies in both horizontal and vertical directions and the native first-bin-best-fit packing very easily gets stuck in local optima. Moreover, from Column $CPU(eval)$, we observe that the major portion time of our algorithm is spent on SP-evaluation (around 73%), even with the applicant of our proposed look-ahead pruning algorithm. Due to 2-D optimization, the runtime of our approach is much longer than 1-D problem comparatively. It takes a few hundred seconds, but is still satisfactory. The design of stencil is only a one-time process before projecting large volume of chips by EBL. Several minutes preprocessing time is relatively very tiny in the whole manufacturing procedure. Next, we will show the effectiveness of our proposed look-ahead sequence pair evaluation technique in Fig. 18. “Without look ahead” applies the extension of longest path based evaluation method in Section IV-B1, while “With look ahead” adopts our proposed look-ahead pruning technique based on longest subcommon sequence computation. For these two versions, we achieve the same solution quality, the throughput. In terms of runtime, our approach can achieve about 48% reduction. When judged by the step of SP evaluation only, the CPU time decreases by 66%. That also demonstrates our statement in Section IV-B2) that although the theoretical timing complexity of Algorithm 3 is as much as naive Algorithm. 2, it performs much better in practice. Similar to the 1-D problem, we also study the individual contributions of different SP perturbation methods applied in our simulated annealing flow. As shown in Fig. 19, the proposed throughput driven and slack base methods perform effectively, which contribute around 80% of total projection time reduction. Their runtimes are comparable, not showing outstanding outlier. As the last experiment, we apply the proposed 2-D simulated annealing based framework to solve 1-D testcases. The result is listed in Table. VI. We added two new columns $CP$ and $VSB$ to show their shot numbers respectively. As we can show, 2-D algorithm generates 24% more shot number and TABLE VI RESULT COMPARISON WHEN APPLYING 2-D METHODS TO 1-D PROBLEM \| cks \| #shot \| Our Approach of 1-D \| 1-D Greedy \| Our Approach of 2-D \| \|-----\|-------\|---------------------\|------------\|---------------------\| \| \| \| #CP \| #VSB \| #char \| CPU(s) \| #shot \| #CP \| #VSB \| #char \| CPU(s) \| #shot \| #CP \| #VSB \| #char \| CPU(s) \| \| 1D-1 \| 10183 \| 6105 \| 3978 \| 951 \| 22.5 \| 4528 \| 1000 \| 7728 \| 90 \| 2.2 \| 16,345 \| 5449 \| 10,896 \| 794 \| 257 \| \| 1D-2 \| 14921 \| 7300 \| 4830 \| 1189 \| 18.8 \| 17929 \| 6262 \| 19667\| 826 \| 2.1 \| 20,986 \| 1611 \| 10,707 \| 678 \| 245 \| \| 1D-3 \| 22503 \| 5334 \| 1769 \| 768 \| 20.6 \| 25155 \| 5043 \| 20112\| 1.9 \| 25,906 \| 4504 \| 21,402 \| 648 \| 240 \| \| 1D-4 \| 26756 \| 4864 \| 21892\| 702 \| 20.1 \| 29462 \| 4611 \| 24851\| 665 \| 1.8 \| 29,157 \| 4372 \| 24,785 \| 589 \| 228 \| \| Total \| 74263 \| 22872\| 51391\| 3301 \| 84.8 \| 86074 \| 21716\| 64358\| 3129 \| 8 \| 92,194 \| 12191\| 73,273 \| 970 \| 1143 \| Fig. 18. Total CPU time with and without look-ahead sequence pair evaluation. (a) (b) Fig. 19. Distribution of CPU time and projection time reduction of different types of SP permutations in our proposed simulated annealing framework (for testcase 2D-1 only). (a) CPU time. (b) Projection time reduction. runs much slower than our proposed 1-D three-step iterative method. Its solution quality is also 7% worse than greedy 1-D DBF packing algorithm. The reasons are two-fold. First, the 2-D algorithm is not aware of the fact that the vertical blank space of 1-D cases is uniform and can be utilized optimally. Moreover, 2-D algorithm is rather heuristic, and not able to handle horizontal overlapping as effectively as Hamilton cycle based compacting proposed in 1-D three phase refinement flow, or DBF packing algorithm in 1-D greedy decomposition. VI. CONCLUSION In the future, to further improve the algorithm runtime, we will try to substitute the annealing-based 2-D stencil design by some nonstochastic packing algorithm, e.g., the work in [24]. We also plan to work on multiple-stencil optimization with massive parallel electronic beam projection as well as electronic beam lithography friendly standard cell design, placement and routing algorithm. In addition, we are interested in expanding our investigation on research opportunity between physical design (e.g., [25], [26]) and other emerging lithography, such as triple patterning, nanoimprint, and EUV. ACKNOWLEDGMENT The authors would like to thank Dr. G.-J. Nam at IBM Austin Research, Austin, TX, for helpful discussions on this problem. REFERENCES [1] A. B. Kahng, C.-H. Park, X. Xu, and H. Yao, “Layout decomposition for double patterning lithography,” in Proc. Int. Conf. Comput.-Aided Des., Nov. 2008, pp. 465–472. [2] K. Yuan, J.-S. Yang, and D. Z. Pan, “Double patterning layout decomposition for reticle memory conflict and stitch minimization,” in Proc. Int. Symp. Phys. Des., Mar. 2009, pp. 105–114. [3] K. Yuan, K. Lu, and D. Z. Pan, “Double patterning lithography friendly detailed routing with redundant consideration,” in Proc. Des. Autom. Conf., Jun. 2009, pp. 63–64. [4] K. Yuan and D. Z. Pan, “WISDOM: Wire spreading enhanced decomposition method in double patterning lithography,” in Proc. Int. Conf. Comput.-Aided Des., Nov. 2010, pp. 32–39. [5] J.-S. Yang, K. Lu, M. Cho, K. Yuan, and D. Z. Pan, “A new graph-theoretic, multi-objective layout decomposition framework for double patterning lithography,” in Proc. Asia South Pac. Des. Autom. Conf., Jan. 2010, pp. 637–644. [6] B. Wang, K. Yuan, and D. Z. Pan, “Layout decomposition for triple patterning lithography,” in Proc. IEEE/ACM Int. Conf. Comput.-Aided Des., Nov. 2011. [7] A. Fujimura, “Beyond light: The growing importance of E-beam,” in Proc. Int. Conf. Comput.-Aided Des., Nov. 2009. [8] A. Fujimura, “Design for E-beam: Getting the best wafers without the exploding mask costs,” in Proc. Int. Symp. Quality Electron. Des., Mar. 2010. [9] A. Fujimura, T. Mitsuhashi, K. Yoshida, S. Matsushita, L. L. Chau, T. D. T. Nguyen, and D. MacMillan, “Stencil design and method for improving throughput and for cell projection in charged particle beam lithography,” U.S. Patent 2009/0083108, Jan. 2010. [10] H. C. Pfeiffer, “New prospects for electron beams as tools for semiconductor lithography,” Proc. SPIE, vol. 7378, p. 737802, May 2009. [11] T. Fujino, Y. Kajiyama, and M. Yoshikawa, “Character-build standard-cell layout technique for high-throughput character-projection EB lithography,” Proc. SPIE, vol. 8190, p. 819016, Jul. 2011. [12] M. Sugihara, T. Takata, K. Nakamura, Y. Matunaga, and K. Murakami, “A CP mask development methodology for MCC systems,” Proc. SPIE, vol. 6283, p. 628331, May 2006. [13] M. Sugihara, K. Nakamura, Y. Matunaga, and K. Murakami, “CP mask optimization for enhancing throughput of MCC systems,” Proc. SPIE, vol. 6349, p. 63494B, Oct. 2006. [14] Y. Matunaga, M. Sugihara, and K. Murakami, “Technology mapping technique for enhancing throughput of multi-column-cell systems,” Proc. SPIE, vol. 6517, p. 65170Z, Mar. 2007. [15] M. Sugihara, “Optimal character-size exploration for increasing throughput of MCC lithographic systems,” Proc. SPIE, vol. 7271, p. 72710L, Feb. 2009. [16] Y. Zhang and C. Chu, “RegularRoute: An efficient detailed router with regular routing patterns,” in Proc. Int. Symp. Phys. Des., Mar. 2011, pp. 45–52. [17] Y. Zhang and C. Chu, “Fastroute 3.0: A fast and high quality global router based on virtual capacity,” in Proc. IEEE/ACM Int. Conf. Comput.-Aided Des., Nov. 2008, pp. 344–345. [18] Y. Zhang and C. Chu, “CROP: Fast and effective congestion refinement of placement,” in Proc. IEEE/ACM Int. Conf. Comput.-Aided Des., Nov. 2009, pp. 344–350. [19] H. Murata, K. Fujiyoshi, S. Nakatake, and Y. Kajitani, “VLSI module placement based on rectangle-packing by the sequence-pair,” IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., vol. 15, no. 12, pp. 1518–1524, Dec. 1996. [20] X. Tang, R. Tian, and M. Wong, “Fast evaluation of sequence pair in block placement by longest common subsequence computation,” in Proc. Des. Autom. Test Eur. Mar. 2000, pp. 106–113. [21] S. H. Adya and I. L. Markov, “Fixed-outline floorplanning: Enabling hierarchical design,” IEEE Trans. Very Large Scale Integr. Syst., vol. 11, no. 6, pp. 1120–1135, Dec. 2003. [22] J. Z. Yan and C. Chu, “Optimal stack-driven block shaping algorithm in fixed-outline floorplanning,” in Proc. Int. Symp. Phys. Design, Mar. 2012. [Online]. Available: http://www.akira.ruc.dk/~keld/research/LKH [23] J. Z. Yan and C. Chu, “DeFer: Deferred decision making enabled fixed-outline floorplanning algorithm,” IEEE Trans. Comput.-Aided Des., vol. 29, no. 3, pp. 367–381, Mar. 2010. [24] J. Z. Yan, T. Vosoughi, and C. Chu, “Handling complexities in modern large-scale mixed-size placement,” in Proc. IEEE/ACM Des. Autom. Conf., Jul. 2009, pp. 436–441. [25] J. Z. Yan and C. Chu, “Handling complexities in modern large-scale mixed-size placement,” IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., vol. 30, no. 7, pp. 1020–1033, Jul. 2011. Kun Yuan received the B.S. degree in electronic engineering information science from the University of Science and Technology of China, Hefei, China, in 2004, and the Ph.D. degree in electrical and computer engineering from the University of Texas, Austin, in 2010. He was with TeraRoute, Austin, TX, in 2007 and with NVIDIA, Santa Clara, CA, in 2009. He is currently a Senior Member of Technical Staff with Cadence Design Systems, San Jose, CA. His current research interests include physical design automation for manufacturability and numerical optimization. Dr. Yuan received the ISPD Routing Contest Award in 2007, and three Best Paper Awards (ASPDAC 2010, ISPD 2011, and the IBM Research 2010 Pat Goldberg Memorial Best Paper Award in CS/EE/Math). Bei Yu received the M.S. degree in computer science from Tsinghua University, Beijing, China, in 2010. He is currently pursuing the Ph.D. degree with the Department of Electrical and Computer Engineering, University of Texas, Austin. His research interests include physical design, design for manufacturability, and optimization algorithms with applications in very large scale integration computer-aided design. David Z. Pan (S’97–M’00–SM’06) received the B.S. degree from Peking University, Beijing, China, and the M.S. and Ph.D. degrees from the University of California, Los Angeles (UCLA). From 2000 to 2003, he was a Research Staff Member with the IBM T. J. Watson Research Center, Yorktown Heights, NY. He is currently an Associate Professor with the Department of Electrical and Computer Engineering, University of Texas, Austin. He has published over 160 papers in international conferences and journals, and holds eight U.S. patents. His current research interests include advanced physical design, design for manufacturability and reliability, vertical integration design and technology, and design/computer-aided design (CAD) for emerging technologies. Dr. Pan has served as an Associate Editor for the IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, the IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION SYSTEMS, the IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS PART I and PART II, the Journal of Computer Science and Technology, and the IEEE CIRCUITS AND SYSTEMS SOCIETY NEWSLETTER. He has served as the Chair of the IEEE CANN Committee and the ACM/SIGDA Physical Design Technical Committee. He is in the IEEE PES/PEMC Working Group on the International Technology Roadmap for Semiconductors. He has served in the technical program committees of major VLSI/CAD conferences, including ASPDAC (Subcommittee Chair), DAC (Subcommittee Chair), DATE, ICCAD (Subcommittee Chair), ISPD (Program Chair), ISCAS (Program Chair), ISCAS (Publication Chair), GLSVLSI, ACISC (Program Chair), ICICDT (Award Chair), and VLSI-DAT (EDA Track Chair). He was the General Chair of ISPD 2008, the General Chair of ACISC 2009, and the Steering Committee Chair of ISPD 2009. He received a number of awards for his research contributions and professional services, including the ACM/SIGDA Outstanding New Faculty Award in 2005, the NSF CAREER Award in 2006, the SRC Young Researcher Award in 2006, 2007, and 2008, the IBM Faculty Award four times in 2004–2006, 2010, the UCLA Engineering Distinguished Young Alumnus Award in 2009, seven Best Paper Awards (SRC Techcon 1998 and 2007, DATE 2009, ICICDT 2009, ASPDAC 2010, ISPD 2010, and the IBM Research 2010 Pat Goldberg Memorial Best Paper Award), CS/EE/Math Best Paper Award, Best Paper Award nominations from DAC/ICCAD/ASPDAC/ISPD, ISPD Routing Contest Awards in 2007, the eASIC Placement Contest Grand Prize in 2009, the IBM Research Bravo Award in 2003, the Dimitris Chorafas Foundation Research Award in 2000, and the ACM Recognition of Service Award in 2007 and 2008. He was an IEEE CAS Society Distinguished Lecturer from 2008 to 2009.
PAROCHIAL NOTES. Sunday School. 10 a.m. and 2.30 p.m. Boys in Parish Hall, Superintendent—Mr. Lovell. Girls in Club Room, Superintendent—Miss M. Harvey. Infants in Church Room at 10.30 a.m. and 2.30 p.m. Superintendent—Miss Watts. Bible Class. For Girls on Sundays in the Club Room at 3 p.m. Band of Hope. In the Club Room on Mondays at 6.15 p.m. Hon. Sec.—Mrs. Grange. Lads’ Club. In Club Rooms, School Road, Mondays and Fridays, 7 p.m. to 9.30 p.m. Hon. Sec.—Mr. Creane. Parochial Church Officer. Mr. R. H. Alford, “Sarum,” Uxbridge Road. Missionary Association. Meetings as announced. Hon. Sec.—Miss Jakeman. Mothers’ Union. Monthly Meetings, First Wednesday in each month in the Club Room, 2.30 p.m., and Quarterly Meetings. Enrolling Member—Mrs. Harvey. St. James’s Young People’s Social Guild. Tuesdays in the Club Room from 8 to 10.30 p.m. Open to Communicants and intending Confirmands between the ages of 16 and 30 years. Hon. Sec.—Mr. E. Gubbins, 36, Links View Road. Mothers’ Meeting. Thursdays in the Club Room, 2.30 p.m. Superintendent—Miss Butler. Girl Guides. Fridays in the Club Room, 7—9 p.m. Captain—Miss R. Le Blanc. Brownies. Thursdays in Club Room, 5.30—7 p.m. Acting Brown Owl—Miss M. Winter. Sea Scouts. Wednesdays in Church Room, 7.30 p.m. Group Scout Master—Mr Ealden. Cub Pack. Wednesdays in Church Room, 5.45 p.m. Cub Master—Miss Newman The Vicar may be seen at the Vicarage on Tuesdays, Thursdays and Saturdays, between 8.45 and 9.45 a.m.; and generally on any day except Tuesday, between 6 and 7 p.m. Sickness. The Vicar will be grateful if cases of sickness are reported to him at once. Churchyard. Contributions towards the upkeep of the Churchyard will be gratefully received by Mr. C. H. Evans, Roseneath, Edward Rd., Hampton Hill. Vicar’s Letter. My Dear People, As is usual at this time of year I have been inundated with requests to insert notices of May meetings in connection with various Church Societies. Unfortunately the space at my disposal will not allow me to give even a line or two to all of them. I have therefore picked out what I think to be the most important as far as we are concerned. I often wonder how many from this parish attend any of the Annual General Meetings that are advertised. Judging from the few who turn up for meetings in the parish on behalf of one or other of the Departments of the Church’s work outside the parish, I should imagine that space at meetings in Town is not overtaxed on account of the attendance of interested people from Hampton Hill. It is rather a pity because the work of such Societies as the Zenana Missionary Society, the British and Foreign Bible Society, to mention only two, are worthy of the interest of all Church people and there should be a representative few willing to spare an evening once a year to learn something about their activities. However, I commend to you the notices you will find elsewhere and shall be glad to know of any who do attend them. This month is one of Fast and Festival. At the beginning comes Rogationtide. The design of these three Fast Days was originally intended to provide a period of fervent prayer and humiliation for averting impending evils, by appeasing the wrath of God, so that He might avert those judgements which the sins of the nation deserved and that He might be pleased to bless the fruits of the earth. We use these days now not only as a time of intercession for a blessing on the crops, but as a time of interceding for the spiritual harvest. Thus the work of the Church becomes prominent in our thoughts. For many years now our Bishop has sent a Pastoral letter which I have read in Church and in which he commended to our prayers and almsgiving the work of the Church in the Diocese as a whole especially with regard to the poorer parishes and, of late, to the needs of the vast population which has been springing up in new areas of Middlesex. This year we shall not have a Pastoral letter. The needs, however, are even greater and we shall plead for these needs before the Throne of Grace on the Monday, Tuesday and Wednesday before Ascension Day. Also on Rogation Sunday the collections will, as usual, go towards our Parochial Quota to the London Diocesan Fund. The Quotas of all parishes have been raised this year, ours of obligation is £83. I would therefore ask you all to be liberal in your offerings on Sunday, May 6th. Following immediately on Rogationtide comes the Feast of the Ascension, in fact the three days before are used by many as a time of preparation for keeping the Festival. It is often felt that it is a pity this great Festival comes in the middle of the week instead of on Sunday when most people are accustomed to go to Church. Personally, I see no reason why the majority of communicants and regular Churchgoers should not pay honour to our Lord on His Coronation Day even if it does come in the middle of the week. I have put a celebration of the Holy Communion at half past six and there are very few communicants who could not attend that service and have time to get their breakfast before going to work. And then there is Evensong at a time when most people have finished their day's work. There is little excuse for the neglect of this important Festival and I hope many more than usual will be present this year. The pity to my mind is that the Festivals seem to fall over one another at this time of the year for Whit Sunday and Trinity Sunday follow so closely that the latter is the most neglected of all. Shall we all try this year to be present at the Services of all these Festivals? I should be grateful for flowers for the decoration of the Church on Whit Sunday. Please send your gifts of flowers on the Saturday between half past nine and half past ten in the Morning. Death has been busy in the ranks of the Mothers' Union during the last few weeks. Two of the members Mrs. Garrett, and Mrs. Jarvis, have been called to their rest and our deepest sympathy goes to their relations in their sad grief. The arrangements for the Annual Fête are progressing. One new member of the congregation has offered to provide a stall. There may be others willing to do the same or help in some other way. I should be very pleased if they would come and see me or write to me on the matter. The Bishop of London has asked all parishes in the Diocese to observe Trinity Sunday as a Day of Prayer for New London. I remain, Your faithful friend and Vicar, FREDK. P. P. HARVEY. Days of Special Remembrance. May 1st, St. Philip and St. James; Holy Communion, 7.30 a.m. May 6th, Rogation Sunday. May 7th, 8th, 9th, Rogation Days; short Intercession Service each day at 8 p.m. May 10th, Ascension Day; Holy Communion, 6.30 a.m., 7.30 a.m. and 11 a.m., Evensong and Sermon, 8 p.m. May 20th, Whit Sunday; Holy Communion, 7 a.m., 8 a.m. and 12 noon, other Services as usual. May 27th, Trinity Sunday; Holy Communion, 8 a.m., Choral Eucharist, 10 a.m., other Services as usual. May 31st, Corpus Christi; Holy Communion, 7.30 a.m. and 11 a.m. Mothers' Union. The Deanery Festival Service held in St. Mary's Church, Twickenham, was exceedingly well attended. There were 720 members present. Canon Gibson gave a most appropriate address based on the story of Martha and Mary as given in St. Luke X, 38—42. He showed how the character of Martha was reproduced to-day and said that although there were also Marys it would be well for the Marthas to become more like the Marys in their spiritual and devotional lives. Specially in this year of the "Call to Renewal" would it be fitting for the members of the Mothers' Union to look into their lives with regard to their worship and particularly their communions. Sixteen Branches took their banners, St. James's, Hampton Hill being amongst them. During tea, which was served in York House, Mrs. Marshall, the Presiding member told the company that two new branches had lately been started and that the Hampton Deanery had the third largest number of members of the Deaneries in the London Diocese. The Monthly Meeting will be held in the Club Room on Wednesday, May 2nd at 3 p.m., when Mrs. Creasy has kindly promised to tell the Mothers something about the Mothers of Hong Kong. District Nursing Association. The Annual General Committee Meeting was held at the Vicarage some little time ago. The report showed that during 1933 one hundred and twenty nine patients were nursed, involving two thousand three hundred and seventy nine visits. The Committee placed on record their great appreciation of Nurse Harvey's ungrudging and skilful service. Amongst those deserving the thanks of the Committee were Mrs. Rosten, Mrs. Glass and her Alexandra Rose Day helpers, the voluntary collectors, the Alexandra Rose Day Committee, the Hampton Flower Show Committee etc. While the Balance Sheet showed a satisfactory year's working, it was reported that an additional £25 a year was needed to keep the finances of the Association in a satisfactory condition. With the steadily increasing population it should not be difficult to raise this extra amount. Jumble Sale. The Annual Jumble Sale will be held during this month. Please look out for the actual date when advertised! Send all the goods you have to dispose of and make a record financial success of this effort. Missionary Association. The Diocesan Missionary Festival Service will be held at St. Paul's on Tuesday, May 8th at 7 p.m., when it is hoped that every parish in the Diocese will be represented. If a sufficient number wish to attend this service a motor coach will be engaged for the evening. The Church Missionary Society. The 135th Anniversary of the above Society is being celebrated on Monday, April 30 and Tuesday, May 1st. The Annual Festival Service will be held in St. Bride's Church, Fleet Street, at 6.30 p.m. on Monday, April 30th, when the preacher will be the Archbishop of York. On May 1st the Annual Meeting will be held in the Central Hall, Westminster, at 11 a.m., the speakers being the Bishop in Kyushu, Canon Stacey Waddy and Canon A. S. J. Thorpe. There will be a Mass Meeting at the Royal Albert Hall on Tuesday, May 1st at 7.15 p.m., the Archbishop of Canterbury, the Bishop of Croydon, Canon E. C. Gore and the Rev. W. H. Murray Walton being the speakers. British and Foreign Bible Society. The Annual Meeting of the above will be held in Queen's Hall, Langham Place, on Wednesday, May 2nd at 11 a.m. The Earl of Athlone, K.G. will be in the Chair and the Archbishop of York, Sir Josiah Stamp and the Rev. A. T. Thompson will address the Meeting. The Church of England Zenana Missionary Society. The Annual Meeting will be held in Caxton Hall, Westminster, on Friday, May 4th, at 2.30 p.m. There will also be an evening Meeting in the same Hall on the same day at 7 p.m. A Diocesan Pageant. One hundred and thirty parishes are taking part in a Pageant Play in Saddlers Wells from May 28th—June 9th. The title of the Pageant is "The Rock." The performances will be given in the evening probably at 8 o'clock. The times of the Matinees are not to hand at present. The prices are 10/6, 7/6, 5/-, 3/6, 2/6, 1/6 and 9d. The Vicar will be glad to organize a party to go to the evening performance on Tuesday, June 5th, providing sufficient names are sent to him by Tuesday, May 15th. The object of this effort is to raise money for the 45 Churches Fund. For further information about this Diocesan venture please read the notice in the Diocesan Leaflet in the April issue of the Parish Magazine and the Bishop's letter in the May number. The Parish Magazine. Two more distributors of the Parish Magazine are urgently needed, one to begin at the end of May, the other at the end of June. The Magazine is published on the last Saturday of the month and it is desirable to take them round as early as possible after that. Miss Jakeman will be very glad to hear of anyone who will undertake this little bit of work for the Church. Acknowledgments. The Treasurers acknowledge with many thanks the following donations:—For the cleaning of the clock, G. F. Stutchbury, Esq., 10/-, Mrs. Swain, 10/-, for the General Fund, Madame Mottu, 10/- and Mrs. Seward, 10/-. Young People’s Social Guild. I am pleased to report that the membership of the Guild continues to increase. On Easter Monday, about twenty members and friends went for a hike. The party tramped from Epsom across the Downs, via Headley, to Box Hill, where a halt was made for lunch. After a rest the "happy band of pilgrims" made their way over Fetcham Downs to Bookham and had a very enjoyable tea at the Old Crown Hotel. From Bookham the party walked across fields to Leatherhead Station, thus concluding a most successful and happy day's ramble. E. GUBBINS, Chairman. BAPTISMS. "Made a Member of Christ." Apr. 8—Myra Rose Turner. ,, 15—Frances Beryl Thorne. ,, 15—Joseph Charles Victor Mitchell. ,, 15—John Raymond Frederick Harvey. ,, 22—Brian Bertram Brown. MARRIAGES. 'Whom God hath joined together let no man put asunder.' Mar. 31—Frank Baldwin and Kathleen Mona Keats. Apr. 1—George Sidney Calvert Lee and Ivy Beatrice Archer. ,, 1—Roland Clifford Adams Baker and Eunice Elizabeth Higgins. ,, 14—Edgar Thomas Hounsome and Katie Evelyn Wood. BURIALS. "I am the Resurrection and the Life." Mar. 24—Dennis Matthew Wilkinson, aged 89 years. ,, 27—Edith Emily Seal, aged 69 years. ,, 31—Ada Charlotte Cheeseman, aged 77 years. ,, 31—John Farnham Laughton, aged 83 years. Apr. 14—Ethel Laura Garrett, aged 56 years. ,, 18—Ellen Key, aged 80 years. ,, 19—Eliza Jarvis, aged 81 years.
Regional NO\textsubscript{x} emission strength for the Indian subcontinent and the impact of emissions from India and neighboring countries on regional O\textsubscript{3} chemistry T. Kunhikrishnan,\textsuperscript{1,2} Mark G. Lawrence,\textsuperscript{1} Rolf von Kuhlmann,\textsuperscript{1,3} Mark O. Wenig,\textsuperscript{4} Willem A. H. Asman,\textsuperscript{1} Andreas Richter,\textsuperscript{5} and John P. Burrows\textsuperscript{5} Received 1 April 2005; revised 16 January 2006; accepted 20 March 2006; published 2 August 2006. [1] This study examines the regional NO\textsubscript{x} emission strength and influence of external emissions on O\textsubscript{3}-related chemistry over the Indian subcontinent based on a three-dimensional chemistry transport model and space-based observations, following up on our earlier work on NO\textsubscript{x} chemistry over southern Asia. NO\textsubscript{x} mass and emissions from Global Ozone Monitoring Experiment (GOME) and Model of Atmospheric Transport and Chemistry—Max Planck Institute for Chemistry Version (MATCH-MPIC) over India are compared, and the uncertainties in the estimates are briefly discussed. This study also employs the concept of small-perturbation tendencies to compute the scaled sensitivities with modified NO\textsubscript{x} emissions for India and the neighboring continents. The sensitivities of NO\textsubscript{x}, O\textsubscript{3}, OH, and reservoir species of NO\textsubscript{x} over India to NO\textsubscript{x} emission from India, Southeast Asia, Africa, China, and the Middle East are discussed with respect to the photochemistry and the regional meteorology. Our study suggests that the mean regional NO\textsubscript{x} emission strength for India is close to 2.5 Tg(N)/yr with a seasonal maximum (~3 Tg(N)/yr) during April and minimum (~1.6 Tg(N)/yr) during winter. The changes in the O\textsubscript{3} concentrations with respect to NO\textsubscript{x} and nonmethane hydrocarbon emissions from India show that southern India is relatively more sensitive to local emission. Citation: Kunhikrishnan, T., M. G. Lawrence, R. von Kuhlmann, M. O. Wenig, W. A. H. Asman, A. Richter, and J. P. Burrows (2006), Regional NO\textsubscript{x} emission strength for the Indian subcontinent and the impact of emissions from India and neighboring countries on regional O\textsubscript{3} chemistry, \textit{J. Geophys. Res.}, 111, D15301, doi:10.1029/2005JD006036. 1. Introduction [2] Industrialization, urbanization, rapid traffic growth and increasing levels of anthropogenic emissions have resulted in a substantial deterioration of air quality over Asia [van Aardtmele et al., 1999; Horowitz and Jacob, 1999; Jaffe et al., 1999; Levy et al., 1999; Streets and Waldhoff, 1999; Yienger et al., 2000]. A recent study by Akimoto [2003] indicates that the NO\textsubscript{x} emission rate from Asia now exceeds the amounts emitted in North America and Europe and the trend is expected to continue over the next two decades. In contrast to the thinning of the O\textsubscript{3} layer in the stratosphere, the O\textsubscript{3} burden in the troposphere is generally increasing because of increasing emissions of precursors such as nitrogen oxides NO\textsubscript{x} (NO + NO\textsubscript{2}) and volatile organic compounds (VOC). NO\textsubscript{x} and O\textsubscript{3} levels in urban air and amounts of nitrate in the aerosol particles over south Asia indicate that NO\textsubscript{x} levels are not negligible [Lal et al., 2000; United Nations Environment Programme (UNEP), 2002]. [3] NO\textsubscript{x} indirectly influences the radiation budget of the atmosphere through O\textsubscript{3}, which possibly represents 10–15% of the total anthropogenic greenhouse radiative forcing in the atmosphere [Fishman et al., 1979; Lacis et al., 1990, Wild et al., 2001]. NO\textsubscript{x} also influences the oxidation capacity of the atmosphere through OH and nitrate. O\textsubscript{3} production in the troposphere is mainly due to the oxidation of CH\textsubscript{4}, CO and hydrocarbons in the presence of NO\textsubscript{x} [Crutzen, 1974; Chameides and Walker, 1973]. A major uncertainty in tropical tropospheric chemistry is the quantification of NO\textsubscript{x} emissions from fossil fuel, bio fuel and biomass burning, soils and lightning. The photochemical transformation and losses of NO\textsubscript{x} are similarly uncertain. [4] Our present knowledge of NO\textsubscript{x} and nonmethane hydrocarbon (NMHC) chemistry and their impact on regional ozone levels is limited because of the lack of observations of various trace gases, especially over India. In situ measurements are limited to certain locations over major cities or from field campaigns. These data are insufficient to assess the overall air quality of the country as a whole. It is reported that the emission of NO\textsubscript{x} due to fossil fuel burning could increase even more rapidly in the... coming years. The uncertainty in our knowledge of the NO\textsubscript{x} chemistry and its influence on O\textsubscript{3} over India is a major problem in the way of developing effective pollution control strategies as well as a better understanding of the air quality of this region. However, a systematic approach to quantify the NO\textsubscript{x} emissions from local and remote sources is also difficult because of nonlinear chemical processes in the atmosphere. As mentioned in our earlier study [Kunhikrishnan et al., 2004a] (hereinafter referred to as K04a), we have proposed the use of the scaled sensitivity with model simulations to examine the impact of local NO\textsubscript{x} emissions over source and downwind regions. K04a indicates that moderate increase in NO\textsubscript{x} over India will not result in a significant change in O\textsubscript{3}, taking into account the nonlinearity with the NO\textsubscript{x} lifetime, and the OH feedback. [3] Recent studies based on in situ observations over certain locations over India indicate that the rural atmosphere is significantly influenced by the urban emission, by transport and mixing [Chand and Lal, 2004; Varsanyi and Agarwal, 1992]. As the region is highly heterogeneous with respect to different climate zones of arid and semiarid features and with respect to emissions from urban and rural areas, the overall atmospheric chemistry is complex. Synergistic use of a chemical transport model and satellite observations can provide a better understanding about the NO\textsubscript{x}-related chemistry over this region. [4] In continuation of our earlier work on Asian NO\textsubscript{x} (K04a), this study examines the regional characteristics of NO\textsubscript{x} induced chemistry over India using outputs from the global model, Model of Atmospheric Transport and Chemistry—Max Planck Institute for Chemistry Version (MATCH-MPIC), retrievals of the tropospheric NO\textsubscript{2} column, and corresponding emission estimates from the Global Ozone Monitoring Experiment (GOME) instrument on board the ERS-II satellite, which are described in the next section, including an examination of the quality of the O\textsubscript{3} simulation with respect to available MOZAIC profiles over India. In section 3, we present a review of NO\textsubscript{x} emissions over India, examine the regional NO\textsubscript{x} emission strengths based on the model, and discuss the uncertainties when using GOME NO\textsubscript{2} observations for deriving source strengths. Following that in section 4, a theoretical approach to examine the nonlinearity in the model runs with modified NO\textsubscript{x} emissions over India is discussed. On the basis of this, the influence of continental emissions on NO\textsubscript{x} chemistry over India in light of the seasonal meteorology and the variation of O\textsubscript{3} with respect to precursor emissions are briefly discussed. 2. GOME Satellite Observations and Global Three-Dimensional Chemical Transport Model [7] This study uses the retrieval of tropospheric NO\textsubscript{2} columns [Richter and Burrows, 2002] (hereinafter referred to as RB02) along with the derived NO\textsubscript{2} emissions [Wong, 2001] (hereinafter referred to as W01) from GOME. GOME is a scanning spectrometer on board the ERS-2 satellite, which was launched by the European Space Agency in April 1995. ERS-2 is in a Sun-synchronous orbit, approximately 800 km above Earth, crossing the equator at 1030 local time. Detailed descriptions of GOME NO\textsubscript{2} retrieval methods are given by Burrows et al. [1999]. For the retrievals used in this work, the spectral window 423–451 nm, where NO\textsubscript{2} has strong absorption features, has been selected. GOME NO\textsubscript{2} emission data used in this study are based on an improved algorithm (W01) following on the study of Leue et al. [2001]. GOME has shown its capability to observe many realistic features in the atmosphere; however, some of the drawbacks are its low resolution and the uncertainties in the retrieval techniques [Kunhikrishnan, 2004]. Generally, the uncertainty in GOME is the smallest for NO\textsubscript{2} source regions and high-albedo regions (RB02) and is typically of the order of $4 \times 10^{14}$ molecules cm$^{-2}$. [8] This study also employs data from a simulation with the three-dimensional global chemistry-meteorology model MATCH-MPIC version 3.1, at a horizontal resolution of T21, 28 vertical levels from surface to 2.7 hPa and a time step of 30 min. Details of the chemical and physical processes in the model are given by Lawrence et al. [1999, 2003], von Kuhlmann et al. [2003a, 2003b], and Rasch et al. [1997]. MATCH is an offline chemistry transport model with driving meteorology from the NCEP/NCAR Reanalysis data [Kalnay et al., 1996]. The model uses the basic meteorological parameters such as pressure, temperature, and wind, which are used to compute the convective transport, cloud microphysics, including cloud fraction for both convective and nonconvective clouds, cloud water and precipitation. The model is able to simulate the full hydrological cycle and determines the water vapor distribution from latent heat fluxes, transport and cloud parameters. The MATCH meteorology simulates transport by advection, convection and dry turbulent diffusion. The chemical component of the model consists of sources, sinks and chemical transformations of important trace gases, which are divided according to their lifetimes into transient and nontransientated species. The model consists of 56 species with 141 gas phase reactions, out of which 33 are photolysis reactions. The chemical scheme in the model deals with photochemical reactions involving major chemical species and the transport related to O\textsubscript{3}, HO\textsubscript{2}, NO\textsubscript{2}, CH\textsubscript{4}, CO and VOCs (up to C5 hydrocarbons). [9] The sources considered in MATCH-MPIC are partly from direct emissions and partly from photochemical transformations. In addition to that, stratospheric intrusions and transport from source to receptor regions also play significant roles as sources of major trace gases. Detailed information on NO\textsubscript{x} emissions used in the model is given in K04a. Carbon monoxide (CO) emission is mainly from industrial and biogenic activities and biomass burning. The oxidation of volatile organic compounds (VOC) also contributes to the total $\sim 2260$ Tg (CO)/yr in the model. Methane (CH\textsubscript{4}) is fixed in the surface layer in the model allowing transport and photochemical oxidation by OH to determine its concentration in the rest of the atmosphere [Lawrence et al., 1999]. [10] Various studies have been made to evaluate MATCH-MPIC with observations from satellites, aircraft data, ozonesondes and various field campaigns over different regions [von Kuhlmann et al., 2003a, 2003b; Lawrence et al., 1999, 2003; Kunhikrishnan et al., 2004a, 2004b]. The model has been extensively compared with observations of various species such as NO\textsubscript{2}, O\textsubscript{3}, and CO and was found to generally reproduce the spatial and temporal variability such as land sea gradients and seasonal cycles. These include Figure 1a. O₃ vertical profiles (green lines) and their mean profile (black line) from MOZAIC and the monthly mean profile (red line) from MATCH-MPIC in parts per billion by volume (ppbv) for Delhi. INDOEX data covering outflow from India [Lawrence et al., 2003], a study of Asian NOₓ, in particular over India along with the GOME NO₂ column (K04a) and observations from the MINOS campaign considering the transport of O₃ precursors from southern Asia to the Mediterranean [Lawrence et al., 2003; Lawrence, 2004]. Over India, the magnitude of the modeled NO₂ column is comparable to the observations from GOME, although the model tends to underestimate the pronounced maximum during the biomass burning periods (K04a). The model is also able to reproduce the occurrence of semiregular NO₂ plumes in the middle troposphere (MT) over the southern Indian Ocean during the monsoon transition periods as observed by the GOME satellite [Kunhikrishnan et al., 2004b]. [11] As a further step toward evaluating the model performance of O₃ chemistry over India, O₃ profiles from Measurement of Ozone by Airbus In-service Aircraft (MOZAIC) [Law et al., 1998; Marencio et al., 1998; Thouret et al., 1998] available over this region which have been preprocessed for 1997–1999 are compared with the monthly mean 24-hour O₃ from MATCH-MPIC. [12] Direct comparison between the model and MOZAIC is limited by the low model resolution and monthly averaged output as compared to the instantaneous MOZAIC profiles available over Indian airports. In this study we use vertical profiles corresponding to takeoff and landing within a horizontal range of 200 km of 3 Indian airports which are frequented by the MOZAIC aircraft: Chennai, Delhi and Mumbai. The monthly mean concentration of O₃ from MATCH-MPIC (red line) is plotted (Figures 1a–1c) with available MOZAIC profiles (green lines) with their mean (black line). Generally, the O₃ concentration increases with respect to height, which is more predominant over Delhi as compared to Mumbai and Chennai. MOZAIC profiles typically show low mixing ratios of O₃ (<40 ppbv) over Mumbai and Chennai, while over Delhi it is 60–80 ppbv within 1–2 km altitude. Generally, MATCH-MPIC tends to overestimate the O₃ concentration as compared to MOZAIC at all levels with some exceptions during October–December for Delhi. However, MATCH-MPIC does reproduce some of the main features. In particular, the lower troposphere (LT) peak, which is frequently seen around 900 hPa, the rough overall vertical shape and the seasonality are reproduced by the model. To analyze the seasonal variations, we examined the time series at three isobaric levels in the LT close to the surface (915 hPa, Figure 2a), the transition layer (690 hPa, Figure 2b), and the midtropospheric layer (500 hPa, Figure 2c) from MOZAIC and MATCH-MPIC shown here in Figure 2 for the example of Chennai. The model simulates a seasonal cycle similar to the one observed by MOZAIC, in particular the O₃ maxima during the dry summer and minima during the summer. monsoon months. Finally, it is interesting to note that there is not much interannual variation in the $O_3$ over these three cities on the basis of either of the data sets. 3. NO$_x$ Emission From India [13] The chemical processes that tend to maintain the chemical composition of the atmosphere and influence climate are initially controlled or modified by the emission of trace gases. Because of the short lifetime of tropospheric NO$_x$, its regional distribution is strongly controlled by its emissions; however, in the midtroposphere (MT) to upper troposphere (UT) NO$_x$ mainly corresponds to the transport by advection and convection from remote and underlying sources. The main sources of the NO$_x$ and hydrocarbons over India are vehicular transport and biofuel burning. Power generation and industrial emissions contribute the remaining fraction. A study by Gadi et al. [2003] has given an estimate of 0.6–1.4 Tg for 1990 from biofuel NO$_x$ emission, which is mainly from the domestic sector of rural India that provides 70% of the total energy requirements. The increasing population and urbanization result in a 4 to 6% annual growth rate of energy consumption in India, which enhances the emissions of NO$_x$ and other trace gases [UNEP, 2002]. Long-term air quality data over the residential and industrial locations of different Indian cities indicate that NO$_2$ levels are increasing steadily over major cities, and some of them already exceed the permissible limits, as high as 100 to 240 $\mu g/m^2$. Different studies based on sector analysis and measurement campaigns show that NO$_x$ emissions over India are expected to increase in the forthcoming years, even though these estimates have large uncertainties for different sectors [Carmichael et al., 1998; Garg et al., 2001; Mitra and Sharma, 2002; Sharma et al., 2002; Streets et al., 2003]. [14] It is clear that more efforts are needed to understand the spatial and temporal evolution of regional NO$_x$ emissions over India with respect to the increasing demand of energy consumptions, rapid urbanization and efforts to stabilize the increasing level of greenhouse gas concentrations as well as improve the air quality of this region. Data from chemistry transport models and satellites provide valuable information to improve estimates of NO$_x$ emissions as well as to identify the source regions and to study the regional $O_3$ chemistry in light of seasonal meteorology. 3.1. Regional NO$_x$ Emission Strength From MATCH-MPIC [15] The industrial source of NO$_x$ used in MATCH-MPIC is based on the Emission Database for Global Atmospheric Research (EDGAR), version 2.0, $1^\circ \times 1^\circ$ emissions inventory [Olivier et al., 1999]. In the EDGAR database, the Indian emissions consist of about 64% from fossil fuel combustion and 30% from biofuel combustion, and the total NO$_x$ emission for 1995 is estimated as 1.63 Tg(N). In addition to this, the model takes into account the emission from soils, lightning, aircraft, biomass burning and transport from the stratosphere, which are discussed in K04a. The NMHC emissions from India consist of about 16.4% from fossil fuel combustion and 69.6% from biofuel combustion with a total national emission of 8.1 Tg(C) for 1995. NMHCs are emitted on a speciated basis, based on EDGAR data for anthropogenic emissions. The industrial and anthropogenic NO$_x$ and NMHC emissions from different sectors especially over India used in the model based on EDGAR emission data are shown in Figure 3. [16] MATCH-MPIC shows an enhancement of NO$_x$ emissions over north to northeast India, with maximum emissions during the dry summer period of India (April) and the least during winter months (October–December). Studies based on ATSR and AVHRR satellite observations show that NE India, especially Orissa and West Bengal, are the regions of most intense biomass burning in India during March–April [Duncan et al., 2003]. The Indian NO$_x$ emissions used in MATCH-MPIC for India are discussed in comparison to GOME in section 3.4. The mean NO$_x$ emission for India used in the model is 2.28 Tg(N) for 1997–1998 with a monthly standard deviation of 0.44 Tg(N). [17] In the following section, we examine the integrated mass of NO$_2$ over India and the seasonal variation of the NO$_x$ lifetime with uncertainties from the model, which are the main parameters that are used for computing the emission strength based on satellite data. ### 3.2. Vertical Distribution of Integrated NO$_2$ Mass Over India [18] Knowing the NO$_2$ mass abundance in the troposphere enables us to estimate the emission strength for a region. For species like NO$_2$, which are present in both the troposphere and the stratosphere, the retrieval of the tropospheric contribution by satellite remote sensing is difficult. As such estimates are not easily available from satellite measurements here we examine the fraction of NO$_2$ mass in different levels of the troposphere over India by using the model. The integrated mass of the NO$_2$ column for five isobaric levels, the boundary layer (BL; surface to 850 hPa), MT (850–500 hPa), UT (500–150 hPa), tropospheric column (surface to 150 hPa) and the stratospheric column (150–2.7 hPa) are computed for the model grids corresponding to the Indian domain. The ratios of mass computed for different levels are summarized in Table 1. This shows that, in winter, $\sim$67% and in summer, $\sim$39% of the NO$_2$ mass in the tropospheric column is in the BL close to the source regions. This is mainly due to subsidence and lack of convective transport during winter and strong convective transport and photochemical activity during... summer. ~30% of the NO$_2$ mass in the tropospheric column is in each of the MT and UT during summer, mainly because of convective transport and the contribution from lightning. The tropospheric mass of the NO$_2$ is less than 40% of the stratospheric NO$_2$ mass over India. As mentioned in K04a, this introduces an uncertainty with the satellite retrieval of the NO$_2$ column over this region. ### 3.3. Uncertainties in the Estimation of Tropospheric Lifetime The primary loss processes of NO$_x$ are the reaction with OH and the hydrolysis of N$_2$O$_5$ during nighttime. These lead to conversion to HNO$_3$, which is then lost to dry and wet deposition as described in detail by Dentener and Crutzen [1993] and Crutzen and Lawrence [2000]. This implies that the photochemical lifetime depends on OH and O$_3$, which induces spatial and temporal variations of the NO$_x$ lifetime as described in K04a. Large seasonal variations of OH associated with photochemistry and moisture transport through the monsoon circulation can be generally seen over India. In the troposphere, the source of NO$_x$ is primarily as NO. There is a rapid cycling between NO and NO$_2$ involving ozone. From the model, it is found that the NO$_2$/NO fraction is about 0.7–0.9 in the LT and 0.3–0.5 in the UT over India (Figure 4) and this is partially the reason for the longer lifetime of NO$_x$ in the UT (the lower OH concentration also contributes). The NO$_x$ lifetime varies spatially and NO$_x$ is recycled through PAN and HNO$_3$. The effective lifetime of NO$_x$ (NO + NO$_2$) is the average time a nitrogen atom spends as a NO$_x$ molecule between when it is emitted and when it is removed from the atmosphere (via deposition of nitrogen to the ground). This effective lifetime is different from the lifetime that is calculated on the basis of the first-order loss reactions and is difficult to quantify because it would require tracking the N through exchanges with NO$_2$, N$_2$O$_5$, HNO$_3$, PAN and other forms of NO$_x$. The most convenient way to estimate the effective lifetime is the mass-emission method (introduced in K04a), by computing the ratio of the total mass abundance of NO$_x$ over the region (taking only the 16 model columns over India) to the mean integrated emission of NO$_x$ on a monthly average basis. The assumption is that total loss rate is equal to the emission under steady state conditions. This would be the correct lifetime if there were no flux of reactive nitrogen into or out of the region. Since there is a nonzero flux, we estimate upper and lower uncertainty bars as follows. The upper limit of the lifetime is computed by taking a broader rectangular region to roughly account for the outflow of NO$_x$ from India. The lower range of the lifetime takes into account only the Figure 3. (a) NO$_x$ and (b) NMHC emissions over India based on EDGAR emissions inventory version 2.0. (c) Sources of NO$_x$ emissions used in MATCH-MPIC. approximate fraction of NO$_x$, which originates from the local NO$_x$ source over India. The range of the lifetime (12–20 hours) computed in this manner is shown in Figure 5; with an estimated average uncertainty ($\delta t$) of approximately ±5 hours from mean value. The minimum values are during February–April, when the sunshine is high over India and the highest values are during October–November. [21] Several approaches for estimating the NO$_x$ lifetime have been discussed in K04a. Here, the lifetimes of tropospheric NO$_x$ and its components computed for India from the mass-emission method and budget analysis based on MATCH-MPIC and from the chemical decay method over the outflow region from India using GOME are summarized in Table 2. The mean lifetime of tropospheric NO$_x$ over India computed from various approaches agrees well between all the methods. However, it varies with respect to seasons on the basis of the photochemistry involving OH, solar radiation, bright sunshine hours and transport. Table 1. Ratio of Integrated Mass of NO$_2$ Column Over the Indian Subcontinent From MATCH-MPIC \| Ratio of Integrated Mass of NO$_2$ \| January \| July \| \|----------------------------------\|---------\|------\| \| BL/TC (PS to 850/PS to 150 hPa) \| 66.5% \| 39.3% \| \| MT/TC (850–500/PS to 150 hPa) \| 23.6% \| 31.1% \| \| UT/TC (500–150/PS to 150 hPa) \| 9.9% \| 29.6% \| \| TC/SC (PS to 150/150–2.7 hPa) \| 40.8% \| 32.5% \| BL, boundary layer; MT, middle troposphere; UT, upper troposphere; TC, tropospheric column; SC, stratospheric column; PS, surface level. 3.4. Regional NO$_x$ Emission Strength From GOME [22] Developing bottom-up emissions inventories is difficult. This is mainly due to the rapidly changing anthropogenic emissions, heterogeneity and uncertainty in detailed statistics, the lack of temporal and spatial continuity as well as the uncertainty in the natural emissions from lightning and soils. A quantitative approach to estimating the regional emission strength from space has been developed in recent years. Estimation of regional NO$_x$ emission from the GOME NO$_2$ column is described in detail by Leue et al. [2001], on the basis of certain assumptions. There have been several encouraging studies in this direction to quantify the NO$_x$ emission from lightning and ships based on GOME satellite observations [Burrows et al., 2004; Richter et al., 2004; Martin et al., 2002; Beirle et al., 2004]. [23] Here we first estimate the NO$_2$ emission for India based on the NO$_2$ retrieval from GOME over India is computed from its number density by taking an areal average. From that, NO$_2$ emission is computed for steady state conditions by taking the temporal variation of lifetime estimated from MATCH-MPIC on the basis of the mass-emission method. However, we cannot rule out biases in the new estimates, because the biases in the model emissions inherently influence the lifetime computed from the model. We found that the mean NO$_2$ source computed in this way from GOME is 0.7 Tg(N)/yr for a broader rectangular region and relatively less (0.53 Tg(N)/yr) for the Indian grids with a maximum source in the range $\sim$0.9–1.2 Tg(N)/yr during February–June (winter–spring). The accuracy of this value is further limited by the uncertainty (up to 50%) in the GOME retrieval especially due to clouds and aerosols during summer monsoon months. [24] Second, we also use an estimate of NO$_x$ emission based on the GOME NO$_2$ retrieval of W01. This estimate is an improved one from GOME following on the work of Leue et al. [2001] with an average uncertainty of about 30–50%. In this method, the vertical column density of NO$_2$ is obtained by multiplying the retrieved NO$_2$ column by a correction factor, which also takes into account the fraction of NO$_2$ to the total NO$_x$. The conversion from NO$_2$ to NO$_x$ emissions is dependent on the photolysis frequencies, which are dependent on the solar zenith angle (SZA). If the SZA increases, the conversion factor decrease, therefore conversion factors are higher in winter than in summer. A basic assumption for computing this fraction is that NO$_x$ as measured by GOME is in photochemical stationary state with NO. The mean NO$_x$ emission from GOME for India, if we were instead to simply assume a constant lifetime of 24 hours, would be estimated at $1.63 \pm 0.24$ Tg(N) for 1996–2001 with a maximum of 1.75 Tg(N) for 1997. Since this estimate is based on a constant lifetime of 24 hours, it is difficult to compare directly, therefore we also compute the emission strength for India with the regionally appropriate lifetime estimated from MATCH-MPIC. The mean annual NO$_x$ emission computed for 1997–1998 from GOME is $2.35 \pm 0.5$ Tg(N)/yr (Figure 6) with a maximum of 3.35 Tg(N)/yr in March–April (dry summer) and a minimum of 1.65 Tg(N)/yr in November–December (winter). The NO$_x$ emission estimated from GOME is close to the mean emission ($2.28 \pm 0.44$ Tg(N)/yr) used in MATCH-MPIC for India with a maximum of 3.03 Tg(N)/yr for the same period. [25] As the lifetime estimate from MATCH is not free from biases due to the uncertainties as mentioned above, we ![Figure 4. NO$_2$/NO$_x$ for India from MATCH-MPIC.](image-url) Figure 5. Estimate of the tropospheric NO$_x$ lifetime over India using the mass-emission method: Uncertainties due to possible imbalances in inflow and outflow of NO$_x$ are given with error bars, the upper limit using a larger, rectangular domain including outflow region and the lower limit based on only the approximate fraction of NO$_2$ that stems from local emissions. examine the column mass of NO$_x$ from GOME emission data (W01) and MATCH-MPIC (Figure 7). The mean NO$_x$ mass computed for India from GOME and MATCH-MPIC for 1996–2001 is $4.6 \pm 0.37 \times 10^6$ kg (N) and $4.47 \pm 0.64 \times 10^6$ kg (N) respectively, in good overall agreement. On the other hand the mass of NO$_x$ for India computed from the RB02 method is around $1.5$ to $2.5 \times 10^6$ kg (N). [26] Regional NO$_x$ emissions estimated from different methods are summarized in Table 3. The average MATCH-MPIC NO$_x$ emission for India during 1997–1998 is $2.28$ Tg(N)/yr and has a maximum of $3.0$ Tg(N)/yr during April, the dry summer period in India. The value is very close to the one from GOME for 1997–1998 ($2.35$ Tg(N)/yr with a maxima of $2.99$ Tg(N)/yr). van Aardenne et al. [1999] have reported a value of $1.5$ Tg(N) for the year 1990 based only on the anthropogenic source of NO$_x$. Garg et al. [2001] have estimated a higher value of $3.5$ Tg(N) for 1995 from sector analysis over India. The average emission computed by Leue et al. [2001] from GOME using the assumption of a constant production rate and a constant lifetime ($\sim 27$ hours) over India and the surrounding region is $2.96$ Tg(N) for 1997. While it is clear that more effort is needed to obtain accurate estimates of the Indian NO$_x$ source, it is at least promising that several different studies and techniques now all indicate it is around $2–3$ Tg(N)/yr. 3.5. Regional NO$_x$ Emission Strength From Space: Possible Uncertainties [27] Estimation of the regional NO$_x$ emission from space is limited by the uncertainties due to many factors such as varying lifetime, external influences on the regional chemistry such as transport, and errors in the retrieval methods with varying clouds, albedo and aerosols. The model sensitivity simulations for India discussed in K04a and section 4 indicate the influence of a nonnegligible contribution of external NO$_x$ sources, which introduces an uncertainty into the estimation of the regional NO$_x$ source strength using GOME as employed by Leue et al. [2001]. In addition to that, the lifetime of NO$_x$ varies with respect to space and time, having an uncertainty of $\sim 5$ hours over India, which implies that we cannot use a constant lifetime to estimate the emissions as long as the variation exceeds the uncertainty. The estimate of NO$_x$ lifetime from GOME is also restricted to only certain outflow regions. [28] There are attempts to improve the emission estimate from GOME by taking different steps. Martin et al. [2003] proposed a new method to estimate the emission by taking top-down information derived from the GOME NO$_2$ column to reduce the uncertainties in NO$_x$, while trying to integrate the information from bottom-up inventories. However, this method does not include varying regional lifetimes and the uncertainty due to export/import as discussed above. Combining these various advances may eventually lead to an improved NO$_x$ emission inventory based on observations from space. 4. Impacts of NO$_x$ Emissions on Regional Tropospheric Chemistry [29] Following up on the previous section, in which we focused on the magnitude of regional NO$_x$ emissions, in this section we examine their impact on the tropospheric chemistry of the region. We first describe the methodology for the sensitivity studies, then several aspects of the impacts of the emissions. Table 2. Lifetime of Tropospheric NO$_x$ Over India From Different Methods Based on Kunhikrishnan et al. [2004a] \| Method \| Lifetime of NO$_x$ \| Basic Assumptions \| Data Source/Remarks \| \|-------------------------------\|--------------------\|----------------------------------------------------------------------------------\|--------------------------------------\| \| Mass-emission method \| 12–22 hours \| steady state conditions; loss rate equals total emission \| MATCH-MPIC \| \| Exponential decay \| t(NO$_2$): 18.2 hours \| $C(t) = C(0) \exp(-t/t_i)$; wind is a function of time; $C$, concentration \| GOME; source region: west coast of India \| \| Sink of NO$_2$ against HNO$_3$ formation by OH \| 11–34 hours \| loss of NO$_2$ equals loss of NO$_x$ \| MATCH-MPIC \| \| Loss rate from budget analysis \| 16–23 hours \| total loss equals loss due to photochemistry and dry deposition \| MATCH-MPIC \| 4.1. Small-Perturbation Tendency: Sensitivity Studies Based on MATCH-MPIC Simulations With Indian Emissions [30] In a previous study (K04a) we have examined the impact of NO$_x$ source perturbation for the outflow regions from India and neighboring continents to the Indian Ocean. In the present study we apply this directly for the source regions to examine their impact on India. We employ here the theory of small-perturbation tendency (SPT as proposed by Kunhikrishnan and Lawrence [2004]) to perform sensitivity studies to assess the impact of neighboring continental emissions on NO$_x$-related chemistry over India. The theoretical approach to the problem is described in detail with respect to perturbations in Indian NO$_x$ emissions in Appendix A. For this method, model sensitivity studies are done by taking a 10% increase of NO$_x$ emissions in a chosen region, which limits the chemical nonlinearity primarily to the first-order feedback via OH. This is then compared to the model base run results to estimate the scaled sensitivity of Indian NO$_x$ from local sources and external sources of neighboring countries. NO$_x$ responds nonlinearly to changes in emissions, since O$_3$ and OH, which depend on NO$_x$, influence its lifetime. The departure from the SPT in the LT and UT after 3 months model spin-up for December 1996 is given in Table 4. This study indicates that the relative percentage difference from the SPT for entirely masking Indian NO$_x$ (100% reduction) for the LT (surface to 500 hPa) is about 22.7% and for the UT (300–150 hPa) is 29%. Thus there will be a deviation of $\sim$20–30% with a test run of 100% reduction in NO$_x$ source strength over India versus the SPT scaled sensitivity. The degree of departure is more prominent when the source strength is reduced by a factor of 50% or more, than when it is increased by the same amount especially for changes of $\geq$50%. This is due to the transition from a typically continental high-NO$_x$ regime to a low-NO$_x$ regime for large negative perturbations. The departure from the SPT for NO$_x$ over India becomes significant only after an increase/decrease of its source strength by a factor of 20% or more. It is clear that smaller perturbations of less than 20% in the source strength will not result in a considerable deviation from the SPT. As employed in K04a, these results show that the 10% perturbation runs can be used to directly examine the current sensitivity of NO$_x$ over India to emissions from local sources as well as to external emissions. 4.2. Sensitivity of NO$_x$ and O$_3$, OH, and PAN Over India to Continental NO$_x$ Emissions From India, Southeast Asia, the Middle East, Africa, and China [31] In this section, we consider the impact of NO$_x$ emissions from India ($60^\circ–95^\circ$E, $7^\circ–35^\circ$N) and the surrounding continental regions, Southeast Asia ($95^\circ–140^\circ$E, $10^\circ$S to $20^\circ$N), Africa ($0^\circ–40^\circ$E, $35^\circ$S to $35^\circ$N), China ($95^\circ–120^\circ$E, $20^\circ–45^\circ$N), and the Middle East ($35^\circ–70^\circ$E, $10^\circ–45^\circ$N) on NO$_x$-related chemistry over India. The surface NO$_x$ emissions during the months representing four Table 3. Regional NO\textsubscript{x} Emission Strength Over India/Extended Indian Domain From Various Sources/Methods \| Source \| Year \| NO\textsubscript{x} Emission, Tg(N)/yr \| Remarks/References \| \|-----------------\|----------\|----------------------------------------\|-------------------------------------\| \| MATCH \| 1997–1998\| 2.28 \| lifetime, 15–20 hours \| \| GOME-MATCH \| 1997–1998\| 2.35 \| lifetime, MATCH; data, GOME emission; Wenig [2001] \| \| GOME \| 1997–1998\| 1.70 \| lifetime, 24 hours; Wenig [2001] \| \| GOME \| 1997 \| 2.96 \| lifetime, 24 hours; extended Indian region; Leue et al. [2001] \| \| RAIN-ASIA \| 1990 \| 3.50 \| based on anthropogenic source; van Aardenne et al. [1999] \| \| Energy Statistics India \| 1995 \| 3.55 \| based on sector analysis, India; Garg et al. [2001] \| major seasons, used in MATCH-MPIC for these regions are given in Table 5. NO\textsubscript{x} emissions are greatest for Africa (partly because of the larger area), and for China, and are lowest for the Middle East. However, annual emission per unit area from India is relatively higher compared to Africa and less than that over China. According to the EDGAR emissions statistics, Africa contributes the majority of NO\textsubscript{x} emissions from biomass burning (~6 Tg(N)/yr) followed by Southeast Asia (~1 Tg(N)/yr) for 1995. NO\textsubscript{x} from fossil and bio fuel emissions together constitute 13.7, 4.7, 4.6 and 4 Tg(N)/yr for China, Africa, the Middle East and Southeast Asia. The scaled sensitivities of NO\textsubscript{x}, O\textsubscript{3}, OH and PAN over India to the emissions from India, Southeast Asia, Africa, China and Middle East are computed for 1997 on the basis of runs with MATCH-MPIC in which the emissions from each region are increased individually by 10%. The results are depicted in Figure 8. Table 4. Relative Deviations From Linearity for Different Isobaric Levels Over India From MATCH-MPIC With Respect to Changes in NO\textsubscript{x} Abundances [(x\textsubscript{i}/x) − 1] Versus Changes in Local Source Strengths [(f − 1)] for December 1996 (After a Three-Month Spin-up) \| f − 1 \| (x\textsubscript{i}/x) − 1 \| Linear Fit \| Deviation \| Relative Deviation, % \| \|-------\|----------------------------\|------------\|-----------\|----------------------\| \| \| \| Surface to 850 hPa \| \| \| \| 1 \| 0.701 \| 0.760 \| 0.059 \| 7.8 \| \| 0.5 \| 0.360 \| 0.380 \| 0.020 \| 5.4 \| \| 0.2 \| 0.148 \| 0.152 \| 0.004 \| 2.7 \| \| 0.1 \| 0.075 \| 0.076 \| 0.001 \| 1.6 \| \| −0.1 \| −0.077 \| −0.076 \| 0.001 \| −1.6 \| \| −0.2 \| −0.156 \| −0.152 \| 0.008 \| −2.9 \| \| −0.5 \| −0.418 \| −0.380 \| 0.038 \| −10.0 \| \| −1 \| −0.944 \| −0.760 \| 0.184 \| −24.1 \| \| \| \| Surface to 500 hPa \| \| \| \| 1 \| 0.622 \| 0.660 \| 0.041 \| 6.2 \| \| 0.5 \| 0.316 \| 0.331 \| 0.015 \| 4.5 \| \| 0.2 \| 0.129 \| 0.133 \| 0.003 \| 2.4 \| \| 0.1 \| 0.065 \| 0.066 \| 0.001 \| 1.5 \| \| −0.1 \| −0.067 \| −0.066 \| 0.001 \| −0.9 \| \| −0.2 \| −0.136 \| −0.133 \| 0.003 \| −2.5 \| \| −0.5 \| −0.361 \| −0.331 \| 0.030 \| −9.1 \| \| −1 \| −0.813 \| −0.663 \| 0.151 \| −22.7 \| \| \| \| 500−150 hPa \| \| \| \| 1 \| 0.080 \| 0.088 \| 0.008 \| 8.9 \| \| 0.5 \| 0.042 \| 0.044 \| 0.002 \| 5.3 \| \| 0.2 \| 0.017 \| 0.018 \| 0.001 \| 2.2 \| \| 0.1 \| 0.009 \| 0.009 \| 0.001 \| 1.1 \| \| −0.1 \| −0.009 \| −0.009 \| 0.000 \| −0.8 \| \| −0.2 \| −0.018 \| −0.018 \| 0.001 \| −2.4 \| \| −0.5 \| −0.047 \| −0.044 \| 0.003 \| −6.8 \| \| −1 \| −0.113 \| −0.088 \| 0.026 \| −29.0 \| Table 5. Mean Surface NO\textsubscript{x} Emission in Tg(N) Used in MATCH-MPIC for India and Neighboring Countries for the Year 1997 \| Month \| India \| Indonesia \| Africa \| China \| Middle East \| \|-------\|-------\|-----------\|--------\|-------\|-------------\| \| January \| 1.38 \| 1.41 \| 5.10 \| 2.84 \| 0.99 \| \| April \| 2.10 \| 2.02 \| 3.88 \| 3.64 \| 1.10 \| \| July \| 1.66 \| 1.86 \| 5.81 \| 4.68 \| 1.59 \| \| October\| 1.14 \| 1.70 \| 5.88 \| 2.66 \| 1.04 \| \| Annual \| 1.55 \| 1.76 \| 5.30 \| 3.46 \| 1.19 \| Figure 8. Sensitivity of Indian NO$_x$, O$_3$, OH, and PAN (percent) to NO$_x$ emissions from India, Africa, Southeast Asia, China, and the Middle East. Atmospheric oxidizing efficiency over India through OH. China and the Southeast Asia contribute a sensitivity of $\sim$50% to OH in the UT over India during summer through their NO$_x$ emissions. However, in the LT, throughout the year especially during the winter season, the major contribution is from India. [18] We also examine the response of the total tropospheric mass of NO$_x$ over India to continental NO$_x$ emissions, which is consistent with the results discussed above, with $\sim$8–10% response to emission from each of Indonesia, China and the Middle East in the LT (Figure 9a). External influence from these regions is largest during the winter season because of the northeasterly trade winds with large-scale subsidence over India. The major influence in the UT is from China and Indonesia (Figure 9b) with a scaled sensitivity of 25–30% each, during the summer monsoon and relatively less (5–10% on average) to emissions from Africa and the Middle East (Figure 9c). [19] The response of peroxy acetyl nitrate (PAN) to tropospheric NO$_x$ from India and other neighboring continents is also examined. PAN acts as a reservoir species of odd nitrogen during transport to remote regions. It is one of the by-products of photochemical oxidation of NMHCs in the presence of NO$_x$. The lifetime of PAN is about one week in the UT and it depends on NO/NO$_2$ and the temperature. The test results in Figure 8 indicate that $\sim$40–50% of PAN in the surface layer and $\sim$30–40% in the MT during summer over India is sensitive to local emissions. PAN over India is 10–20% responsive to NO$_x$ emissions from the Middle East and Africa during winter due to SWJ. Twenty to twenty-five percent of PAN over India in the UT is influenced by the NO$_x$ emissions from Southeast Asia and China during the summer monsoon. About 50% of tropospheric PAN over India is sensitive to local NO$_x$ emissions, and this would imply an influence of approximately 50% (mainly in the MT and UT) due to external NO$_x$ emissions. It is found that PAN over India in the MT to UT is insensitive to local NO$_x$ emission from India during winter, indicating the possible influence of external NO$_x$ to India as PAN. During summer, an enhanced sensitivity of PAN over India to emissions from China and Southeast Asia can be seen in the UT. 4.3. Regional Variation of O$_3$ With Respect to NO$_x$ and Nonmethane Hydrocarbon Emissions [20] The details of NO$_x$ and NMHC emissions employed in the model were explained in section 2, and their individual emission contributions from different sectors over India based on the EDGAR inventory were shown in Figure 3. O$_3$ production depends on NO$_x$ and NMHC emissions in a complex and nonlinear manner depending on their source locations [Sillman and Samson, 1999; Duncan and Chameides, 1998]. The sensitivity runs for NO$_x$ and NMHCs over India as a whole indicate that regional ozone is strongly Figure 9. Sensitivity of mass abundance of NO$_x$ (percent) in the (a) lower troposphere (surface to 500 hPa) and (b and c) upper troposphere (500–150 hPa) over India to NO$_x$ emissions from Indonesia, China, Africa, and the Middle East (after four-month model spin-up). sensitive neither to NO$_x$ nor to NMHCs. However, there are large variations in NO$_x$ and NMHC emissions in different parts of the country from rural to urban regions. This suggests that the generalization of NO$_x$ or NMHC limited regimes as a whole is not very valid over a country like India where the emissions are highly heterogeneous. This may also be influenced by the import of O$_3$ from outside the region. A recent study by Luo et al. [2000] suggested that O$_3$ over China is sensitive to both NO$_x$ and VOC over the region but it varies differently from urban to rural areas. An attempt has been made here to understand the regional variation of the response of O$_3$ to emissions by dividing India into different regimes such as North, Central and south India. [38] The sensitivity of O$_3$ (percent) over these regimes with respect to NO$_x$ emissions (in contours) and NMHC emissions (shaded) are shown in Figure 10. This figure shows that the sensitivity of O$_3$ over southern India with respect to both NO$_x$ and NMHC emissions is relatively stronger than that over northern India. The runs also indicate that the response of O$_3$ to NO$_x$ is larger than to NMHC over all the sub regions. The response is generally more during the summer monsoon, when convection is stronger and most of the LT NO$_x$ is carried into the UT where the NO$_x$ lifetime is longer, resulting in more O$_3$ formation in the troposphere. The difference in the sensitivity of O$_3$ over south to north is factor of 2 or more with respect to its two precursor emissions (Figure 10). This is partly due to stronger photochemistry with high insolation and humidity over the south as the region is close to the equatorial belt. It is also due to greater present emissions in northern India, so that the chemical system is more saturated with NO$_x$ and responds less to changes in emissions. 5. Discussion and Conclusions [39] The Indian region is characterized by increasing NO$_x$ emissions with a high energy demand from the increasing population, in addition to natural emissions, which are highly uncertain. As the emissions are heterogeneous with... respect to rural and urban India with distinct climatological zones, the limited in situ measurements and field campaigns cannot give an overall picture of the regional NO\textsubscript{x} chemistry. In this study, an attempt has been made to understand the NO\textsubscript{x} emission and its impact on O\textsubscript{3} over India in light of meteorological features based on GOME satellite observations and model results from MATCH-MPIC. Model estimates show that the tropospheric mass of NO\textsubscript{2} is ~35% of the stratospheric column over India and varies with respect to seasons, which introduces a difficulty for the satellite estimates of tropospheric columns over the region. The regional NO\textsubscript{x} emission strength estimated for India from GOME is 2.35 Tg(N)/yr and is close to the value (2.28 Tg(N)/yr) used in MATCH-MPIC for 1997–1998 with the maximum emissions during April and minimum during the winter months. These values are close to those reported on the basis of sector analysis for India. [40] The largest response of NO\textsubscript{x} to local emissions is 60–70% (in terms of scaled sensitivity) in the LT throughout all seasons and the least response is in the UT during winter. Sensitivity simulations suggest that ~15–25% of NO\textsubscript{x} in the MT and UT over India is sensitive to emissions each from Africa, Southeast Asia and China. Fifty to seventy percent of NO\textsubscript{x} in the MT is influenced by local sources during summer. In the LT, the influence from the Middle East is about 10 to 15%. These figures are comparable to the budget of tropospheric mass of NO\textsubscript{2} computed for India. [41] In order to put the simulated O\textsubscript{3} sensitivity to NO\textsubscript{x} emissions into a better perspective, the model simulated O\textsubscript{3} was compared with MOZAIC data for 3 Indian airport locations: Delhi, Chennai and Mumbai. Although a direct comparison is difficult between the monthly average model output and the means of available MOZAIC profiles because of different time sampling and the low resolution of the model output, we found that the model generally overestimates the O\textsubscript{3} concentration, while it simulates the seasonal trends well. [42] O\textsubscript{3} over the southern part of India is considerably more sensitive (~25–30%) than northern India with respect to NO\textsubscript{x} and NMHC emissions, largely because of stronger present emissions over northern India. Twenty to twenty-five percent of the BL O\textsubscript{3} is sensitive to local NO\textsubscript{x} irrespective of the seasonality. The external NO\textsubscript{x} influences on Indian O\textsubscript{3} are predominantly seen in the MT with a sensitivity of ~10% each from Southeast Asia and Africa. The oxidizing capacity through OH in the MT and UT over India is significantly influenced by external NO\textsubscript{x} emissions from China and Southeast Asia, while the impact of local sources is mainly limited to the LT except in summer. The influence of local NO\textsubscript{x} on PAN during winter and the premonsoon is less than 50% and comparatively more (50–80%) during the summer monsoon and the postmonsoon periods. [43] In conclusion, this study has probed the way to use satellite observations and model output to understand the regional NO\textsubscript{x} emission strength and influences on O\textsubscript{3}-related chemistry over India. However, low fractions of tropospheric mass of NO\textsubscript{2} and the influence of external emissions limit the accuracy of this estimate from satellites in addition to the uncertainty with retrievals. O\textsubscript{3}-related chemistry is very sensitive to southern Asian emissions, but shows large spatial and temporal variation because of varying lifetime and photochemistry with respect to the seasonal meteorology associated with large-scale transport and mixing. Appendix A:* Small-Perturbation Tendency [44] The relation between the NO\textsubscript{x} concentration and perturbation in NO\textsubscript{x} source strength is examined by using the MATCH sensitivity simulations for the Indian region. [45] Let $X = [\text{NO}_x]$ be the concentration (mole/m\textsuperscript{3}) of NO\textsubscript{x} and $dX/dt$ be the rate of change of the concentration with respect to time. [46] For a steady state, $dX/dt = \partial X/\partial t + L\nabla X = 0$, i.e., $X = S/L$, assuming the concentration of NO\textsubscript{x} is constant, where $S$ is the source and $L$ is the loss frequency (assumed constant, for the first part of this discussion; changes in $L$ are discussed below). [47] The source strength $$S = S_l + S_r,$$ where $S_r$ is the remote source due to flux transport from outside the region and $S_l$ is the direct emission contribution, which is the difference between the local emission and flux out from the region. $$X = S_l/L + S_r/L.$$ (A1) [48] Let $f$ be the factor that corresponds to different perturbations on the NO\textsubscript{x} base source strength $S_l$ and $S^l_i$ be the new source strength of NO\textsubscript{x} in the test run after applying the perturbation. Then $$S^l_i = fS_l$$ (A2) where $f = 0, 0.5, 0.8, 0.65, 0.9, 0.99, 1.01, 1.1, 1.35, 1.2, 1.5,$ and $2$ for the runs here. [49] The ratio of NO\textsubscript{x} concentration in the test run to base run can be explained in terms of the local and remote source strength and can be written from equations (A1) and (A2) as $$X^l/X = (S^l_i + S^r_i)/(S_l + S_r) = (fS_l + S^r_i)/(S_l + S_r)$$ (A3) where $X^l$ is the test run NO\textsubscript{x} and $S^l_i$ is the perturbation in the remote source strength of the test run. $X$ is the base run NO\textsubscript{x}. For this introductory discussion, we assume that the remote source is not changing in the test compared to the base run ($S^r_i = S_r$), although it is important to note that $S_r$ is not really constant and introduces an important uncertainty in the results, in addition to the uncertainty in the local source strength of the Indian region as compared to remote region source strength. [50] To derive the relation between the NO\textsubscript{x} concentration and perturbation in the NO\textsubscript{x} base source strength, let $f_1$ and $f_2$ be the factors of two different NO\textsubscript{x} perturbations corresponding to the ratios of the test run to base run $X_1/X$ and $X_2/X$ respectively. [51] Let $k_i = X_i/X$ (from equation (A3)). $$k_i = (f_iS_l + S_r)/(S_l + S_r), \text{ where } i = 1, 2$$ (A4) From equations (A3) and (A4), \[ \frac{(k_1 - 1)/(k_2 - 1)}{[(f_1 S_l + S_r) - (S_l + S_r)]} = \frac{(f_1 S_l + S_r) - (S_l + S_r)}{[(f_2 S_l + S_r) - (S_l + S_r)]} \] \[ \frac{(k_1 - 1)/(k_2 - 1)}{= (f_1 - 1)/(f_2 - 1)} \] (A5) Equation (A5) indicates that this ratio is independent of the relation between the local ($S_l$) and remote ($S_r$) base source strength and depends only on the factor $f$ by which the source strength varies with respect to the base source strength, as long as $S_r$ does not change much. Since $S_r$ is only changed because of the changes in gradient, this is expected to hold for small perturbations provided only the local direct emissions are perturbed. The discussion so far applies only to linear tracers, for which $L$ is constant. For NO$_x$, however, $L$ depends on OH and O$_3$, which in turn depend on NO$_x$, so that it will not be constant. Thus, in contrast to Wild et al. [2001], we do not even expect a small perturbation (e.g., 10%) to produce linear effects. However, we do expect the change in OH and O$_3$, and thus in $L$, to be approximately linear for small perturbations: If a reduction in NO$_x$ emissions by 1% causes a reduction in OH by 0.5%, then we would expect that a reduction of 2% in the emissions results in an OH reduction of 1%. For large changes, however, the entire photochemical regime can shift (e.g., low NO$_x$ versus high NO$_x$) leading to a departure from this small-perturbation tendency (SPT). The model simulated NO$_x$ column abundance for the period December 1996 (after a three-month spin-up) is analyzed by increasing/decreasing the source strength by 1%, 10%, 20%, 35%, 50% and 100% and the ratio of the new abundance with respect to base run is computed for the period. The difference of the ratio with unity $(x_1/x - 1)$ and the perturbation factor with unity $(f - 1)$ is analyzed for different vertical columns to see the departure from the SPT (Table 4). Acknowledgments. The first author is thankful to Joe Lelieveld and Rüdiger Lang, MPIK, Mainz, for useful discussions and Phil Rasch, NCAR, Boulder, for supporting MATCH. We also express our appreciation to Jim Crawford, NASA Langley Research Center, USA, for providing facilities to finish the final version of the paper. Thanks are due to the Editor at Journal of Geophysical Research and the two anonymous referees for valuable comments. Funding for this work has been provided by the BMBF, Germany, project 07-AFC-42. References Akimoto, H. (2003), Global air quality and pollution, Science, 302, 1716–1719. Beirle, S., U. Platt, M. Wenig, and T. Wagner (2004), NO$_x$ production by lightning, in Proceedings with GOME-2A, pp. 79–81. Burrows, J. P., M. Wohlfahrt, M. Buchwitz, V. V. Rozanov, A. Ladstätter-Weilermann, A. Richter, R. DeBeek, R. Hoogen, K. Brustvedt, and K. U. Eichmann (1999), The Global Ozone Monitoring Experiment (GOME): Mission concept and first scientific results, J. Atmos. Sci., 56, 13–31. Burrows, J. P., A. Richter, and L. Elid (2004), Studies of NO$_x$ from lightning and convective uplifting using GOME data, in Sounding the Troposphere from Space: A New Era in Atmospheric Chemistry, edited by P. Burges, J. J. Remedios, and U. Platt, pp. 293–306, Springer, New York. Carnielli, G. R., I. Uno, M. I. Phadnis, Y. Zhang, and Y. Sunwoo (1998), Tropospheric ozone production and transport in the springtime east Asia, J. Geophys. Res., 103, 10,649–10,671. Chameides, W. L., and J. C. G. Walker (1973), A photochemical theory of tropospheric ozone, J. Geophys. Res., 78, 8751–8760. Chand, D., and S. Lal (2004), High ozone at rural sites in India, Atmos. Chem. Phys. Discuss., 4, 3359–3380. Crutzen, P. J. (1974), Photochemical reactions initiated by and influencing ozone in the lower tropical stratosphere, Tellus, 26, 48–57. Crutzen, P. J., and M. J. Lawrence (1998), The impact of precipitation scavenging on the transport of trace gases: A 3-dimensional model sensitivity study, J. Atmos. Chem., 37, 81–112. Dentener, F. J., and P. J. Crutzen (1993), Reaction of N$_2$O$_x$ on tropospheric aerosols: Implications for vertical distributions of NO$_x$, O$_3$, and OH, J. Geophys. Res., 98, 7149–7163. Duncan, B. N., and W. L. Chameides (1998), Effects of urban emission control strategies on the export of ozone and ozone precursors from the urban atmosphere to the troposphere, J. Geophys. Res., 103, 28,159–28,180. Duncan, B. N., R. V. Martin, A. C. Staudt, R. Yevich, and J. A. Logan (2003), Interannual and seasonal variability of biomass burning emissions constrained by satellite observations, J. Geophys. Res., 108(D2), 4100, doi:10.1029/2002JD002379. Fishman, J., S. Solomon, and P. J. Crutzen (1979), Observational and theoretical evidence in support of a significant in-situ photochemical source of tropospheric ozone, Tellus, 31, 432–446. Garg, A. R., C. Kulshrestha, A. K. Sarkar, S. C. Garg, and D. C. Panwar (2003), Emissions of SO$_2$ and NO$_x$ from biofuels in India, Tellus, Ser. B, 55(3), 787–795. Garg, A., P. R. Shukla, S. Bhattacharya, and V. K. Dadhwal (2001), Subregion (district) and sector level SO$_2$ and NO$_x$ emissions for India: Assessment of inventories and mitigation flexibility, Atmos. Environ., 35, 703–713. Horowitz, L. W., and D. J. Jacob (1999), Global impact of fossil fuel combustion on atmospheric NO$_x$, J. Geophys. Res., 104, 23,823–23,840. Jaffe, D., et al. (1999), Transport of Asian air pollution to North America, Geophys. Res. Lett., 26, 711–714. Kalnay, E., et al. (1996), The NCEP/NCAR 40-Year Reanalysis Project, Bull. Amer. Meteorol. Soc., 77, 437–471. Kunkulienkar, T. (2004), Synthetic Use of a Global 3D Chemistry and Meteorology Model and Satellite Observations for the Study of Tropospheric NO$_x$-Related Chemistry Over Asia and the Indian Ocean, Ber. Inst. Physik, Univ. Berlin, 23, 1–109, Akademie Verlag, Berlin. Kunkulienkar, T., and M. J. Lawrence (2004), Impact of NO$_x$ over the Indian Ocean to emissions from the surrounding continents and nonlinearities in atmospheric chemistry responses, Geophys. Res. Lett., 31, L15102, doi:10.1029/2004GL020210. Kunkulienkar, T., M. G. Lawrence, R. von Kuhlmann, A. Richter, A. Ladstätter-Weisenmayer, and P. Burrows (2004a), Analysis of tropospheric NO$_x$ over Asia using the model of atmospheric transport and chemistry (MATCH-MPIC) and GOME-satellite observations, Atmos. Environ., 38, 581–592. Kunkulienkar, T., M. G. Lawrence, R. von Kuhlmann, A. Richter, A. Ladstätter-Weisenmayer, and J. P. Burrows (2004b), Semiannual NO$_x$ plumes during the monsoon transition periods over the central Indian Ocean, Atmos. Res., 71, 10810, doi:10.10129/2003GL019269. Lawrence, A. A. D., J. W. Wooldridge, and J. A. Logan (1990), Radiative forcing of climate by changes in the vertical distribution of ozone, J. Geophys. Res., 95, 9971–9981. Liu, S., M. Naja, and B. H. Subbaraya (2000), Seasonal variations in surface concentrations of its precursors over an urban site in India, Atmos. Environ., 34, 2713–2724. Law, K. S., P. H. Platevin, D. E. Shallcross, H. L. Rogers, J. A. Pyle, C. Grouhel, V. Thouret, and M. Marsden (1998), Evaluation of modeled O$_3$ using measurements of O$_3$ and O$_3$/NO$_2$ by the Stratospheric Aircraft (MOZAIC) data, J. Geophys. Res., 103, 25,721–25,737. Lawrence, M. G. (2004), Export of air pollution from southern Asia and its large-scale effects, in Handbook of Environmental Chemistry, vol. 4, edited by A. Stohl, pp. 131–172, Springer, New York. Lawrence, M. G., J. P. Burrows, R. von Kuhlmann, R. A. Eaton, and N. M. Mahowald (1999), A model for studies of tropospheric photochemistry: Description, global distributions and evaluation, J. Geophys. Res., 104, 26,245–26,277. Lawrence, M. G., et al. (2003), Global chemical weather forecasts for field campaign planning: Predictions and observations of large-scale features during INDOEX, MINOS, and CONTRACE, Atmos. Chem. Phys., 3, 267–289. Levin, C. M., M. Wenig, T. Wagner, O. Klimm, U. Platt, and B. Jähn (2001), Quantitative analysis of NO$_x$ emissions from Global Ozone Monitoring Experiment satellite image sequences, J. Geophys. Res., 106, 5493–5506. Levy, H., W. J. Moimix, A. A. Kloniceki, and P. S. Kasibhatla (1999), Simulated tropospheric NO$_x$: Its evolution, global distribution and individual source contributions, J. Geophys. Res., 104, 26,279–26,306. Luo, C., J. John, Z. Xiuji, K. Lam, T. Wang, and W. Chameides (2000), A nonurban ozone air pollution episode over eastern China: Observations and model simulations, J. Geophys. Res., 105, 1889–1908. Manzello, D. P., et al. (1999), The Measurement of Ozone and Water Vapor by Airborne In-situ Sensors: The MOZAIC airborne program, an overview, J. Geophys. Res., 103, 25,631–25,642. Martin, R. V., et al. (2002), An improved retrieval of tropospheric nitrogen dioxide from GOME, J. Geophys. Res., 107(D20), 4437, doi:10.1029/2001JD001018. Martin, R. V., D. J. Jacob, K. Chance, T. P. Kurosu, P. L. Palmar, and M. J. Evans (2003), Global inventory of nitrogen oxide emissions constrained by space-observed observations of NO$_x$ columns, J. Geophys. Res., 108(D17), 4337, doi:10.1029/2002JD003453. Mitra, A. P., and C. Sharma (2002), Indian aerosols: Present status, Chemosphere, 49, 1175–1190. Olivier, J. G. J., J. A. J. Bleeker, J. J. M. Borodowski, A. J. H. Visserdijk, and A. F. H. Zuidema (1998), A 1995 global inventory of anthropogenic tropospheric sources of carbon monoxide on $1^\circ \times 1^\circ$ developed in the framework of EDGAR(GEIA), Chemosphere Global Change Sci., 1, 1–17. Rasch, P. J., N. M. Mahowald, and B. E. Eaton (1997), Representations of transport, convection, and the hydrologic cycle in chemical transport models: Implications for the simulation of short lived and soluble species, J. Geophys. Res., 102, 28,127–28,138. Richter, A., and J. P. Burrows (2002), Retrieval of tropospheric NO$_2$ from GOME measurements, Adv. Space Res., 29, 1673–1683. Richter, A., V. Eyring, J. P. Burrows, H. Bovery, S. Lauer, B. Sierk, and P. Crutzen (2004), Satellite measurements of NO$_x$ from international shipping emissions, Geophys. Res. Lett., 31, L23110, doi:10.1029/2004GL020852. Sharma, C., S. Dasgupta, and A. P. Mitra (2002), Future scenarios of inventories of GHGs and urban pollutants from Delhi and Calcutta, paper presented at IGES/APN Mega-City Project Workshop, Inst. for Global Environ. Strategies, Kitakyushu, Japan, 23–25 Jan. Sillman, S., and D. J. Samson (1999), Impact of temperature on oxidant photochemistry in urban, polluted, rural and remote environments, J. Geophys. Res., 100, 11,497–11,508. Streets, D. G., and S. T. Waldhoff (1999), Biofuel use in Asia and acidifying emissions, Environ. Sci., 23, 1035–1042. Streets, D. G., K. F. Yuen, T.-H. Lin, and G. R. Carmichael (2003), Biomass burning in Asia: Annual and seasonal estimates and atmospheric emissions, Global Biogeochem. Cycles, 17(4), 1099, doi:10.1029/2003GB002040. Thouret, V., A. Marenco, J. Logan, P. Nedelec, and C. Grohal (1998), Comparisons of ozone measurements from MOZAIC airborne program and the ozone sounding network at eight locations, J. Geophys. Res., 103, 25,695–25,720. United Nations Environment Programme (UNEP) (2002), Air pollution and health, in The Asian Brown Cloud: Climate and Other Environmental Impacts, Assess. Rep. UNEP/DEWA/R5/02-3, chap. 3, pp. 16–20, Nairobi. van Aardenne, J. A., G. R. Carmichael, H. Levy II, D. Streets, and I. Horowitz (2003), Anthropogenic NO$_x$ emission in Asia in the period 1990–2010, Atmos. Environ., 37, 633–644. Varshney, C. K., and M. Agarwal (1992), Ozone levels in the urban atmosphere of Delhi, Atmos. Environ. Part B, 26, 291–294. von Kuhlmann, R., M. G. Lawrence, P. J. Crutzen, and P. J. Rasch (2003a), A model for the impact of tropospheric ozone and methane hydrocarbons: Model description and ozone results, J. Geophys. Res., 108(D9), 4294, doi:10.1029/2002JD002893. von Kuhlmann, R., M. G. Lawrence, Crutzen, and P. J. Rasch (2003b), A model for the impact of tropospheric ozone and nonmethane hydrocarbons: Model evaluation of ozone related species, J. Geophys. Res., 108(D23), 4729, doi:10.1029/2002JD003348. Wenig, M. (2001), Satellite measurements of long-term global tropospheric trace gas distributions and source strengths: Algorithm development and data analysis, Ph.D. thesis, Rupertus Carola Univ. of Heidelberg, Heidelberg, Germany. Wild, O., M. J. Prather, and H. Akimoto (2001), Indirect long-term global radiative cooling from NO$_x$ emissions, Geophys. Res. Lett., 28, 1719–1722. Yingzer, J. J., M. Galanter, T. A. Holloway, M. J. Phadnis, S. K. Gutikunda, G. R. Carmichael, W. J. Moxim, and H. Levy II (2000), The episodic nature of air pollution transport from Asia to North America, J. Geophys. Res.*, 105, 26,951–26,946. W. A. H. Asman and M. G. Lawrence, Department of Atmospheric Chemistry, Max Planck Institute for Chemistry, Postfach 3060, D-55020 Mainz, Germany. J. P. Burrows and A. Richter, Department of Environmental Physics and Remote Sensing, University of Bremen, D-28359 Bremen, Germany. T. Kunhkrishnan, Atmospheric Science Directorate, NASA Langley Research Center, 100 NASA Road, Hampton, VA 23681, USA, (email@example.com). R. von Kuhlmann, German Aerospace Center, Königswinterer Strasse 522-524, D-53227 Bonn, Germany. M. O. Wenig, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA.
A Chiral Pentadecanuclear Metallamacrocycle with a Sextuple Twisted Möbius Topology Rohith P. John,† Mira Park,† Dohyun Moon,† Kyungjin Lee,† Seunghhee Hong,† Yang Zou,† Chang Seop Hong,‡ and Myoung Soo Lah*,† Department of Chemistry and Applied Chemistry, College of Science and Technology, Hanyang University, Ansan, Kyunggi-Do 426-791, Korea, and Department of Chemistry, Korea University, Seoul 136-701, Korea Received August 8, 2007; E-mail: firstname.lastname@example.org Construction of cyclic molecules has received much attention in recent years because of both scientific curiosity and their potential use in applications such as magnetic materials, sensors for small molecules, and new classes of catalysts. Although the possibility of forming molecules with a twisted cyclic band structure (Möbius molecules) was predicted theoretically among organic molecules such as polyenes and other conjugated aromatic systems, only a few of such molecules have been realized experimentally. Recently, an example of an inorganic Möbius strip of NbSe$_3$ crystal based on the coordination chemistry of inorganic elements has been reported. A suitable design of the ligands and an appropriate choice of metal ions acting as connecting nodes can also lead to twisted cyclic architectures with a Möbius topology. The only metallamacrocycle of Möbius topology was obtained from the self-assembly of [L$_{10}$]$_{15}$[Au$_{10}$]$_{15}$ (L = 2,6-diaminopyridine), a $N,N,N',N'$-tetracarboxyethylene diamine) and a Au$^{III}$-diphosphine complex. While several strategies exist to introduce multiple twists in conjugated cyclic polyenes or other aromatic systems, none of these exhibits a twist greater than double twists ($360^\circ$ twists), which is probably due to the large strain that the cyclic system needs to accommodate. Metallamacrocycles that may experience lower strains around the metal centers due to the twist may be a convenient alternative for the generation of multiple twisted cyclic systems. The pentadentate N-acylsalicylhydrazide can serve as an asymmetric ditopic bridging ligand between octahedral metal ions using its tridentate and bidentate binding modes to form metallamacrocycles (Scheme 1). Depending on the influence of the N-terminal steric domain during self-assembly, the ligand chooses either $\Delta$ or $\Lambda$ configuration around each metal center (Figure S1). In addition, it is also possible to modify the nuclearity of the macrocycle and the stereochemistry of the metal centers. A decrease in the steric volume near the Ca—Cβ region may increase the possibility of three or more successive metal centers having the same chiral configuration in the macrocycle that can induce multiple twists in the cycle. Herein, we report on the self-assembly and characterization of a unique manganese-based pentadecanuclear metallamacrocycle from a manganese ion and a pentadentate ligand, N-phenylpropionyl salicylhydrazide (H$_2$L), which has the smallest steric volume near the Ca—Cβ region (Scheme 1). X-ray quality single crystals of the complex were grown from a DMF solution of the ligand, N-phenylpropionyl salicylhydrazide, and manganese acetate by layering with diethyl ether. The pentadecanuclear metallamacrocycle with $C_3$ symmetry, [Mn$_{15}$L$_{15}$S$_{15}$] I (Figure 1), where S is either a DMF or a water molecule, crystallizes in the chiral space group $P6_3$, and the asymmetric unit contains five ligand units bound to five metal centers (Figure S2). The steric requirements of the ethynyl phenyl group in the N-terminal steric domain restrain the chiral sequence to $\cdots \Delta \Delta \Delta \Delta \Delta \cdots$ around the metal center. The occurrence of successive metal centers of the same $\Delta$ configuration produces double twists ($360^\circ$ twists) in the metallamacrocycle along the direction of the propagation (Figure 2 and Figure S3). To obtain an insight into the nature and extent of the twist and its contribution to the overall geometry of the metallamacrocycle, we calculated the twist resulting from the chiral configuration at each metal center with respect to the ligand orientation. The consecutive twist angles calculated using the dihedral angle between the neighboring ligand planes were +90, +81, +81, +82, and −92°, where the direction and the extent of an individual twist is determined by the chiral sequence of four consecutive metal centers (Table S1). The five consecutive (+ + + + −) twists contribute a net twist of 360° (double twist) along the direction of propagation. The macrocycle makes a sharp turn at the metal center with a $\Lambda$ configuration (Figure 2). Three such turns in the backbone lead to a $C_3$ symmetric pentadecanuclear metallamacrocycle. This results in a sum of six 180° twists in the overall cycle, creating a sextuple twisted metallamacrocycle (Figure S4). The double twist in the asymmetric unit orients the three salicyl moieties toward the inner core of the macrocycle on one face of the cycle (Figure S5). The chiral nature of the $C_3$ symmetric macrocycle is exemplified by the fact that the inner core of the other face of the macrocycle has different residues, that is, three ethynyl phenyl groups. Depending on the nature of the N-acyl residue, the chiral sequence of the metal centers is determined to minimize the unfavorable steric repulsion and to maximize the favorable weak interactions, such as the C—H···π and π···π stacking interactions. The macrocyclic assembly has a total of nine intramolecular C—H···O interactions between the N-terminal aromatic C—H's and oxygen atoms in the bridging domains of the different ligand units (Table S2). Three intramolecular C—H···π interactions between... rigid ethynyl phenyl N-tail to the ligand minimized the steric interaction in the Cα—Cβ region and provided a moderate steric volume beyond this region. This led to the formation of four successive metal centers of the same Δ configuration, followed by a Λ, which resulted in a sextuple twisted chiral macrocycle. To the best of our knowledge, this is the first sextuple twisted high-nuclearity metallamacrocycle that has a Möbius topology. Acknowledgment. This work was supported by KRF (KRF-2005-070-C00068), KOSEF (R01-2007-000-10167-0), and CBMH. The authors also acknowledge PAL for beamline use (2007-2041-17). Supporting Information Available: Synthesis of the ligand and 1, crystallographic details of 1. This material is available free of charge via the Internet at http://pubs.acs.org. Figure 1. A stick diagram of chiral pentadecauclear metallamacrocycle 1 with $C_3$ symmetry. The —N—N— connectivities are shown in green. The metal ions with Δ configuration appear as blue spheres, and the metal ions with Λ configuration appear as red spheres. Figure 2. A section of the macrocyclic backbone that shows the double twist and the turn in the direction of propagation. The metal ions are shown as spheres and the ligand binding moieties as planes. The blue spheres have Δ configuration, and the red spheres have Λ configuration. The blue arrows represent a clockwise (+) rotation, and the red arrows represent an counterclockwise (−) turn for successive ligand planes. The N-terminal phenyl groups and three other interactions between the salicyl C—H’s and ethynyl phenyl groups are identified in the assembly. In addition, three intramolecular π···π stacking interactions between the ethynyl phenyl ring and the salicyl ring of neighboring ligand units, and extensive van der Waals interactions, support the assembly. In addition, three intramolecular π···π stacking interactions between the ethynyl phenyl ring and the salicyl ring of neighboring ligand units, and extensive van der Waals interactions, support the assembly. One coordinated DMF molecule per crystallographic asymmetric unit is lodged in a small pocket created between the ligand frames of D- and B-labeled residues using van der Waals interactions (Figure S7 and Table S2). We believe that a rigid steric domain of the ligand with a narrow Cα—Cβ region near the bridging domain and a bulky aromatic phenyl end located away from the bridging domain induces the four consecutive Δ configurations for a double twist and the final Λ configuration for a turn in the propagation direction in the asymmetric unit. The combination of weak interactions, such as C—H···π and π···π stacking interactions, and extensive van der Waals interactions may be the reason for the stabilization of the sextuple twisted pentadecauclear assembly forming. In conclusion, we have constructed a unique $C_3$ symmetric pentadecauclear metallamacrocycle using a pentadentate bridging ligand, N-phenylpropiolyl salicylhydrazide. The introduction of a References (1) (a) Farrell, J. R.; Mirkin, C. A.; Guazz, I. A.; Liible-Sands, L. M.; Rispensold, A. L. Angew. Chem., Int. Ed. 1998, 37, 465–467. (b) Holliday, B. J.; Mirkin, C. A. Angew. Chem., Int. Ed. 2001, 40, 2022–2043. (c) Szleiti, J. Chem. Rev. 1998, 98, 1743–1753. (d) Spurlin, E. L.; Johnson, C. A.; Mirkin, C. A. Macromolecules 2002, 35, 5336–5346. (e) Olenyuk, B.; Stang, P. J. Chem. Rev. 2000, 100, 853–908. (f) Tsaiopoulos, G.; Herges, R.; Malava, T.; Wernsdorfer, W.; Ahmad, K. A.; Chiang, Q. Angew. Chem., Int. Ed. 2004, 43, 2117–2121. (g) Fujita, M. Chem. Soc. Rev. 1998, 27, 417–425. (h) Giannaccini, N. C.; Bertin, P. A.; Mirkin, S. Y.; Mirkin, C. A.; Zakharov, N. N.; Rispensold, A. L. J. Am. Chem. Soc. 2001, 123, 10400–10401. (i) Yamamoto, K.; Yamamoto, S.; Ito, R.; Kashi, Y.; Fujiwiro, A.; Kubota, N.; Fukun, T.; Arakawa, R. Angew. Chem., Int. Ed. 2001, 40, 2268–2271. (j) Minnasi, L.; Cordier, C.; Cingolani, D.; Duhayon, C.; Mann, B. E.; Amouri, H. Organometallics 2007, 26, 860–864. (2) (a) Heilbronner, E. Tetrahedron Lett. 1964, 5, 1923–1928. (b) Rzepa, H. S. Chem. Commun. 1998, 10, 1011–1012. (3) (a) Ajami, D.; Ockler, O.; Simon, A.; Herges, R. Nature 2003, 426, 819–821. (b) Herges, R. Chem. Rev. 2006, 106, 4820–4842. (c) Ajami, D.; Hess, K.; Köhler, F.; Ockler, O.; Ockler, Q.; Simon, A.; Yamamoto, C.; Zhang, X.; Herges, R. Chem. Eur. J. 2006, 12, 5434–5445. (4) Kui, S. C. F.; Huang, J.-S.; Sun, R. W.-Y.; Zhu, N.; Che, C.-M. Angew. Chem., Int. Ed. 2006, 45, 4661–4664. (5) (a) Taniguchi, T.; Ohkubo, Y.; Inagaki, K.; Yamaya, K.; Hatukamaka, N. Nature 2002, 417, 397–398. (b) Patzek, G. R. Angew. Chem., Int. Ed. 2003, 42, 975–974. (6) (a) Ajami, D.; Ockler, O.; Prabhala, V.; Chandrasekhar, T. K.; Joshi, B. S.; Roy, R. Chem. Commun. 2005, 3343–3345. (b) Sriruta, N.; Latos-Grażyński, L. Chem.—Eur. J. 2001, 7, 5099–5112. (c) Hinrichs, H.; Boyer, J. A.; Jorgensen, G.; Hess, K.; Herges, R.; Haase, M.; Hopf, H. Chem. Eur. J. 2006, 12, 7110–7116. (d) Ajami, D.; Kashi, Y.; Izumi, N. J. Am. Chem. Soc. 1999, 121, 8957–8958. (e) Collins, S. K.; Yap, S. P. A.; Falk, A. G. Org. Lett. 2006, 8, 3183–3186. (7) (a) John, R. P.; Lee, K.; Kim, B. J.; Suh, B. J.; Rhee, H.; Lah, M. S. Inorg. Chem. 2005, 44, 1109–1115. (b) Kim, S. Y.; Moon, D.; Liu, B.-Z.; Lang, C. J.; Hwang, J.-Q. Chem. Eng. Int. Ed. 2001, 40, 1081–1087. (c) John, R. P.; Park, J.; Moon, D.; Lee, K.; Lah, M. S. Chem. Commun. 2008, 3699–3701. (8) (a) Moon, D.; Lee, K.; Lah, M. S. Chem. Commun. 2004, 2660–2664. (b) Moon, D.; Lee, K.; John, R. P.; Kim, G. H.; Suh, B. J.; Lah, M. S. Inorg. Chem. 2006, 45, 7991–7993. (9) The total sum of the two angles calculated from the dihedral angles between the two consecutive ligand planes in one asymmetric unit ($\sim 280^\circ$) does not make up the $360^\circ$ for double twist in the asymmetric unit along the propagation direction, as shown in Figure 2. This discrepancy is from the fact that the calculations were performed using the ligand planes that were not parallel to the direction of the propagation of the macrocyclic backbone. For a more accurate calculation, we need to calculate the twist-in-angle that results from the intersection of the normal vector residing at the ligand planes, and the twist angles obtained using this information will form the required $360^\circ$ for a double twist comprising four consecutive $-120^\circ$ clockwise twists and a $+120^\circ$ counterclockwise twist. (10) The molecular structure is nearly possible in a mirror plane, which does not contain any S$_{4}$ symmetry. Since the metallamacrocycle having the chiral sequence related by an S$_{4}$ symmetry leads to the same extent of twist but in an opposite direction, hence the net twist in an achiral metallamacrocycle is always zero regardless of the handedness of the ligand. (11) The Flack parameter of the crystal was found to be 0.20(3). However, the bulk crystals dissolved in chloroform did not show any CD activity, which suggests that metallamacrocycle 1 in bulk is not homochiral but racemic.
SECURITIES AND EXCHANGE COMMISSION FORM 425 Filing under Securities Act Rule 425 of certain prospectuses and communications in connection with business combination transactions Filing Date: 2013-01-14 SEC Accession No. 0001193125-13-011617 (HTML Version on secdatabase.com) SUBJECT COMPANY PPG INDUSTRIES INC CIK: 79879 \| IRS No.: 250730780 \| State of Incorp.: PA \| Fiscal Year End: 1231 Type: 425 \| Act: 34 \| File No.: 001-01687 \| Film No.: 13527300 SIC: 2851 Paints, varnishes, lacquers, enamels & allied prods Mailing Address ONE PPG PL 40 EAST PITTSBURGH PA 15272 Business Address ONE PPG PL 40 EAST PITTSBURGH PA 15272 4124343131 FILED BY PPG INDUSTRIES INC CIK: 79879 \| IRS No.: 250730780 \| State of Incorp.: PA \| Fiscal Year End: 1231 Type: 425 SIC: 2851 Paints, varnishes, lacquers, enamels & allied prods Mailing Address ONE PPG PL 40 EAST PITTSBURGH PA 15272 Business Address ONE PPG PL 40 EAST PITTSBURGH PA 15272 4124343131 Copyright © 2013 www.secdatabase.com. All Rights Reserved. Please Consider the Environment Before Printing This Document The following press release, issued by Georgia Gulf Corporation on January 14, 2013, was distributed to all commodity chemicals employees of PPG Industries, Inc. on January 14, 2013. Axiall Corporation To Be New Company Name Following Georgia Gulf Merger with PPG Commodity Chemicals ATLANTA – January 14, 2013 – Georgia Gulf Corporation (NYSE: GGC) today announced that when its pending merger with PPG’s commodity chemicals business is completed, it will change its name to Axiall Corporation. The company will trade on the New York Stock Exchange under the ticker symbol AXLL. “The new name is the result of more than five months of effort and creativity by a group of employees from both Georgia Gulf and PPG,” said Paul Carrico, president and chief executive officer. “The new and distinctive brand represents a clear break from the legacy names of the two businesses, and gives us the flexibility to enter new markets, make future acquisitions, and diversify our portfolio without the need to rebrand our organization.” More information about the Axiall brand story can be viewed online at www.axiallcorp.com. About Georgia Gulf Georgia Gulf Corporation is a leading, integrated North American manufacturer of two chemical lines, chlorovinyls and aromatics, and manufactures vinyl-based building and home improvement products. The company’s vinyl-based building and home improvement products, marketed under Royal Building Products and Exterior Portfolio brands. Georgia Gulf, headquartered in Atlanta, Georgia, has manufacturing facilities located throughout North America to provide industry-leading service to customers. For more information, visit www.ggc.com. Cautionary Statements Regarding Forward-Looking Information This communication contains certain statements relating to future events and our intentions, beliefs, expectations, and predictions for the future. Any such statements other than statements of historical fact are forward-looking statements within the meaning of the Securities Act of 1933 and the Securities Exchange Act of 1934. Words or phrases such as “will likely result,” “are expected to,” “will continue,” “is anticipated,” “we believe,” “we expect,” “estimate,” “project,” “may,” “will,” “intend,” “plan,” “believe,” “target,” “forecast,” “would” or “could” (including the negative or variations thereof) or similar terminology used in connection with any discussion of future plans, actions, or events, including with respect to the proposed separation of PPG’s commodity chemicals business from PPG and the merger of the PPG commodity chemicals business and Georgia Gulf (the “Transaction”), generally identify forward-looking statements. These forward-looking statements include, but are not limited to, statements regarding the expected timing of the closing of the Transaction, the expected benefits of the Transaction and of the Company’s new name, Georgia Gulf’s anticipated future financial and operating performance and results, and the anticipated financial and operating performance of the combined company. These statements are based on the current expectations of the management of Georgia Gulf. There are a number of risks and uncertainties that could cause Georgia Gulf’s actual results to differ materially from the forward-looking statements included in this communication. These risks and uncertainties include risks relating to (i) PPG being unable to obtain any remaining regulatory approvals required to complete the Transaction, or such required approvals delaying the Transaction or resulting in the imposition of conditions that could have a material adverse effect on the combined company or causing the companies to abandon the Transaction, (ii) other conditions to the closing of the Transaction not being satisfied, (iii) a material adverse change, event or occurrence affecting Georgia Gulf or the PPG commodity chemicals business prior to the closing of the Transaction delaying the Transaction or causing the companies to abandon the Transaction, (iv) problems arising in successfully integrating the businesses of the PPG commodity chemicals business and Georgia Gulf, which may result in the combined company not operating as effectively and efficiently as expected, (v) the possibility that the Transaction may involve other unexpected costs, liabilities or delays, (vi) the businesses of each respective company being negatively impacted as a result of uncertainty surrounding the Transaction, (vii) disruptions from the Transaction harming relationships with customers, employees or suppliers, and (viii) uncertainties regarding future prices, industry capacity levels and demand for Georgia Gulf’s products, raw materials and energy costs and availability, feedstock availability and prices, changes in governmental and environmental regulations, the adoption of new laws or regulations that may make it more difficult or expensive to operate Georgia Gulf’s businesses or manufacture its products before or after the Transaction, Georgia Gulf’s ability to generate sufficient cash flows from its business before and after the Transaction, future economic conditions in the specific industries to which its products are sold, and global economic conditions. In light of these risks, uncertainties, assumptions, and factors, the forward-looking events discussed in this communication may not occur. Other unknown or unpredictable factors could also have a material adverse effect on Georgia Gulf’s actual future results, performance, or achievements. For a further discussion of these and other risks and uncertainties applicable to Georgia Gulf and its business, see Georgia Gulf’s Annual Report on Form 10-K for the fiscal year ended December 31, 2011 and subsequent filings with the Securities and Exchange Commission (the “SEC”). As a result of the foregoing, readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date of this communication. Georgia Gulf does not undertake, and expressly disclaims, any duty to update any forward-looking statement whether as a result of new information, future events, or changes in its expectations, except as required by law. This communication does not constitute an offer to buy, or solicitation of an offer to sell, any securities of Georgia Gulf, and no offer or sale of such securities will be made in any jurisdiction where it would be unlawful to do so. In connection with the Transaction, Georgia Gulf has filed with the Securities and Exchange Commission (“SEC”) a registration statement on Form S-4 relating to the Transaction. INVESTORS AND SECURITY HOLDERS ARE URGED TO READ THE PROSPECTUS FORMING PART OF THE REGISTRATION STATEMENT, AND ANY OTHER RELEVANT DOCUMENTS BECAUSE THEY CONTAIN IMPORTANT INFORMATION ABOUT GEORGIA GULF, PPG’S COMMODITY CHEMICALS BUSINESS AND THE TRANSACTION. Investors and security holders will be able to obtain these materials and other documents filed with the SEC free of charge at the SEC’s website, www.sec.gov. In addition, copies of the registration statement may be obtained free of charge by accessing Georgia Gulf’s website at www.GGC.com by clicking on the “Investors” link and then clicking on the “SEC Filings” link, or upon written request to Georgia Gulf at 115 Perimeter Center Place, Suite 460, Atlanta, Georgia 30346, Attention: Investor Relations. Shareholders may also read and copy any reports, statements and other information filed by Georgia Gulf with the SEC, at the SEC public reference room at 100 F Street, N.E., Washington D.C. 20549. Please call the SEC at 1-800-SEC-0330 or visit the SEC’s website for further information on its public reference room. CONTACTS: Investor Relations Martin Jarosick 770-395-4524 Media Alan Chapple 770-395-4538 email@example.com Rule 425 Legend Forward-Looking Statements Statements in this press release relating to matters that are not historical facts are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 reflecting PPG Industries’ current view with respect to future events or objectives and financial or operational performance or results. These matters involve risks and uncertainties as discussed in PPG Industries’ periodic reports on Form 10-K and Form 10-Q, and its current reports on Form 8-K, filed with the Securities and Exchange Commission (the “SEC”). Accordingly, many factors could cause actual results to differ materially from the company’s forward-looking statements. This press release also contains statements about PPG Industries’ agreement to separate its commodity chemicals business and merge it with Georgia Gulf Corporation or a subsidiary of Georgia Gulf (the “Transaction”). Many factors could cause actual results to differ materially from the company’s forward-looking statements with respect to the Transaction, including, the parties’ ability to satisfy the conditions of the Transaction; the parties’ ability to complete the Transaction on anticipated terms and schedule, including the ability of PPG to successfully complete the exchange offer and the ability of the parties to obtain any remaining regulatory approvals; risks relating to any unforeseen liabilities, future capital expenditures, revenues, expenses, earnings, synergies, economic performance, indebtedness, financial condition, losses and future prospects; business and management strategies for the management, expansion and growth of Georgia Gulf’s operations; and Georgia Gulf’s ability to integrate PPG’s commodity chemicals business successfully after the closing of the Transaction and to achieve anticipated synergies; and the risk that disruptions from the Transaction will harm PPG’s or Georgia Gulf’s business. Consequently, while the list of factors presented here is considered representative, no such list should be considered to be a complete statement of all potential risks and uncertainties. Unlisted factors may present significant additional obstacles to the realization of forward-looking statements. Consequences of material differences in results as compared with those anticipated in the forward-looking statements could include, among other things, business disruption, operational problems, financial loss, legal liability to third parties and similar risks, any of which could have a material adverse effect on PPG’s consolidated financial condition, results of operations or liquidity. Forward-looking statements speak only as of the date of their initial issuance, and PPG does not undertake any obligation to update or revise publicly any forward-looking statement, whether as a result of new information, future events or otherwise, except as otherwise required by applicable law. Additional Information and Where to Find It This communication does not constitute an offer to buy, or solicitation of an offer to sell, any securities of Georgia Gulf, PPG’s commodity chemicals business or PPG. In connection with the Transaction, Georgia Gulf has filed with the SEC a proxy statement on Schedule 14A and a registration statement on Form S-4 which includes a prospectus of Georgia Gulf relating to the Transaction. In addition, Eagle Spinco, Inc., a subsidiary of PPG, has filed with the SEC a registration statement on Form S-4 and S-1 which includes a prospectus of the PPG chlor-alkali and derivatives business relating to the Transaction. INVESTORS AND SECURITY HOLDERS ARE URGED TO READ THE REGISTRATION STATEMENTS AND PROXY STATEMENT/PROSPECTUS, AND ANY OTHER RELEVANT DOCUMENTS BECAUSE THEY CONTAIN IMPORTANT INFORMATION ABOUT GEORGIA GULF, PPG’S CHLOR-ALKALI AND DERIVATIVES BUSINESS AND THE TRANSACTION. Investors and security holders may obtain these materials and other documents filed with the SEC free of charge at the SEC’s website, www.sec.gov. In addition, copies of the registration statements and proxy statement/prospectus may be obtained free of charge by accessing Georgia Gulf’s website at www.ggc.com by clicking on the “Investors” link and then clicking on the “SEC Filings” link, or upon written request to Georgia Gulf, Georgia Gulf Corporation, 115 Perimeter Center Place, Suite 460, Atlanta, Georgia 30346, Attention: Investor Relations, or from PPG upon written request to PPG, PPG Industries, Inc., One PPG Place, Pittsburgh, Pennsylvania 15272, Attention: Investor Relations. Shareholders may also read and copy any reports, statements and other information filed by Georgia Gulf or PPG with the SEC, at the SEC public reference room at 100 F Street, N.E., Washington D.C. 20549. Please call the SEC at 1-800-SEC-0330 or visit the SEC’s website for further information on its public reference room. Participants in the Solicitation Georgia Gulf, PPG, and certain of their respective directors, executive officers and other members of management and employees may be deemed to be participants in the solicitation of proxies from shareholders in respect of the Transaction under the rules of the SEC. Information regarding Georgia Gulf’s directors and executive officers is available in its 2011 Annual Report on Form 10-K filed with the SEC on Feb. 24, 2012, and in its definitive proxy statement filed with the SEC on April 16, 2012, in connection with its 2012 annual meeting of stockholders. Information regarding PPG’s directors and executive officers is available in its 2011 Annual Report on Form 10-K filed with the SEC on Feb. 16, 2012, and in its definitive proxy statement filed with the SEC on March 8, 2012, in connection with its 2012 annual meeting of stockholders. Other information regarding the participants in the proxy solicitation and a description of their direct and indirect interests, by security holdings or otherwise, will be contained in the registration statement and proxy statement/prospectus and other relevant materials to be filed with the SEC.
Arthur D. Little (ADL) has been at the forefront of sustainability for many decades, helping clients and organizations to understand and meet sustainability imperatives. Building on this tradition, ADL recently ran a global study exploring organizations’ current maturity in integrating sustainability into their business models. This two-part Executive Update series analyzes the results of that survey and offers insight into operationalizing sustainability. Here in Part I, we explore the current state of sustainability as well as challenges to effective business sustainability. The State of Sustainability in Organizations Driven by public and stakeholder pressure, sustainability has risen to the top of the business agenda for CEOs in all sectors. Demonstrating the focus from investors, nearly a third of shareholders (32%) supported social and environmental proposals at US company shareholder meetings in 2021 — up from 21% in 2017, according to the Sustainable Investments Institute. The Securities and Exchange Commission (SEC) has proposed new rules that mandate that all publicly traded companies disclose climate change risks in their regular filings. Similar regulatory and reporting initiatives are in place in the EU and Asian countries. This has led to a wealth of commitments covering reporting against sustainability metrics and setting targets for improvement. For example, 45 of the 50 largest companies in the S&P 500 Index now report their carbon disclosures to the CDP (formerly the Carbon Disclosure Project) and 31 have committed to achieving net-zero carbon emissions. While acting on sustainability initially was perceived as part of being a “good corporate citizen,” it is now evident that properly adopting sustainability within an organization can deliver both top-line and bottom-line benefits, leading to higher shareholder returns. Among benefits leaders in sustainability have experienced recently are faster growth, reduced cost of debt financing, lower employee turnover, and increased resilience in the face of business risks and market challenges, as well as improved corporate reputation and branding. However, the road to introduce and implement sustainability within an organization is certainly not without challenges. Many companies have made commitments — but are they able to deliver on them successfully? The ADL study went beyond the buzz to see what is actually being put in place — and what is not. The headline findings are of a business world in transition. On the positive side, companies see the benefits of sustainability across multiple dimensions and how it increases their attractiveness to customers, employees, and investors. However, the majority have yet to see a direct impact on financial results. This is due to a disconnect between business operations and sustainability, with many companies unable to link sustainability performance to financial or innovation metrics and reporting systems. Only 40% of respondents apply any externalities/social ROI or creating shared value evaluation models when making business and strategic decisions. Despite understanding the benefits, a fifth of companies (20%) still do not have a sustainability strategy in place — and less than 30% believe the impact of that strategy is clear to all employees (see Figure 1). The majority are struggling to embed sustainability within the organization, and it is thus currently disconnected from staff, strategy, and operations. This failure to take meaningful action and embed sustainability at the center of the organization does not only cause potential environmental and social harm; it also holds back business performance. Understanding the Challenges to Effective Business Sustainability Achieving sustainability is not a straightforward task. In today’s complex value chains, being able to move beyond making commitments to successfully implement sustainability programs requires a transformative approach based on overcoming six key challenges, described below. 1. Sustainability Strategy Not Understood Most sustainability strategies are not yet mature when it comes to being understood by those that are on the front line of delivering them: the employees. Strategies are either not in place or are not understood by staff within 71% of responding companies. Even among those with a strategy in place, nearly two-thirds admit its impact and direction are not clear to employees. Three factors stand out as causes for this gap: a lack of commonly understood internal sustainability standards, poor communication, and lack of incentives. Unclear Internal Standards Undermine Decision Making Currently, many organizations suffer from a lack of commonly understood internal standards when it comes to measuring sustainability performance. Companies need to clearly articulate what they are going to measure, what the KPIs are, and what “good enough” performance looks like for the organization. If this is unclear, employees, particularly those who are not in central sustainability teams, will not be able to make informed decisions in their day-to-day activities and will not see how they can contribute to overall sustainability targets. Poor Communication Holds Back Understanding More than three-quarters (77%) of responding organizations have not communicated their strategy at all, have communicated irregularly, or have shared it only with specific organizational levels. Becoming sustainable is a major cultural change for many organizations, meaning that it requires an ongoing dialogue with staff to ensure they understand objectives, their role in achieving them, and can measure progress toward success. Interestingly, alignment is much more prevalent when organizations have made individual business units (BUS) responsible for their own sustainability initiatives (see Figure 2). Over three-quarters (77%) of this group said their employees understood their sustainability strategy. However, the way sustainability is organized needs to be well-defined and anchored into the wider organization. It needs to fit with how the organization operates and governs itself. Otherwise, activities Companies that answered: “We have a sustainability strategy, and its impact is clear to all employees” \| Organizational Structure \| Percentage \| \|---------------------------------------------------------------\|------------\| \| Companies where there is not a dedicated sustainability team \| 6% \| \| Companies where there is a dedicated central (corporate) team that is administering only sustainability-related projects \| 21% \| \| Companies where there is a dedicated central team actively enforcing sustainability compliance across different business units (BUs)/functions \| 33% \| \| Companies where there is a small central (corporate) team plus additional staff supporting the BUs \| 29% \| \| Companies where each BU is responsible for its own sustainability-related initiatives \| 77% \| Figure 2. Employee understanding of sustainability based on organizational structure (source: Arthur D. Little) will not be aligned, particularly around standards on what is “good” performance (i.e., which indicators to follow and what thresholds need to be met) and how management incentives reinforce the right behaviors. It is possible that BU activities could even counteract or undermine corporate sustainability initiatives. Two-Thirds Do Not Incentivize Sustainability Along with effective communication and strong governance structures that anchor sustainability in the organization, linking sustainability progress to incentives is proven to change behavior and drive change. This starts at the top and then cascades down into the objectives of all staff. However, two-thirds (65%) of surveyed companies do not currently link senior management incentives to sustainability performance. Fixing the issues of communication and incentives is at the heart of engaging employees at all levels with sustainability, ensuring they both understand the strategy and are enthusiastically involved in its execution. 2. Failure to Change Business Models Sustainability is an opportunity for organizations, but to be successful it must be central to both new and existing business models. Simply seeing it as another cost to be borne or regulation to be met will not deliver its real, transformative benefits. However, just 8% of respondents say that they have changed their business model due to sustainability. Most organizations have simply tweaked existing ways of doing business. While some businesses (such as in digital industries) may not need to change their models, the majority (59%) of survey respondents are in sectors such as automotive, energy, chemicals, engineering, and transport that are heavily impacted by sustainability, and are required to make significant changes. This lack of a profound systemic change further demonstrates the current lack of maturity around sustainability. To gain its advantages, organizations must change models and ways of operating, often dramatically, rather than continuing business as usual with minor modifications. Sustainability targets, such as around carbon emissions, cannot be met without starting to take major action now. 3. No Clear Balance Between Short- & Long-Term Plans For sustainability strategies to deliver, it is vital to set a mix of long-term (20, 30, or 40 years) and short-term goals. Simply putting a stake in the ground and saying you will achieve net-zero by 2050 is not enough — you need a plan that maps the short-term steps that will get you there. It is critical to start now with a concrete strategy and detailed plans and to involve your wider value chain. For example, in areas such as carbon, emissions are cumulative, meaning companies need to make a big effort now to invert the trend and start to lower emissions if they are to hit science-based targets by 2050. Insufficient Detail & Scope in Planning A third of responding organizations have not yet put long-term plans and goals in place, with a further third defining plans but not yet deploying them. Despite the need to balance long and short time frames, less than half (48%) of organizations have considered both when creating business plans (see Figure 3). Even here there is a worrying lack of detail — just half of these (24%) have defined and implemented structured long-term plans with clear objectives for every initiative. This echoes other research findings. For example, the “Corporate Climate Responsibility Monitor 2022,” from the New Climate Institute and Carbon Market Watch reports that headline pledges for emissions reductions are often ambiguous or --- Figure 3. Quantifying future sustainability goals (source: Arthur D. Little) limited, with just 12% of companies surveyed clearly committing to deep decarbonization of over 90% of their full value chain emissions by their set target dates. Failure to Set Challenging Goals for GHG & Carbon Emissions Reduction Achieving net-zero for greenhouse gas (GHG) emissions should be a clear first target for most businesses. Yet 43% of surveyed companies have not yet set a strategy. Targets vary. While many are long-term, reaching up to 2050, the nature of science-based targets means they need to be backed up by short- and medium-term plans to show how companies will achieve their overall net-zero aims. For example, companies that adopt the Science-Based Target initiative (SBTi) Net-Zero Standard are required to specify both short/medium-term and long-term science-based targets. To meet their targets, organizations need to start by estimating their current GHG emissions. This is challenging for most industries as those emissions occur across their value chain, upstream and downstream of company production facilities. Therefore, it is crucial to involve the wider ecosystem (and its emissions) in improving performance, as we discuss below. 4. Not Seeking Sustainability Partners Across Wider Ecosystems Given its interconnected nature, sustainability requires an ecosystem approach. To unlock the opportunities it brings, organizations must move beyond traditional partners. The ADL survey reveals increasing collaboration between organizations and an expanding range of stakeholders. Most sustainability collaboration remains with existing partners (81%), although a growing number of organizations now collaborate Successful change requires sufficient resources, accurate data, and strong governance to ensure that the entire organization, at every level, moves in the same direction, speaks the same language, and shares a collective understanding of goals. Slow Progress on Embracing the Circular Economy Research from Material Economics identifies a strong potential for reducing emissions by adopting circularity. However, the vast majority of companies are not yet embracing these opportunities, instead taking a linear approach that does not have concrete circularity initiatives or targets. 5. Wide Range of Governance Hampers Control Successful change requires sufficient resources, accurate data, and strong governance to ensure that the entire organization, at every level, moves in the same direction, speaks the same language, and shares a collective understanding of goals. It is vital that organizations define and measure standards across the business, with everyone clear on what “good” looks like, if governance is to be effectively embedded within daily activities. There is no single approach organizations take to governance models (see Figure 4). What is vital, however, is that sustainability governance is organically embedded in the existing corporate organization and structure with some level of corporate control or oversight. Targets and objectives must be cascaded down (and validated bottom-up) to meet corporate objectives. 6. Inability to Track Performance with the Right Tools Clearly, organizations need to be able to effectively track and demonstrate progress toward their commitments both internally and with external stakeholders such as investors, regulators, and customers. However, a company’s activities are complex and are continually changing, due to a variety of situations including: - Products and divisions can be sold. - New products are bought or introduced. - Growth in particular markets or regions impacts emissions. - Low-carbon products may be retired due to market changes. - Value chain partners can implement changes that affect a company’s emissions. Consequently, without the ability to track performance it is difficult to understand exactly where you stand and to measure sustainability performance and progress against targets. This will become much more important in the near term when regulators mandate greater transparency around climate change-related emissions and risks. Yet, in this area companies are held back by a lack of flexible, usable measurement tools. In fact, among respondents, 43% are not confident that they can track performance, while a further 47% have tools but feel they are too labor-intensive to deliver full value. The Business Impact of Poor Measurement & Reporting There is a growing volume of investment capital looking for sustainable investment opportunities. Attracting this requires organizations to be able to demonstrate their sustainable credentials and progress toward achieving commitments. Yet sustainability performance reporting remains the poor cousin of financial reporting. Just 17% of surveyed organizations believe that the two areas share the same rigor, while only 13% integrate sustainability and financial reporting into the same document. Often this lack of rigor comes from companies being reactive rather than proactive. They begin with the goal of creating a sustainability report rather than starting with the bigger objective of setting a strategy and governance that will allow the proper steering of sustainability inside the organization. As a result, these companies fail to articulate their strategies and develop the necessary confidence around their sustainability initiatives, which are essential in discussions with investors. Of respondents, 30% say they do not discuss sustainability with investors at all, and only 22% are using their performance metrics to attract green investors. Essentially, 78% of the organizations are not fully exploiting the growing availability of green investment and the better terms it offers. Part II of this Update series will set out key recommendations to unlock the benefits of sustainability by adopting an ecosystem approach, integrating sustainability reporting, redefining culture, and focusing on innovation. About the Authors Stefano Milanese is a Partner at Arthur D. Little (ADL), leads ADL’s Risk practice, and is a member of ADL’s AMP open consulting network. His more than 15 years’ experience in consulting has focused on risk management and sustainability for Italian and international companies, especially in the fields of energy, transportation, and manufacturing, in Europe, the Middle East, America, and Africa. Mr. Milanese earned a bachelor’s degree in engineering from the Polytechnic University of Torino, Italy, and a master’s degree from Erasmus University Rotterdam, the Netherlands. Martijn Eikelenboom is a Cutter Expert, Managing Partner at Arthur D. Little (ADL), global head of ADL’s Sustainability practice, and a member of ADL’s AMP open consulting network. He is also a member of ADL’s Strategy & Organization and Technology & Innovation Management global practices. Mr. Eikelenboom focuses on corporate, growth, and innovation strategies, with an emphasis on media, food, energy, clean tech, chemical, private equity, and the services industries. His expertise includes sustainability strategy, digital strategy, data analytics, marketing and sales excellence, and transaction support. Mr. Eikelenboom is former CEO of ECI/Bertelsmann, Benelux’s leading media publisher, distributor, retailer, and e-commerce player, where he worked on transforming the company into an online player with award-winning customer service and logistics performances, supported by predictive analytics systems. He is also founder and chairman of a nonprofit timber trading company that helps protect tropical forests. Mr. Eikelenboom earned a BBA from Nyenrode University, the Netherlands, and an MBA from IESE Business School. Carlo Stella is a Partner at Arthur D. Little (ADL), a member of ADL’s Utilities & Alternate Energy and Strategy & Organization practices, and a member of ADL’s AMP open consulting network. His main areas of expertise cover strategy definition, organizational design, transformation programs, and sustainability, across several industries, including postal, travel, private equity, fast-moving consumer goods, manufacturing, and financial services. Mr. Stella supports case managers in the end-to-end management of complex projects in the Gulf Cooperation Council (GCC), with responsibility over the team, deliverables, client relationships, and project profitability. In addition, he holds responsibility over proactive development of business opportunities, as well as improvement of company processes, where he plays a key role in the Intellectual Capital Management initiative and the Collaboration initiative. Mr. Stella earned a master of science degree in chemical engineering from University of Illinois, Chicago; and a master of science degree in engineering from the Polytechnic University of Torino, Italy. Stefano Decadri is a Manager at Arthur D. Little (ADL), a member of ADL’s Utilities & Alternate Energy practice and Sustainability Competence Center, and a member of ADL’s AMP open consulting network. His focus areas include sustainability and environmental, social, and governance (ESG) risk with particular expertise in integrating sustainability and ESG risk management into business; externalities, social return on investment (SROI), and local content assessment; sustainable procurement; and risk assessment. Mr. Decadri works mainly in the transportation, oil and gas, and infrastructure sectors, but has past experience in food, real estate, telecom, and utilities. He earned a bachelor’s degree and a master’s degree in engineering physics from the Polytechnic University of Milan, Italy. Cutter Consortium, an Arthur D. Little community, is dedicated to helping organizations leverage emerging technologies and the latest business management thinking to achieve competitive advantage and mission success. Cutter helps clients address the spectrum of challenges disruption brings, from implementing new business models to creating a culture of innovation, and helps organizations adopt cutting-edge leadership practices, respond to the social and commercial requirements for sustainability, and create the sought-after workplaces that a new order demands. Since 1986, Cutter has pushed the thinking in the field it addresses by fostering debate and collaboration among its global community of thought leaders. Coupled with its famously objective “no ties to vendors” policy, Cutter’s Access to the Experts approach delivers cutting-edge, objective information and innovative solutions to its community worldwide. For more information, visit www.cutter.com or call us at +1 781 648 8700.
THE CIRCUMSTELLAR ENVELOPES OF CEPHEIDS AND THEIR IMPACT ON THE PERIOD-LUMINOSITY RELATIONSHIP IN THE JWST AND ELT ERA. V. Hocdé\textsuperscript{1}, N. Nardetto\textsuperscript{1}, E. Lagadec\textsuperscript{1}, G. Niccolini\textsuperscript{1}, A. Domiciano de Souza\textsuperscript{1}, A. Mérand\textsuperscript{2}, P. Kervella\textsuperscript{3} and A. Gallenne\textsuperscript{4} Abstract. Cepheids are the keystone of the extragalactic distance ladder since their pulsation periods correlate directly with their luminosity, through the Period-Luminosity (PL) relation (Leavitt & Pickering 1912). The discovery of the accelerated expansion of the Universe (Riess et al. 1998; 2011 Nobel prize) is largely based on the Cepheid distance ladder. However, the calibration of the PL relation is still suffering from systematics errors of at least 2% and it is the largest contributor on the Hubble constant $H_0$ (Riess et al. 2016). These systematics could be partly due to the CircumStellar Envelopes (CSEs) of Cepheids discovered in the last decade by interferometry. Using Spitzer Space Telescope observations, we reconstruct the spectral energy distribution of 5 Cepheids and we report the observation of an infrared (IR) excess continuum. We show for the first time that the IR excess can be modeled by a free-free emission due to a thin circumstellar shell of ionized gas in the chromospheric region. Keywords: Techniques : Spectrometry, Photometry – Infrared : CSE, ISM – Stars : Cepheids – 1 Introduction The extragalactic distance ladder is still largely based on Cepheids PL relations, whose uncertainties on both zero point and slope are today one of the largest contributors to the error on $H_0$ (Riess et al. 2019). One possible bias could be due to IR excesses from CSEs such as the ones discovered using near- and mid-infrared interferometry around nearby Cepheids (Kervella et al. 2006; Mérand et al. 2006). These studies determined a CSE radius of about 3 stellar radii and a flux contribution in the K band ranging from 2% to 10% of the continuum, for medium- and long-period Cepheids. However, we still do not know how these CSEs are formed, neither their nature, nor their characteristics (density and temperature profiles, chemical composition...). This work aims at understanding the nature of these CSEs by reconstructing and modeling the IR excess of Cepheids. 2 Building the infrared excess using the SPIPS algorithm SpectroPhoto-Interferometric modeling of Pulsating Stars (SPIPS) is a model-based parallax-of-pulsation code which includes photometric, interferometric, effective temperature and radial velocity measurements in a robust model fit (Mérand et al. 2015). SPIPS uses a grid of ATLAS9 atmospheric models\textsuperscript{} (Castelli & Kurucz 2003) to compute synthetic photometry to match those from the dataset. While the visible domain up to $\sim 1 \mu m$ are well described by the pulsational model, an IR excess, increasing with wavelength, is observed (see Fig. 1-left). This IR excess has been modeled by an \textit{ad-hoc} analytic law (see green line in Fig. 1-left): $$\text{IR}_{\text{ex}} = \Delta \text{mag} = m_{\text{obs}} - m_{\text{kurucz}} = \begin{cases} 0, & \text{for } \lambda < 1.2 \mu m \\ \alpha (\lambda - 1.2)^{\beta}, & \text{for } \lambda > 1.2 \mu m \end{cases} \quad (2.1)$$ \textsuperscript{1} Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, France, email : email@example.com \textsuperscript{2} European Southern Observatory, Karl-Schwarzschild-Str. 2, 85748 Garching, Germany \textsuperscript{3} LESIA (UMR 8109), Observatoire de Paris, PSL, CNRS, UPMC, Univ. Paris-Diderot, 5 place Jules Janssen, 92195 Meudon, France \textsuperscript{4} European Southern Observatory, Alonso de Córdova 3107, Casilla 19001, Santiago 19, Chile \textsuperscript{}\url{http://wwwuser.oats.inaf.it/castelli/grids.html} with two parameters, $\alpha$ and $\beta$. In the next section we reconstruct the IR excess up to 30$\mu$m owing to Spitzer space telescope observations. Our study aims to physically explain the behaviour of this IR excess, which is likely due to CSE. ### 3 Spitzer data In order to study the IR excess of Cepheids we selected a sample of Galactic Cepheids with Spitzer observations (Werner et al. 2004). The spectroscopic observations were made with the InfraRed Spectrograph IRS (Houck et al. 2004) onboard the Spitzer telescope and the full spectra were retrieved from the CASSIS atlas (Lebouteiller et al. 2011). We derive the IR excess of each star in the sample at the specific phase of Spitzer using: $$\Delta \text{mag} = m_{\text{Spitzer}} - m_{\text{kurucz}}[\phi_{\text{Spitzer}}]$$ \hspace{1cm} (3.1) where $m_{\text{Spitzer}}$ is the magnitude of the Spitzer observation and $m_{\text{kurucz}}[\phi_{\text{Spitzer}}]$ is the magnitude of the ATLAS9 atmospheric model interpolated at the phase of Spitzer observations ($\phi_{\text{Spitzer}}$). The $T_{\text{eff}}(\phi)$ and $\log(g)$ values of the star at the phase of Spitzer are provided by the SPIPS algorithm, while the interpolation is then done in a ATLAS9 grid of models with steps of 250K in effective temperature and 0.5 in log $g$, respectively. The angular diameter derived by SPIPS is then used to calculate $m_{\text{kurucz}}[\phi_{\text{Spitzer}}]$. For V Cen we obtained the IR excess presented in Fig. 1-right. The observed discontinuity at 14$\mu$m is due to the different aperture sizes of Spitzer short and long wavelengths detectors. We also identify a silicate absorption around 10$\mu$m which could obscures a silicate emission from CSE. Therefore it is necessary to correct for the spectra from interstellar silicate absorption for studying the IR excess from CSEs. ### 4 Correcting for the interstellar silicate absorption in Spitzer data In order to correct for the Spitzer spectra from interstellar silicate absorption we first derived the visible absorption $A_v$ assuming an extinction law $A_v = R_v E(B-V)$ with a ratio of total-to-selective extinction of $R_v = 3.1$, which corresponds to a diffuse ISM along the line of sight (Savage & Mathis 1979). Then we used the relation $A_v/\tau_{9.7} = 18.5$ (Roche & Aitken 1984) which is suited to the diffuse ISM in the solar vicinity, in order to derive $\tau_{9.7}$. We obtain the following equation for the diffuse ISM: $$A^{\text{ISM}}_{9.7} = 1.086 \tau_{9.7} = 1.086 \frac{3.1}{18.5} E(B-V) = 0.182 E(B-V),$$ \hspace{1cm} (4.1) Once we derived the specific absorption at 9.7$\mu$m, we can use this value to normalize a synthetic silicate absorption model in order to correct the entire Spitzer observations. Since we assumed an average ISM temperature of 20K, the dust emission is negligible in the Spitzer wavelength range according to Wien's law. Thus, we simply derive the absorption $A^{\text{ISM}}_\lambda$ analytically using Mie theory. Hence we adopted the following expression for $\lambda$ between 5 and 30$\mu$m: $$A^{\text{ISM}}_\lambda \propto \kappa_\lambda = \int C^{\text{abs}}_\lambda(a) \pi a^2 n(a) da$$ \hspace{1cm} (4.2) We first derived $C^{\text{abs}}_\lambda$ using complex refractive index for silicates from Draine & Lee (1984) (hereafter DL84) assuming an uniform distribution of ellipsoidal shapes given by Bohren & Huffman (1983). Then we derived the absorption coefficient $\kappa_\lambda$ by taking into account a standard grain size distribution $n(a) \propto a^{-3.5}$ (Mathis et al. 1977). Finally we normalize $A^{\text{ISM}}_\lambda$ using its specific value $A^{\text{ISM}}_{9.7}$ at 9.7$\mu$m. Correction is presented in Fig. 1. ### 5 The IR excess from a thin shell of ionized gas The shape of the mid-IR excess, saturating to a constant flux ratio at large wavelengths (see Fig. 2), suggests an opacity source increasing with wavelength. We used the free-free and bound-free opacities for a pure H shell presenting such behaviour. We consider the emission of a thin gas shell around the star with constant density and temperature for the sake of simplicity. The combined absorption coefficient (in m$^{-1}$; SI(MKS) unit system) for these two opacities sources is given by (e.g. Rybicki & Lightman 2008) $$\kappa_\lambda = 3.692 \times 10^{-2} \left[ 1 - e^{-\frac{\lambda c}{3kT_e}} \right] T_e^{-1/2} \times (\lambda/c)^3 (\gamma \rho/m_H)^2 [g_{ff}(\lambda, T_e) + g_{bf}(\lambda, T_b)]$$ \hspace{1cm} (5.1) Fig. 1. Left: IR excess analytic law (Eq. 2.1) for V Cen together with the measurements. For each photometric band, red dots with error bars are the mean excess value over the cycle of the Cepheid and the corresponding standard deviation. The green zone is the error on the magnitude obtained using the covariance matrix of SPIPS fitting result. Right: Spitzer IR excess derived from Eq. 3.1. Red points are SPIPS interpolated IR excesses at the specific Cepheid phase of Spitzer observation. The transition between SL and LL detectors is indicated by a dashed line at $\sim 14\mu m$. The cycle-averaged ad-hoc analytic laws from SPIPS are represented by a dashed green for comparison only. Fig. 2. Left: The IR excess of V Cen is presented, including (1) the interpolated IR excess model from SPIPS, (see Sect. 2), (2) the Spitzer observations cleaned from different camera effects and also corrected from the silicate absorption due to the ISM (orange curve see Sect. 4). Right: IR excess fitting results of a shell of ionized gas presented with residuals. Yellow region is the error on the magnitude obtained using the covariance matrix of the fitting result. where $h$, $c$, and $k$ have their usual meanings, $\gamma$ is the degree of ionization (between 0 and 1), $T_s$ the temperature of the shell, $m_H$ the hydrogen mass and $g_{ff}$ and $g_{bf}$ are the free-free and bound-free Gaunt factors respectively. These factors were computed mainly from approximation formulas given by Brussaard & van de Hulst (1962), Hummer (1988), and references therein. We computed the SED of the star plus the gas shell taking into account the latter absorption coefficient $\kappa_\lambda$. In order to match the SPIPS photometries plus corrected Spitzer spectra we performed a $\chi^2$ fitting using the Levenberg-Marquardt method. We fitted three parameters from the gas shell \textit{i.e.} ionized shell mass $\gamma M_\star$, its temperature $T_s$ and radius $R_s$. In addition, since the SPIPS fitting assumes that there is no excess in the visible it is necessary to relax this assumption to allow the data to present deficit or excess in the visible depending on the physical behaviour of the ionized shell. We fitted a fourth parameter corresponding to the IR excess offset corresponding to $\Delta m \neq 0$ for $\lambda < 1.2\mu m$. Results are presented in Fig. 2 in the case of V Cen only. We show that we can reproduce the IR excess with a thin shell of ionized gas with a temperature ranging from 3500 to 4500K depending on the Cepheid considered, with a width of typically $\simeq 15\%$ of the radius of the star. In the case of V Cen, we obtain $T_{\text{shell}} = 4353 \pm 106$K, $R_{\text{shell}} = 1.156 \pm 0.012$ $R_{\text{star}}$, $\gamma M_{\text{shell}} = 3.61.10^{-9} \pm 2.0.10^{-10} M_\odot$ and $\Delta m=0.057 \pm 0.004$. 6 Conclusion and perspectives For the five Cepheids we studied, we report a continuum IR excess increasing up to $\sim 0.1/-0.2$ magnitudes at $30\mu m$, which cannot be explained by a hot or cold dust model of CSE. We show for the first time that IR excesses of Cepheids can be explained by free-free emission from a thin shell of ionized gas with a thickness of $\simeq 8-17\%$ star radius, an ionized mass of $10^{-9} - 10^{-7} M_\odot$ and a temperature of $3500-4500 K$. In this simple model, density and temperature have a constant radial distribution. Interferometric observations have resolved CSEs around Cepheids. These CSEs were modeled with a ring at a distance of 2 to $3 R_\star$, i.e. close to the star, in a region where the temperature is high enough ($> 2000 K$) to prevent dust condensations. Thus, these observations are more likely explained by a shell of partially ionized gas. In parallel, it is interesting to compare Cepheids to very long period Mira stars for which a radiosphere near $2 R_\star$ due to free-free emission has been reported (Reid & Menten 1997). Also, extensive studies of H$\alpha$ profiles in the atmosphere of Cepheids have shown that strong increases of turbulence occurs when the atmosphere is compressed during its infalling movement, or because of shock waves dynamics (see for instance Breitfellner & Gillet (1993)). Moreover the analytical work of Neilson & Lester (2008) have shows that mass loss is enhanced by pulsations and shocks in the atmosphere. We suggest this mass loss could be in the form of partially ionized gas. The model of the shell of ionized gas could also be linked to the chromospheric activity of Cepheids. For example, Sasselov & Lester (1994) report HeI$\lambda 10830$ observation on seven Cepheids providing the evidence of a high temperature plasma and steady material outflow in the highest part of the atmosphere. However our model does not take into account temperature nor density gradients in the star’s atmosphere, and in particular compression and/or shock waves which could also heat up the shell and ionize the gas. Thus, a spatial and chromatic analysis of the shell including interferometric constrains in all available bands with in particular VEGA/CHARA (visible), PIONIER/VLTI (infrared) and MATISSE/VLTI (L, M, N bands) is still necessary to better understand the environment of Cepheids, and eventually, check the impact on the PL relation. The authors acknowledge the support of the French Agence Nationale de la Recherche (ANR), under grant ANR-15-CE31-0012-01 (project UnlockCepheids). References Bohren, C. F. & Huffman, D. R. 1983, Absorption and scattering of light by small particles (New York: Wiley, 1983) Breitfellner, M. G. & Gillet, D. 1993, A&A, 277, 553 Brusaard, P. J. & van de Hulst, H. C. 1962, Reviews of Modern Physics, 34, 507 Castelli, F. & Kurucz, R. L. 2003, in IAU Symposium, Vol. 210, Modelling of Stellar Atmospheres, ed. N. Piskunov, W. W. Weiss, & D. F. Gray, A20 Draine, B. T. & Lee, H. M. 1984, ApJ, 285, 89 Houck, J. R., Roellig, T. L., van Cleve, J., et al. 2004, ApJS, 154, 18 Hummer, D. G. 1988, ApJ, 327, 477 Kervella, P., Mérand, A., Perrin, G., & Coude du Foresto, V. 2006, A&A, 448, 623 Lebouteiller, V., Barry, D. J., Spoon, H. W. W., et al. 2011, ApJS, 196, 8 Mathis, J. S., Rumpl, W., & Nordsieck, K. H. 1977, ApJ, 217, 425 Mérand, A., Kervella, P., Breitfelder, J., et al. 2015, A&A, 584, A80 Mérand, A., Kervella, P., Coude du Foresto, V., et al. 2006, A&A, 453, 155 Neilson, H. R. & Lester, J. B. 2008, ApJ, 684, 569 Reid, M. J. & Menten, K. M. 1997, ApJ, 476, 327 Riess, A. G., Casertano, S., Yuan, W., Macri, L. M., & Scolnic, D. 2019, arXiv e-prints Roche, P. F. & Aitken, D. K. 1984, MNRAS, 208, 481 Rybicki, G. B. & Lightman, A. P. 2008, Radiative processes in astrophysics (John Wiley & Sons) Sasselov, D. D. & Lester, J. B. 1994, ApJ, 423, 777 Savage, B. D. & Mathis, J. S. 1979, ARA&A, 17, 73 Werner, M. W., Roellig, T. L., Low, F. J., et al. 2004, ApJS, 154, 1
HOW YOUR PERSONALITY MAY BE AFFECTING YOUR PRODUCTIVITY How much time do you spend using communication tools such as email and messaging, and what is it doing to your productivity? While you might not realize how much you bounce back and forth between tasks, every time you switch your focus, you’re fragmenting your workday. Communication tools in particular can be a productivity killer. Recent studies show that the average office employee spends almost a third of the workday reading and responding to email. That number is 10-15% higher for remote employees. One analysis discovered that when we toss in other communication tools such as Slack or Microsoft Teams that number jumps to 40%. In another study, researchers found that, in general, workers average only 3 minutes on any given task before switching and about 2 minutes using any digital tool before switching. While many of us think we are multitasking ninjas, neuroscience tells us that there really is no such thing as multitasking. As incredible as the human brain is, it has a tough time focusing on two demanding tasks simultaneously. Rather, your brain just bounces back and forth between tasks. There is a wealth of research surrounding “task switching.” The University of Utah’s Department of Psychology dove deep into this problem in their study on how task switching impacts performance. They found that, in general, when we try to focus on more than one thing at a time, performance suffers. Specifically, the team studied people who drive while on the phone. Over the course of a decade, he and his colleagues had demonstrated that drivers using cell phones—even hands-free devices—were at just as likely to get into an accident as intoxicated drivers. Reaction time slowed, attention decreased to the point where they’d miss more than half the things they’d otherwise see such as a change in speed limit, a billboard or a child by the road. They maintain that there is a tiny but persistent subset of the population—about 2% —who do not lose productivity, focus, and cognitive ability and in some cases they can even improve when multiple demands are placed on their attention. According to the team multitasking it is more like I.Q.: most people cluster in an average range, but there is a long tail where only a tiny fraction—single digits among thousands—will ever find themselves. The problem is that we all like to think that we’re in the 2% club. Ironically, people who multitask the least are the best at it, and the people who multitask the most are generally the worst at it. The super multitaskers are true outliers. For the vast majority of us, interrupting that deck you’re working on to stop and check your inbox does more than just slow you down on the presentation. It actually zaps your cognitive resources. Constant task switching changes the structure of the brain physically and chemically impacting concentration, recall, and overall mental performance. The more you bounce back and forth between tasks and email throughout the day, the more inefficient you get at each task. According to a University of California-Irvine study, we lose 20 minutes every time we shift our focus from the current task to our inbox. Further research shows that the stress of email overload generates cortisol and that increase of cortisol makes it harder for the prefrontal cortex – the rational, thinking brain that weighs actions against consequences – to do its job effectively. We’ve all been there…pound out a response to an email and hit send before stopping to think about how that message may be received or whether that message should be sent at all. Research shows the single most effective way to boost your productivity is to manage your inbox the same way you manage other tasks in your day. Rather than attending to it every time a message pops up, turn off your notifications. Schedule blocks of time on your calendar specifically for email and treat it as you would any other meeting. (article continued on page 3) FROM JEFF’S DESK As you know, the Chamber has been leading an effort to promote the livability of Sidney and the surrounding area through a placemaking initiative launched last year. It is our feeling and that of others collaborating with us that a promotional strategy like this will assist the city with growing its population and aide all Shelby County employers in their pursuit of new talent to join their companies. Let’s face it, employers in Shelby County and throughout the country are facing a monumental challenge to attract new talent. We know that the employers here are some of the best in the world and offer career opportunities unmatched by many. As they recruit in a highly competitive environment, they can sell the merits of their organization as a great place to grow a career. Our role is to sell the merits of this community as a great place to live. Together we can accomplish much. In late 2020 we introduced a new community and destination brand. Earlier this year we unveiled a brand story video that captures the essence of our community lifestyle. Late last month we published a microsite to gain an online presence where those considering a new job or relocation can be introduced to Sidney as an attractive option to make their home. To see our strategy in action, we encourage you to view www.ExperienceSidney.com. Here you will find an information hub where those interested can familiarize themselves with our lifestyle. The site has been fashioned in such a way that it will appeal to a variety of preferences be they singles or couples, young families with school aged children, empty nesters, or those later in their careers. Over the next several weeks we will be reaching out to area employers and human resources professionals to familiarize them with this new resource and to encourage their sharing of our site link in the employment related section of their company’s web site. Similarly, we will be contacting area real estate agents to encourage their sharing of this site with prospective home seekers from outside the area. As we go forward, our plan is to do much more to promote site visits to ExperienceSidney.com. Outdoor advertising, digital strategies that include search engine marketing and targeted online advertising tactics will be utilized. In the end we hope to influence the decision of many to join us as a resident, an employee, and a contributing member to the growth and prosperity of our community. NEW CHOICES TO HOST ANNUAL “GIVE A LITTLE, LAUGH A LOT” FUNDRAISER New Choices, Inc. invites you to attend their annual comedy night fundraiser, “Give A Little, Laugh A Lot”, on Friday, April 23, 2021 beginning at 6:00 p.m. Your $30.00 ticket purchase will include dinner and entertainment. New Choices, Inc. is excited to host comedian, Mack Dryden. Throughout the evening, guests will have the opportunity to participate in a live and silent auction for some amazing prizes! The mission at New Choices is to provide emergency shelter, crisis intervention, education and advocacy for adults and children who are victims of domestic violence; to assist victims through the criminal justice process; to raise awareness in the community through education; and to provide assistance through personal education and training to help victims break the cycle of domestic violence and regain control of their lives. New Choices, Inc. is the ONLY agency in Shelby County that provides services to victims of Domestic Violence. Their crisis hotline and emergency shelter are available to those who need assistance 24 hours a day, 7 days a week. 100% of the comedy night proceeds will go toward supporting New Choices’ mission and help provide services and support to the many families who live with abuse in the Shelby County community. For tickets, call (937) 498-7261 or email firstname.lastname@example.org. TUESDAY, MAY 25, 2021 Shelby Oaks Golf Course 10:30 a.m. shot-gun start Details for the 2021 Chamber Golf Classic will be sent to members in the next week or two. Sponsorship information and a registration form are included in April’s Chamber Mail packet. It’s never too early to get your team together! Contact Dawn Eilert with questions, 492-9122. SIDNEY ALIVE EASTER EGG HUNT AND COLORING CONTEST DETAILS — THERE’S STILL TIME TO PARTICIPATE Egg Hunt—Participating businesses have numbered eggs in their storefront windows. Grab a hunt sheet from the Sidney Alive office, print one from home or pick one up in a downtown business. Sheets can be dropped off to the Sidney Alive office anytime before 4:00 p.m. on Thursday, April 1st. The winner of a great prize pack will be drawn from the sheets turned in. Coloring Contest—Coloring sheets can be picked up from the Sidney Alive office or printed at home. There are three age groups: 1-3, 4-6 and 7-11. Winning coloring sheets will be displayed in Sidney Alive’s office window. Winners of each age group will receive an Easter prize pack. Coloring sheets can be dropped off to the Sidney Alive office anytime before 4:00 p.m. on Thursday, April 1st. For more information, visit www.sidneyalive.org. DISCOVER MAGAZINE We are thrilled to announce the next edition of Discover Shelby County magazine is officially at the printer! Each year in mid-April, we receive thousands of copies of the magazine to distribute to each member, libraries, hotels, HR departments, the hospital, banks, village offices and hotspots, schools. We want to share this publication anywhere and everywhere. Let us know if you’d like additional copies! WELCOME NEW MEMBERS DAVE NAGEL EXCAVATING, LLC BUSINESS CATEGORY: Construction/Building Supplies Anna Nagel 10491 State Route 47 Sidney, Ohio 45365 Phone: (937) 726-0790 Web Site: www.davenaglexcavating.com WATERDEEPMEDIA, LLC BUSINESS CATEGORY: Media 9650 State Route 119 West Anna, Ohio 45334 Phone: (937) 538-6948 Web Site: www.waterdeepmedia.com BWC APRIL TRAINING OPPORTUNITIES Webinars 4/13—Cost-Control Strategies: A Collaborative Approach (11:00 a.m.—12:30 p.m.) 4/20—Engaging Millennials in the Workplace (2:15—3:15 p.m.) If you miss a webinar, just search “replay” in the BWC Learning Management System. Virtual Classes 4/5-6—Electrical Safety in the Workplace through Insight and Implementation of NFPA 70E (9:30 a.m.—12:45 p.m.) 4/8—Series I: Intro to OSHA and Safety Culture Basics (9:30 a.m.—1:00 p.m.) 4/14—Noise and Hearing Half Day Workshop (1:00—4:30 p.m.) 4/19-23—Electrical Hazard Recognition and Abatement (9:30 a.m.—2:00 p.m.) 4/27—Wellness in the Workplace (1:00—4:30 p.m.) 4/28—Personal Protective Equipment Selection Criteria (9:30 a.m.—1:00 p.m.) Online E-Courses On-demand courses cover topics including Bloodborne Pathogens, Developing a Safety Culture, OSHA Recordkeeping 101, and others. Online course times range from 30 minutes to approximately two hours and provide the ultimate convenience for learning at your pace and at your desired time. Check out the schedule at: www.bwclearningcenter.com/CServer/Document/BF85C8D3052D4250B4E56E60B44EC1AE/ONLINE%20COURSE%20FLYER.pdf To register for webinars or virtual classes, go to www.bwclearningcenter.com. You will need an account to use the Learning Center. HOW YOUR PERSONALITY MAY BE AFFECTING YOUR PRODUCTIVITY (cont’d) The Four Most Common Multi-Tasking Personalities Some of us have a harder time staying focused for extended periods of time than others. But, new studies indicate that the tendency to multitask may be baked into our personalities. There are essentially four types of people with a greater tendency to get caught in the multitasking trap: 1. Approach-oriented or Reward-focused: Persons with a strong approach orientation, that is, a strong reward or gain focused motivational orientation, may be especially enticed to take on multiple tasks because of the high potential rewards. In contrast, persons who are avoidance oriented, that is, who are risk averse and sensitive to losses or punishments, may be more inclined to focus on a singular task rather than multitask because of the higher potential losses and greater effort associated with trying to do more. 2. High-sensation Seeker: High sensation seekers may be especially apt to multitask for the sake of the more varied and complex sensations that are afforded by multiple vs. singular tasks. Impulsivity is a complex construct that is commonly defined as a predisposition toward rapid, unplanned reactions to internal or external stimuli without regard to the negative consequences of these reactions. 3. 2% Club Member: These people are convinced they are in that tiny sliver of master multitaskers. People who chronically multitask are not those who are the most capable of multitasking effectively. To the contrary, task performance is negatively correlated with self-reported multitasking activity. 4. Easily Distracted: Across all analyses, multitasking was most strongly associated with attentional impulsivity. Thus, the people who are most likely to multitask appear to be those who have difficulty focusing attention or concentrating on a single task. The Bottom Line: According to researchers, people don’t multitask because they are good at it. They do it because they are more distracted. GET REIMBURSED UP TO $2,000 PER EMPLOYEE FOR TRAINING Tech Cred is here with more opportunities to help you provide technical training for your employees! Employers that are interested in participating in Ohio’s TechCred program have a new opportunity to apply for funding to upskill their employees. The new application period begins April 1 and ends April 30, 2021. Employers can receive up to $2,000 for each current or incoming employee, and up to $30,000 in total per round. Businesses of all sizes are eligible. Edison State offers numerous certificates which qualify for the TechCred program in the areas of: Apprenticeships Business Technology Customized Training Healthcare Technology Information Technology IT and Cybersecurity Manufacturing Robotics/Automation Find out more at: https://files.constantcontact.com/b6a89c3b001/4ce6e4d34-f9af-4c96-aa70-271696d46dfc.pdf Don’t see the program you’re looking for? Edison State works closely with employers to develop and provide in-demand training. Certificates completed through Ohio’s TechCred program also set participants on a direct path to later complete an associate degree. Completing the application process is easy, but please let Edison help! Find out more at https://www.edisonohio.edu/techcred/ or contact Brandi Olberding at email@example.com. WELCOME APRIL! • 04/02: CLOSING at Noon for Good Friday • 04/09: Board Meeting, 9:00a • 04/13: Luncheon 11:30a; Meals available, call by April 9 to pre-register. Sponsored by Citizens for Sidney City Schools. • 04/26: Newsletter Mailing, 9:00a No Spring Card Party, however please put your loose change in the Jingle Jangle Jar. Every donation helps! THANK YOU. 304 S. West Avenue, Sidney, OH 45365 P: (937) 492-5266 F: (937) 492-2134 www.srcentersidney-shelby.com firstname.lastname@example.org 101 South Ohio Avenue Floor 2 Sidney, Ohio 45365 (937) 492-9122 CHECK US OUT ON FACEBOOK, LINKEDIN & TWITTER SOLLMANN ELECTRIC INDUSTRIAL • COMMERCIAL • RESIDENTIAL OH LIC 26835 310 East Russell Road • Sidney, Ohio 45365 Ph: 937-492-0346 Fax: 937-492-4447 Web: www.sollmannelectric.com Electricians Needed! No experience required, we will provide training. Apply in person at Sollmann Electric Company 310 E. Russell Rd., Sidney, OH 45365 INTRODUCING... THE VAULT A secure off site co-location for your company's important information! Battery UPS Generator Backup 24/7 Secure Access 24/7 Video Surveillance FULL RACKS AVAILABLE CALL 1-888-NKTELCO FOR MORE INFORMATION www.nktelco.com HAPPY EASTER SIDNEY-SHELBY COUNTY CHAMBER OF COMMERCE GOLF CLASSIC Event Date: Tuesday, May 25, 2021 Location: Shelby Oaks Golf Course, 9900 Sidney-Freyburg Road, Sidney Time: 10:30 a.m. start time (please arrive by 10:00 a.m.) Attendance: 2020 – 156 golfers; 2019 – 164 golfers SPONSORSHIP OPPORTUNITIES Gold Sponsorship - $750.00 1. Company logo on tag attached to participant gift. 2. Golf foursome (includes two carts, green fees, lunch, hors d’oeuvres, drinks). 3. Company name on sign located at one tee box or green. 4. Company name on signage displayed in the morning as players arrive as well as in the Oaks Club after golf. 5. Company name in program and recognition from podium at after golf event. 6. Company name listed in the Chamber newsletter, Perspective (circulation 600+) and on web site. Silver Sponsorship - $550.00 1. Golf foursome (includes two carts, green fees, lunch, hors d’oeuvres, drinks). 2. Company logo tents on Clubhouse tables and Oaks Club tables. 3. Company logo sign on leader board before and after tournament. 4. Company name on sign located at one tee box or green. 5. Company name on signage displayed in the morning as players arrive as well as in the Oaks Club after golf. 6. Company name in program and recognition from podium at after golf event. 7. Company name listed in the Chamber newsletter, Perspective (circulation 600+) and on web site. Bronze Sponsorship - $400.00 1. Golf foursome (includes two carts, green fees, lunch, hors d’oeuvres, drinks). 2. Company name on table tents at dining tables in Oaks Club. 3. Company name on sign located at one tee box or green. 4. Company name on signage displayed in the morning as players arrive as well as in the Oaks Club after golf. 5. Company name in program and recognition from podium at after golf event. 6. Company name listed in the Chamber newsletter, Perspective (circulation 600+) and on web site. Hole Sponsorship - $100.00 1. Company name on sign located at one tee box or green 2. Company name on signage displayed in the morning as players arrive as well as in the Oaks Club after golf. 3. Company name in program and recognition from podium at after golf event. 4. Company name listed in the Chamber newsletter, Perspective (circulation 600+) and on web site. Proximity Prizes - Value of $25.00 or greater We are seeking approximately 12-15 proximity prizes valued at $25.00 or more to award to participants with the Longest Putt, Longest Drive, and Closest to the Pin on each nine. All donations will be recognized in the event program, in the Chamber newsletter, Perspective, and on our web site.
Evidence of Spin-Orbital Angular Momentum Interactions in Relativistic Heavy-Ion Collisions Original Evidence of Spin-Orbital Angular Momentum Interactions in Relativistic Heavy-Ion Collisions / Acharya, S; Adamova, D.; Adolfson, J; Aggarwal, M.; Rinella, G.; Agnello, M.; Agrawal, N; Bufalino, S.; Concas, M.; Grosa, F.; Ravaenga, I.; Catalano, F.; Fecchio, P.. - In: PHYSICAL REVIEW LETTERS. - ISSN 0031-9007. - STAMPA. - 125:1(2020), p. 012301. [10.1103/PhysRevLett.125.012301] Availability: This version is available at: 11583/2848372 since: 2020-10-14T09:50:01Z Publisher: American Physical Society Published DOI:10.1103/PhysRevLett.125.012301 Terms of use: openAccess This article is made available under terms and conditions as specified in the corresponding bibliographic description in the repository (Article begins on next page) Evidence of Spin-Orbital Angular Momentum Interactions in Relativistic Heavy-Ion Collisions S. Acharya et al.* (The ALICE Collaboration) (Received 21 November 2019; revised 25 February 2020; accepted 27 May 2020; published 30 June 2020) The first evidence of spin alignment of vector mesons ($K^{0}$ and $\phi$) in heavy-ion collisions at the Large Hadron Collider (LHC) is reported. The spin density matrix element $\rho_{00}$ is measured at midrapidity ($\|y\| < 0.5$) in Pb-Pb collisions at a center-of-mass energy ($\sqrt{s_{NN}}$) of 2.76 TeV with the ALICE detector. $\rho_{00}$ values are found to be less than $1/3$ ($1/3$ implies no spin alignment) at low transverse momentum ($p_T < 2$ GeV/$c$) for $K^{0}$ and $\phi$ at a level of $3\sigma$ and $2\sigma$, respectively. No significant spin alignment is observed for the $K_S^0$ meson (spin = 0) in Pb-Pb collisions and for the vector mesons in $pp$ collisions. The measured spin alignment is unexpectedly large but qualitatively consistent with the expectation from models which attribute it to a polarization of quarks in the presence of angular momentum in heavy-ion collisions and a subsequent hadronization by the process of recombination. DOI: 10.1103/PhysRevLett.125.012301 Ultrarelativistic heavy-ion collisions create a system of deconfined quarks and gluons, called the quark-gluon plasma (QGP) [1–3] and provide the opportunity to study its properties. In collisions with nonzero impact parameter, a large angular momentum of $O(10^7)\hbar$ [4] and magnetic field of $O(10^{14})$ T [5] are also expected. While the magnetic field is short lived (a few fm/$c$), the angular momentum is conserved and could affect the system throughout its evolution. Experimental observables like correlations in azimuthal angle [6,7] could be used to study the influence of these initial conditions on the properties and the dynamical evolution of the QGP and its subsequent hadronization. Spin-orbit interactions have wide observable consequences in several branches of physics [8–10]. In the presence of a large angular momentum, the spin-orbit coupling of quantum chromodynamics (QCD) could lead to a polarization of quarks followed by a net-polarization of vector mesons ($K^{0}$ and $\phi$) [11–15] along the direction of the angular momentum. The spin state of a vector meson is described by a $3 \times 3$ Hermitian spin-density matrix [15]. Its trace is 1 and $\rho_{11}$ and $\rho_{-1,-1}$ cannot be measured separately in two-body decays to pseudoscalar mesons. Consequently, there is only one independent diagonal element, $\rho_{00}$. The elements of the spin-density matrix can be studied by measuring the angular distributions of the decay products of vector mesons with respect to a quantization axis. Here two different quantization axes are used: (i) a vector perpendicular to the production plane (PP) of the vector meson and (ii) the normal to the reaction plane (RP) of the system. The PP is defined by the flight direction of the vector meson and the beam direction. The spin-density matrix element $\rho_{00}$ is determined from the distribution of the angle $\theta^$ between the kaon decay daughter and the quantization axis in the decay rest frame [16,17], $$\frac{dN}{d\cos\theta^} \propto [1 - \rho_{00} + \cos^2\theta^(3\rho_{00} - 1)]. \quad (1)$$ $\rho_{00}$ is $1/3$ in the absence of spin alignment and the angular distribution in Eq. (1) is uniform. The experimental signature of spin alignment is a nonuniform angular distribution ($\rho_{00} \neq 1/3$). The direction of the angular momentum in noncentral heavy-ion collisions is perpendicular to the reaction plane (subtended by the beam axis and impact parameter) [12]. The spin-orbit interaction is expected to lead to spin alignment with respect to the RP. The reaction plane orientation cannot be measured directly, but is estimated from the final state distributions of particles. This experimentally measured plane is called the event plane (EP) [18]. The deviation of the EP with respect to the RP is corrected using the EP resolution ($R$) and observed $\rho_{00}^{\text{obs}}$ [19], $$\rho_{00} = \frac{1}{3} + \left(\rho_{00}^{\text{obs}} - \frac{1}{3}\right) \frac{4}{1 + 3R}. \quad (2)$$ There are specific qualitative predictions for the spin alignment effect [13]: (a) $\rho_{00} > 1/3$ if the hadronization of a polarized parton proceeds via fragmentation and less than $1/3$ for hadronization via recombination, (b) $\rho_{00}$ is expected to have a smaller deviation from $1/3$ for both central (impact parameter $\lesssim 3$ fm) and peripheral (impact parameter $\gtrsim 11$ fm) heavy-ion collisions, and a maximum deviation for mid-central collisions, where the angular momentum is also maximal, (c) the $\rho_{00}$ value is expected to have maximum deviation from $1/3$ at low $p_T$ and reach the value of $1/3$ at high $p_T$ in the recombination scenario, and (d) the effect is expected to be larger for $K^{0}$ compared to $\phi$ due to their constituent quark composition. The initial large magnetic field might also affect the $\rho_{00}$ values [15]. This leads to $\rho_{00} > 1/3$ for neutral and $\rho_{00} < 1/3$ for charged vector mesons. Hence magnetic field and angular momentum could have opposite effects on electrically neutral $K^{0}$, $\phi$. All of these features are probed for $K^{0}$ and $\phi$ mesons in Pb-Pb collisions presented in this Letter. As a cross check, a control measurement is carried out using $pp$ collisions, which do not possess large initial angular momentum, and the same analysis is done in Pb-Pb collisions for $K^0_S$ meson, which has zero spin. In addition, the measurements are carried out by randomizing the directions of the event (RNDEP) and production planes. The analyses are carried out using 43 million minimum bias $pp$ collisions at $\sqrt{s} = 13$ TeV, taken in 2015 and 14 million minimum bias Pb-Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV, collected in 2010. The minimum bias event selection in Pb-Pb collisions require at least one hit in any of V0A, V0C, and silicon pixel detectors while in $pp$ collisions at least one hit in both V0A and V0C is required. The events are further required to have a primary vertex position within $\pm 10$ cm of the detector center along the beam axis. The events were classified by collision centrality classes based on the amplitude measured in the V0 counters [20]. The measurements are performed at mid-rapidity ($\|y\| < 0.5$) as a function of $p_T$ and are reported for $pp$ collisions as well as for different centrality classes in Pb-Pb collisions. The $K^0_S$ analysis is performed only for Pb-Pb collisions in the 20–40% centrality class which corresponds to the top 20–40% of V0 amplitude distribution. The details of the ALICE detector, trigger conditions, centrality selection, and second order event plane estimation using the V0 detectors at forward rapidity, can be found in [20–23]. The $K^{0}$ and $\phi$ candidates are reconstructed via their decays into charged $K\pi$ and $KK$ pairs, respectively, while the $K^0_S$ is reconstructed via its decay into two charged pions. The time projection chamber (TPC) [24] and time-of-flight (TOF) detector [25] are used to identify the decay products of these mesons via specific ionization energy loss and time-of-flight measurements, respectively. The $K^{0}$ and $\phi$ yields are determined via the invariant mass technique [26–28]. The background coming from combinatorial pairs and misidentified particles is removed by constructing the invariant mass distribution from so-called mixed events for the $K^{0}$ and $\phi$ [26,27]. The combinatorial background for the $K^0_S$ candidates is significantly reduced by selecting the distinctive $V$-shaped decay topology [28]. The invariant mass distributions are fitted with a Breit-Wigner and Voigtian (convolution of Breit-Wigner and Gaussian distributions) function for the $K^{0}$ and $\phi$ signals, respectively, along with a second-order polynomial that describes the residual background [26,27]. Extracted yields are then corrected for the reconstruction efficiency and acceptance in each $\cos \theta^+$ and $p_T$ bin [26,27]. The reconstruction efficiency is determined from Monte Carlo simulations of the ALICE detector response based on GEANT3 simulation [26,27]. The signal extraction procedures for the vector mesons and $K^0_S$ are identical to those used in earlier publications reporting the $p_T$ distribution of the mesons [26–28]. The mass peak positions and widths of the resonances across all the $\cos \theta^+$ bins for various $p_T$ intervals in $pp$ collisions and in different centrality classes of Pb-Pb collisions are consistent with those obtained from earlier analyses [26–28] and no significant dependence on $\cos \theta^+$ is seen. The resulting efficiency and acceptance corrected $dN/d\cos \theta^+$ distributions for selected $p_T$ intervals in minimum bias $pp$ collisions and in 10–50% central Pb-Pb collisions are shown in Fig. 1. These distributions are fitted with the functional form given in Eq. (1) to determine $\rho_{00}$ for each $p_T$ bin in $pp$ and Pb-Pb collisions. For the EP results, the resolution values $R$ are 0.71, 0.53, 0.72, 0.66, and 0.40 for 10–50%, 0–10%, 10–30%, 30–50%, and 50–80% collision centralities, respectively [29]. There are three main sources of systematic uncertainties in the measurements of the angular distribution of vector meson decays. (i) Meson yield extraction: this contribution is estimated by varying the fit ranges for the yield extraction, the normalization range for the signal + background and background invariant mass distributions, the procedure to integrate the signal function to get the yields, and by leaving the width of the resonance peak free or keeping it fixed to the PDG value as discussed in Refs. [26,27]. The uncertainties for $\rho_{00}$ is at a level of 12(8)% at the lowest $p_T$ and decrease with $p_T$ to 4(3)% at the highest $p_T$ studied for the $K^{0}(\phi)$. (ii) Track selection: this contribution includes variations of the selection on the distance of closest approach to the collision vertex, the number of crossed pad rows in the TPC [24], the ratio of found clusters to the expected clusters, and the quality of the track fit. The systematic uncertainties for $\rho_{00}$ are 14(6)% at the lowest $p_T$ and about 11(5)% at the highest $p_T$ for $K^{0}(\phi)$. (iii) Particle identification: this is evaluated by varying the particle identification criteria related to the TPC and TOF detectors. The corresponding uncertainty is 5(3)% at the lowest $p_T$ and about 4(4.5)% at the highest $p_T$ studied for $K^{0}(\phi)$. Systematic uncertainties due to different variations are considered as uncorrelated and the total systematic uncertainty on $\rho_{00}$ is obtained by adding all the contributions in quadrature. Several consistency checks are carried out and details can be found in the Supplemental Material [17]. The final measurement is reported for the average yield of particles ($K^{0}$) and antiparticles ($\bar{K}^{0}$) as results for $K^{0}$ and $\bar{K}^{0}$ were consistent. Figure 2 shows the measured $\rho_{00}$ as a function of $p_T$ for $K^{0}$ and $\phi$ mesons in $pp$ collisions and Pb-Pb collisions, along with the measurements for $K_S^0$ in Pb-Pb collisions. In mid-central (10–50%) Pb-Pb collisions, $\rho_{00}$ is below 1/3 at the lowest measured $p_T$ and increases to 1/3 within uncertainties for $p_T > 2$ GeV/$c$. At low $p_T$, the central value of $\rho_{00}$ is smaller for $K^{0}$ than for $\phi$, although the results are compatible within uncertainties. In $pp$ collisions, $\rho_{00}$ is independent of $p_T$ and equal to 1/3 within uncertainties. For the spin zero hadron $K_S^0$, $\rho_{00}$ is consistent with 1/3 within uncertainties in Pb-Pb collisions. The results with random event plane directions are also compatible with no spin alignment for the studied $p_T$ range, except for the smallest $p_T$ bin, where $\rho_{00}$ less than 1/3 but still larger than for EP and PP measurements. The results for the random production plane (the momentum vector direction of each vector meson is randomized) are similar to RNDEP measurements. These results indicate that a spin alignment is present at lower $p_T$, which is a qualitatively consistent with predictions [13]. Figure 3 shows $\rho_{00}$ for $K^{0}$ and $\phi$ mesons as a function of average number of participating nucleons ($\langle N_{\text{part}} \rangle$) [20,22] for Pb-Pb collisions at $\sqrt{s_{\text{NN}}} = 2.76$ TeV. Large $\langle N_{\text{part}} \rangle$ correspond to central collisions and small $\langle N_{\text{part}} \rangle$ correspond to peripheral collisions (see Table I of the Supplemental Material [17]). In the lowest $p_T$ range, $\rho_{00}$ shows maximum deviation from 1/3 for intermediate centrality and approaches 1/3 for both central and peripheral collisions. This centrality dependence is qualitatively consistent with the dependence of the initial angular momentum on impact parameter in heavy-ion collisions [4]. At higher $p_T$, $\rho_{00}$ is consistent with 1/3 for all centrality classes. For the low-$p_T$ measurements in 10–30% (20–40% for $\phi$ meson with respect to PP) mid-central Pb-Pb collisions, the maximum deviations of $\rho_{00}$ from 1/3 with respect to the PP (EP) are 3.2 (2.6) $\sigma$ and 2.1 (1.9) $\sigma$ for $K^{0}$ and $\phi$ mesons, respectively. The errors ($\sigma$) are calculated by adding statistical and systematic uncertainties in quadrature. The relation between the $\rho_{00}$ values with respect to different quantization axes can be expressed using Eq. (2). and calculating the corresponding factor $R$. This gives $\Delta \rho_{00}(RNDEP) = \Delta \rho_{00}(EP) \times \frac{1}{4} (R = 0$ for random plane$)$ and $\Delta \rho_{00}(PP) = \Delta \rho_{00}(EP) \times (1 + 3v_2)/4$ ($R = v_2$ for production plane, where $v_2$ is the second Fourier coefficient of the azimuthal distribution of produced particles relative to the event plane angle). Here $\Delta \rho_{00} = \rho_{00} - 1/3$. This is further confirmed using a toy model simulation with the PYTHIA 8.2 event generator [30] by incorporating $v_2$ and spin alignment (see the Supplemental Material [17] for further details). In the past, spin alignment measurements in $e^+e^-$ [31–33], hadron-proton [34] and nucleon-nucleus collisions [35] were carried out to understand the role of spin in the dynamics of particle production, finding $\rho_{00} > 1/3$ and off-diagonal elements close to zero with respect to the PP. For $pp$ collisions at $\sqrt{s} = 13$ TeV, we find $\rho_{00} \sim 1/3$ within the studied $p_T$ range (see Fig. 2). New preliminary results from RHIC have found deviations of $\rho_{00}$ from $1/3$ indicating spin alignment for vector mesons at lower $\sqrt{s_{NN}}$ [36,37]. The $\rho_{00}$ for $\phi$ mesons in mid-central Pb-Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV is less than $1/3$ while the preliminary finding for mid-central Au-Au collisions at $\sqrt{s_{NN}} = 200$ GeV is $\rho_{00}$ greater than $1/3$. The $\rho_{00} > 1/3$ for $\phi$ mesons has been interpreted as evidence for a coherent $\phi$ meson field [38]. Similar conclusions cannot be easily applied to $K^{0}$ as it consists of valence quarks of unequal mass ($s$ and $\bar{d}$), which makes it impossible to separate the effects of vorticity and due to electromagnetic and mesonic fields. Significant polarization of $\Lambda$ baryons (spin = 1/2) was reported at low RHIC energies. The polarization is found to decrease with increasing $\sqrt{s_{NN}}$ [39,40]. At the LHC, the global polarization for $\Lambda$ baryon is compatible with zero within uncertainties [$P_\Lambda(\%) = 0.01 \pm 0.06 \pm 0.03$] [41]. The spin alignment for vector mesons in heavy ion collisions could have contributions from angular momentum [12,13], electromagnetic fields [15] and mesonic fields [38]. While no quantitative theoretical calculation for vector meson polarization at LHC energies exists, the expected order of magnitude can be estimated and the measurements for vector mesons and hyperons can be related in a model dependent way. Considering only the angular momentum contribution and recombination as the process of hadronization [13], the $\rho_{00}$ of vector mesons are related to quark polarization as $\rho_{00} = (1 - P_q P_{\bar{q}})/(3 + P_q P_{\bar{q}})$ where $P_q$ and $P_{\bar{q}}$ are quark and antiquark polarization, respectively. Assuming $P_u = P_u = P_d = P_d$ and $P_s = P_s$, the measured $p_T$ integrated $\rho_{00}$ values for $K^{0}$ and $\phi$ mesons in 10–50% Pb-Pb collisions could translate to light quark polarization of $\sim 0.8$ and strange quark polarization of $\sim 0.2$. Using a thermal and nonrelativistic approach as discussed in [42], vorticity ($\omega$) and temperature ($T$) are related to $\rho_{00}$ and vector meson polarization ($P_V$) as $\rho_{00} \approx \frac{1}{3} \left\{ 1 - \left[ (\omega/T)^2/3 \right] \right\}$ and $P_V \simeq (2\omega/3T)$, respectively. Also in this approach, the measured $\rho_{00}$ for $K^{0}$ would correspond to $K^{0}$ polarization of $\sim 0.6$ and the $\rho_{00}$ for $\phi$ mesons would give $\phi$ meson polarization of $\sim 0.3$. In the recombination model, $\Lambda$ polarization depends linearly on quark polarization whereas vector meson polarization depends quadratically on it. One would therefore expect the polarization for $K^{0}$ to be of the same order or smaller than the one measured for the $\Lambda$ at LHC [41], i.e., vanishing small [$O(0.01\%)$] rather than order 1. The large effect observed for the $\rho_{00}$ in mid-central Pb-Pb collisions at low $p_T$ is therefore puzzling. This result should stimulate further theoretical work in order to study which effects could make such a huge difference between $\Lambda$ and $K^{0}$ polarization. Possible reasons may include the transfer of the quark polarization to the hadrons (baryon vs meson), details of the hadronization mechanism (recombination vs fragmentation), rescattering, regeneration, and possibly the lifetime and mass of the relevant hadron. Moreover, the vector mesons are predominantly directly produced whereas the hyperons have large contributions from resonance decays. In conclusion, for the first time, evidence has been found for a significant spin alignment of vector mesons in heavy-ion collisions. The effect is strongest at low $p_T$ with respect to a vector perpendicular to the reaction plane and for mid-central (10–50%) collisions. These observations are qualitatively consistent with expectations from the effect of large initial angular momentum in noncentral heavy-ion collisions, which leads to quark polarization via spin-orbit coupling, subsequently transferred to hadronic degrees of freedom by hadronization via recombination. However, the measured spin alignment is surprisingly large compared to the polarization measured for $\Lambda$ hyperons where, in addition, a strong decrease in polarization with $\sqrt{s_{NN}}$ is observed. In future measurements, the difference in the polarization of $K^{\ast\pm}$ and $K^{\ast 0}$, due to their difference in magnetic moment, would be directly sensitive to the effect of the large initial magnetic field produced in heavy-ion collisions. The ALICE Collaboration would like to thank all its engineers and technicians for their invaluable contributions to the construction of the experiment and the CERN accelerator teams for the outstanding performance of the LHC complex. The ALICE Collaboration gratefully acknowledges the resources and support provided by all Grid centres and the Worldwide LHC Computing Grid (WLCG) collaboration. The ALICE Collaboration acknowledges the following funding agencies for their support in building and running the ALICE detector: A. I. Alikhanyan National Science Laboratory (Yerevan Physics Institute) Foundation (ANSL), State Committee of Science and World Federation of Scientists (WFS), Armenia; Austrian Academy of Sciences, Austrian Science Fund (FWF); [M 2467-N36] and Nationalstiftung für Forschung, Technologie und Entwicklung, Austria; Ministry of Communications and High Technologies, National Nuclear Research Center, Azerbaijan; Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Financiadora de Estudos e Projetos (Finep), Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) and Universidade Federal do Rio Grande do Sul (UFRGS), Brazil; Ministry of Education of China (MOEC), Ministry of Science & Technology of China (MSTC) and National Natural Science Foundation of China (NSFC), China; Ministry of Science and Education and Croatian Science Foundation, Croatia; Centro de Aplicaciones Tecnológicas y Desarrollo Nuclear (CEADEN), Cubaenergía, Cuba; Ministry of Education, Youth and Sports of the Czech Republic, Czech Republic; The Danish Council for Independent Research \| Natural Sciences, the VILLUM FONDEN and Danish National Research Foundation (DNRF), Denmark; Helsinki Institute of Physics (HIP), Finland; Commissariat à l’Energie Atomique (CEA), Institut National de Physique Nucléaire et de Physique des Particules (IN2P3) and Centre National de la Recherche Scientifique (CNRS) and Région des Pays de la Loire, France; Bundesministerium für Bildung und Forschung (BMBF) and GSI Helmholtzzentrum für Schwerionenforschung GmbH, Germany; General Secretariat for Research and Technology, Ministry of Education, Research and Religions, Greece; National Research, Development and Innovation Office, Hungary; Department of Atomic Energy Government of India (DAE), Department of Science and Technology, Government of India (DST), University Grants Commission, Government of India (UGC) and Council of Scientific and Industrial Research (CSIR), India; Indonesian Institute of Science, Indonesia; Centro Fermi—Museo Storico della Fisica e Centro Studi e Ricerche Enrico Fermi and Istituto Nazionale di Fisica Nucleare (INFN), Italy; Institute for Innovative Science and Technology, Nagasaki Institute of Applied Science (IIIST), Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) and Japan Society for the Promotion of Science (JSPS) KAKENHI, Japan; Consejo Nacional de Ciencia (CONACYT) y Tecnología, through Fondo de Cooperación Internacional en Ciencia y Tecnología (FONCICYT) and Dirección General de Asuntos del Personal Académico (DGAPA), Mexico; Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), Netherlands; The Research Council of Norway, Norway; Commission on Science and Technology for Sustainable Development in the South (COMSATS), Pakistan; Pontificia Universidad Católica del Perú, Peru; Ministry of Science and Higher Education and National Science Centre, Poland; Korea Institute of Science and Technology Information and National Research Foundation of Korea (NRF), Republic of Korea; Ministry of Education and Scientific Research, Institute of Atomic Physics and Ministry of Research and Innovation and Institute of Atomic Physics, Romania; Joint Institute for Nuclear Research (JINR), Ministry of Education and Science of the Russian Federation, National Research Centre Kurchatov Institute, Russian Science Foundation and Russian Foundation for Basic Research, Russia; Ministry of Education, Science, Research and Sport of the Slovak Republic, Slovakia; National Research Foundation of South Africa, South Africa; Swedish Research Council (VR) and Knut & Alice Wallenberg Foundation (KAW), Sweden; European Organization for Nuclear Research, Switzerland; Suranaree University of Technology (SUT), National Science and Technology Development Agency (NSDTA) and Office of the Higher Education Commission under NRU project of Thailand, Thailand; Turkish Atomic Energy Agency (TAEK), Turkey; National Academy of Sciences of Ukraine, Ukraine; Science and Technology Facilities Council (STFC), United Kingdom; National Science Foundation of the United States of America (NSF) and United States Department of Energy, Office of Nuclear Physics (DOE NP), United States of America. [1] J. Adams et al. (STAR Collaboration), Experimental and theoretical challenges in the search for the quark gluon plasma: The STAR Collaboration’s critical assessment of the evidence from RHIC collisions, Nucl. Phys. A757, 102 (2005). [2] M. Gyulassy and L. McLerran, New forms of QCD matter discovered at RHIC, Nucl. Phys. A750, 30 (2005). [3] A. Andronic, P. Braun-Munzinger, K. Redlich, and J. Stachel, Decoding the phase structure of QCD via particle production at high energy, Nature (London) 561, 321 (2018). [4] F. Becattini, F. Piccinini, and J. Rizzo, Angular momentum conservation in heavy ion collisions at very high energy, Phys. Rev. C 77, 024906 (2008). [5] D. E. Kharzeev, L. D. McLerran, and H. J. Warringa, The effects of topological charge change in heavy ion collisions: ‘Event by event P and CP violation’, Nucl. Phys. A803, 227 (2008). [6] R. J. Fries, G. Chen, and S. Somanathan, Initial Angular Momentum and Flow in High Energy Nuclear Collisions, Phys. Rev. C 97, 034903 (2018). [7] V. Voronyuk, V. D. Toneev, W. Cassing, E. L. Bratkovskaya, V. P. Konchakovski, and S. A. Voloshin, Electromagnetic field evolution in relativistic heavy-ion collisions, Phys. Rev. C 83, 054911 (2011). [8] J. D. Jackson, Classical Electrodynamics, 3rd ed. (Wiley, New York, 1982), Secs. 11–8 and 11–11. [9] V. B. Berestetskii, E. M. Lifshitz, and L. P. Pitaevskii, Quantum Electrodynamics, Volume 4 of Course of Theoretical Physics, 2nd ed. (Butterworth-Heinemann, Oxford, 1982). [10] M. G. Mayer, On closed shells in nuclei. II, Phys. Rev. 75, 1969 (1949). [11] S. A. Voloshin, Polarized secondary particles in unpolarized high energy hadron-hadron collisions?, arXiv:nucl-th/0410089. [12] Z.-T. Liang and X.-N. Wang, Globally Polarized Quark-Gluon Plasma in Non-Central A + A Collisions, Phys. Rev. Lett. 94, 102301 (2005); Erratum, Phys. Rev. Lett. 96, 039901 (2006). [13] Z.-T. Liang and X.-N. Wang, Spin alignment of vector mesons in non-central A + A collisions, Phys. Lett. B 629, 20 (2005). [14] Z.-T. Liang, Global polarization of QGP in non-central heavy ion collisions at high energies, J. Phys. G 34, S323 (2007). [15] Y.-G. Yang, R.-H. Fang, Q. Wang, and X.-N. Wang, Quark coalescence model for polarized vector mesons and baryons, Phys. Rev. C 97, 034917 (2018). [16] K. Schilling, P. Seyboth, and G. E. Wolf, On the analysis of vector meson production by polarized photons, Nucl. Phys. B15, 397 (1970); Erratum, Nucl. Phys. B18, 332 (1970). [17] See the Supplemental Material at http://link.aps.org/supplemental/10.1103/PhysRevLett.125.012301 for complete angular distribution, consistency checks, analytical relation between EP and PP, toy model simulation to show the relation between EP, PP and RNDEP, and centrality classes. [18] A. M. Poskanzer and S. A. Voloshin, Methods for analyzing anisotropic flow in relativistic nuclear collisions, Phys. Rev. C 58, 1671 (1998). [19] A. H. Tang, B. Tu, and C. S. Zhou, Practical considerations for measuring global spin alignment of vector mesons in relativistic heavy ion collisions, Phys. Rev. C 98, 044907 (2018). [20] B. Abelev et al. (ALICE Collaboration), Centrality determination of Pb–Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV with ALICE, Phys. Rev. C 88, 044909 (2013). [21] K. Aamodt et al., The ALICE experiment at the CERN LHC, J. Instrum. 3, S08002 (2008). [22] K. Aamodt et al. (ALICE Collaboration), Centrality Dependence of the Charged-Particle Multiplicity Density at Mid-Rapidity in Pb–Pb Collisions at $\sqrt{s_{NN}} = 2.76$ TeV, Phys. Rev. Lett. 106, 032301 (2011). [23] B. B. Abelev et al. (ALICE Collaboration), Elliptic flow of identified hadrons in Pb–Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV, J. High Energy Phys. 06 (2015) 190. [24] J. Alme et al., The ALICE TPC, a large 3-dimensional tracking device with fast readout for ultra-high multiplicity events, Nucl. Instrum. Methods Phys. Res., Sect. A 622, 316 (2010). [25] G. Dellacasa et al. (ALICE Collaboration), ALICE technical design report of the time-of-flight system (TOF), Report No. CERN-LHCC-2000-012 (2000). [26] J. Adam et al. (ALICE Collaboration), $K^(892)^0$ and $\phi(1020)$ meson production at high transverse momentum in $pp$ and Pb–Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV, Phys. Rev. C* 95, 064606 (2017). [27] B. B. Abelev et al. (ALICE Collaboration), $K^(892)^0$ and $\phi(1020)$ production in Pb–Pb collisions at $\sqrt{s_{NN}} = 2.76$ TeV, Phys. Rev. C* 91, 024609 (2015). [28] B. B. Abelev et al. (ALICE Collaboration), $K^0_S$ and $\Lambda$ Production in Pb–Pb Collisions at $\sqrt{s_{NN}} = 2.76$ TeV, Phys. Rev. Lett. 111, 222301 (2013). [29] B. B. Abelev et al. (ALICE Collaboration), Performance of the ALICE Experiment at the CERN LHC, Int. J. Mod. Phys. A 29, 1430044 (2014). [30] T. Sjöstrand, S. Ask, J. R. Christiansen, R. Corke, N. Desai, P. Ilten, S. Mrenna, S. Prestel, C. O. Rasmussen, and P. Z. Skands, An Introduction to PYTHIA 8.2, Comput. Phys. Commun. 191, 159 (2015). [31] K. Ackerstaff et al. (OPAL Collaboration), Spin alignment of leading $K^{0}(892)$ mesons in hadronic $Z^0$ decays, Phys. Lett. B* 412, 210 (1997). [32] K. Ackerstaff et al. (OPAL Collaboration), Study of $\phi(1020)$, $D^{\pm\pm}$ and $B^$ spin alignment in hadronic $Z^0$ decays, Z. Phys. C* 74, 437 (1997). [33] P. Abreu et al. (DELPHI Collaboration), Measurement of the spin density matrix for the $\rho^0$, $K^{0}(892)$ and $\phi$ produced in $Z^0$ decays, Phys. Lett. B* 406, 271 (1997). [34] M. Barth et al. (Brussels-Genoa-Mons-Nijmegen-Serpukhov-CERN Collaborations), Inclusive resonance production in $K^+ p$ interactions at 70-GeV/c, Nucl. Phys. B223, 296 (1983); Erratum, Nucl. Phys. B232, 547 (1984). [35] A. N. Aleev et al. (EXCHARM Collaboration), Spin alignment of $K^{\pm}(892)$ mesons produced in neutron carbon interactions, Phys. Lett. B* 485, 334 (2000). [36] C. Zhou, $\phi$ meson and $K^{0}$ global spin alignment at STAR, Nucl. Phys.* A982, 559 (2019). [37] B. I. Abelev et al. (STAR Collaboration), Spin alignment measurements of the $K^{0}(892)$ and $\phi(1020)$ vector mesons in heavy ion collisions at $\sqrt{s_{NN}} = 200$ GeV, Phys. Rev. C* 77, 061902 (2008). [38] X.-L. Sheng, L. Oliva, and Q. Wang, What can we learn from global spin alignment of $\phi$ meson in heavy-ion collisions?, Phys. Rev. D 101, 096005 (2020). [39] L. Adamczyk et al. (STAR Collaboration), Global Λ hyperon polarization in nuclear collisions: Evidence for the most vortical fluid, Nature (London) 548, 62 (2017). [40] J. Adam et al. (STAR Collaboration), Global polarization of Λ hyperons in Au +Au collisions at $\sqrt{s_{NN}} = 200$ GeV, Phys. Rev. C 98, 014910 (2018). [41] S. Acharya et al. (ALICE Collaboration), Global polarization of Λ Λ hyperons in Pb-Pb collisions at $\sqrt{s_{NN}} = 2.76$ and 5.02 TeV, Phys. Rev. C 101, 044611 (2020). [42] F. Becattini, I. Karpenko, M. Lisa, I. Upsal, and S. Voloshin, Global hyperon polarization at local thermodynamic equilibrium with vorticity, magnetic field and feed-down, Phys. Rev. C 95, 054902 (2017). L. Görlich, S. Gotovac, V. Grabski, L. K. Graczykowski, K. L. Graham, L. Greiner, A. Grelli, C. Grigoras, V. Grigoriev, A. Grigoryan, S. Grigoryan, O. S. Groettvik, F. Grosa, J. F. Grosse-Oetringhaus, R. Grosso, R. Guernane, M. Guittiere, K. Gulbrandsen, T. Gunji, A. Gupta, R. Gupta, I. B. Guzman, R. Haake, M. K. Habib, C. Hadjidakis, H. Hamagaki, G. Hamar, M. Hamid, R. Hannigan, M. R. Haque, A. Harlenderova, J. W. Harris, A. Harton, J. A. Hasenbichler, H. Hassan, D. Hatzifotiadou, P. Hauer, S. Hayashi, S. T. Heckel, E. Hellbär, H. Helstrup, A. Hergelegiu, T. Herman, E. G. Hernandez, G. Herrera Corral, F. Herrmann, K. F. Hetland, T. E. Hilden, H. Hillemanns, C. Hills, B. Hippolyte, B. Hohlweiger, D. Horak, A. Hornung, S. Hornung, R. Hosokawa, P. Hristov, C. Huang, C. Hughes, P. Huhn, T. J. Humanic, H. Hushnud, L. A. Husova, N. Hussain, S. A. Hussain, D. Hutter, J. P. Iddon, R. Ilkaev, M. Inaba, G. M. Innocenti, M. Ippolito, A. Isakov, M. S. Islam, M. Ivanov, V. Ivanov, V. Izucheev, B. Jacak, N. Jacazio, P. M. Jacobs, S. Jadlovská, J. Jadlovsy, S. Jaelani, C. Jahne, M. J. Jakubowska, M. A. Janik, T. Janson, M. Jeric, O. Jevons, M. Jin, P. G. Jones, Jung, M. Jung, A. Jusko, P. Kalinak, A. Kalweit, V. Kaplin, S. Kar, A. Karasu Uysal, O. Karavichev, T. Karavicheva, P. Karczmarczyk, E. Karpechev, A. Kazantsev, U. Kebischull, Re. Keidel, M. Keil, C. Ketzer, Z. Khabanova, A. M. Khan, S. Khan, A. Khaznadeev, Y. Kharlov, A. Khatun, A. Khuntia, B. Kileng, B. Kim, B. Kim, D. Kim, D. J. Kim, E. J. Kim, H. Kim, J. Kim, J. S. Kim, J. Kim, J. Kim, J. Kim, J. Kim, M. Kim, S. Kim, T. Kim, T. Kim, S. Kirsch, I. Kisel, S. Kiselev, A. Kisiel, J. L. Klay, C. Klein, J. Klein, S. Klein, C. Klein-Bösing, M. Kleiner, A. Kluge, M. L. Knichel, A. G. Knospe, C. Kobdaj, M. K. Köhler, T. Kollegger, A. Kondratyev, N. Kondratyeva, E. Kondratyuk, J. Konig, P. J. Konopka, L. Koska, O. Kovalenko, V. Kovalenko, M. Kowalski, I. Králík, A. Kravčáková, L. Kreis, M. Krivda, F. Krizek, K. Krizkova Gajdosova, M. Krüger, E. Kryshen, M. Krzewicki, A. M. Kubera, V. Kučera, C. Kuhn, P. G. Kuijer, L. Kumar, S. Kumar, S. Kundu, P. Kurashvili, A. Kurepin, A. B. Kurepin, A. Kuryakin, S. Kushpi, J. Kvapil, M. J. Kwon, Y. Kwon, S. L. La Pointe, P. La Rocca, Y. S. Lai, R. Langoy, K. Lapidos, A. Lardeux, P. Larionov, E. Laudi, R. Lavicka, Z. Lazareva, R. Lea, L. Leardini, J. Lee, S. Lee, F. Lehas, S. Lehner, J. Lehrbach, R. C. Lemmon, I. León Monzón, E. D. Lesser, M. Lettrich, P. Léval, X. Li, X. L. Li, J. Lien, R. Lietava, B. Lim, V. Lindenstruth, S. W. Lindsay, C. Lippmann, M. A. Lisa, V. Litichevskyi, A. Liu, S. Liu, W. J. Llope, I. M. Lofnes, V. Loginov, C. Loizides, P. Loncar, X. Lopez, E. López Torres, J. R. Luhder, M. Lunardon, G. Luparello, Y. Ma, A. Maevskaya, M. Mager, S. M. Mahmood, T. Mahmoud, A. Maire, R. D. Majka, M. Malaev, Q. W. Malik, L. Malinina, D. Mal’Kevich, P. Malzacher, G. Mandaglio, V. Manko, F. Manso, V. Manzari, Y. Mao, M. Marchisone, J. Mareš, G. V. Margagliotti, A. Margotti, J. Margutti, A. Marín, C. Markert, M. Marquard, N. A. Martin, P. Martinengo, J. L. Martinez, M. I. Martínez, G. Martínez García, M. Martinez Pedreira, S. Masciocchi, M. Masera, A. Masoni, L. Massacrier, E. Masson, A. Mastroserio, A. M. Mathis, O. Matonoha, P. F. T. Matuoka, A. Matyja, C. Mayer, M. Mazzilli, M. A. Mazzoni, A. F. Mechler, F. Meddi, Y. Melikyan, A. Menchaca-Rocha, C. Mengke, E. Meninno, M. Meres, S. Mhlanga, Y. Miake, L. Micheletti, D. L. Mihaylov, K. Mikhaylov, A. Mischke, A. N. Mishra, D. Miśkowiec, A. Modak, N. Mohammadi, A. P. Mohanty, B. Mohanty, M. Mohisin Khan, C. Mordasini, D. A. Moreira De Godoy, L. A. P. Moreno, I. Morozov, A. Morsch, T. Mrnjavac, V. Muccifora, E. Mudnic, D. Mühlheim, S. Muhuri, J. D. Mulligan, M. G. Munhoz, R. H. Munzer, H. Murakami, S. Murray, L. Musa, C. J. Myers, J. W. Myrcha, B. Naik, R. Nair, B. K. Nandi, R. Nania, E. Nappi, M. U. Naru, A. F. Nassirpour, C. Nattrass, R. Nayak, T. K. Nayak, S. Nazarenko, A. Neagu, R. A. Negrao De Oliveira, L. Nellen, S. V. Nesbo, G. Neskovic, D. Nesterov, L. T. Neumann, B. S. Nielsen, S. Nikolaev, S. Nikulin, V. Nikulin, F. Noferini, P. Nomokonov, J. Norman, N. Novitzky, P. Nowakowski, A. Nyanin, J. Nystrand, M. Ogino, A. Ohlson, J. Oleniacz, A. C. Oliveira Da Silva, M. H. Oliver, C. Oppedisano, R. Orava, A. Ortiz Velasquez, A. Oskarsson, J. Otwinowski, K. Oyama, Y. Pachmayer, V. Pacik, D. Pagano, G. Paić, J. Pan, A. K. Pandey, S. Panebianco, P. Pareek, J. Park, J. E. Parkkila, S. Parmar, S. P. Pathak, R. N. Patra, B. Paul, H. Pei, T. Peitzmann, X. Peng, L. G. Pereira, H. Pereira Da Costa, D. Peresunko, G. M. Perez, E. Perez Lezama, V. Peskov, Y. Pestov, V. Petráček, M. Petrovici, R. P. Pezzci, S. Piano, M. Pikna, P. Pillot, O. Pinazza, L. Pinsky, C. Pinto, S. Pisano, D. Pistone, M. Ploskon, M. Planinic, F. Plquett, J. Pluta, S. Pochybova, M. G. Poghosyan, B. Polichtchouk, N. Poljak, A. Pop, H. Poppenborg, S. Porteboeuf-Houssais, V. Pozdniakov, S. K. Prasad, R. Preghenella, F. Prino, C. A. Pruneau, I. Pshenichnov, M. Puccio, J. Putschke, R. E. Quishpe, S. Ragoni, S. Raha, S. Rajput, J. Rak, A. Rakotozafindrabe, L. Ramello, F. Rami, R. Raniwala, S. Raniwala, S. S. Räsänen, R. Rath, V. Ratza, I. Ravasenga, K. F. Read, K. Redlich, A. Rehman, P. Reichelt, F. Reidt, X. Ren, R. Renfordt, Z. Rescakova, J.-P. Revol, K. Reygers, V. Riabov, T. Richert, M. Richter, P. Riedler, W. Riegler, C. Ristea, S. P. Rode, M. Rodriguez Cahuantzi, K. Røed, R. Rogalev, E. Rogochaya, D. Rohr, D. Röhrich, P. S. Rokita, F. Ronchetti, E. D. Rosas, K. Roslon, A. Rossi, A. Rotondi, A. Roy, P. Roy, O. V. Rueda, R. Rui, B. Rumyantsev, A. Rustamov, E. Ryabinkin, Y. Ryabov, A. Rybicki, H. Rytkonen, O. A. M. Saarimaki, S. Sadhu, S. Sadovsky, K. Šafařík, S. K. Saha, B. Sahoo, P. Sahoo, R. Sahoo, S. Sahoo, P. K. Sahu, J. Saini, S. Sakai, S. Sambyal, V. Samsonov, D. Sarkar, N. Sarkar, P. Sarma, V. M. Sarti, M. H. P. Sas, E. Scapparone, B. Schaefer, J. Schambach, H. S. Scheid, C. Schiaua, R. Schicker, A. Schmah, C. Schmidt, H. R. Schmidt, M. O. Schmidt, M. Schmidt, N. V. Schmidt, A. R. Schmier, J. Schukraft, Y. Schutz, K. Schwarz, K. Schweda, G. Scioli, E. Scomparin, M. Šefčík, J. E. Seger, Y. Sekiguchi, D. Sekihata, I. Selyuzhenkov, S. Senyukov, D. Serebryakov, E. Serradilla, A. Sevcenco, A. Shabanov, A. Shabetai, R. Shahoyan, W. Shaikh, A. Shangaraev, A. Sharma, H. Sharma, M. Sharma, N. Sharma, A. I. Sheikh, K. Shigaki, M. Shimomura, S. Shirinkin, Q. Shou, Y. Sibiriak, S. Siddhanta, T. Siemiarczuk, G. Simatovic, G. Simonetti, R. Singh, R. Singh, R. Singh, V. K. Singh, V. Singhal, T. Sinha, M. Sitta, T. B. Skaali, M. Slupecki, N. Smirnov, R. J. M. Snellings, T. W. Snellman, C. Soncco, J. Song, A. Songmoolnak, F. Soramel, S. Sorensen, I. Sputowska, J. Stachel, I. Stan, P. Stankus, P. J. Steffanic, E. Stenlund, D. Stocco, M. M. Storteveld, L. D. Stritto, A. A. P. Suaide, T. Sugitate, M. Suleymannov, M. Suljic, R. Sultanov, M. Šumbera, S. Sumowidagdo, S. Swain, A. Szabo, I. Szarka, U. Tabassam, G. Tailepleid, J. Takahashi, G. J. Tambave, S. Tang, M. Tarhini, M. G. Tarzila, A. Tauro, G. Tejeda Muñoz, A. Telesca, C. Terrevoli, D. Thakur, S. Thakur, D. Thomas, F. Thoresen, R. Tieulent, A. Tikhonov, A. R. Timmins, A. Toia, N. Topilskaya, M. Toppi, F. Torales-Acosta, S. R. Torres, S. Tripathy, T. Tripathy, S. Trogolo, G. Trombetta, L. Tropp, V. Trubnikov, W. H. Trzaska, T. P. Trzcinski, B. A. Trzeciak, T. Tsuji, A. Tunkin, R. Turrisi, T. S. Tytler, K. Ullaland, E. N. Umaka, A. Uras, G. L. Usai, A. Utrobicic, M. Vala, N. Valle, S. Vallero, N. van der Kolk, L. V. R. van Doremalen, M. van Leeuwen, P. Vande Vyvre, D. Varga, Z. Varga, M. Varga-Kofarago, A. Vargas, M. Vasileiou, A. Vasiliev, O. Vázquez Doce, V. Veccherini, A. M. Veen, E. Vercellin, S. Vergara Limón, L. Vermunt, R. Vernet, R. Vértesi, L. Vickovic, Z. Vilakazi, O. Villalobos Baillie, A. Villatoro Tello, G. Vino, A. Vinogradov, T. Virgili, V. Vislavicius, A. Vodopyanov, B. Volkel, M. A. Völk, K. Voloshin, S. A. Voloshin, G. Volpe, B. von Haller, I. Vorobyev, D. Voscek, J. Vrláková, B. Wagner, M. Weber, S. G. Weber, A. Wegrzynek, D. F. Weiser, S. C. Wenzel, J. P. Wessels, J. Wiechula, J. Wikne, J. Wilkinson, G. A. Willems, E. Willsher, B. Windelband, M. Winn, W. E. Witt, Y. Wu, R. Xu, S. Yalcin, K. Yamakawa, S. Yang, S. Yano, Z. Yin, H. Yokoyama, I.-K. Yoo, J. H. Yoon, S. Yuan, A. Yuncu, V. Yurchenko, V. Zaccolo, A. Zaman, C. Zampolli, H. J. C. Zanolli, N. Zardoshti, A. Zarochentsev, P. Závada, N. Zaviyalov, H. Zbroszczyk, M. Zhalov, S. Zhang, X. Zhang, Z. Zhang, V. Zherebchevskii, D. Zhou, Y. Zhou, Z. Zhou, J. Zhu, Y. Zhu, A. Zichichi, M. B. Zimmermann, G. Zinovjev, and N. Zurlo (The ALICE Collaboration) 1 A.I. Alikhanyan National Science Laboratory (Yerevan Physics Institute) Foundation, Yerevan, Armenia 2 Bogolyubov Institute for Theoretical Physics, National Academy of Sciences of Ukraine, Kiev, Ukraine 3 Bose Institute, Department of Physics and Centre for Astroparticle Physics and Space Science (CAPSS), Kolkata, India 4 Budker Institute for Nuclear Physics, Novosibirsk, Russia 5 California Polytechnic State University, San Luis Obispo, California, United States 6 Central China Normal University, Wuhan, China 7 Centre de Calcul de l'IN2P3, Villeurbanne, Lyon, France 8 Centro de Aplicaciones Tecnológicas y Desarrollo Nuclear (CEADEN), Havana, Cuba 9 Centro de Investigación y de Estudios Avanzados (CINVESTAV), Mexico City and Mérida, Mexico 10 Centro Fermi—Museo Storico della Fisica e Centro Studi e Ricerche “Enrico Fermi,” Rome, Italy 11 Chicago State University, Chicago, Illinois, United States 12 China Institute of Atomic Energy, Beijing, China 13 Comenius University Bratislava, Faculty of Mathematics, Physics and Informatics, Bratislava, Slovakia 14 COMSATS University Islamabad, Islamabad, Pakistan 15 Creighton University, Omaha, Nebraska, United States 16 Department of Physics, Aligarh Muslim University, Aligarh, India 17 Department of Physics, Pusan National University, Pusan, Republic of Korea 18 Department of Physics, Sejong University, Seoul, Republic of Korea 19 Department of Physics, University of California, Berkeley, California, United States 20 Department of Physics, University of Oslo, Oslo, Norway 21 Department of Physics and Technology, University of Bergen, Bergen, Norway 22 Dipartimento di Fisica dell’Università “La Sapienza” and Sezione INFN, Rome, Italy 23 Dipartimento di Fisica dell’Università and Sezione INFN, Cagliari, Italy 24 Dipartimento di Fisica dell’Università and Sezione INFN, Trieste, Italy 25 Dipartimento di Fisica dell’Università and Sezione INFN, Turin, Italy 26 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Bologna, Italy 27 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Catania, Italy 28 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Padova, Italy 29 Dipartimento di Fisica “E.R. Caianiello” dell’Università and Gruppo Collegato INFN, Salerno, Italy 30 Dipartimento DISAT del Politecnico and Sezione INFN, Turin, Italy 31 Dipartimento di Scienze e Innovazione Tecnologica dell’Università del Piemonte Orientale and INFN Sezione di Torino, Alessandria, Italy 32 Dipartimento Interateneo di Fisica “M. Merlin” and Sezione INFN, Bari, Italy 33 European Organization for Nuclear Research (CERN), Geneva, Switzerland 34 Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia 35 Faculty of Engineering and Science, Western Norway University of Applied Sciences, Bergen, Norway 36 Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague, Czech Republic 37 Faculty of Science, P.J. Šafárik University, Košice, Slovakia 38 Frankfurt Institute for Advanced Studies, Johann Wolfgang Goethe-Universität Frankfurt, Frankfurt, Germany 39 Fudan University, Shanghai, China 40 Gangneung-Wonju National University, Gangneung, Republic of Korea 41 Gauhati University, Department of Physics, Guwahati, India 42 Helmholtz-Institut für Strahlen- und Kernphysik, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany 43 Helsinki Institute of Physics (HIP), Helsinki, Finland 44 High Energy Physics Group, Universidad Autónoma de Puebla, Puebla, Mexico 45 Hiroshima University, Hiroshima, Japan 46 Hochschule Worms, Zentrum für Technologie transfer und Telekommunikation (ZTT), Worms, Germany 47 Horia Hulubei National Institute of Physics and Nuclear Engineering, Bucharest, Romania 48 Indian Institute of Technology Bombay (IIT), Mumbai, India 49 Indian Institute of Technology Indore, Indore, India 50 Indonesian Institute of Sciences, Jakarta, Indonesia 51 INFN, Laboratori Nazionali di Frascati, Frascati, Italy 52 INFN, Sezione di Bari, Bari, Italy 53 INFN, Sezione di Bologna, Bologna, Italy 54 INFN, Sezione di Cagliari, Cagliari, Italy 55 INFN, Sezione di Catania, Catania, Italy 56 INFN, Sezione di Padova, Padova, Italy 57 INFN, Sezione di Roma, Rome, Italy 58 INFN, Sezione di Torino, Turin, Italy 59 INFN, Sezione di Trieste, Trieste, Italy 60 Inha University, Incheon, Republic of Korea 61 Institut de Physique Nucléaire d’Orsay (IPNO), Institut National de Physique Nucléaire et de Physique des Particules (IN2P3/CNRS), Université de Paris-Sud, Université Paris-Saclay, Orsay, France 62 Institute for Nuclear Research, Academy of Sciences, Moscow, Russia 63 Institute for Subatomic Physics, Utrecht University/Nikhef, Utrecht, Netherlands 64 Institute of Experimental Physics, Slovak Academy of Sciences, Košice, Slovakia 65 Institute of Physics, Homi Bhabha National Institute, Bhubaneswar, India 66 Institute of Physics of the Czech Academy of Sciences, Prague, Czech Republic 67 Institute of Space Science (ISS), Bucharest, Romania 68 Institut für Kernphysik, Johann Wolfgang Goethe-Universität Frankfurt, Frankfurt, Germany 69 Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, Mexico City, Mexico 70 Instituto de Física, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, Brazil 71 Instituto de Física, Universidad Nacional Autónoma de México, Mexico City, Mexico 72 iThemba LABS, National Research Foundation, Somerset West, South Africa 73 Jeonbuk National University, Jeonju, Republic of Korea 74 Johann-Wolfgang-Goethe Universität Frankfurt Institut für Informatik, Fachbereich Informatik und Mathematik, Frankfurt, Germany 75 Joint Institute for Nuclear Research (JINR), Dubna, Russia 76 Korea Institute of Science and Technology Information, Daejeon, Republic of Korea 77 KTO Karatay University, Konya, Turkey 78 Laboratoire de Physique Subatomique et de Cosmologie, Université Grenoble-Alpes, CNRS-IN2P3, Grenoble, France 79 Lawrence Berkeley National Laboratory, Berkeley, California, United States 80 Lund University Department of Physics, Division of Particle Physics, Lund, Sweden 81 Nagasaki Institute of Applied Science, Nagasaki, Japan 82 Nara Women’s University (NWU), Nara, Japan 83 National and Kapodistrian University of Athens, School of Science, Department of Physics, Athens, Greece 84 National Centre for Nuclear Research, Warsaw, Poland 85 National Institute of Science Education and Research, Homi Bhabha National Institute, Jatni, India 86 National Nuclear Research Center, Baku, Azerbaijan 87 National Research Centre Kurchatov Institute, Moscow, Russia 88 Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark 89 Nikhef, National institute for subatomic physics, Amsterdam, Netherlands 90 NRC Kurchatov Institute IHEP, Protvino, Russia 91 NRC «Kurchatov Institute»—ITEP, Moscow, Russia 92 NRNU Moscow Engineering Physics Institute, Moscow, Russia 93 Nuclear Physics Group, STFC Daresbury Laboratory, Daresbury, United Kingdom 94 Nuclear Physics Institute of the Czech Academy of Sciences, Řež u Prahy, Czech Republic 95 Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States 96 Ohio State University, Columbus, Ohio, United States 97 Petersburg Nuclear Physics Institute, Gatchina, Russia 98 Physics department, Faculty of science, University of Zagreb, Zagreb, Croatia 99 Physics Department, Panjab University, Chandigarh, India 100 Physics Department, University of Jammu, Jammu, India 101 Physics Department, University of Rajasthan, Jaipur, India 102 Physikalisches Institut, Eberhard-Karls-Universität Tübingen, Tübingen, Germany 103 Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 104 Physik Department, Technische Universität München, Munich, Germany 105 Politecnico di Bari, Bari, Italy 106 Research Division and ExtreMe Matter Institute EMMI, GSI Helmholtzzentrum für Schwerionenforschung GmbH, Darmstadt, Germany 107 Rudjer Bošković Institute, Zagreb, Croatia 108 Russian Federal Nuclear Center (VNIIEF), Sarov, Russia 109 Saha Institute of Nuclear Physics, Homi Bhabha National Institute, Kolkata, India 110 School of Physics and Astronomy, University of Birmingham, Birmingham, United Kingdom 111 Sección Física, Departamento de Ciencias, Pontificia Universidad Católica del Perú, Lima, Peru 112 St. Petersburg State University, St. Petersburg, Russia 113 Stefan Meyer Institut für Subatomare Physik (SMI), Vienna, Austria 114 SUBATECH, IMT Atlantique, Université de Nantes, CNRS-IN2P3, Nantes, France 115 Suranaree University of Technology, Nakhon Ratchasima, Thailand 116 Technical University of Košice, Košice, Slovakia 117 Technische Universität München, Excellence Cluster ‘Universe’, Munich, Germany 118 The Henryk Niewodniczanski Institute of Nuclear Physics, Polish Academy of Sciences, Cracow, Poland 119 The University of Texas at Austin, Austin, Texas, United States 120 Universidad Autónoma de Sinaloa, Culiacán, Mexico 121 Universidade de São Paulo (USP), São Paulo, Brazil 122 Universidade Estadual de Campinas (UNICAMP), Campinas, Brazil 123 Universidade Federal do ABC, Santo André, Brazil 124 University of Cape Town, Cape Town, South Africa 125 University of Houston, Houston, Texas, United States 126 University of Jyväskylä, Jyväskylä, Finland 127 University of Liverpool, Liverpool, United Kingdom 128 University of Science and Technology of China, Hefei, China 129 University of South-Eastern Norway, Tønsberg, Norway 130 University of Tennessee, Knoxville, Tennessee, United States 131 University of the Witwatersrand, Johannesburg, South Africa 132 University of Tokyo, Tokyo, Japan 133 University of Tsukuba, Tsukuba, Japan 134 Université Clermont Auvergne, CNRS/IN2P3, LPC, Clermont-Ferrand, France 135 Université de Lyon, Université Lyon 1, CNRS/IN2P3, IPN-Lyon, Villeurbanne, Lyon, France 136 Université de Strasbourg, CNRS, IPHC UMR 7178, F-67000 Strasbourg, France, Strasbourg, France 137 Université Paris-Saclay Centre d’Etudes de Saclay (CEA), IRFU, Département de Physique Nucléaire (DPhN), Saclay, France 138 Università degli Studi di Foggia, Foggia, Italy 139 Università degli Studi di Pavia, Pavia, Italy 140 Università di Brescia, Brescia, Italy 141 Variable Energy Cyclotron Centre, Homi Bhabha National Institute, Kolkata, India 142 Warsaw University of Technology, Warsaw, Poland 143 Wayne State University, Detroit, Michigan, United States 144 Westfälische Wilhelms-Universität Münster, Institut für Kernphysik, Münster, Germany 145 Wigner Research Centre for Physics, Budapest, Hungary 146 Yale University, New Haven, Connecticut, United States 147 Yonsei University, Seoul, Republic of Korea a Deceased. b Dipartimento DET del Politecnico di Torino, Turin, Italy. c M.V. Lomonosov Moscow State University, D.V. Skobeltsyn Institute of Nuclear Physics, Moscow, Russia. d Department of Applied Physics, Aligarh Muslim University, Aligarh, India. e Institute of Theoretical Physics, University of Wrocław, Poland.
"THE INCREDIBILITY OF A STABLE" TEXT: "God has chosen what the world calls foolish to shame the wise; He has chosen what the world calls weak to shame the strong". (I Corinthians 1: 28) INTRODUCTION Among his various achievements in recent years in behalf of the American consumer, Ralph Nader has been instrumental in getting Truth in Packaging Laws on the books in many states. In size and in printed information, the box and what's inside must correspond. Yesterday's NY Times carried word of his latest attempt to have the Federal Trade Commission ban ads on products that are not backed by scientific tests. Truth in packaging - this is what it's all about. This is his concern. For his efforts, we are grateful. DEVELOPMENT I cannot help but wonder if there is a sense in which the circumstances of the birth of Jesus misrepresented the importance of what was going on? Misrepresented - not by overstatement as is so often the case, but rather by under-statement. A Messiah, a Saviour, a King -- born to the commonest of parents; welcomed by shepherds, but unrecognized by professional divines; and nursed by his mother not in the temple, not in the palace, not in a clean, well-ordered home, but tenderly cared for by his mother in - would you believe, a stable? By faith, we are led to believe that the birth of Jesus was a well-planned happening, that no detail was left to chance, that all of the props were just as God wanted them. Yes - even down to the stable. Suppose we reflect for a few minutes this morning on the appropriateness of the stable, and what it suggests. OUTSIDE ESTABLISHED CHANNELS To begin with, at the very least that stable suggests to us that life's creative forces often flow outside the established channels. The inn in Bethlehem that night housed the conventional, the customary, the ordinary, the usual. But Jesus was born in a cave, in a stable, as an outsider. I wonder if God is trying to say something to us. I was interested and glad to read only the other day in the doctor's office while I was waiting my turn that all progress in the business world is not generated by the giant corporations and the huge conglomerates. I came across these interesting facts in an old copy of a business magazine: 1. Hilton or Statler did not create the motel revolution. It was small, undercapitalized local entrepreneurs who had no previous experience in the lodging business. 2. Neither Ward Baking nor National Biscuit created the frozen cake revolution. It was Sara Lee, a local baker in Chicago. 3. Similarly, it was not Ward or Bond that created the quality revolution in packaged white bread. That was done by two companies absolutely new to the baking business. Pepperidge Farm and Arnold Bakers. 4. It was not RCA or GE that created the mass market for transistors. It was a remote and tiny company in the geophysical business, now called Texas Instruments. 5. Armstrong Cork and Congoleum-Nairn did not bring on the vinyl tile revolution in floor covering and care. It is a little known firm called Delaware Floor products. 6. Metro-Goldwyn Mayer and Paramount did not create the feature TV series which made television an entertainment success. It was entirely new companies like Four Star, Revue and Desilu. 7. General Foods and Campbell Soup did not create the new wave in the convenience foods revolution. It was completely new companies like Swanson. 8. Neither General Motors nor Standard Oil of NJ developed the first mass production efficiency 24-hour automotive repair and maintenance center. It was the small Canadian Tire Company of Toronto. 9. Federated Department Stores and Allied Stores did not develop the mass merchandising, one-floor suburban shopping revolution. It was people never before in the retailing business, like Two Guys from Harrison. 10. Nor finally, in the 1930's was it the A & P or Jewel or Kroger chain that created the supermarket. It was undercapitalized merchandising upstarts like King Kullen on Long Island and Big Bear in Ohio". In the business world some creative forces flow outside the set and established channels. What shall we say of the church? Enamored of the truth she holds and the power she possesses, the church, at times, is apt to become puffed-up with a sense of pride and self-importance as though all truth must be received through her and interpreted by her. Often in the gospels the line of the future seems to be forming outside the establishment. Jesus went to the town of Jericho where there was an accredited school for prophets, but he did not visit the school. Instead, he made his way to the house of Zacchaeus, a tax collector. And on that first Palm Sunday when the faithful were gathered in the temple, the real wave of the future rested in what was going on outside, in the hosannas the people were shouting because of one who was entering their city riding on a donkey. When the church gets soft on justice and its passion for peace, God is likely to speak His word through an aroused militant wing, or through demonstrations and marches in the streets. When we are obsessed by status symbols and all of the popular notions of success and the things that money can buy, God is likely to speak from without through flower children who warm to human values and spurn our dress and titles and styles and nations of success. When we tend to become stuffy and stifled in worship, he is likely to speak through soul music from the outside and through the gut rhythms of rock and roll. At the great insistence of my 12 year old, I spent two hours last Sunday night listening to this new rock opera from England, Jesus Christ, Super-Star. I confess I liked it; it was a real treat and something of an experience. I can see how people can be drawn to it. Sit down and listen to it from beginning to end, with script in hand. It may turn you on. What I am trying to say is that God is not limited to sources that are dear to us. Let us be careful not to over-estimate the inn, nor to under-estimate what's happening in the stable. GOD MEETS US IN THE STABLE Let me go on to the second thought I wish to put before you. More important, the stable reminds us that God meets men in the earthiness of life which that stable is a symbol of. This is, after all, the meaning of the Incarnation - God in the midst of life. Now when you come right down to it, there's just no way that we can pretty-up that stable - no matter how many creches we place on our mantels or tables, no matter how many prints of a clean stable we find on our Christmas cards. Chances are that stable back in Bethlehem was a primitive and pungent place alive with all the sounds and smells and what have you of everyday life. And any one intent on going there would be well advised to take along a can of lysol spray and wear his high-top shoes - for that's the kind of place it is. The stable stands as a testimony and a reminder to the earthiness of our Lord. I think we fail to appreciate this. So often we try to deny the existence of that earthy side of life and of our own nature - the wild, boisterous, elemental, basic, unrefined side. Believe it or not, there is a direction sign over in the Hayden Planetarium that says: "To the Solar System and Rest Rooms". It reminds us that even in our loftiest pursuits we are not delivered from our most basic needs. It reminds me of a comment that was made by two spinsters in our church a number of years ago after announcement was made in the bulletin regarding the adoption of a child by our former associate minister and his wife. As the prelude was being softly played, one turned to the other and said in a whisper that was overheard: "Now, that's the right way for a minister to have a child". Any treasure we humans hold is held in earthen vessels, whether it be a treasure of art, of culture, of religious insight, and yet we sometimes fail to appreciate this and create the impression that this part of our nature is strange, foreign, unmentionable. We would like to deny that Christ can be there on that seamy, underside of life. We have an aversion to the rough and a strong preference for the smooth. We like to sing about "The Old Rugged Cross", but the crosses we know and love have their surfaces planed and highly polished. When we baptize, we do not use the muddy waters of a river, but the clean waters and make use of a beautiful baptismal font. When we have communion, we do not celebrate with chunks of bread torn from a loaf and wine taken from a bottle, but rather we use bread that is neatly cut into cubes and where would we be without Welch's grape juice. The unlettered crudity of the men that Jesus gathered to himself - men like James, John, and rough, tough Peter - offend us. We can't take them until we've embedded them in stained glass or made them over into saints. What I am concerned about here is that we are guilty in our churches of overly identifying Jesus with the "upper" or "proper" side of life. And maybe this is what some of those rebellious voices on the outside (many of whom have been reared in our churches and Sunday Schools) are trying to tell us. Whenever the Christian church or community tries to identify itself only with exaltation it's in danger of becoming false and counterfeit. It may be that our inability to recognize Christ's presence in the underside of life is what makes the church so inept in the city. The words of George McCleod, that great spirit of the church of Scotland in our time, speak a sharp and urgent word in this direction: "I simply argue that the cross be raised again at the center of the market place, as well as on the steeple of the church. I am recovering the claim that Jesus was not crucified in a cathedral between two candles, but on a cross between two thieves; on the town garbage heap - at a cross roads so cosmopolitan that they had to write his title in Hebrew, Latin and Greek. At the kind of place where cynics talk smut and thieves curse, and soldiers gambled because that is where he died and that is what he died about, and that is where more of us should be and this is what churchmanship should be about". Said Emerson: "Tis not in the high stars alone, Nor in the red breast's mellow tone, But in the mud and scum of things, There, always - always something sings". A SUGGESTION OF SUFFERING LOVE Finally, and most important of all, the stable is a hint, a suggestion to us of God's suffering love. Far from misleading us as a mode of revelation of divine truth, it corresponds to what we believe to be God's nature and his way of doing things. It hints of a patient and suffering love... The stable does not mis-represent him; it is no gimmick, no attempt to contrive a rustic effect that might play upon our sentiments. This is the way God works. As Paul reminds us in the words of our text: (nothing spectacular about a stable) "God has chosen what the world calls foolish to shame the wise; He has chosen what the world calls weak to shame the strong!" The stable does not overwhelm us, indeed - it underwhelms us. In a way it is suggesting that force is not an attribute of God. Samuel Miller, in one of his books, writes: "The logos that was powerful enough to make the world, became flesh - was born a baby, absolutely helpless. This is the message of Christianity - that the permanent structure of reality is not coercive, but persuasive, of the nature of grace, and not of law." And if we have the courage to come a little closer to that manger scene, we will discover that this love of God that we speak of is essentially an activity of "self-expenditure". The stable, as a symbol and a mode of revelation of truth, brings us face to face with the fact that there are two kinds of power operative in our world: first, the power of self-expansive (if you will) that makes use of force, that coerces to achieve, that seeks to dominate, that drives rather than draws, and second, there is the power of self-expenditure which lifts, raises, draws rather than drives and which we are given to believe is God's kind of power. HAVE WE BEEN THERE I think what we all deeply want at Christmas is to be able to say that we've really been there, that our lives have been significantly touched and that we have brushed up against eternal truth. How can we know that we've been there. The Wise Men, you will recall, after they were there at the scene, "went back by another way". And therein is our answer. How can we know if we've been there. It's tied in with how we go back after it's all over. On the deeper level, it seems, the birth of Christ is raising a very issue. It asks us which kind of power we believe in and which kind of power we will support and identify ourselves with. Is it the power of self-expansiveness that is linked with force, or is it the power of self-expenditure which is symbolized by the stable and linked with service, with love, with giving. I heard a story this week that I pass on to you. It seems a young, rather innocent girl from a small mid-western town came to seek work here in the city. She was given one of those massive forms to fill out: "Name, address, family history, etc." When she came to the question, "In case of an emergency whom should we notify?" she paused, called the personnel man over and said, "I don't understand. What do you mean?" He explained it to her: "If some accident, you know, befell you, or some emergency arose, whom should we call?" She replied, "Why, the nearest human being I would hope". She would be safe in her answer - if the world were totally committed to the kind of power that seeks to expend itself for others. Our world, so it seems, has a long way to go before it reaches that point. And it will reach that point only as individuals brush up against this eternal truth, and depart - "willing to go back by another way". Yes, the Wise Men went back by another way. May this be the way it is with us as we celebrate the deep joy of this Holy Season. The choice, dear friends, is laid gently at the doorsteps of our hearts. Even here, God will not coerce, nor force. "Though Christ a thousand times In Bethlehem be born, If he's not born in thee, Thy soul is still forlorn" PRAYER Lord, we need an experience of Christmas this year that brings us face to face with divine truth. Help us to trust the kind of power that prefers service to domination, and is prepared to suffer for that which it would win. May we live - not after the conventional wisdom of those who gathered that night in the Inn, but rather by the light of that love which rose from the stable. In the name of Christ, we pray. Amen
Secure Outsourced Biometric Authentication with Performance Evaluation on Smartphones Jaroslav Šeděnka†, Sathya Govindarajan, Paolo Gasti, Kiran S. Balagani New York Institute of Technology {jsedenka, sgovin, pgasti, firstname.lastname@example.org Abstract—We design privacy-preserving protocols for Scaled Manhattan and Scaled Euclidean verifiers, secure against malicious clients and honest-but-curious server. We then augment our protocols with principal component analysis (PCA), which can help improve authentication accuracy. We evaluate the performance of our protocols on an emerging application—namely, continuous authentication of smartphone users. We compare the performance of protocols secure under the malicious client model, with three protocols secure in the honest-but-curious model. We report tradeoffs between computation overhead, communication cost, and authentication accuracy. Our key observations are: 1) Scaled Manhattan without PCA gives the best tradeoff between security, accuracy, and overhead; and 2) with PCA, memory availability on current smartphones limits the number of features that can be used with Scaled Manhattan, and prevents the Scaled Euclidean protocol from running. Our extended evaluation on a laptop client shows that PCA with both Scaled Manhattan and Scaled Euclidean verifiers is feasible given sufficient memory. Index Terms—Privacy-preserving authentication, behavioral biometrics, cryptographic protocols, secure multiparty computation, garbled circuits, homomorphic encryption I. INTRODUCTION Outsourcing biometric authentication involves delegating authentication to third party service providers (e.g., AdmitOne Security [1] and BehavioSec [4]), who specialize in deploying and maintaining biometric systems. Despite the advantages that outsourcing could offer in terms of convenience and cost savings, it raises privacy and security concerns. Biometric data is sensitive, and disclosing it to a third party is undesirable. In this paper, we address the problem of securely outsourcing biometric authentication. We characterize outsourced authentication as a two-party problem which involves: (1) a client (a device in the hands of a user); and (2) an authentication server. Our goal is to design protocols that allow both parties to perform authentication without disclosing any information—besides the outcome of the authentication (to the client) and the authentication score (to the server). Specifically, our protocols conceal the authentication sample collected by the client from the server, and the user’s template from both parties. A. Our Contributions Protocols Secure Against Malicious Clients. We design protocols for Scaled Manhattan and Scaled Euclidean verifiers, secure against malicious clients. Previous work in this area (e.g., [19], [8]) mostly focused on providing security against honest-but-curious (HBC) adversaries. However, security in the HBC model may be inadequate for authentication because it assumes that the client and the server faithfully follow all protocol specifications. For instance, the HBC model does not guarantee that a malicious client cannot get authenticated by “cheating” or deviating from the protocol. On the other hand, there are generic ways to transform HBC to fully malicious protocols (e.g., cut-and-choose [20], [33]). However, these transformations increase the overall protocol cost by at least two orders of magnitude [23]. To achieve a balance between security and cost, we designed protocols that are secure against malicious clients and HBC server. We believe our hybrid model is more practical for authentication: unlike HBC, our model ensures that a malicious client does not benefit from deviating from the protocol. At the same time, it is reasonable to assume that the authentication server can be audited, to expose any misbehavior not captured by the HBC model. Augmenting Authentication Protocols with PCA. We augment our authentication protocols with principal component analysis (PCA), mainly for the following two reasons: (1) the verifiers considered in this paper assume that features are independent. PCA is commonly used as a preprocessing step to suppress covariance among features, to bring the data closer to the model assumption and thus possibly improve performance; and (2) although PCA has been used many times in biometric authentication (e.g., [48], [28]), to our knowledge no study has introduced secure protocols for verification in PCA space. Protocol Evaluation for Smartphone Authentication. We evaluate the performance of the new protocols for continuous authentication of smartphone users. We believe this is both timely and important. The former, because more and more companies have started to offer smartphone user authentication as a service (e.g., [4]). The latter—because in order to make biometric outsourcing viable in the long term, privacy issues must be fully addressed. Another motivation for choosing smartphones for evaluating our protocols is that even though they have become ubiquitous, limited number of studies evaluated secure protocols on this platform. In fact, to our knowledge, only three studies ([19], [43], [8]) have explicitly addressed privacy-preserving continuous authentication in the smartphone domain. Through our evaluation, we aim to mainly address three questions: (1) are our protocols, secure in the hybrid model, practical for continuous authentication on commodity smartphones? (2) How do these protocols compare in terms of computation and communication overhead with protocols secure in the HBC model? And (3) how does PCA impact the cost of the protocols? Our findings can be summarized as follows: (1) Our Scaled Manhattan and Scaled Euclidean protocols in the hybrid model are feasible on current commodity smartphones. Moreover, Scaled Manhattan outperforms Scaled Euclidean in terms of both computation and communication costs. (2) Our protocols secure in the hybrid model have lower communication cost than their HBC counterparts; however, they are more computationally expensive. (3) Although PCA provides a notable improvement in verification accuracy, it also significantly increases the computation and communication overhead. Scaled Manhattan in PCA space can be run on commodity smartphone devices, albeit with limited number of features. However, Scaled Euclidean in PCA space requires more memory than what the OS provides to apps. Comparison with a New Lightweight Protocol.* As a further point of comparison between protocols secure in our hybrid model versus protocols in the HBC model, we design a simple protocol for privacy-preserving 1-Prob verifier, secure in HBC. To our knowledge, our 1-Prob verifier is the fastest protocol for continuous authentication on smartphone devices. B. Organization In Section II, we review related research. In Section III, we present the required biometric and cryptographic background for our protocols. In Section IV, we introduce privacy-preserving protocols for authentication, and present their PCA counterparts in Section V. We evaluate the security of our protocols in Section VI and present their performance in Section VII. We conclude in Section VIII. ## II. RELATED WORK Since the seminal work on garbled circuit evaluation [51], [18], it has been shown that any function can be securely evaluated by representing it as a boolean circuit. Similar results exist for secure evaluation of any function using secret sharing techniques, e.g., [40], or homomorphic encryption, e.g., [9]. In recent years, a number of tools have been developed for automatically creating a secure protocol from its function description written in a high-level language. Examples include Fairplay [34], VIFF [12] and TASTY [21]. However, “custom” optimized protocols for specific applications are often more efficient than such general techniques. Garbled circuits offer security in the honest-but-curious model. However, a technique called cut-and-choose can be used to make garbled circuits secure in the malicious model. With cut-and-choose, the server creates multiple garblings of a circuit. The client randomly selects a subset of these garblings, and asks the server to reveal these circuits’ input keys. The client verifies that all circuits are constructed properly, and evaluates the remaining circuits to obtain the result of the computation. Then, both parties switch roles and repeat this process. There are several approaches to implement cut-and-choose, each requiring a different number of circuits to achieve a given level of security. (See, e.g., [20] or [33].) In particular, recent results by Huang [23] achieve $\kappa$-bit security with approximately $\kappa$ circuits. Another approach to privacy-preserving computation is fully homomorphic encryption (FHE), first constructed by Gentry in [15]. FHE allows computation of arbitrary ring operations in the encrypted domain, and as such can be used for outsourcing biometric authentication. Despite advancements in the area, current implementations are still too slow, and impractical for our setting. (See for example [16] and [39].) A number of publications address the problem of privacy-preserving biometric authentication and identification. Bringer et al. [7] were the first to introduce a general security model for biometric user authentication. The model assumes low trust between the involved parties, and formalizes privacy for biometric authentication. Furthermore, the paper introduces a privacy-preserving protocol that computes the Hamming distance of two bit strings, representing a biometric sample and a template. Barbosa et al. [2] extend the framework of Bringer et al. [7] with a classifier to improve authentication accuracy and propose an instantiation based on Support Vector Machine (SVM) using homomorphic encryption. Schoemakers et al. [45] introduced a protocol for secure privacy-preserving iris matching. The protocol is implemented using threshold ElGamal, and computes (encrypted) Hamming distance between two bit strings representing a template and a user sample, encoded using IrisCode. The result of the Hamming distance is then compared, in encrypted form, with a threshold. Secure face recognition was first addressed by Erkin et al. [13]. In this paper, the authors designed a privacy-preserving face recognition protocol based on Eigenfaces. Sadeghi et al. [42] subsequently improved the performance of the protocol of Erkin et al. More recently, Osadchy et al. [36] designed a new face recognition algorithm together with its privacy-preserving realization called SCiFI. SCiFI simultaneously improves robustness and efficiency of [42]. Blanton et al. [5] focused on privacy-preserving iris and fingerprint matching. The authors rely on a hybrid approach based on garbled circuits and homomorphic encryption for optimal performance. Barni et al. [3] presented a privacy-preserving protocol for fingerprint identification using FingerCodes [25], which is not as discriminative as techniques based on location of minutiae points, but is particularly suited for efficient privacy-preserving implementations. Govindarajan et al. [19] present two protocols for smartphone user authentication. They implemented Scaled Manhattan and Scaled Euclidean verifiers secure in the HBC model. In Section VII, we compare the performance of our protocols secure against malicious clients with the protocols in [19]. Safa et al. [43] present a protocol for outsourcing continuous authentication with a Scaled Manhattan verifier on smartphones, and consider security against malicious clients. However, the security argument presented in their HBC protocol does not take into account information disclosed by order-preserving encryption, even in ciphertext-only scenarios. (See, e.g., the analysis of Boldyreva et al. [6].) As such, there is substantial amount of information leaked by the client during the protocol. In contrast, our scheme provably leaks no information. Chun et al. [8] introduced a garbled-circuit based protocol for outsourcing biometric authentication. Their protocol is secure in the honest-but-curious model and the performance reported by the authors is highly impractical: the lowest reported protocol execution time is approx. 39 minutes. Techniques based on fuzzy commitments (e.g., [26], [49], [27] and [50]) are commonly used to provide template protection and to implement access control on encrypted documents. However, such techniques require biometric comparisons to be performed in a feature space different from that of the original biometrics, possibly increasing equal error rate (EER) [31]. In contrast, our protocols do not affect the EER of the underlying biometric modality, since the comparison between the user sample and the template is functionally the same as the comparison in the unencrypted domain. III. BACKGROUND A. Biometric Verification We implemented privacy-preserving protocols for three verifiers: (1) Scaled Euclidean, (2) Scaled Manhattan, and (3) first-order probabilistic (1-Prob). Let $\{Y_1, Y_2, \ldots, Y_M\}$ denote $M$ $n$-dimensional training data points. Let $x = \{x_1, \ldots, x_n\}$ denote an $n$-dimensional test point. A brief description of the verifiers follows. Scaled Euclidean and Scaled Manhattan Verifiers. Let $y = \{y_1, \ldots, y_n\}$ be the mean vector computed from the feature vectors in the training set of a user. The Scaled Euclidean verifier calculates the verification score as: \[ D_E(x, y) = \sqrt{\sum_{i=1}^{n} \frac{(x_i - y_i)^2}{\sigma_i^2}} \] and the Scaled Manhattan verifier calculates the score as: \[ D_M(x, y) = \sum_{i=1}^{n} \frac{\|x_i - y_i\|}{\sigma_i}, \] where $\sigma_i$ is the standard deviation of the $i$-th feature. First-order Probabilistic Verifier (1-Prob). This verifier outputs the probability score $P(U_j, x)$, which is the joint probability of the $j^{th}$ user $U_j$ and the discrete test vector $x$. The score is calculated using the multiplication rule of probability as follows: \[ P(U_j, x) = P(U_j) \cdot \prod_{i=1}^{n} P(x_i \| U_j) \] where $x_i$ is the $i$-th feature and probabilities of the type $P(x_i \| U_j)$ are estimated from $j^{th}$ user’s training set using: \[ P(x_i \| U_j) = \frac{r_i + mp}{r_j + m}, \] where $r_j$ is the number of training samples that belong to $U_j$, $r_i$ is the number of samples in the $i^{th}$-bin of $x_i$ for user $U_j$, and $m$ is the $m$-estimate. We assumed all users are equally likely, so $P(U_j)$ is same for all users. To estimate probabilities of the type $P(x_i \| U_j)$, we discretized each user’s features into equal-width bins and created a probability histogram. Because Equation (3) is a product of probabilities, if the probability of one bin is zero, the entire score becomes zero. To avoid this, we used $\sum_{i=1}^{n} \log P(x_i \| U_j)$ instead of $\prod_{i=1}^{n} P(x_i \| U_j)$, and treated $\log 0$ as 0. Choice of Verifiers. Scaled Manhattan and Scaled Euclidean are fairly popular verifiers in the biometric authentication literature (see for example, [25], [44], [28], [52] and [46]). The 1-Prob verifier implements a 1-class version of the Naive Bayes classifier and was chosen because of its simplicity and the surprisingly good performance of Naive Bayes in classification tasks despite its feature independence assumption [41]. Verification in PCA Space. We performed PCA to decorrelate features, thereby making the features compatible with the feature independence assumption of Scaled Euclidean, Scaled Manhattan, and 1-Prob verifiers. Furthermore, there is compelling evidence that PCA improves authentication accuracy for several biometric modalities (see, e.g., [48], [47]). Though the feature decorrelation property of PCA is well known, we briefly review it for completeness. Let $Y_{M \times n}$ be a data-matrix containing $M$ $n$-dimensional zero-centered training samples of a user. The covariance matrix of $Y_{M \times n}$ is $Y^T Y$. PCA transforms $Y_{M \times n}$ to $Y'_{M \times n}$ ($YA = Y'$), where $A_{n \times n}$ is a matrix containing $n$ eigenvectors of $Y^T Y$. It is easy to show that $Y'^T Y'$, the covariance matrix of the transformed data is a diagonal matrix with eigenvalues $\lambda_1, \ldots, \lambda_n$ as diagonal elements. B. Cryptographic Preliminaries In this section, we review the cryptographic tools used in our constructions. Garbled Circuit Evaluation. Originally proposed in [51], garbled circuits allow two parties to securely evaluate any function represented as a boolean circuit. The basic idea is that given a circuit composed of gates connected by wires, the server “garbles” the circuit by assigning two randomly chosen encryption keys, $\omega_{j,0}$ and $\omega_{j,1}$, to each wire $w_j$. These keys represent, respectively, 0 and 1. (In the garbled circuits literature, keys are usually referred to as labels.) The server then encrypts a truth table corresponding to each gate. Values in the table are also represented using labels, and each label is encrypted with two keys, $\omega_{j,b_j}$ and $\omega_{l,b_l}$, corresponding to the values on the gate’s input wires. Therefore, computing the “output label” of each gate requires knowing two of its input labels. The output of the circuit is encoded in its output wires, constructed and interpreted as follows. Let $\omega_{i,b} \in \{0,1\}^\kappa$ be the label of output wire $i$ corresponding to output bit $b$, and $s$ the number of output wires. The server selects a pair of random labels $(\omega_{i,0}, \omega_{i,1})$ for each output wire $i \in [1,s]$. Then, it builds a table $T = (\omega_{1,0}, \omega_{2,0}, \ldots, \omega_{s,0})$ and sends it to the client as part of the circuit. The client uses $T$ to interpret the output of the circuit as follows: after evaluating the circuit, the client decrypts exactly one label $\omega_{i,b_i}$ for each output wire $i$, where $b_i \in \{0,1\}$ is $i$-th bit of the protocol’s output. For each $i$, if $\omega_{i,b_i}$ appears in $T$ then $b_i$ is 0, otherwise 1. Recent literature provides optimizations that reduce computation and communication overhead associated with circuit construction and evaluation. Kolesnikov et al. [30] describe a modification that permits XOR gates to be evaluated for free, i.e., there is no communication overhead associated with such gates and their evaluation does not involve cryptographic functions. With this modification, elements of a pair $(\omega_{i,0}, \omega_{i,1})$ might not be selected independently. Instead, $\omega_{i,0} = \omega_{i,1} \oplus r$ for some value $r$, constant throughout the circuit. Pinkas et al. [38] additionally give a mechanism for reducing communication complexity of binary gates by 25%; now each gate can be specified by encoding only three outcomes of the gate instead of all four. Finally, Kolesnikov et al. [29] improve the complexity of certain commonly used operations such as addition, multiplication, comparison, etc. by reducing the number of non-XOR gates. Before garbled circuit evaluation, the client engages in oblivious transfer (OT) protocol, described next. Oblivious Transfer. In 1-out-of-2 Oblivious Transfer ($OT_1^2$), one party (denoted as sender) has two strings $m_0, m_1$, and the other party (the receiver) has one bit ($b$) as its input. At the end of the protocol, the receiver learns $m_b$ and the sender learns nothing. Similarly, in 1-out-of-$N$ OT the receiver obtains one of the $N$ strings held by the sender. In this paper we use an efficient implementation of $OT_1^2$ from [35] as well as techniques from [24] that reduce a large number of OT protocol executions to $\kappa$ (where $\kappa$ is the security parameter). Homomorphic Encryption. Our constructions use a semantically secure additively homomorphic encryption scheme. In an additively homomorphic encryption scheme, $\text{Enc}(m_1) \cdot \text{Enc}(m_2) = \text{Enc}(m_1 + m_2)$, which also implies that $\text{Enc}(m)^a = \text{Enc}(a \cdot m)$. While any encryption scheme with the above properties (such as the well known Paillier encryption scheme [37]) suffices for the purposes of this work, we use the construction due to Damgård et al. [11], [10] (DGK) because it is fast and it produces small ciphertexts. In the rest of the paper, we use $\llbracket m \rrbracket$ to refer to the DGK encryption of message $m$ under the server’s public key. (The server is assumed to have access to the corresponding decryption key.) Our privacy-preserving protocol for computing Scaled Manhattan distance requires privacy-preserving comparison of two encrypted values. For this task, we rely on the comparison protocol of Erkin et al. [13]. Representation of Values in the Encrypted Domain. All inputs to our privacy-preserving protocols are integer values, so we mapped real-valued inputs to $2^e - 1$ equal-width bins corresponding to integers in the interval $[-2^{e-1} + 1, 2^{e-1} - 1]$. We represented integers in two’s-complement notation throughout our protocols. The size of the discretized domain was more than $e$ bits, to account for addition and multiplication operations throughout the protocol. We used the following formula for discretization: $$\text{discretize}_{c,F}(x_i) = \left\lfloor \frac{(2^e - 1) \cdot x_i}{\max_F} \right\rfloor$$ (5) where $F$ is the feature being discretized, $x_i$ is an instance in $F$ and $\max_F$ is the maximum of absolute values in $F$. ## IV. PRIVACY-PRESERVING PROTOCOLS We now introduce our privacy-preserving protocols for secure outsourcing of authentication. Our protocols are divided in two phases: enrollment and verification. During enrollment, a mobile device (client) is used to acquire biometric signals from the user. With the help of an enrollment server, the biometric signals are processed to build the user’s encrypted template. Processing is performed either: (a) by the client—here, the enrollment server simply binds the user’s identity with the encrypted template; or (b) by the enrollment server—in this case, the computational burden is offloaded from the client at the cost of revealing the template to the server. During enrollment, all parties involved (the user, the mobile device, and the enrollment server) are trusted. In practice, this is necessary to ensure that data collection is performed correctly. In all our protocols based on garbled circuits, the client needs an additive share of user’s template for authentication. To lower the amount of information stored on the client, the user’s additive share $r$ can be encrypted under the client’s symmetric key $c$ during enrollment, and stored as $\text{AES}_c(r)$ on the authentication server. Before each authentication attempt, the authentication server sends $\text{AES}_c(r)$ to the client. This allows the client to store only one key, $c$, instead of the user’s share. We do not further elaborate on this modification, as its implementation is straightforward and the cost is insignificant compared to the rest of the protocol. ### A. Modification of the Garbled Circuits Protocol to Achieve Security Against Malicious Clients To achieve security against malicious client, we implement three modifications to the garbled circuits protocol: (1) Output Labels Are Selected Independently. The security of our protocol against malicious clients relies on the security of the garbled circuits in the HBC model. To our knowledge, the only available proof of security for garbled circuits against HBC adversaries is that of Lindell et al. [32]. This proof requires that output labels are selected independently at random—we modified [22] to do this. (2) Client Does Not Learn $T$. The security of our protocol relies on the assumption that the client learns only one output label for each pair $(\omega_{i,0}, \omega_{i,1})$. For this reason, we modify garbled circuits so that the server does not send $T$ to the client. Instead, the client sends the output labels to the server. (3) Server Aborts on Invalid Output Labels. In the HBC model, the labels returned by the client are always valid because all parties faithfully follow the protocol. Therefore, the behavior of the server when labels are invalid need not be defined. However, in our setting we assume that the client can arbitrarily deviate from the protocol. Let $M = (\omega'_1, \ldots, \omega'_s)$ be a $s$-tuple of strings in $\{0, 1\}^n$. $M$ is a valid tuple of output labels iff there exist $(b'_1, \ldots, b'_s) \in \{0, 1\}^s$ s.t. $M = (\omega_{1,b'_1}, \ldots, \omega_{s,b'_s})$. The server verifies if the labels returned by the client are valid. If not, the server aborts. The server incurs negligible overhead for checking the validity of the labels—for each label $\omega'_i$ that is not in $T$ (i.e., $\omega'_i$ does not encode 0), the server verifies that $\omega'_i = \omega_{i,1}$. Modifications (2) and (3) are implemented by at least one existing garbled circuits instantiations [22], although not for improving the security of the protocol. ### B. Scaled Manhattan and Scaled Euclidean Protocols Secure Against Malicious Client We implemented the Scaled Manhattan and Scaled Euclidean verifiers with our modifications to garbled circuits. Enrollment Phase. A user’s scaled and discretized template $y = (y_1, ..., y_n)$ is computed from the mean template $y'$, as is described Figure 1. Standard deviations $\sigma = (\sigma_1, ..., \sigma_n)$ used for scaling are taken as the mean of standard deviations of all users in the training set. The same $\sigma$ is used for all users and is made public. The additive share $r$ of the user’s template can be stored either on the client, or encrypted as AES$_c(r)$ and stored on the server. Verification Phase. The server’s input is the additive share of the biometric template $-(y + r)$ and the client’s input is $x + r$, where $x$ is the current biometric (test) sample and $r$ is the additive share of the template. The layout of the verification circuit is in Figure 2. The client and server run the verification circuit on their inputs. The server obtains the Manhattan/Euclidean distance between $x$ and $y$. The server then makes an authentication decision based on the distance. In case of Scaled Euclidean, the circuit outputs the square of the distance, to avoid the costly computation of square root. Because square root is a monotonous function, this does not impact the outcome of authentication. One-bit Output Instead of Distance. Our protocols can be modified to output an authentication decision (i.e., whether or not the distance between the biometric template and the client’s input is smaller than a threshold). To achieve this, we added a subcircuit that computes the “smaller than” functionality. This subcircuit takes the (unencrypted) threshold and the (garbled) output of the sum subcircuit of Figure 2 as input. This modification improves client’s privacy because the server does not learn any information besides the authentication decision. ![Fig. 2: Generic authentication circuit for Scaled Manhattan/Euclidean distance. The distance component is instantiated using Manhattan or Euclidean distance subcircuit. Output of the circuit is the verification score.](image) ### C. Protocols Secure in the HBC Model Scaled Manhattan and Scaled Euclidean. We summarize the privacy-preserving protocols for Scaled Manhattan and Scaled Euclidean in the Appendix (Figures 6 and 7, respectively). Full description and analysis of the protocols is available in [19]. Next, we discuss our 1-Prob protocol. 1-Prob. We designed a new 1-Prob protocol as a way to implement a fast, yet accurate verifier. Our protocol is secure in the HBC model, and is implemented using homomorphic encryption instead of garbled circuits. Reasons for not implementing a 1-Prob protocol with garbled circuits to secure against a malicious client follow. Under this setting, it is difficult to guarantee both template protection and verification correctness. In 1-Prob, the client must select a set of probabilities from a matrix $\mathcal{Y}'$, which represents the user template. Clearly $\mathcal{Y}'$ cannot be shared with the client, since it would disclose the entire biometric information stored in the system for a particular user. Similarly, $\mathcal{Y}'$ cannot be revealed to the authentication server. As an alternative, the server could store an additive share of $\mathcal{Y}'$ (say, $\overline{\mathcal{Y}}$), where a different random value $r_{i,j}$ is added to each element $\mathcal{Y}'_{i,j}$. The input to the authentication protocol would include $\overline{\mathcal{Y}}$ for the server, and $r_{i,j}$-s for the client. Although this approach provides template protection, it also allows a malicious client to trivially authenticate without knowing the user’s biometric. The client could, for example, replace each random value $r_{i,j}$ with $r'_{i,j} = r_{i,j} - \delta$ for some $0 < \delta \leq 1/n$. The output of the protocol would therefore be $n\delta + \sum_{i=0}^{n} \mathcal{Y}_{i,j}$ instead of $\sum_{i=0}^{n} \mathcal{Y}_{i,j}$, allowing the malicious client to authenticate with overwhelming probability. Although techniques based on zero-knowledge proofs can be used to prevent this attack, the additional overhead from computation, communication, and number of rounds would make the 1-Prob protocol too expensive for mobile devices (and thereby defeat our primary motivation for using 1-Prob). Next, we discuss our design of 1-Prob protocol. In the enrollment phase, biometric samples are collected and processed. Each sample is represented as an $n$-dimensional feature vector. For each feature, we define $\ell$ bins, which correspond to different ranges of values for that feature. For each feature, bins contain the logarithm of the probability associated with the corresponding range. The output of this \| Biometric template \| $y' = (y'_1, \ldots, y'_n)$ \| Vector $y'$ is the mean of user’s biometric measurement from the training session. \| \|-------------------\|-------------------------------\|----------------------------------------------------------------------------------\| \| Scaled and Discretized template \| $y = (d_{e,1}(y'_1/\sigma_1), \ldots, d_{e,n}(y'_n/\sigma_n))$ \| Discretization ranges are the same for all users, per feature. \| \| Final template for enrollment \| $y + r = (d_{e,1}(y'_1/\sigma_1) + 1, \ldots, d_{e,n}(y'_n/\sigma_n) + r_n)$ \| Addition is modulo $2^\ell$. The vector $r$ is selected uniformly at random. \| Fig. 1: Step by step overview of processing user’s biometric template $y' = (y'_1, \ldots, y'_n)$ during enrollment for the Scaled Manhattan and Scaled Euclidean protocols without PCA. Randomization is omitted for protocols secure in the HBC model. Input: Client: sample $x = (x_1, \ldots, x_n)$, server’s public key and decryption key for $\text{AES}_c(\cdot)$; Server: encrypted $n \times m$ probability template matrix $\text{AES}_c(Y) = \text{AES}_c([\![ -\log p_{0,1} ]\!], \ldots, [\![ -\log p_{n,\ell} ]\!])$ and decryption key for the homomorphic encryption scheme. Output: The server learns $\text{Score}_{NB}(x, Y)$. Protocol steps: 1) The server sends $\text{AES}_c(Y)$ to the client, which decrypts it, obtaining $[\![ -\log p_{0,0} ]\!], \ldots, [\![ -\log p_{n,\ell} ]\!]$. 2) For $i = 1, \ldots, n$ Client selects $[\![ -\log p_{i,x_i} ]\!]$ from $Y$ and computes: $$[d] = \left\lfloor \left( \sum_{i=1}^{n} -\log p_{i,x_i} \right) \right\rfloor = \prod_{i=1}^{n} [\![ -\log p_{i,x_i} ]\!]$$ 3) Client randomizes $[d]$ as $[\![ d ]\!] = [d][\![ 0 ]\!]$. 4) Client sends $[\![ d ]\!]$ to the server, which decrypts it and outputs $\text{Score}_{NB}(x, Y)$ as $d$. Fig. 3: Computation of privacy-preserving 1-Prob score Input: Client: sample $x = (x_1, \ldots, x_n)$, server’s public key, and decryption key for $\text{AES}_c(\cdot)$; Server: encrypted $n \times n$ Eigen matrix $\text{AES}_c(A) = \text{AES}_c([\![ a_{1,1} ]\!], \ldots, [\![ a_{n,n} ]\!])$, encrypted $n$-dimensional template vector in the PCA space $\text{AES}_c(y') = \text{AES}_c([\![ y'_1 ]\!], \ldots, [\![ y'_n ]\!])$ and decryption key for the homomorphic encryption scheme. Output: The client learns encryption of $x$ transformed in PCA space. Protocol steps: 1) The server sends $\text{AES}_c(y')$ and $\text{AES}_c(A)$ to the client, which decrypts it, obtaining $[\![ y'_1 ]\!], \ldots, [\![ y'_n ]\!]$ and $[\![ a_{1,1} ]\!], \ldots, [\![ a_{n,n} ]\!]$. 2) Let $([\![ x'_1 ]\!], \ldots, [\![ x'_n ]\!])$ denote the component-wise encryption of $Ax$. For $i = 1, \ldots, n$ the client computes: $$[\![ x'_i ]\!] = \left\lfloor \left( \sum_{j=1}^{n} a_{i,j} \cdot x_j \right) \right\rfloor = \prod_{j=1}^{n} [\![ (a_{i,j}) ]\!]^{\pi_j}$$ 3) For $i = 1, \ldots, n$ the client computes $$[\![ x''_i ]\!] = [\![ x'_i ]\!] \cdot [\![ y'_i ]\!]^{-1} = [\![ x'_i - y'_i ]\!]$$ 4) Client outputs $x''$, which corresponds the encryption of $x$ transformed in PCA space. Fig. 4: Computation of privacy-preserving PCA transformation process is the following $n \times \ell$ matrix: $$Y' = \begin{bmatrix} -\log p_{1,1} & \cdots & -\log p_{1,\ell} \\ \vdots & \ddots & \vdots \\ -\log p_{n,1} & & -\log p_{n,\ell} \end{bmatrix} \quad (6)$$ where $\log p_{i,j}$ is the logarithm of the probability that feature $i$ falls within bin $j$. (If computation is performed in PCA space, feature vectors are pre-processed as discussed in Section III-A.) Matrix $Y'$ is computed by encrypting each element of $Y'$ separately using homomorphic encryption. That is, the element on the $i$-th row, $j$-th column of $Y'$ is $[\![ -\log p_{i,j} ]\!]$. Finally, $Y'$ is encrypted as $\text{AES}_c(Y')$ ($c$ known to the client) and sent to the authentication server. Since the server does not have access to the decryption key for $\text{AES}_c(\cdot)$, it cannot extract any information from $\text{AES}_c(Y')$ (besides its size). Verification phase is illustrated in Figure 3. V. PRIVACY-PRESERVING VERIFICATION IN PCA SPACE In addition to the results reported in this paper, PCA’s positive impact on authentication accuracy has been evidenced in many behavioral modalities—see, e.g., [47]. For the Scaled Manhattan and for the 1-Prob verifier, PCA de-correlates features, thus bringing them closer to 1-Prob’s assumption that features are independent. From a privacy perspective, however, the eigenmatrix $A$ obtained with PCA reveals important information about a user’s template, such as: (a) the relative importance of features, especially in eigenvectors corresponding to top eigenvalues and (b) the structure of the feature covariance matrix. Note that concealing the content of $A$ while revealing the authentication attempt in eigenspace will not resolve the privacy issue: if the adversary obtains multiple instances of the authentication vector and its counterpart in eigenspace, it can estimate $A$. Scaling in the PCA space has to be done differently compared to feature space: 1) The PCA transformation matrix is different for each user, so the average of the standard deviations in PCA space is not meaningful. Therefore, we use standard deviation calculated user-wise. 2) The eigenvalues corresponding to individual Principal Components (PCA space dimensions) are again different for each user, and therefore are calculated user-wise. The above items are not published, unlike in the Scaled Manhattan/Euclidean verifiers in the feature space. Instead, these are “embedded” in the PCA matrix rows/columns, respectively. The number of basis vectors/elements in the PCA space $m$ is lower than or equal to the number of vectors in the feature space (now denoted as $n$). ### A. PCA with protocols based on Garbled Circuits The PCA subcircuit is the same for Scaled Manhattan and Scaled Euclidean verifiers, and is illustrated in Figure 5. The transformation matrix $A$ and the biometric template $y$ are additively shared between the server and client as follows. During enrollment, the client generates a random matrix $R \leftarrow \mathbb{Z}_2^{m \times n}$ and a random vector $s \leftarrow \mathbb{Z}_2^n$. The client stores $(A + R)$ and $s$, while the server stores $-R$ and $-(y + s)$. During authentication, the client acquires biometric sample $x$. It then evaluates the circuit in Figure 5. The client input is $(x + s)$ and $(A + R)$ while the server’s is $-(y + s)$ and $-R$. The circuit computes $\hat{A} = (A + R) - R$ and the zero-centered biometric sample $x - y = (x + s) - (y + s)$; then, it maps $x - y$ to PCA space, obtaining $A(x - y)$, and uses the circuit in Figure 2 to compute the Manhattan/Euclidean distance between $A(x - y)$ and the 0-vector. ![Figure 5: Circuit for authentication using Scaled Manhattan/Euclidean distance in PCA space. The Scaled Manhattan and Scaled Euclidean circuits differ in the weight subcircuit.](image) ### B. PCA with protocols based on Homomorphic Encryption To prevent information leakage via $A$ or $Ax$, we modify our protocols so that the client’s input $Ax$ is provided in encrypted form. The unencrypted version of $Ax$ is never observed by any of the parties. Additionally, we design a short protocol, shown in Figure 4, to compute $Ax$ in encrypted form. Below, we detail the changes to the verification protocols that must be implemented to allow distance computation in PCA space. To zero-center user’s biometric sample $y = (y_1, \ldots, y_n)$ using the corresponding biometric sample $x = (x_1, \ldots, x_n)$ and transform it to PCA space defined by transformation matrix $A$, we need to compute $A(x - y)$. In the encrypted domain, it is more convenient to use the linearity of matrix multiplication and compute $Ax - Ay$ instead. $A$ and $Ay$ is stored encrypted on the server and sent to client. Scaled Euclidean. In order to compute $\llbracket -2\alpha_i x_i y_i \rrbracket$ in Figure 7, Step 2, client and server perform a short interactive protocol [13] to calculate the product of $-x_i$ and $2\alpha_i y_i$ in the encrypted domain. Scaled Manhattan. No change is required to the Scaled Manhattan protocol because it operates on encrypted inputs for both client and server. 1-Prob. During enrollment, probability matrix $\mathcal{Y}'$ is computed using biometric samples in PCA space. Each row is randomly shuffled, and two values $l_{i,j}$ and $h_{i,j}$ (representing the lower and higher limit of the bin corresponding to $\log p_{i,j}$) are associated with each element of $\mathcal{Y}'$. $l_{i,j}$ and $h_{i,j}$ are encrypted separately using homomorphic encryption. Ciphertexts $\llbracket l_{i,j} \rrbracket$ and $\llbracket h_{i,j} \rrbracket$ are then encrypted, together with $\mathcal{Y}$ ($\mathcal{Y}$ is the encryption of $\mathcal{Y}'$ as discussed in Section IV-C), using AES$_k(\cdot)$. During verification, the client transforms vector $x$ to PCA space using the protocol in Figure 4, obtaining $Ax = (x'_1, \ldots, x'_n)$. Then, the client decrypts AES$_k(\mathcal{Y}, l_{i,j}, h_{i,j})$ and, for all $i, j$, it interacts with the server to compute $c_{i,j} = (x'_i < h_{i,j}) + (x'_i > l_{i,j}) - 1$. Note that $c_{i,j}$ is 1 if $x'_i$ falls within bin $i, j$, and 0 otherwise. Then the client computes $\llbracket \sum_{j=1}^n (-\log p_{i,x'_i}) \rrbracket = \prod_{j=1}^n \llbracket -\log(p_{i,j}) \cdot c_{i,j} \rrbracket$. In order to compute $-\log(p_{k,j}) \cdot c_i$ in the encrypted domain, client and server perform the short interactive multiplication protocol in [13]. ## VI. Security Analysis In this section, we analyze the security of our protocols based on garbled circuits. We show that the protocols are secure against a malicious client. Security against a HBC server is shown in [32]. We argue that our modifications to garbled circuits, presented in Section IV-A, do not invalidate the proof in [32]. In particular, modification (1) does not affect the analysis in [32], which assumes that labels are selected independently. Modification (2) also does not affect the security of the protocol: to the server, knowing the protocol’s output is equivalent to knowing the corresponding output labels. Finally, modification (3) is irrelevant when the client follows the protocol faithfully. Security Model. We use the term adversary to refer to insiders, i.e., protocol participants. This allows us to model both misbehaving and compromised clients and servers. We do not consider outside adversaries, since their actions can be mitigated via standard secure protocols, such as TLS. Informally, a HBC party follows the prescribed protocol behavior, and might try to learn additional information from the protocol transcripts. Formally [17]: Definition 1. Let $P_1$ and $P_2$ participate in protocol $\pi$ that computes function $f(\text{in}_1, \text{in}_2) = (\text{out}_1, \text{out}_2)$, where Let $ \text{VIEW}_\pi(P_i) $ denote the view of participant $ P_i $ during the execution of protocol $ \pi $. $ P_i $'s view is formed by its input, internal random coin tosses $ r_i $, and messages $ m_1, \ldots, m_t $ passed between the parties during protocol execution: $ \text{VIEW}_\pi(P_i) = (\text{in}_i, r_i, m_1, \ldots, m_t) $. We say that protocol $ \pi $ is secure against HBC party $ P_j $, $ j \in \{1, 2\} $ if for party $ P_j $ there exists a probabilistic polynomial time simulator $ S_j $ such that $ \{S_j(\text{in}_j, f_j(\text{in}_1, \text{in}_2))\} \equiv \{\text{VIEW}_\pi(P_j), \text{out}_j\} $, where “$ \equiv $” denotes computational indistinguishability. We use the standard “ideal world” vs. “real world” definition of security for malicious adversaries, which we briefly summarize here. (Complete definition is available in Section 7.2.3 of [17].) A malicious party can use any PPT algorithm during protocol execution. The protocol is deemed secure if all attacks that are possible during protocol executions (i.e., in the “real world”) can also be executed when the protocol is replaced by a trusted third party (in the “ideal world”): Definition 2. We say that protocol $ \pi $ securely computes $ f $ in the presence of malicious party $ P_2 $, if for every PPT algorithm $ A $ that is admissible for the real model, there exists a PPT algorithm $ B $ that is admissible in the ideal model such that \[ \{\text{IDEAL}_{f,(P_1(z),B(z))}(x,y)\} \equiv \{\text{REAL}_{\pi,(P_1(z),A(z))}(x,y)\} \] Security against malicious clients is stated in Theorem 1. Theorem 1. Assuming that the oblivious transfer protocol is secure against malicious adversaries, and that the encryption scheme has indistinguishable encryptions under chosen plaintext attacks, our modified garbled circuit protocol is secure in the random oracle model with respect to a malicious client. Proof (sketch). We first show that, with our modified garbled circuits, a malicious client can construct a valid message for the server only by faithfully executing the circuit. Then, we show that the client cannot extract any information on the server’s inputs and outputs from the protocol execution. We conclude by showing that these two properties allow us to build a simulator that uses a malicious client to implement an attack in the ideal world. Valid Message by the Client Can Only be Constructed by Faithfully Executing the Protocol. Recall that the message sent from the client to the server is $ M = (\omega'_1, \ldots, \omega'_s) $. All encryption keys are selected independently. All $ \omega_{i,h_i} $ are also selected independently, and the encryption scheme is semantically secure; therefore, the client can obtain valid output labels only by decrypting the corresponding ciphertexts. Keys used for encrypting output labels are also encrypted, and so are all previous “layers” of the circuit. Hence, to obtain the decryption keys for the output labels, decryption must start at the topmost layer of the circuit, using the input wire keys. The client needs one key corresponding to the server’s input and one key corresponding to client input for each input gate. Server’s keys are made public, while client’s keys are exchanged using OT. Because OT is secure in the malicious model, the client only learns one client key per input gate. Therefore, the client can only decrypt one input label per each gate/wire in the circuit, including output wires. Because the labels were chosen uniformly at random by the server, the probability of guessing a valid label without decrypting the output label is negligible. When $ M $ was not constructed according to the protocol, it will be invalid with overwhelming probability, and the server will abort the protocol. Server Input Privacy. If the client is faithfully following the protocol, security in HBC model of garbled circuits implies that the client does not learn any information about the server’s input. We have shown that the client can either follow the protocol, or the server aborts with overwhelming probability and without sending additional messages to the client. Therefore, our modified garbled circuit protocol provides server input privacy against malicious client. Correctness. Correctness follows from the fact that the client either follows the protocol (then, the result is correct), or the server aborts with overwhelming probability and there is no output. Server Output Privacy. The client either follows the protocol or the server aborts. In the first case, the security of garbled circuits in HBC implies the server output privacy. Real vs. Ideal World Security. Let $ C^* $ be a malicious client in the real world. From server input/output privacy and correctness, we know that $ C^* $ does not learn any information about the server’s input and output. $ C^* $ can influence the server’s output only via its input. For any malicious client $ C^* $, we show that it is possible to construct a simulator $ S $ that uses $ C^* $ to implement the same attack in the ideal world. Using the OT in [35], $ S $ can extract the input of $ C^* $ exploiting the random oracle. $ S $ can then present the same input with the trusted third party in the ideal world. From the correctness of the protocol in the real world, the output of the trusted third party in the ideal world must be the same as the output of the protocol. We conclude that $ C^{} $'s strategy is not an attack on the protocol, as it can be simulated in the ideal world. $ \square $ ### VII. Performance Analysis We evaluated the overhead of our protocols for continuous authentication of smartphone users. Running the protocols requires choosing parameters that impact communication and computation (for example, number of features and level of discretization). Optimal parameters are dataset- and features-specific. Because these parameters were not disclosed in existing smartphone authentication literature, we determined them by performing biometric experiments on two publicly available datasets, Dataset-LTU [46] and Dataset-Frank [14]. Dataset Description. Dataset-LTU was collected from 190 users, in two separate sessions for each user. We used the first session for training, and the second for testing. The dataset is divided into four partitions, based on phone orientation (landscape or portrait) and the orientation of strokes on the screen (horizontal or vertical). Although we ran experiment on all four subsets, for the sake of presentation we include two most representative settings: landscape-horizontal swipes (646 genuine scores) and portrait-vertical swipes (2748 genuine scores). These settings illustrate where PCA has a significant impact on error rates and where it does not, respectively. Dataset-Frank was collected in one session only, so we split the feature vectors belonging to each user into two sets: training and testing. We used the first 90 percent of a user’s vectors to perform 1-class training. We used the remaining 10 percent (188 genuine scores) for zero-effort testing (i.e., impostor attempts of one user are genuine attempts of the remaining users). In Supplement [53], we give further details on our biometric experiments. In the rest of this section, we report biometric accuracies, protocol costs, and tradeoffs. A. Biometric Accuracy Here we report the equal error rates (EERs) achieved on Dataset-LTU and Dataset-Frank. Note that the main purpose of our experiments was to observe the impact of discretization and PCA on ERRs. Impact of PCA. In Table I, we summarized EERs with and without PCA, with no discretization. The use of PCA improves the EER by about 3% in the ‘portrait-vertical’ subset of Dataset-LTU, as well as in Dataset-Frank. For the ‘landscape-horizontal’ subset of Dataset-LTU, PCA does not improve EERs. These results are in agreement with existing research in other behavioral biometrics (e.g., [48], [47]), and confirm that PCA is worth considering and can improve the authentication accuracy—although not in all circumstances. Impact of Discretization. Our cryptographic building blocks require that all operations are performed on integers rather than floating point numbers. Therefore, we evaluated the biometric accuracy of our protocols with different discretization ($e$) values, corresponding to number of bits used for discretization of protocol inputs. In order to achieve roughly the same authentication accuracy as with floating point arithmetics with the Dataset-LTU, $e = 8$ was sufficient for 1-Prob without PCA, while $e = 12$ was needed for all other verifiers with and without PCA. With Dataset-Frank, floating-point-arithmetic-precision was achieved with the same discretization parameters in all configurations, except for Scaled Manhattan and Scaled Euclidean in PCA space, which required $e = 20$. B. Protocol Performance Precomputation for Garbled Circuits. In continuous authentication, an authentication window represents the time slot in which a biometric sample is collected. The size of the window (plus time taken by the verification algorithm) specifies the unavoidable delay in the authentication process. The precomputation phase of our protocols runs during the authentication window. Therefore, precomputation does not increase the authentication delay, as long as the time taken for precomputation is within the authentication window. The most expensive part of the OT protocol can be performed ahead of time by the client and the server [5]. Moreover, the server can construct the garbled circuit independently from the client. This reduces the cost of the protocols once authentication data is available. In the subsequent online phase, the parties complete the OT and the client evaluates the circuit. We report the time and bandwidth overhead for the precomputation and authentication phases separately in Tables II, III and IV. Communication required for the precomputation phase is constant and consist of 22 KB, as is required for the OT. We only include total communication time in the tables. Asymptotic Complexity. The protocols secure against malicious clients are implemented using garbled circuits, so their complexity (both time and space) is linear in the number of the circuit gates. Let $n$ denote the number of features, $m$ the number of PCA features (in case PCA is used) and $e$ the number of bits per feature or element used for discretization. The Scaled Manhattan protocol without PCA has complexity $\mathcal{O}((e + \log(n)) \cdot n)$, and our Scaled Euclidean protocol has higher complexity $\mathcal{O}((e + \log(n))^2 \cdot n)$ due to the multiplication required for computing Euclidean distance. The Scaled Manhattan protocol with PCA has complexity $\mathcal{O}((e + \log(n) + \log(m))^{2} \cdot mn)$ and the Scaled Euclidean protocol with PCA has complexity $\mathcal{O}((e + \log(n) + \log(m))^{4} \cdot mn)$. The computational complexity of our Scaled Manhattan protocol in PCA space based on homomorphic encryption is $\mathcal{O}(n)$ for both client and server, where $n$ is the number of features. The complexity of our Scaled Euclidean protocol is $\mathcal{O}(n)$ for the client and $\mathcal{O}(1)$ for the server, i.e., the cost for the server does not depend on the number of features. In particular, the server only needs to decrypt one ciphertext regardless of the number of features used for authentication. In terms of communication, both protocols require the server to send $\text{AES}_c(\mathcal{Y})$ to the client. Additionally, the protocol that computes Scaled Manhattan distance also requires the parties to run multiple instances of the comparison and multiplication protocols, both of which exchange constant number of messages. As an optimization, if the client caches a copy of $\text{AES}_c(\mathcal{Y})$, then the communication cost of computing the Scaled Euclidean distance is reduced to $\mathcal{O}(1)$. This latter optimization can be used for all protocols, therefore reducing the amount of data exchanged between client and server—or entirely removing the need for further communication, in case of the protocol in Figure 4. The cost of our privacy-preserving 1-Prob protocol is $\mathcal{O}(1)$ for the server, and $\mathcal{O}(n)$ for the client. When performing the same protocol in PCA space, the cost increases to $\mathcal{O}(n\ell)$ for both client and server, where $\ell$ is the number of bins. For this reason, the protocol, although feasible for the wide range of parameters tested in this paper, is practical only for a relatively small number of features/bins when run on a commodity smartphone. Protocols Implementation and Evaluation.** We used the Huang et al. [22] implementation of Garbled Circuits, written in Java, on a Samsung Galaxy S4 smartphone (1.9 GHz Krait 300 CPU) and Macbook Pro laptop as server (2.6 GHz Intel Core i5 CPU). The protocols based on homomorphic encryption were implemented in Java using the BigInteger library. All experiments were run in a single thread. Performance results in this section are reported, for all \| Dataset-LTU portrait-vertical \| Scaled Manhattan no PCA 0.314 (26) \| Scaled Manhattan PCA 0.262 (22/28) \| Scaled Euclidean no PCA 0.305 (26) \| Scaled Euclidean PCA 0.269 (22/28) \| 1-Prob no PCA 0.344 (28) \| 1-Prob PCA 0.261 (23/24) \| \|-------------------------------\|----------------------------------\|-------------------------------------\|----------------------------------\|----------------------------------\|------------------------\|------------------------\| \| Dataset-LTU landscape-horizontal \| 0.231 (27) \| 0.231 (15/28) \| 0.243 (23) \| 0.245 (15/28) \| 0.215 (26) \| 0.248 (12/24) \| \| Dataset-Frank \| 0.213 (3) \| 0.183 (26/27) \| 0.210 (8) \| 0.183 (9/12) \| 0.267 (15) \| 0.241 (26/27) \| TABLE I: Equal error rates for three verifiers when features are not discretized. For each verifier, we report the best EER achieved across different parameter settings (see Supplement [53] for details). We give the number of features in parenthesis. For PCA, we report the number of features as number of PCA features/number of original features used to generate PCA matrix. Except for the ‘landscape-horizontal’ subset of Dataset-LTU, PCA improves the authentication accuracy by at least 2.5%. \| Protocols secure against malicious client precomp. auth. comm. \| Honest-but-curious model protocols auth. comm. \| \|---------------------------------------------------------------\|-----------------------------------------------\| \| Smartphone client \| Smartphone client \| \| Manhattan (n=8) \| 1.6s 1.5s 0.04 MB \| \| Manhattan (n=16) \| 2.5s 3.2s 0.06 MB \| \| Manhattan (n=28) \| 3.5s 5.6s 0.09 MB \| \| Euclidean (n=8) \| 13.2s 39s 0.46 MB \| \| Euclidean (n=16) \| 27.9s 105.5s 0.95 MB \| \| Euclidean (n=28) \| 52.8s 237.6s 1.66 MB \| \| 1prob (n=8) \| n/a \| \| 1prob (n=16) \| n/a \| \| 1prob (n=28) \| n/a \| \| 34ms 16.2 MB \| 35ms 32.5 MB \| \| 36ms 57.5 MB \| \| TABLE II: Performance of our protocols without PCA. We benchmarked the protocols on a smartphone client and Macbook Pro server. All results are for $e = 12$ bits discretization. We only implemented the 1-prob verifier in the HBC model, as its main purpose is low-latency authentication. \| Protocols secure against malicious client precomp. auth. comm. \| Honest-but-curious model protocols auth. comm. \| \|---------------------------------------------------------------\|-----------------------------------------------\| \| Smartphone client \| Smartphone client \| \| Manhattan (n=4, m=3) \| 16.4s 60.2s 0.7 MB \| \| Manhattan (n=8, m=7) \| 151.1s 838s 3.6 MB \| \| Manhattan (n=10, m=9) \| 363.6s 2051s 6.1 MB \| \| 1prob (n=8, m=10, 10 bins) \| n/a \| \| 1prob (n=16, n=27, 10 bins) \| n/a \| \| 1prob (n=28, n=27, 10 bins) \| n/a \| \| 12.2s 18.2 MB \| 18.7s 40.5 MB \| \| 45.9s 174.0 MB \| \| TABLE III: Performance of our protocols with PCA. As the amount of memory available to applications is limited on Android, we evaluated our Scaled Euclidean protocol and the Scaled Manhattan protocol for higher amount of features on a laptop client. \| Protocols secure against malicious client precomp. auth. comm. \| Honest-but-curious model protocols auth. comm. \| \|---------------------------------------------------------------\|-----------------------------------------------\| \| Laptop client \| Laptop client \| \| Manhattan (n=10, m=9) \| 7.1s 3.7s 6.2 MB \| \| Manhattan (n=16, m=15) \| 20.0s 12.0s 17.3 MB \| \| Manhattan (n=28, m=27) \| 87.2s 71.2s 61.1 MB \| \| Euclidean (n=10, m=9) \| 31.0s 27.1s 21.4 MB \| \| Euclidean (n=16, m=7) \| 83.4s 70.3s 57.1 MB \| \| Euclidean (n=28, m=27) \| 642s 655s 187.3 MB \| \| 23ms 14.2 MB \| 37ms 40.3 MB \| \| 65ms 129.4 MB \| \| TABLE IV: Generalization of our results—performance for protocols with PCA on a laptop client. The performance gain for the Scaled Manhattan protocol with PCA secure against malicious client is more than two orders of magnitude faster on the laptop compared to the smartphone, because there are no memory restrictions and no need for garbage collection. The performance of HBC Scaled Manhattan protocol with PCA on laptop is less than five times faster compared to the smartphone. Protocols, when input was discretized with $e = 12$ bits, both with and without PCA. We evaluated the performance of our protocols for $n$ varying from 8 to 28 features. Performance of our protocols is summarized in Tables II, III and IV. Performance without PCA. For hybrid-model protocols, Scaled Manhattan without PCA is practical for continuous user authentication of smartphones (running time for authentication is from 1.5 to 5.6 seconds with less than 100 KB data transfer, for 8 to 28 features). The Scaled Euclidean protocol without PCA is feasible, but not as practical as Scaled Manhattan, with running time of the authentication protocol of 39 seconds to 238 seconds and 0.5 MB to 1.6 MB communication for the same number of features. This difference is mainly due to the expensive multiplication operations in the Euclidean distance computation. While the protocols secure in the HBC model have very low running time for authentication (between 0.85 and 2.2 seconds) the communication cost is higher. With Scaled Manhattan, the communication overhead is over 50 times higher compared to our hybrid-model protocol. The difference is not as large with Scaled Euclidean protocols. As a sidenote, we provide the performance results without PCA on a laptop client in the Supplement [53]. Performance with PCA. The number of features in PCA space was chosen as $m = n - 1$. In practice, we expect applications to use less than $m$ PCA features. Therefore, our measurements provide an upper bound on the computation and communication cost of our garbled circuit PCA protocols. Smartphone devices have limited amount of memory compared to desktop computers and the Android platform limits the amount of memory available to applications. The per-application memory limit on our client device is 512 MB; when the amount of memory used by the application approaches this limit, garbage collection consumes a significant portion of the application’s running time. Even though the available memory suffices for all our experiments without PCA, we were unable to complete experiments with more than 9 PCA features (computed from 10 original features) with Scaled Manhattan verifier and to complete any experiments with the Scaled Euclidean verifier. It is important to note that the amount of memory available in smartphones has been increasing steadily, often doubling every two years. By extrapolating from this trend, we can assume that protocols limitations due to memory availability will become less relevant in the near future. As an additional point of comparison, we evaluated the performance of our protocols using the MacBook Pro laptop also as client. The Scaled Manhattan protocol with PCA is roughly four times faster on the laptop, compared to the smartphone, in the honest-but-curious model. The performance difference between smartphone and laptop client is much higher for the protocol secure against malicious clients, mainly due to memory constraints on the smartphone and high cost of repeated garbage collection. One-bit Output Instead of Distance. The server’s output of our protocols is the distance between the template and the authentication sample. The cost of modifying our protocols, secure against malicious client, to output only one bit as the authentication result is less than 1 second of computation. Amount of communication does not increase significantly, as the additional cost of the smaller than subcircuit is compensated by the reduction in the number of output wires to one. Performance in Malicious Model. The cut-and-choose technique allows transformation of garbled circuits to achieve security in the malicious model (in this model, either the client or the server can be malicious). This technique increases both the time and communication cost roughly $\kappa$-fold to achieve $\kappa$-bit security. ### VIII. Conclusions We introduced the first efficient privacy-preserving protocols for securely outsourcing authentication using Scaled Euclidean and Scaled Manhattan verifiers, in feature space and PCA space. Our protocols are secure in the hybrid model, where the client is assumed to act maliciously and the server to be HBC. We also designed a light-weight protocol, secure in the HBC model, for 1-Prob verifier. We performed experiments to demonstrate the accuracy and practicality of our protocols on an emerging area—outsourcing of continuous smartphone user authentication. Our results can be summarized as follows: - The choice of discretization parameter $e$ is crucial for all our privacy-preserving protocols. For all protocols on Dataset-LTU and for most protocols on Dataset-Frank, $e = 12$ is sufficient and provides similar accuracy to using floating point arithmetics. - When security against malicious clients or limited communication is required, Scaled Manhattan without PCA gives the best tradeoff between cost and EER. - If HBC security is sufficient, privacy-preserving Scaled Manhattan and Scaled Euclidean protocols in PCA space offer the best tradeoff between accuracy and performance (with a slight performance edge for Scaled Manhattan). - Privacy-preserving 1-Prob verifier without PCA in HBC model provides remarkably low overhead. This is, to the best of our knowledge, the fastest protocol designed for secure outsourcing of continuous authentication on smartphones. ### References [1] AdmitOne Security. http://www.admitonesecurity.com. [2] M. Barbosa, T. Brouard, S. Cauchie, and S. de Sousa. Secure biometric authentication with improved accuracy. In Proc. of the 13th Australasian conference on Information Security and Privacy, 2008. [3] M. Barni, T. Bianchi, D. Catalano, M. Di Raimondo, R. Labati, P. Falla, D. Fiore, R. Lazzoretti, V. Piuri, F. Scotti, and A. Piva. Privacy-preserving fingercode authentication. In Proc. of the 12th ACM Workshop on Multimedia and Security, 2010. [4] Behaviosec. http://www.behaviosec.com. [5] M. Blanton and P. Gasti. Secure and efficient protocols for iris and fingerprint identification. In Proc. of the 16th European Conf. on Research in Computer Security, 2011. [6] A. Boldyreva, N. Chenette, and A. O’Neill. Order-preserving encryption revisited: Improved security analysis and alternative solutions. In CRYPTO, 2011. [7] J. Bringer, H. Chabanne, M. Izabachene, D. Pointcheval, Q. Tang, and S. Zimmer. An application of the Goldwasser-Micali cryptosystem to biometric authentication. In Proc. of the 12th Australasian Conference on Information Security and Privacy, 2007. [8] H. Chun, Y. Elmehdwi, F. Li, P. Bhattacharya, and W. Jiang. Outsourcable two-party privacy-preserving biometric authentication. In ASIACCS, 2014. [9] R. Cramer, I. Damgård, and J. Nielsen. Multiparty computation from threshold homomorphic encryption. In Proc. of the Intl. Conf. on the Theory and Application of Cryptographic Techniques, 2001. [10] I. Damgård, M. Geisler, and M. Kroigaard. A correction to efficient and secure comparison for on-line auctions. Cryptology ePrint Archive, Report 2008/321, 2008. [11] I. Damgård, M. Geisler, and M. Kroigaard. Homomorphic encryption and secure comparison. Journal of Applied Cryptology, 1(1), 2008. [12] I. Damgård, M. Geisler, and M. Kroigaard. Asynchronous multiparty computation: Theory and implementation. In PKC, 2009. [13] Z. Erkin, M. Franz, J. Guajardo, S. Katzenbeisser, I. Lagendijk, and T. Toft. Privacy-preserving face recognition. In PETS, 2009. [14] M. Frank, R. Biedert, E. Ma, I. Martinovic, and D. Song. Touchalytics: On the applicability of touchscreen input as a behavioral biometric for continuous authentication. TIFS, 8(1), 2013. [15] C. Gentry. Fully homomorphic encryption using ideal lattices. In STOC, 2009. [16] C. Gentry and S. Halevi. Implementing gentry’s fully-homomorphic encryption scheme. In EUROCRYPT, 2011. [17] O. Goldreich. Foundations of Cryptography: Volume 2, Basic Applications. Cambridge University Press, 2004. [18] O. Goldreich, S. Micali, and A. Wigderson. How to play any mental game or a completeness theorem for protocols with honest majority. In STOC, 1987. [19] S. Govindarajan, P. Gasti, and K. Balagani. Secure privacy-preserving protocols for outsourcing continuous authentication of smartphone users with touch data. In BTAS, 2013. [20] V. Goyal, P. Mohassel, and A. Smith. Efficient two party and multi party computation against covert adversaries. In Nigel Smart, editor, EUROCRYPT, 2008. [21] W. Henecka, S. Kogl, A.-R. Sadeghi, T. Schneider, and I. Wehrenberg. TASTY: Tool for Automating Secure Two-partY computations. In CCS, 2010. [22] Y. Huang, D. Evans, J. Katz, and L. Malka. Faster secure two-party computation using garbled circuits. In USENIX, 2011. [23] Y. Huang, J. Katz, and D. Evans. Efficient secure two-party computation using symmetric cut-and-choose. In CRYPTO, 2013. [24] Y. Ishai, J. Kilian, K. Nissim, and E. Petrank. Extending oblivious transfers efficiently. In CRYPTO, 2003. [25] A. Jain, S. Prabhakar, L. Hong, and S. Pankanti. Filterbank-based fingerprint matching. IEEE Trans. on Image Processing, 9(5), 2000. [26] A. Juels and M. Sudan. A fuzzy vault scheme. In ISIT, 2002. [27] A. Juels and M. Wattenberg. A fuzzy commitment scheme. In CCS, 1999. [28] K. Killourhy and R. Maxion. Comparing anomaly-detection algorithms for keystroke dynamics. In Proc. of the Annual IEEE/IFIP Intl. Conf. on Dependable Systems and Networks, 2009. [29] V. Kolesnikov, A.-R. Sadeghi, and T. Schneider. Improved garbled circuit building blocks and applications to auctions and computing minima. In CANS, 2009. [30] V. Kolesnikov and T. Schneider. Improved garbled circuit: Free XOR gates and applications. In International Colloquium on Automata, Languages and Programming, 2008. [31] G. Kumar, S. Tulyakov, and V. Govindaraju. Combination of symmetric hash functions for secure fingerprint matching. In Proc. of the 20th Intl. Conf. on Pattern Recognition, 2010. [32] Y. Lindell and B. Pinkas. A proof of security of Yao’s protocol for two-party computation. Journal of Cryptology, 22(2), 2009. [33] Y. Lindell and B. Pinkas. Secure two-party computation via cut-and-choose oblivious transfer. Journal of Cryptology, 25(4), 2012. [34] D. Malkhi, N. Nisan, B. Pinkas, and Y. Sella. Fairplay – a secure two-party computation system. In USENIX, 2004. [35] M. Naor and B. Pinkas. Efficient oblivious transfer protocols. In ACM-SIAM Symposium On Discrete Algorithms (SODA), 2001. [36] M. Osadchy, B. Pinkas, A. Jarrous, and B. Moskovich. SCIFI – A system for secure face identification. In IEEE Symp. on Security and Privacy, 2010. [37] P. Paillier. Public-key cryptosystems based on composite degree residuosity classes. In EUROCRYPT, 1999. [38] B. Pinkas, T. Schneider, N. Smart, and S. Williams. Secure two-party computation is practical. In ASIACRYPT, 2009. [39] T. Plantard, W. Susilo, and Z. Zhang. Fully homomorphic encryption using hidden ideal lattice. TIFS, 8(12), 2013. [40] T. Rabin and M. Ben-Or. Verifiable secret sharing and multiparty protocols with honest majority. In STOC, 1989. [41] I. Rish. An empirical study of the Naive Bayes classifier. In IJCAI-01 workshop on Empirical Methods in AI, 2001. [42] A.-R. Sadeghi, T. Schneider, and I. Wehrenberg. Efficient privacy-preserving face recognition. In Intl. Conf. on Information Security and Cryptology, 2009. [43] N. Safa, R. Safavi-Naini, and S. Shahandashti. Privacy-preserving implicit authentication. In ICT Systems Security and Privacy Protection, 2014. [44] C. Sanchez-Avilaa and R. Sanchez-Reillob. Two different approaches for iris recognition using Gabor filters and multiscale zero-crossing representation. Pattern Recognition, 38(2), 2005. [45] B. Schoenmakers and P. Tuyls. Security with Noisy Data: Private Biometrics, Secure Key Storage and Anti-Counterfeiting, chapter ‘Computationally secure authentication with noisy data’. Springer-Verlag, 2007. [46] A. Serwadda, V. Phoha, and Z. Wang. Which verifiers work?: A benchmark evaluation of touch based authentication algorithms. In BTAS, 2013. [47] C. Shen, Z. Cai, X. Guan, Y. Du, and R. Maxion. User authentication through mouse dynamics. TIFS, 8(1), 2013. [48] Matthew Turk and Alex Pentland. Eigenfaces for recognition. Journal of cognitive neuroscience, 3(1), 1991. [49] U. Uludag, S. Pankanti, and A. Jain. Fuzzy vault for fingerprints. In Audio-and Video-Based Biometric Person Authentication, 2005. [50] J. Šedénka, K. Balagani, V. Phoha, and P. Gasti. Privacy-preserving population-enhanced biometric key generation from free-text keystroke dynamics. In IJCB, 2014. [51] A. Yao. How to generate and exchange secrets. In FOCS, 1986. [52] X. Zhao, T. Feng, and W. Shi. Continuous mobile authentication using a novel graphic touch gesture feature. In BTAS, 2013. [53] Supplementary material to this paper. Available on http://ieeexplore.ieee.org/. APPENDIX A. HBC Scaled Manhattan and Scaled Euclidean In Section VII, we compared our new protocols, secure against malicious clients, with the protocols secure in the HBC model from [19]. For reference, we include the Scaled Manhattan and Scaled Euclidean protocol in Figures 6 and 7, respectively. Security proofs and detailed description can be found in [19]. B. Security Analysis of l-Prob Verifier in HBC The semantic security of AES_c(.) prevents the server from extracting any information from AES_c(Y), besides its size. The server’s view of the protocols consists of the encryption and decryption key for the homomorphic encryption scheme, encrypted matrix AES_c(Y), and ciphertext $\llbracket d \rrbracket$ from the client. The server’s output is $d$. Simulator $S_s$ provides the server with AES_c(Y) and the decryption key for the homomorphic encryption scheme as input. It then uses $d$ to construct $\llbracket d \rrbracket$, and sends it to the server. Since $\llbracket d \rrbracket$ is properly distributed, i.e., is a valid encryption of $d$, the server cannot distinguish between the simulation and a real execution of the protocol. Therefore, the protocol is secure against a curious server. The client’s view of the protocol consists in the server’s public key, the symmetric key for AES_c(.), $x$ and AES_c(Y). The client has no output. Simulator $S_c$ selects a random set of values $y_{1,1}, \ldots, y_{n,\ell}$, with $0 \leq y_{i,j} \leq 2^{\ell-1} - 1$. It then constructs $Y$ and sends AES_c(Y) to the client. The semantic security of the homomorphic encryption scheme prevents the client from determining that $Y$ corresponds to the encryption of random values. Therefore, AES_c(Y) is properly distributed. For this reason, the client cannot distinguish between interaction with the $S_c$ and with a honest server. Hence the protocol is secure against a curious client. Jaroslav Šedénka received his M.Sc. in Mathematics and B.Sc. in Computer Science from Masaryk University Brno, Czech Republic, in 2010 and 2007, respectively. He is currently pursuing a Ph.D. degree at the same university. His current research interests include both mathematical and applied cryptography, cryptographic protocols, algebraic number theory and algebra. Input: Client: sample $x = (x_1, \ldots, x_n)$, server’s public key and decryption key for AES$_c(\cdot)$; Server: encrypted template AES$_c(y) = \text{AES}_c([\alpha_1 y_1], \ldots, [\alpha_n y_n])$ (where $\alpha_i = 1/\sigma_i$) and decryption key for the homomorphic encryption scheme. Output: The server learns $D_M(x, y)$. Protocol steps: 1) The server sends AES$_c(y)$ to the client, which decrypts it, obtaining $[\alpha_1 y_1], \ldots, [\alpha_n y_n]$. 2) For $i = 1, \ldots, n$, the client and the server interact in a privacy-preserving comparison protocol. At the end of the protocol the client learns the encryption of bit $b_i = (\alpha_i x_i < \alpha_i y_i)$ 3) For $i = 1, \ldots, n$, the client computes: \[ [d_i] = [\alpha_i x_i - \alpha_i y_i] = \left[ \max(\alpha_i x_i, \alpha_i y_i) - \min(\alpha_i x_i, \alpha_i y_i) \right] = \left[ (b_i \cdot (\alpha_i y_i - \alpha_i x_i) + \alpha_i x_i) - (b_i \cdot (\alpha_i x_i - \alpha_i y_i) + \alpha_i y_i) \right] \] as: \[ [d_i] = \left[ b_i \cdot \alpha_i y_i \right]^2 \cdot \left[ b_i \cdot \alpha_i x_i \right]^{-2} \cdot \left[ \alpha_i x_i \right] \cdot \left[ \alpha_i y_i \right]^{-1} \] The computation of $[d_i]$ requires client and server to perform a short interactive protocol [13] for computing $[b_i \cdot \alpha_i y_i]$. 4) Then, the client computes: \[ [d] = \left[ \left( \sum_{i=1}^{n} d_i \right) \right] = \prod_{i=1}^{r} [d_i] \] 5) The client sends $[d]$ to the server, which decrypts it and outputs $D_M(x, y)$ as $d$. --- Fig. 6: Computation of privacy-preserving Scaled Manhattan distance secure in the HBC model Input: Client: sample $x = (x_1, \ldots, x_n)$, decryption key for AES$_c(\cdot)$ and server’s public key; Server: encrypted template AES$_c(y) = \text{AES}_c([\alpha_1 y_1^2], [\alpha_1 2y_1], \ldots, [\alpha_n y_n^2], [\alpha_n 2y_n])$ (where $\alpha_i = 1/\sigma_i^2$) and decryption key for the homomorphic encryption scheme. Output: The server learns $D_E(x, y)$. Protocol steps: 1) The server sends AES$_c(y)$ to the client, which decrypts it as $[\alpha_1 y_1^2], [\alpha_1 2y_1], \ldots, [\alpha_n y_n^2], [\alpha_n 2y_n]$. 2) For $i = 1, \ldots, n$, the client computes: \[ [d_i] = \left[ (\alpha_i \cdot (x_i - y_i)^2) \right] = \left[ (\alpha_i x_i^2) \right] \cdot \left[ (\alpha_i y_i^2) \right] \cdot \left[ (\alpha_i 2y_i) \right]^{-\pi}; \] 3) Then, the client computes \[ [d] = \left[ \left( \sum_{i=1}^{n} d_i \right) \right] = \prod_{i=1}^{r} [d_i] \] 4) The client sends $[d]$ to the authentication server, which computes $d$ and outputs $D_E(x, y)$ as $\sqrt{d}$. --- Fig. 7: Computation of privacy-preserving Scaled Euclidean distance secure in the HBC model --- Sathya Govindarajan received his B.Tech. degree in electronics and communication engineering from the Jawaharlal Nehru Technological University, Hyderabad, India, in 2009, and M.S. degree in computer science from the New York Institute of Technology, Old Westbury, New York, U.S.A., in 2013. His research interests include data analytics, digital image and video processing, human computer interaction, and new advancements in computer graphics that include simulation, rendering, and scientific visualization. Dr. Kiran Balagani is an assistant professor of computer science at the New York Institute of Technology. His research interests are in cyber-behavioral anomaly detection (e.g., unauthorized user-access behaviors), behavioral biometrics, and privacy-preserving biometrics. Balagani’s work has appeared in several peer-reviewed journals, including the IEEE Transactions on Pattern Analysis and Machine Intelligence, the IEEE Transactions on Information Forensics and Security, the IEEE Transactions on Knowledge and Data Engineering, the IEEE Transactions on Systems, Man, and Cybernetics, and Pattern Recognition Letters. He holds three U.S. patents in network-centric attack detection. His teaching interests include development of graduate and undergraduate courses in network security and biometrics. Balagani received the Ph.D. degree and two M.S. degrees from Louisiana Tech University; and the B.S. degree from Bangalore University, India. Dr. Paolo Gasti is an assistant professor of Computer Science at the New York Institute of Technology (NYIT), School of Engineering and Computing Sciences. Dr. Gasti’s research focuses on behavioral biometrics, privacy-preserving biometric authentication and identification, secure multi-party protocols and network security. Before joining NYIT, he worked as a research scholar at University of California, Irvine. His research has been sponsored by the Defense Advanced Research Project Agency and the U.S. Air Force. He received his B.S., M.S., and Ph.D. degrees from University of Genoa, Italy. He is a Fulbright scholar, and member of IEEE.
The Dawning Day for the Lamanites Author(s): A. Merlin Steed Source: The Latter-Day Saints' Millennial Star, Vol. 95, No. 21 (1 June 1933) Published by: The Church of Jesus Christ of Latter-day Saints Page(s): 355–359, 363 Abstract: This article describes a time when Native American students visited the Alberta, Canada Temple. They saw wall paintings that depicted Lehi offering a sacrifice after landing on American soil, and a picture of Christ administering the sacrament to a Lamanite at his coming after his resurrection. THE DAWNING DAY FOR THE LAMANITES ELDER A. MERLIN STEED STAKE GENEALOGICAL REPRESENTATIVE, LOS ANGELES STAKE On the fourth of August, 1932, I arrived in Cardston, Alberta, Canada, and arranged with President Wood and John Smith to go out to a stray horse sale on the Indian reservation. Just as we were leaving, Jack Galbraith, an old friend of mine, joined the party. Upon our return, as we were nearing Cardston, Jack invited me to accompany him to his ranch on the Blackfoot Indian Reservation, which I did. On the following day, we went to Browning, Montana. There Mr. Galbraith took me to the home of his brother-in-law, who had married a half-breed Indian. I met her mother and talked to her for several hours on the Gospel, telling her of the Book of Mormon and of the work that lay before their people. She seemed deeply interested and promised to read the Book of Mormon. From here I went with Jack to the home of his uncle, Frank Mountain Chief, who is the last of the Hereditary Tribal Chiefs, as far as is known, in America. I had a long and most interesting talk with this venerable old Chief, who is the historian of his people; and many hours were spent with him. He told me that the Bloods, the Blackfeet and the Peigans were all of the same tribe of Indians. He related several instances that caused their separation into the three tribes. He sang many of his old war songs to me. We know that this Chief is at least 98 years old, for in 1854 he signed a treaty between the United States Government and his tribe of Indians. He would have to be at least 21 years old at that time. He now stands over six feet tall, and, notwithstanding his advanced age, is erect and powerful. He has had many wonderful experiences. He witnessed the inauguration of both President William McKinley and President Woodrow Wilson, being called to Washington, D.C., especially for these occasions. Once, in company with his nephew, Jack Galbraith, Mr. Frank Sander- ville and Mr. Clark, he was called to Washington to adjust the dif- ference between the Indians and the Government. These four repre- sented the five Indian reservations in Montana. After I had talked to him several hours, he asked me to sing a song for him, which I did. Then I commenced to teach him the Gospel. He said to me, "My God is not good to me. For many years I have been a Christian Indian, and if my God were good to me He would not allow me to suffer as I do, for I am now blind and must be led everywhere I go by my son or grandson." Then I said to him, "I had a very dear friend in Cardston. His name was Brother Nuby. He was blind for many years and yet every time I heard him speak, he would say, 'My God is good to me.' Notwithstanding being blind, he was happy because of the bless- ings he received from his Heavenly Father." Then I said to the old chief, "If you worship the right God, and do what is right, He will cause you to be happy regardless of your loss of sight." He asked me to come and see him again before returning to my home in California. This I was happy to do. On September 5th, I spent several hours talking to him again. I told him of the Book of Mormon, which I have since mailed to him. He said, "Whenever you come, you make my heart glad, for a sweet spirit comes over me." Then I told him this was the Spirit of God who was the Father of his spirit, which he was en- titled to receive by serving the Lord. He then told me to take him by the hand and lead him into the house. I did so, after taking his picture with my daughter and with his son Walter and little grandson. After reaching the house, he gave me a pair of moccasins that he had worn when going to Washington to the inauguration of President Woodrow Wilson. He said to me, "Remember me often, and when you wear these moccasins think of me." After leaving the old Chief's home, we returned to Mr. Gal- braith's ranch where we had supper. Then we spent the entire evening preaching the Gospel to those who had assembled during the day. The Spirit of the Lord was poured out in rich abun- dance and we all rejoiced in the same. YOUNG INDIANS VISIT THE ALBERTA TEMPLE On Saturday, August 6th, 1932, upon the invitation of President Wood I went to the Cardston Temple where 220 Indian students, boys and girls from 12 to 17 years of age, met with their parents. These young Indians were all winners in a two-day contest of the Indians belonging to the 4-H clubs of the five reservations in the State of Montana. Mr. Campbell, who was the Indian Agent for forty-two years, had planned for many years to teach and educate the Indians to be more useful. He had arranged the 4-H clubs among the Indians of the five reservations, and was doing all in his power to teach them to be more useful to society. He had had the privi- lege of going through the Temple several times prior to its dedication and had felt the inspiration and the Spirit of the Lord that permeated this holy place. When his forty years of service were up and he was entitled to a pension, he asked for two additional years as agent that he might realize some of his plans. Early in the spring of 1932 he wrote to President Wood, asking if it would be possible to have these young Indians visit the Temple. After much corresponding, the privilege was granted by the Authorities of the Church that they might make the visit. On the 4th and 5th of August, the try-outs were held near Browning, Montana, and on the 6th, the winners were brought to the House of the Lord in forty-seven cars. I shall never forget nor cease to be thankful for the privilege of attending this service, and seeing and feeling the marvellous things I experienced on this occasion. That all may better understand what this meant and will yet mean to those present, I will describe as nearly as possible the proceedings. PICTURES ON THE WALLS OF THE TEMPLE After all the Indians were seated, or standing in the aisles of the Chapel room, the opening song and invocation were given. Then President Wood welcomed the Indians in our midst. On the north wall of this room is the painting representing Lehi after landing upon American soil. There he offered sacrifice, and the paintings of him and his family are in full life size. Lehi has his hands raised toward heaven, offering a burnt offering to God for the safe arrival of himself and family. His sons and daughters are depicted kneeling in humble devotion around the altar. Then on the east wall is the scene, also in life size, of Jesus, the Risen Lord, passing the sacrament to a great assembled throng. Christ is pictured in the very act of handing the sacrament to a Lamanite. The congregation in this painting is about equally mixed, Nephites and Lamanites. Imagine the thoughts that must have been in the minds of these boys and girls and the deeper understanding that they must have felt as they witnessed that Christ is no respecter of persons, and that the repentant Lamanite was entitled to the sacrament and blessings of Jesus the same as the Nephites. Every heart was touched as President Wood described these paintings and talked to them in this room. Brother Joseph Y. Card gave a very splendid brief history of the settlement of Cardston by his father in 1887. Mr. Campbell made a few remarks and was followed by Mr. Frank Sanderville, an Indian, who is a graduate of Carlisle University and who has spent many years in the service of the Government. Brother Jack Galbraith was next asked to speak and I hope every heart was touched as mine was when he arose and said: "My brothers and sisters, and friends, you of my people, how happy I am to meet you here in the House of the Lord. You are here because you are winners of the contests you have entered in your various fields of endeavours. I hope you can understand that if you will continue to learn of God and to walk in His ways, some day you may have the privilege of returning to this House of the Lord and here receive the greatest prize God has for His faithful sons and daughters, winning the prize of Eternal Life, which is the prize given for lives of worthiness and service to God." After listening to a beautiful rendition of the sextet from Lucia, given by John S. Smith and company, the Indians were allowed to go into the baptismal room where again hearts were touched to almost the melting point. Here in this marvellous room it was explained to these visitors that the baptismal font was set upon the backs of twelve oxen representing the twelve tribes of Israel. In each corner of the room are three shining shafts of granite representing the twelve apostles, holding up the Celestial Kingdom of God. This was beautifully explained by Brother Jacobs. On the north wall of this room is the picture of Adam three years before his death, when he gathered all his righteous posterity together in the valley of Adam-ondi-Ahman and blessed them, prophesying what should befall his posterity even to the latest time. On the east wall is a picture of Moses after crossing the Red Sea, and after the murmurings of the children of Israel, when hordes of serpents came among them. Many of the people were bitten and died. Then the Lord told Moses to raise a serpent upon a stick and all who would look upon it would be healed, explaining that even as Moses raised the serpent upon the stick, so would Christ be raised up on the cross in the meridian of time, that they who would come unto Him might also live. In this marvellous picture is the thought of the serpent destroying by biting the flesh. How typical it is of Satan's attacks upon us today, destroying our lives of usefulness by tempting us to do wrong through the flesh. On the south wall is the picture of John the Baptist standing in the River Jordan, with Jesus by his side. His hand is raised toward heaven, ready to immerse the Christ in baptism. Then on the west wall is the picture of Joseph Smith and Oliver Cowdery humbly kneeling in prayer to God, asking for an explanation of 3rd Nephi, the 11th chapter, regarding baptism. John the Baptist is portrayed as standing with a hand on either head, restoring the Aaronic Priesthood, conferring upon these two men the power and authority, in the name of Jesus Christ, to baptize members into the Church of Jesus Christ. INDIAN BOYS AND GIRLS SHED TEARS OF JOY As these things were beautifully explained, tears of joy coursed down the faces of many present. I noticed many of the little Indian boys and girls gathered around the baptismal font, reaching out their fingers and gently and lovingly touching a nose or a horn of the oxen and then kissing their fingers. I imagined the things that were filling their souls with joy as the Spirit of God touched their little hearts and made them understand His goodness. Then the entire group sang with much feeling, "We Thank Thee, O God, for a Prophet," and "O My Father, Thou That Dwellest," etc. The meeting was dismissed and the visitors were taken to the lawns outside, where they were served ice cream and cake. For many years it has been my privilege to work with the youth of Zion, and I find they are anxious to learn of God and to do all in their power to push this great work forward. As I thought of these things, my mind went out to the five reservations in Montana, and I could imagine these boys and girls mingling with their associates, telling them of the things they had seen and heard in the Temple of our God; and telling them with a spirit and power that will cause a greater desire to come among them, to know of the Gospel and to learn of its ways. I can see much good coming from this visit, for I feel it will cause many of our Lamanite friends to see the light of the Gospel. Three days later, it was my privilege to meet Mr. Campbell and General Scott of the United States army in Cardston and talk to them about the visit. Mr. Campbell was very well satisfied with the results, feeling that he had accomplished the things that he had desired. Mr. Scott was very enthusiastic with the spirit possessed by the people in the "Mormon" communities of Alberta. How gratifying it is to know that, now when Mr. Campbell's official duties have been terminated by his retirement, his successor in the work is Elder Murdock, a "Mormon" brother who has filled an honourable mission. Surely the blessings of God promised Lehi's descendants are being fulfilled. And we, as Latter-day Saints, should be ready by the time they are ready, to do our part as they are being prepared to do theirs. A MARVELLOUS MISSION IN LIFE On January 3rd, I wrote to Jack Galbraith in Browning, Montana. From that letter I quote the following passages: You know that I am deeply interested in your people. I see the dawn breaking for them, and because of my knowledge of the Gospel, I see and understand their great mission in these latter-days. If the dear old Chief could but understand that it is the goodness of the Lord that has preserved the righteous seed of Joseph to perform a marvellous work in the latter-days, he would feel better. When you take the quilt to them, have a good talk with them. Tell the old Chief that I am his true friend and if he would but have power to be baptized, he would go into the Spirit World, a heralded Chief to stand as a servant of God and a redeemer of his people, fitted and prepared to teach unto those departed souls the true message of salvation; to warn and prepare their hearts for the work that shall surely be done in the Temples of our Lord. I am enclosing a letter for the Chief in with yours, then I will be sure he gets it. I do not believe you realize the mission you have in life, and I am very desirous to have you do all on your part to fulfil your life's mission, to lay up treasures in heaven which will be everlasting and enduring. Money will not buy happiness nor contentment nor that peace of mind which comes from service in the Church and Kingdom of God. You are being watched by hundreds of your people. Your actions speak far louder than words. Help your children to see the bigness of the Gospel, and hold out to them the blessings that come from the Priesthood of God. You know the joy you felt as we talked of your relatives and friends, and I shall never, nor can you forget the Spirit of the Lord that filled your heart as you talked to those 220 young boys and girls in the Temple, and you told them how to become winners of the greatest things God has for His children who serve Him. Your brother and friend, (Signed) A. M. Steed. In January of this year I received this interesting letter from my friend, Jack Galbraith, whose mother was a pure Blackfoot Indian, and the sister of Mountain Chief: "Dear Brother Steed: I received your letter and was indeed pleased to know that you still have the Lamanites at heart. I am very anxious to tell you what happened to the letters you wrote to Mountain Chief and me. I was in Browning the day the letter arrived and I worked a little ahead of your instructions. The quilt did not arrive at the time. After reading your letter I walked up to the little log cabin to interview the old Chief. There sat beside him two old Indians, all with their heads bowed, and they seemed to be in deep meditation. I shook hands with them. They seemed to be visiting. One was an Indian who lives away up on the northeastern part of the Blood Indian Reservation in Canada. It was quite a sight to see these three old men with heads bowed. On the other bed sat blind Walter. Close beside Mountain Chief sat, in a chair, a fine looking Indian woman, wife of the old Canadian Indian. I said to Walter, ‘Do you know me?’ He answered me, ‘No.’ I told him I had a letter I’d like to read to his father. ‘Walter, you interpret this just as I read it and see if either of you know who I am.’ I read the letter that you wrote me first, leaving out my name. Of course, soon as I mentioned the quilt for my cousin Walter’s wife, Walter said, ‘It is 100 Bears (the Indian name for me), and I also know who wrote the letter, because no other man I know of would bother to make Indians like us a quilt, unless it is Brother Steed.’” I wasn’t interrupted by anyone while I read your letter to me, and the half of your letter to Mountain Chief. I thought I’d ask him if he really understood it, and what it was all about. One of the old warriors, Bull Calf, who is blind and could neither speak nor write English, sat and listened with great interest. “I need no interpreter to understand that kind of a letter, I understand every word in English, 100 Bears. The man who wrote that letter is blessed with the Great Spirit,” he said. I thought this was remarkable. Brother Warner, the Indian, and I had the same kind of experience once before in Browning, some sixteen years ago. The interpretation of tongues was exhibited among the poor and lowly. I know these things happen. “THE DAY IS AWAKENING FOR THE INDIANS” Mountain Chief spoke up then, and said, “That letter sounds like the instructions we used to get from our leaders when I was a young lad. It does my heart good to hear it read, we need our Brother’s advice to each other. Tell Steed his heart is right, the advice is good, it will give me good ideas to talk to my people, when they need me. It is encouraging. I want to say more later to this man.” When he had finished talking, the Blood Indian rose to his feet and said, “Eighty winters have passed me and I know my God and this is the kind of message His representatives speak and deliver. I know this man is a member of the ‘many Wives Church.’” (This is the way these Indians have of expressing the name of our Church.) He went on speaking, “I, too, am impressed with this man’s letter. I’ve just decided after hearing 100 Bears explain these letters, that these ‘Mormons’ are good people. I am ready to be baptized into God’s church.” I asked him if he was sure he was converted and understood the Gospel. Before he could answer, this Indian woman (his wife), who is a Catholic, spoke up and said, “I give my consent for him to join.” I asked him what he thought of President Wood, at Cardston, and he said, “He’s a good man, strict, honest and not misleading. I am glad my friends are in favour of this good Church and enjoy hearing 100 Bears read these letters.” Your kind letters with good advice not only apply to me, they had a purpose and served it well. I feel that the day is awakening for the Indians. I am going to tell you now, Brother Merlin, that the one who has had the greatest influence on my life is my good wife, Susan Alder Hudson, whom I married, at the age of 24, in 1908, at Salt Lake City. She was of the Aetna Ward, Alberta Stake, and a daughter of George M. Hudson. I’ll now relate a few of the things which converted me, and why I came into the Church. I was baptized into the Catholic Church when I was seven years old, and was brought up with that in my mind until I was twenty-one years old. My mother died when I was seven years old. She was buried in a lonely spot, out in the hills, but I had always felt her presence near me. Soon I learned that I had parted with my best friend, my mother. I was disliked among the white folks in the community where I was brought up, because of my Indian blood, and I was disliked among the Indians because of my white blood. It was a hard predicament to be in. I soon thought that Catholicism was a hindrance to me. Naturally I was impressed to think of my mother (Last Kills—her maiden name), and she, poor soul, unable to speak, read or write English. I must find some way out for her. I kept this within myself, knowing my father was popular among the citizens and Indians. Once I asked him why he didn’t join the Catholic Church, and he answered me thus: “I have all the religion I need, which is the cream of all religions, and that is Masonry.” In the meantime I had such men as Bill (Wm. R.) Sloan and his brothers talk to me, besides what my wife had talked and lived, which was the Gospel taught by the Church of Jesus Christ of Latter-day Saints. I thought my father knew best, and I inquired as much as possible about Masonry. Finally, in 1912, I was initiated, but I wasn’t long in noticing that this order lacked the main spoke of the wheel, which was the women. Still no hopes for my mother’s salvation! I was not long deciding that my wife’s example and her true teachings of the Gospel were what I had been eight years looking for. With two years’ studying and trying to believe that Masonry was right, I was still not convinced. My wife taught me what the “Mormon” Church called the Word of Wisdom, and why there were such laws instituted in the Church. Now I began to see it was for improving the body as well as the mind. On the face of all this I could still see my father’s influence as a Mason and his example. It was heartrending for me to tell him that either he or the “Mormons” were wrong. I saw it was either my wife and family or my father and friends. It was a serious problem. At this point I became converted to prayer. EXEMPLARY CONDUCT OF HIS WIFE Still my wife in her meek way carried on with her prayers, paid some tithing, and lived an exemplary life which could not help being noticed. She was clean in mind and expression; this was getting a deep root in my mind. All of the time uppermost in her mind was her religion. I could write pages on the mental suffering I went through. I was becoming ill. I thought I’d go east to Chicago for nose and eye treatment, to be alone and away from the antagonisms which tormented me. Our family doctor, Doctor Lynn of Cardston, persuaded me to let him make a preliminary operation. What he at first believed would be a very simple operation proved so serious, and I was so weak from the loss of blood, that I was sure I was going to pass out of this life. At this time an experience came to me which fully convinced me of the truth of the Gospel. I made a pledge to my Maker that I would join the Church and live the truth. I recovered and walked several blocks to my wife. I confessed to her that I was fully converted, and we had prayer together. I was impressed to study the Book of Mormon, for I would be needed some day, and there would be plenty of work for me to do. On July 5th, 1914, I walked alone into the Tabernacle at Cardston and took a seat in the rear. It was the evening meeting, about 7 p.m., and I planned on seeing President Wood who at this time became a dear friend to me. Like the old Indian, I learned that “he seemed so earnest and sincere in all his dealings” and he treated me with such reverence. His strictness, his great ability for doing good, morning, noon and night, his attention to the sick, his presence in the homes and at funerals to console, and his wonderful exemplary life played a big part in my change of heart and in that of many others. He has remained up to the present time my close friend, and now is, in very deed, a brother. (Continued on page 363) the spiritual leader of a group of seekers after the Saviour of Men, though his time is mainly given to the business which he has built up. Recently, he was impelled again to inquire about "Mormonism;" he was ill at ease spiritually. He read into the teachings of the Lord the doctrine of an every-day religion, of a social system in which all should be brethren and sisters, caring for and being cared for by one another. Again he wrote to Salt Lake City; again, guided only by the printed word and by letters, he is pondering the truths revealed in the latter-days. Again, "Mormonism" seems to comply with the requirements laid down by the Master. We were the first Latter-day Saints he had seen. He asked for more visits. The course of his life had been turned by his casual knowledge of the Church of Jesus Christ of Latter-day Saints. Mr. Gagliano's experience is that of every honest, earnest, sincere seeker after eternal truth.—W. --- THE DAWNING DAY FOR THE LAMANITES (Concluded from page 359) I thought that after the meeting I'd ask President Wood to baptize me. Before the meeting commenced President Wood tapped me on the shoulder and said, "Good evening, Jack." I told him I wanted to see him after meeting. "I know what you want, Jack; you're ready for baptism." I saw in his eye that spiritual knowledge that you don't see in the eyes of ordinary men. I asked him when he could do the job. "Yes, Jack, you are ready for baptism, and I'll see the bishop about it." I protested, "No, I want you to baptize me." "I'll gladly do it," he said. I told him, "The sooner the better, after this meeting." After the meeting he came to me and said, "You go to the river at a certain cottonwood tree and I'll be there." I went there and waited on the banks of Lee's Creek and, to my surprise, Brother Wood, a choir, and a nice crowd came to see the first Lamanite in Canada thoroughly converted to the Gospel of Jesus Christ of Latter-day Saints, go into the waters of baptism. I must say that it was the happiest hour of my life, and the crowd made me feel overjoyed at their presence. They sang some beautiful hymns. What followed, Meelin, would fill a book. I hope I've been as mild as possible in relating the truth of my conversion into the Church. This is the first time in eighteen years I've made an attempt to write, in my own hand, my experience. I want to state that I kept the Word of Wisdom strictly two years before I was baptized, so I knew the fruits of "Mormonism" were good for the soul and essential to human understanding of God's kingdom. Wishing you and your family many happy days, Your brother, (Signed) J. J. Galbraeth, (Keepipokayo, alias 100 bears). —The Utah Genealogical and Historical Magazine for April, 1933. --- Who knows but what our trials are the circumstances that make us find our way home and back to God?—Levi Edgar Young.
Welcome to the Distance Education Centre Victoria. The Distance Education Centre Victoria (DECV) is Victoria’s leading virtual school. With over 4300 students from Foundation to Year 12, more students study at DECV than any other school in Victoria. We provide education to a range of students with varying personal circumstances who are unable to attend mainstream schooling. Central to our mission of providing flexible learning is a commitment to engagement and inclusion, regardless of each student’s background or circumstances. Our teachers are dedicated to building the confidence and resilience of each and every student. We take pride in helping our students to prepare for university, higher learning or employment beyond their schooling. At the end of the 2018 school year, DECV will have developed a comprehensive teaching and learning program which includes the re-writing of all our 137 subjects, throughout Foundation to Year 12, in DECVOnline - our online learning environment. This allows for flexible, self-paced learning, and is enhanced by multimedia and collaborative spaces, as well as access to a wide range of online resources and activities. Whatever your reason for considering an enrolment with us, I encourage you to read the information in this handbook, as well as the subject information available on our website. If you or your family would like to learn more about DECV and online learning, or have any other questions, our enrolment staff are here to help and provide confidential advice online, over the phone or in person. Best wishes for your studies with us, Bretton New Principal What is DECV? As the state’s leading virtual school, the Distance Education Centre Victoria (DECV) provides flexible learning opportunities to a diverse and unique cohort of Victorian Foundation - Year 12 (F-12) students. We foster a safe, engaging and inclusive learning environment by coupling state-of-the-art technology with innovative and adaptive teaching practices. Our teachers are trained and experienced in best-practice online learning techniques, providing students with regular interaction and individualised support. We are focused on challenging students to pursue excellence and achieve their individual learning goals. In 2016, DECV launched Australia’s first F-12 pedagogical model for online learning. We are a system leader in online, virtual and blended learning, constantly evolving our practice, based on the latest research. We adopt a whole-school approach to student wellbeing and have a resolute commitment to health promotion and child safety. We are dedicated to do all that we can in order to build the confidence and resilience of every student. What is it like learning by distance? Learning by distance is very different to mainstream schooling. DECV students have the flexibility to learn in a way that is not always possible in a “bricks and mortar” classroom. Students can work at their own pace to complete set learning activities, without distraction. They also collaborate with peers in our secure online environment. DECV students have access to a range of structured learning materials available online. Online classes and face-to-face seminars are held regularly in many subjects, providing further opportunities to connect and collaborate with peers. Teachers are available to speak to students in person, over the phone or online. The time needed to set aside each week depends on how many subjects are being studied. If enrolled in a full-time learning program, students can expect to be devoting the same amount of time to study as students in mainstream schools. DECV students set up their own weekly timetable and complete set weekly work. What will I learn? In Years F-10, the DECVs learning programs are in line with the Victorian Curriculum. In Years 11-12, the DECV follows the curriculum set by the Victorian Curriculum and Assessment Authority (VCAA) in our wide range of VCE subjects. Learning programs are organised into manageable segments. Students are guided through the subjects they are studying and provided with tasks to complete, which are then assessed by teachers and returned with feedback. Individual learning programs are developed where appropriate so all students, regardless of their age or ability, can be engaged and challenged. Will I require internet access? All students enrolling at the DECV require regular and reliable access to a computer/device and the internet. Students will need to login to the school's online portal to communicate and submit work online regularly. They are expected to participate in multiple areas of the online courses. This includes online lessons, forums and interactive activities. All student reports are only available online. The subject information pages on the DECV website have more information about the technological and other requirements for individual subjects. Please read these pages carefully before making subject selections. What role do parents/carers play? Teachers at the DECV understand the essential role parents and carers play in supporting and encouraging their children to learn. DECV teachers provide up-to-date learning and engagement information to parents and carers to support them to take an active role in their child’s education. The DECV Portal is an area on our website where parents, carers and supervisors can view information about a student’s progress at DECV. Once registered and logged in, parents/carers will be able to find details about the: - subjects the student is completing - recent work the student has submitted and the feedback received - teachers’ contact details - individual student reports. To access the DECV Portal, parents/carers need to enter a valid email address in the Primary Family Information section of the enrolment application. Contact with the DECV Family School Action Team can be made by emailing us at firstname.lastname@example.org Will textbooks need to be purchased? Some of our subjects require either digital or physical textbooks and other resources to be purchased. Our Resource List for 2019 is available at http://www.distance.vic.edu.au/resourcelist/ Our recommended textbook supplier is Campion Education. Campion Education 94 McEwan Road, Heidelberg West VIC 3081 Phone: 1300 433 982 Website: http://www.campion.com.au/ Campion delivers throughout Victoria and has provision for interstate and overseas delivery. It is important that textbooks are ordered as soon as possible upon enrolment. Textbooks can be ordered online at https://mi/cmpio.com.au/ebooklists The DECV school code is RWCX Our recommended electronic device and software supplier is JB Hi-Fi Education. Parents and students can browse a selection of devices and software on the JB Hi-Fi Education Portal. Students may choose to use alternative devices to those listed on the portal. JB Hi-Fi Education Portal Website: https://www.jbeducation.com.au/byod Password: decv2019 All orders through the Education Portal can be picked up at any Australian JB Hi-Fi store. Can I study a language other than English? Languages other than English are offered by the Victorian School of Languages (VSL). The VSL operates separately to the DECV, but many of our students choose to study languages with them. The VSL is also located at 315 Clarendon St, Thornbury. The languages application form is part of the DECV enrolment application. The form needs to be completed and submitted to the DECV along with the rest of your application. Once your enrolment with the DECV has been approved, we will send your language application to the VSL. The languages available include Arabic, Chinese, Classical Greek, French, German, Greek, Hindi, Indonesian, Italian, Japanese, Latin and Spanish. For face-to-face language classes provided at 40 VSL campuses around Victoria after hours, visit the VSL website www.vsl.vic.edu.au Will I need to sit exams? Students completing VCE Unit 3 and 4 studies are expected to sit the VCAA examinations. The VCAA General Achievement Test (GAT) will be held on 12 June 2019. All students studying a scored Unit 3 and 4 subject are expected to sit the GAT. The VCAA Extended Investigation exam will be held on 31 July 2019. Students studying the Extended Investigation VCE subject will be required to sit this exam. Please note the requirements stated on the pre-enrolment form. The VCAA Oral and Performance exams will be held from 7 October until 3 November 2019. Students studying any VCE subject that has a performance component will be required to sit this exam, which must be completed in Victoria. The VCAA end of year examinations will be held from 30 October until 20 November 2019. These are subject specific exams for students studying Unit 3 and 4 subjects. Some students may require Special Exam Arrangements and/or interstate/overseas examinations. VCAA have strict requirements for these arrangements and students/families will need to fill out application forms and supply supporting documentation. Further information will be sent to all students in Term One and re-sent in Term Three. Any questions can be emailed to email@example.com When can I start? The DECV operates on the same term dates as all other Victorian Government Schools. The dates for 2019 are: - Term 1: 30 January to 5 April - Term 2: 23 April to 28 June* - Term 3: 15 July to 20 September - Term 4: 7 October to 20 December The school year is broken into two semesters. For F-10 students, Semester One covers the first and second terms and Semester Two covers third and fourth terms. For 11-12 students, Semester One commences on 30 January and Semester Two commences on Monday 17 June, two weeks before the end of Term Two. F-10 Enrolments* For students enrolling in Years F-10, 2019 enrolments are open from 15 October 2018 until 26 July 2019. Students intending to start the school year at the DECV are encouraged to submit their enrolment application as early as possible to enable them to commence at the start of Term One. 11-12 Enrolments For Year 11 and 12 students, enrolment applications are open for the following periods. - Semester 1: - Enrolments open: Monday 15 October 2018 - Semester starts: Wednesday 30 January 2019 - Late enrolments will not be accepted after Friday 8 February 2019 - Semester 2: - Year 11 subject enrolments only - Enrolments open: Monday 29 April 2019 - Semester starts: Monday 17 June 2019 - Late enrolments will not be accepted after Friday 21 June 2019 Transfers from other schools may be accepted outside these dates in some circumstances. Please email firstname.lastname@example.org to contact the DECV enrolments team for more information. Who to contact? If there are questions or concerns about the enrolment process, please contact the DECV Enrolments Office. For subject selection advice, contact the relevant Year Level Coordinator. - Phone: (03) 8480 0000 - Free call: 1800 133 511 - Fax: (03) 9416 8487 - Email: email@example.com Eligibility criteria for direct enrolment with the DECV Who can enrol at the DECV? The DECV exists to meet the educational needs of Victorian students whose circumstances prevent them from accessing mainstream schooling. In order to directly enrol with us, students must meet the criteria of one of the five enrolment categories outlined below. Enrolment under each category requires documented evidence to be submitted with the application. The evidence required for each category varies and is outlined below. Enrolment applications without all of the required documentation cannot be reviewed or approved. Enrolment at the DECV is granted only for the duration of the school year. Students enrolled in 2018 who want to continue their enrolment with the DECV in 2019 are required to re-enrol for the 2019 school year by completing a new application form and supplying updated supporting documentation to the DECV. Students in mainstream schools Students wishing to maintain their current enrolment in a mainstream school and complete one, or (for 11-12 students) two subjects not available at their school, can apply by downloading the Students in Schools Application Form at [http://www.distance.vic.edu.au/students-enrolled-in-other-schools/](http://www.distance.vic.edu.au/students-enrolled-in-other-schools/). Alternatively, Year 11 and 12 subject applications can apply by completing an online enrolment. Pre-enrolment interview In certain circumstances, in order to optimise student engagement and learning outcomes, the DECV may deem that a face-to-face interview is required prior to approving an enrolment application. If this is the case, the student or their parent/carer will be contacted after they have submitted their application to schedule a time. This may result in a delay in the approval of the enrolment at DECV. CATEGORY 1: MEDICAL 1a. Medical (Physical) Students enrolling in this category must have a specialist diagnosis of a physical disability that restricts their ability to attend a mainstream school for a period of six months or more. Evidence required: - A School Referral Form The student’s most recent school must complete this form, outlining the student’s prior schooling experience, the supports that have been put in place for the student, and a recommendation regarding the suitability of an enrolment with the DECV. If the student is not at school, an Agency Referral Form is required confirming their situation. - A Practitioner Referral Form A medical practitioner in the relevant field must complete this form, outlining how the student’s diagnosis and/or treatment prevents their attendance at a mainstream school, how this may impact upon their ability to engage with study, and what continuing treatment or support will be provided. In situations where an agency is also involved with a student, applicants are asked to provide an Agency Referral Form. Requirements for all students 1. Victorian residency Enrolment at the DECV is subject to Department of Education and Training (DE&T) guidelines and is only available to Victorian residents. A resident is a person with a permanent residential address in Victoria. The onus is on the enrolling student’s family to provide proof of their residency. Victorian residents temporarily travelling overseas or interstate are eligible to apply. 2. Minimum enrolment period Enrolment is available to all eligible Victorian residents who are unable to access regular schooling for a minimum of two consecutive school terms or six consecutive months within a single school year. 3. Proof of age All students must submit their proof of age with their enrolment application. This can include a copy of a birth certificate, passport, or driver’s licence. In some circumstances, certified copies may be requested. 1b. Medical (Social/Emotional) Students enrolling in this category must have a diagnosed mental health or social emotional difficulty, meaning it is not possible for them to attend mainstream school for an extended period of time. The DECV recognises that an important prerequisite for successful engagement with education is the establishment of effective ongoing treatment and support for mental health and social emotional difficulties. Students must demonstrate they are receiving ongoing professional treatment and/or support for their condition/s. Evidence required: - A School Referral Form The student’s most recent school must complete this form, outlining the student’s prior schooling experience, the supports that have been put in place for the student, and a recommendation regarding the suitability of an enrolment with the DECV. If the student is not at school, an Agency Referral Form is required confirming their situation. - A Practitioner Referral Form A specialist practitioner (psychologist, psychiatrist, paediatrician or social worker) must complete this form, outlining all relevant primary conditions that prevent the student’s attendance at a mainstream school, any other issues that may impact on the student’s ability to engage with study, and summary details of the ongoing treatment and/or support plan prepared for the student. And/or - An Agency Referral Form Where there has been agency involvement in the care and support of a student (e.g. DHHS, Anglicare, Salvation Army) an Agency Referral Form must be completed by the relevant support worker. This should outline all relevant primary reasons why the student is unable to attend a mainstream school, any issues that may impact on the student’s ability to engage with study, and summary details of the ongoing treatment and/or support plan prepared for the student. CATEGORY 2: TRAVEL Students enrolling in this category must be Victorian residents who are unable to attend a mainstream school because they are travelling within Victoria, Australia or overseas. Travel may be required due to a parent’s work commitments or an extended family holiday. Following the period of travel, the student must recommence enrolment at a mainstream school in Victoria. Evidence required: - A letter from the student’s most recent school A letter from the student’s most recent school is required confirming: (a) the student was/is enrolled at the school; (b) the student has left or is leaving for the purpose of travel; and (c) the expected date that the student will return from travel and recommence enrolment at a mainstream school. - A letter from a parent’s/carer’s employer This needs to outline the period of employment where the parent/carer is expected to be out of Victoria and/or travelling. Or - A certified Statutory Declaration This needs to be signed in front of an authorised witness, and must detail: (a) the proposed itinerary of travel/work; (b) the nature of the travel (e.g. work or family holiday); and (c) the expected date of return. A Statutory Declaration is also required when travel is due to parent’s work and they are self-employed. Generally, it is expected that students who have been, or expect to be, based in a location interstate or overseas for a period of more than two years should enrol in a mainstream school in the location where they are based. Students who have been enrolled at the DECV under the Travel category for two or more years are required to seek approval from the DECV to continue their enrolment by resubmitting the supporting documentation. Approval is at the discretion of the DECV Principal. CATEGORY 3: SPORTS/PERFORMANCE Students enrolling in this category must have elite sporting or performing arts commitments which prevent them from attending mainstream school. These commitments may include performance or training commitments during regular school times, or extended interstate/international travel for training, competition or performance. Evidence required: - A letter from the student’s most recent school A letter from the student’s most recent school is required confirming: (a) the student was/is enrolled at the school; (b) the student has left or is leaving for the purpose of sports/performing arts commitments; and (c) that in their view, Distance Education is an appropriate alternative educational experience. - A letter from the sporting/performing arts association/agent A letter from the student’s sporting/performing arts association/agent is required confirming: (a) the student is involved or competing in elite level sport or performing arts; (b) the student’s performance/training schedule prevents them from attending mainstream school; (c) that in their view, Distance Education is an appropriate alternative educational experience and (d) contact details of a sporting/performance supervisor that is authorised to liaise with the DECV regarding student progress including name, organisation, phone, email and address. CATEGORY 4: DISTANCE Students enrolling in this category must be Victorian residents who live in a remote area where it is not feasible to travel to a local mainstream Government School. The requirements for enrolment in this category vary depending on the age of the student. The following minimum distance requirements must be met: Students over nine years of age who live more than 4.8 kilometres from the nearest Government School and the nearest bus route. Students under nine years of age who live more than 3 kilometres from the nearest Government School and the nearest bus route. Evidence required: - A letter from the closest Government School This letter should confirm the student's current residential address, and that this address meets the minimum distance requirements outlined above. CATEGORY 5: YOUNG ADULTS Students aged at least 17 years of age at the beginning of the semester they wish to commence in and less than 21 years on 1 January 2019, may enrol to study a VCE course of study with the DECV in 2019. Evidence required: - A Young Adult Referral Form This form needs to be completed by either: a) the student’s previous school; b) a treating specialist medical practitioner; or c) an agency working with the student or their family. This form should outline the student’s prior schooling experience, the supports that have and will be put in place for the student and a recommendation regarding the suitability of an enrolment with the DECV. Students enrolling in this category are also required to provide proof of age such as a birth certificate, passport or driver’s licence, and must complete the Student Enrolment information page in their application, outlining why they feel the DECV and online learning is the most appropriate educational option for them rather than studying at a mainstream school, TAFE or other educational provider. Primary Enrolments It is important that all F-6 students visit the DECV with their parents for an interview to meet their teachers and have their learning needs assessed so we can select an appropriate individual learning program for the student. Before the interview please complete and return your child’s enrolment forms (including any required referrals, reports and supporting documentation) to the DECV Enrolment Office. Once processed, you will be contacted in order to arrange a mutually convenient time to meet the F-6 Coordinator and (if possible) your child’s future class teacher. For the interview please bring: - your child(ren) - your child’s most recent school report. At the conclusion of the interview, each child enrolled will be given learning materials that are set at an appropriate level. If it is not possible to physically visit the DECV, it is important that you speak to the F-6 Coordinator and send a copy of your child’s last school report with your enrolment application. This will ensure that an appropriate learning program can be selected to suit your child’s individual needs. Children commencing primary school in Foundation (Prep) must be five years of age or older by 30 April of the year they start school. Victorian children between the ages of six and 17 must be enrolled in a school. Are there any social opportunities at DECV? DECV’s Student Voice team is an excellent way to get involved with DECV, connect with other students and improve your confidence and teamwork skills. Student Voice runs a number of projects throughout the year such as the Jumper Design Competition and the Social Entrepreneurs’ Club, amongst other things! This is a space for you to have a voice in your school. DECV also offers a variety of social opportunities for students. Our Games Club is very popular with students and is celebrated on Games Day. We offer an annual camp for students in Years 7-12, subject seminars as well as Connect with Distance days each term so that students can meet their peers and build friendships. To find out more, go to the Student Voice page on DECV online. If you would like to join the team, please send an email letting us know your name and year level to firstname.lastname@example.org DECV Online Acceptable Use Agreement This agreement applies to all students undertaking study at the DECV. It is a condition of enrolment that the Student Enrolment Agreement on page 19 is signed, agreeing to this. When engaging with my school work, either at the Distance Education Centre Victoria or at home, I agree to be a safe, responsible and ethical user at all times by: - respecting others and communicating with them in a supportive manner; never writing or participating in online bullying (e.g. forwarding messages and supporting others in harmful, inappropriate or hurtful online behaviours) - protecting my privacy; not giving out personal details, including my full name, telephone number, address, passwords and images - protecting the privacy of others; never posting or forwarding their personal details or images without their consent - contacting a teacher or a trusted adult if I personally feel uncomfortable or unsafe online - contacting a teacher or a trusted adult if I see others participating in unsafe, inappropriate or hurtful online behaviours - contacting a teacher or a trusted adult if someone writes something I don’t like, makes me and/or my friends feel uncomfortable or requests information from me I know is private - carefully considering the content that I upload or post online; knowing that this is a personal reflection of who I am and what people may think of me - investigating the terms and conditions of use of the websites I sign up to as part of the learning program (e.g. age restrictions, parental consent requirements) and if my understanding is unclear seeking further explanation from a trusted adult - confirming that I meet the stated terms and conditions and completing the required registration processes - abiding by copyright and intellectual property regulations; requesting permission to use images, text, audio and video and citing references where necessary - not downloading unauthorised programs, including games while using the school’s ICT equipment - not interfering with network systems and security, the data of another user or attempting to log into the network with a user name or password of another student, parent/carer or school staff member - not revealing my password to anyone including teachers or administrators at the school - using social networking sites for education purposes and directed by the school - not deliberately entering or remaining in any site that has obscene language or offensive content. In addition, when I use my personal mobile devices (including my phone) while engaging with my school work I agree to be a safe, responsible and ethical user at all times by: - respecting the privacy of others; only taking photos or recording sound or video when others are aware and formal consent has been provided as part of an approved learning program - respecting others and communicating with them in a supportive manner, including outside school hours and in social contexts by not making harassing phone calls/text messages or forwarding on messages - obtaining appropriate (written) consent from individuals who appear in images or sound and video recordings before forwarding them to other people or posting/uploading them to online spaces. Privacy Information for Parents and Carers During the ordinary course of your child’s attendance at Distance Education Centre Victoria (DECV), school staff will collect your child’s personal and health information, when necessary, to educate your child and/or to support your child’s social and emotional wellbeing or health. Such information will also be collected when required to fulfil a legal obligation, including duty of care, anti-discrimination law and occupational health and safety law. If that information is not collected, the school may be unable to provide optimal education or support to your child or fulfil those legal obligations. For example, health information may be collected through the school’s social workers, psychologist or Student Wellbeing Coordinators. If your child is referred to a specific health service at school, such as a Student Wellbeing Coordinator, or school-engaged psychologist, the required consent will be obtained. Our school may use online tools, such as apps and other software, to collect and manage information about your child. When our school uses these online tools, we take steps to ensure that your child’s information is secure. These online tools enable our school to efficiently and effectively manage important information about your child and also to communicate with you. If you have any concerns about the use of these online tools, please contact us. School staff will only share your child’s personal or health information with other staff who need to know, enabling the school to educate or support your child, or fulfil a legal obligation. When our students transfer to another Victorian government school, personal and health information about that student will be transferred to the next school. Transferring this information is in the best interests of our students and assists that next school to provide optimal education and support to students. In some limited circumstances, information may be disclosed outside of the school (and outside of the Department of Education and Training). The school will seek your consent for such disclosures unless the disclosure is allowed or mandated by law. Our school values the privacy of every person. When collecting and managing personal and health information, all school staff must comply with Victorian privacy laws. For more information about privacy, including how to access personal and health information held by the school about you or your child, see DET’s information privacy policy at: http://www.education.vic.gov.au/Pages/privacypolicy.aspx The term “staff” refers to principals, teachers, Learning Advisors, Student Wellbeing Coordinators, social workers and any other allied health practitioners, as well as all other staff at our school. This includes employees, agents and service providers (contractors) of the Department, whether paid or unpaid. How do I enrol? In order to enrol at the DECV, the following four tasks must be completed: 1. Identify an enrolment category Students must identify which category they intend to enrol under, as outlined on the previous pages. 2. Gather supporting documentation Different supporting evidence is required, depending on the enrolment category. Referral forms can be found at the end of this Enrolment Handbook. Schools, specialist medical practitioners and outside agencies can return completed referral forms to the student so they can submit all enrolment documentation at once. Alternatively, they can send them directly to the DECV. All enrolment applications require the necessary supporting documentation: - Completed Referral Forms/Evidence (see the Eligibility Criteria section for information about which referrals/evidence are required) - A copy of the student’s Immunisation History Statement from the Australian Immunisation Register (if enrolling in a Primary Level F-6) - A copy of a Concession or Health Care Card (if claiming a concession). New enrolment applications also require: - A copy of the student’s most recent school report (important enrolment may be delayed if a report is not provided) - Proof of student’s age (e.g. birth certificate or passport) - A copy of the student’s most recent NAPLAN student report (if available) - A copy of the student’s VCE details transcript (if applicable; for students who have already completed a VCE/VET subject at another school). 3. Complete the Enrolment Application Form All students enrolling need to complete the 2019 Enrolment Application Form, including the Subject Selection and Personal Details sheets relevant to the year level they are applying for. - Completed 2019 Enrolment forms - Completed Subject Selection forms (Years 7-12 only). 4. Send your completed application to the DECV. Once completed, the application needs to be sent to the DECV enrolments team. There are a number of ways this can be done: 1. Email: Scan and email the completed application to email@example.com Please note: email attachments must be less than 10MBs per email. 2. Fax: Fax the completed application to: (03) 9416 8487 3. Post: Post the completed application to: Enrolments Office Distance Education Centre Victoria 315 Clarendon Street Thornbury, VIC, 3071 4. In person: Drop off the completed application in person between 9am-4pm during term times at our Thornbury address: 315 Clarendon Street Thornbury, VIC, 3071 What will happen once you submit your application? Step 1 – Determining eligibility Once your enrolment application has been submitted, it will be reviewed by our Enrolment Officers in order to ensure you meet our eligibility criteria. If it is determined that further evidence is required in order for the enrolment to be processed, you or your parents/carers will be informed of this. Complex enrolment cases will be referred to our Enrolment Advisors who will contact you or your parents/carers directly. Step 2 – Confirming an appropriate learning program Once we have determined that you meet our enrolment eligibility criteria, we will then consider your individual circumstances and determine the most appropriate learning program for you. It may be necessary for us to call you or your parent/carer in order to explain/negotiate a program. Step 3 – Providing online access Once your learning program has been determined, your details will be entered onto our system and a Student ID number will be generated for you. Once your details are entered, it takes another 48 hours for an email to be sent out confirming your enrolment. Primary Family Information Primary Family is considered the family the student mostly lives with. Enrolment contact details may be used by DECV to communicate general information regarding the DECV and its programs. ### Parent/Carer 1 (Parent/Carer 1 will be the main contact) \| Title: \| Dr \| Mr \| Mrs \| Ms \| Miss \| \|--------\|----\|----\|-----\|----\|------\| \| Family Name \| ________________________________ \| \| First Name \| ________________________________ \| \| Carer’s Relationship to Student: (tick one) \| \| □ Parent \| □ Step-Parent \| □ Foster Parent \| □ Grandparent \| □ Host Family \| \| □ Relative \| □ Friend \| □ Self \| □ Other (please specify) ___________________________ \| \| Home Phone Number \| ________________________________ \| \| Work Phone Number \| ________________________________ \| \| Mobile Phone Number \| ________________________________ \| \| Email Address (required in order to register for access to the DECV Portal) \| ________________________________ \| ### Parent/Carer 2 \| Title: \| Dr \| Mr \| Mrs \| Ms \| Miss \| \|--------\|----\|----\|-----\|----\|------\| \| Family Name \| ________________________________ \| \| First Name \| ________________________________ \| \| Carer’s Relationship to Student: (tick one) \| \| □ Parent \| □ Step-Parent \| □ Foster Parent \| □ Grandparent \| □ Host Family \| \| □ Relative \| □ Friend \| □ Self \| □ Other (please specify) ___________________________ \| \| Home Phone Number \| ________________________________ \| \| Work Phone Number \| ________________________________ \| \| Mobile Phone Number \| ________________________________ \| \| Email Address (required in order to register for access to the DECV Portal) \| ________________________________ \| ### List any other family members attending the DECV in 2019 \| Full Name/s \| Relation \| \|-------------\|----------\| \| • \| • \| \| • \| • \| \| • \| • \| \| • \| • \| ### Student Emergency Contacts This section should include emergency contacts other than the Primary Carer. List in order of importance. \| Name \| Relationship \| Telephone Contact \| \|------\|--------------\|-------------------\| \| \| \| \| \| \| \| \| ## Additional Family Details ### Parent 1/Carer 1 \| Field \| Details \| \|--------------------------------------------\|-------------------------------------------------------------------------\| \| Family Name \| \| \| First Name \| \| \| Occupation \| \| \| Current Employer (if applicable) \| \| \| In which country were they born? \| \| Does the Parent 1/Carer 1 mainly speak a language other than English at home? - [ ] No, English only - [ ] Yes, other - please specify What is the highest year of primary or secondary school the parent/carer has completed? - [ ] Year 12 or equivalent - [ ] Year 10 or equivalent - [ ] Year 11 or equivalent - [ ] Year 9 or equivalent or below What is the level of the highest qualification the parent/carer has completed? - [ ] Bachelor Degree or above - [ ] Advanced Diploma/Diploma - [ ] Certificate I to IV (including trade certificate) - [ ] No non-school qualification What is the current occupation of the parent/carer? Please select the occupation group letter from the list below. \| Group \| \| \|-------\|---\| ### Parent 2/Carer 2 \| Field \| Details \| \|--------------------------------------------\|-------------------------------------------------------------------------\| \| Family Name \| \| \| First Name \| \| \| Occupation \| \| \| Current Employer (if applicable) \| \| \| In which country were they born? \| \| Does the Parent 2/Carer 2 mainly speak a language other than English at home? - [ ] No, English only - [ ] Yes, other - please specify What is the highest year of primary or secondary school the parent/carer has completed? - [ ] Year 12 or equivalent - [ ] Year 10 or equivalent - [ ] Year 11 or equivalent - [ ] Year 9 or equivalent or below What is the level of the highest qualification the parent/carer has completed? - [ ] Bachelor Degree or above - [ ] Advanced Diploma/Diploma - [ ] Certificate I to IV (including trade certificate) - [ ] No non-school qualification What is the current occupation of the parent/carer? Please select the occupation group letter from the list below. \| Group \| \| ### Parental Occupation Group Codes (Choose the code that best reflects your current situation) - Group A: Senior management in large business organisation, government administration and defence, and qualified professionals - Group B: Other business managers, arts/media/sportspersons and associate professionals - Group C: Tradesmen/women, skilled office, sales and service staff - Group D: Machine operators, hospitality staff, assistants, labourers and related workers - Group N: If the person has not been in paid work in the last 12 months Note: If person is not currently in paid work but has had a job in the last 12 months or has retired in the last 12 months, please use the person's last occupation. Medical Information Is the student deaf or hearing impaired? ☐ No ☐ Yes Is the student blind or vision impaired? ☐ No ☐ Yes Has the student been diagnosed with ASD/Asperger’s? ☐ No ☐ Yes Does the student have an intellectual disability? ☐ No ☐ Yes Does the student have a physical disability? ☐ No ☐ Yes Does the student have a severe behavioural disorder? ☐ No ☐ Yes Does the student have a severe language disorder? ☐ No ☐ Yes Does the student have a diagnosed mental health condition? ☐ No ☐ Yes (please specify): ____________________________ Does the student have a history of allergies? ☐ No ☐ Yes (please specify): ____________________________ Has the student been diagnosed as at risk of anaphylaxis? ☐ No ☐ Yes If yes, please include a copy of the student’s Individual Anaphylaxis Management Plan (IAMP) and a colour copy of their ASCIA Action Plan for Anaphylaxis. A blank template can be found at [www.distance.vic.edu.au/iamp/](http://www.distance.vic.edu.au/iamp/). Students attending a DECV approved activity who require an EpiPen® must provide their own Adrenaline Autoinjector and report to the supervising teacher upon arrival. Has the student been diagnosed with any other condition? ☐ No ☐ Yes If Yes, what is the name of the condition/s? _______________________________________________________ Are there any other medical issues the DECV should be aware of? _______________________________________ _____________________________________________________ Provide details of any assessments undertaken by the following specialists (Provide copies or use an additional page if necessary.) \| Specialist \| Name of Specialist \| Year \| \|------------------\|--------------------\|------\| \| Paediatrician \| \| \| \| Optometrist/Ophthalmologist \| \| \| \| Audiologist \| \| \| \| Psychologist \| \| \| \| Psychiatrist \| \| \| \| Speech Therapist \| \| \| \| Occupational Therapist \| \| \| \| Other \| \| \| Has your child ever received support from any of the following programs or services? ☐ Program for Students with Disabilities (PSD) ☐ Public Hospital Education Setting ☐ Home Based Education Support Program (HBESB) ☐ Visiting Teacher Service ☐ DHHS ☐ Child First ☐ Child and Adolescent Mental Health Service (CAMHS) ☐ DET Social worker ☐ DET Psychologist ☐ DET Speech Pathology ☐ Navigator ☐ Lookout Centre ☐ Other (Please specify): _______________________________________________________ Please list people who can be contacted to support your child’s enrolment at the DECV (e.g. teacher, counsellor, psychologist etc.) \| Name \| Position \| Phone \| \|---------------\|----------\|-------\| \| \| \| \| \| \| \| \| \| \| \| \| Parent/Carer Rights and Responsibilities Parent/Carer Rights All parents/carers have the right to: - meaningful feedback from and communication with teaching staff - be notified of any pertinent information relating to the student - access any information regarding the student within the confines of any legal requirement. Parent/Carer Responsibilities All parents/carers are responsible for: - liaising with members of the school community in a safe, positive and respectful manner - providing the school with all relevant information pertaining to the student - working with the school in a variety of forums and responding to communications in a timely manner - supporting students to engage and participate in their learning program and the wider school community - ensuring their child submits work in accordance with the prescribed or negotiated submission timetable. --- Consent to Access Student Records and Information The DECV aims to work together with students and their families to provide the best possible teaching and support. We use the information we collect on each student in order to establish supports that are required and to provide students with appropriate services. I give consent for the DECV to access and share any existing relevant personal or health information with specialist practitioners or agencies that have been listed in this enrolment application. --- Consent to use student work: At times, DECV staff may wish to publish student work outside the DECV online learning environment for informational and/or promotional purposes. (It is DECV policy to only publish the first name of students.) You will be informed if, when and where your work will be published. I give permission to share my work on the DECV external website, on-site displays, printed materials, and in our e-newsletter. Signature __________________________________________ Date: ______ / ______ / ______ Student Signature __________________________________________ Date: ______ / ______ / ______ Parent/Carer (if student under 16) --- Parent/Carer Agreement - I accept and agree to the parent/carer rights and responsibilities outlined on this page. - I will support my child to complete and submit their work regularly in accordance with the submission schedules for each subject and will help them to contact the DECV if situations arise which may prevent this. - I accept the DECV will initiate the Student Management Action Plan (SMAP) for students who do not submit work in accordance with their submission schedule. This may result in my child’s enrolment being reviewed and/or cancelled. - To the best of my knowledge, there are no Family Court orders that would prevent this enrolment from being processed. - I declare that the information provided on this application is, to the best of my knowledge, true and accurate. - I understand that my child will require reliable internet access in order to access their learning. Please tick one of the following options: - [ ] I am applying for a full enrolment with the DECV - [ ] I am applying for a shared enrolment with the DECV and an appropriate mainstream school. (Please have the school complete the shared enrolment details on page 39.) Signature __________________________________________ Date: ______ / ______ / ______ Parent/Carer/Legal Guardian Payment of Fees The DECV sets a materials and services charge for students enrolling with us. This charge is used to cover the cost of learning materials, online curriculum access and other consumables. For students enrolling directly with us, the charge is levied equally, regardless of the number of subjects undertaken. Current Concession or Health Care Card holders are entitled to receive a $30 discount on the full materials and services charge. There is no discount if enrolling for one semester only. If applying for a concession rate, a photocopy of your current Concession or Health Care Card must be attached to this application. Students must list an Australian postal address on their enrolment application. Standard rate: $160 per year; $80 per semester Concession rate: $130 per year; $80 per semester All payments are GST free. Payment can be made by cheque, money order, or credit card. Cheques or money orders should be made out to the “Distance Education Centre Victoria” and attached to this form. Credit card provision is available below. No cash should be sent through the mail. Payment by instalments may be considered in certain circumstances. To discuss instalment options please contact the Bursar. Credit Card Payments Please debit my: □ Visa □ MasterCard Name on Card Credit Card Number Expiry Date (mm/yyyy) $ Amount Signature Date (dd/mm/yyyy) Refund Policy Applications for refunds will be accepted if they are received in writing at least 7 days prior to the date of the course commencement. A standard handling fee of $30 will apply. After the course has commenced, no refunds will apply. The DECV cannot accept responsibility for changes in personal circumstances after instruction has commenced. Can I receive a Government Allowance while at the DECV? Students and their parents/carers may be eligible for government allowances while studying at the DECV. Centrelink administers student assistance payments for Youth Allowance, Austudy, Abstudy, Assistance for Isolated Children (AIC) and Pensioner Education Supplement (PES) for eligible students. Parents or students who hold a Commonwealth Pensioner Health Benefits Card, a Commonwealth Health Care Card or a Commonwealth Health Benefits Card on the first day of Term One may be entitled to some Government allowances. These allowances are also payable to legal guardians, foster parents and T&PI pensioners. If you believe that you are entitled to any Commonwealth Government payments, contact your Centrelink office: - Family and Parents Line 136 150 - Youth and Student Line 132 490 - Abstudy Line 132 317 - Assistance for Isolated Children 132 318 (Youth and Students Line) Students and their parents/carers should be aware that the DECV is required by law to supply an audit report to Centrelink each term on student enrolment status and work submission rates. Students receiving Centrelink payments who fail to submit work according to the prescribed timetable may be reviewed by Centrelink, which may affect their entitlements and create a debt to repay. We recommend speaking to Centrelink directly for further details or visiting the Centrelink website for more information: [www.humanservices.gov.au](http://www.humanservices.gov.au) Subject selection prior to enrolment is not compulsory. Once your enrolment has been processed, you will be contacted by your Learning Advisor. Your Learning Advisor will work with you for the rest of the year. He/she will begin with an intake interview and help you to complete Launch Pad (orientation/assessment). Following that, your Learning Advisor will work with you to choose your subjects and a subject load that suits you. If you have particular needs, please complete the “Requested modifications to a standard learning program” box and your Learning Advisor will be in touch after your enrolment to discuss this with you. Year 9 Subject Information - A standard full-time learning program consists of six subjects in each semester. They will also complete the Launch Pad orientation program. - Subject descriptions for all subjects are available online at [www.distance.vic.edu.au/subjects](http://www.distance.vic.edu.au/subjects) - Students are required to choose one elective subject per semester. If enrolling for the whole year you must enrol in one Arts and one Technology subject. I am intending to enrol for: □ Semester One only □ Semester Two only □ Both Semester One and Semester Two \| Subject \| All Year \| Office use only \| \|--------------------------\|----------\|-----------------\| \| English 301/302 \| ✓ \| \| \| Mathematics 343/344 \| ✓ \| \| \| Science 355/356 \| ✓ \| \| \| Health & Physical Education 415/416 \| ✓ \| \| \| Humanities \| \| \| \| Civics and Citizenship 404 (Semester 1) \| ✓ \| \| \| Geography 311 (Semester 2) \| \| \| The subjects listed below run for one semester only. You cannot complete the same subject in both Semester One and Semester Two. \| Subject Group \| Subject \| Semester 1 \| Semester 2 \| \|--------------------------\|--------------------------\|------------\|------------\| \| Electives \| \| \| \| \| Arts \| Media 327 \| \| \| \| \| Dance 428 \| \| \| \| Technology \| Design Technology (Food) 332 \| \| \| \| \| Digital Technology 425 \| \| \| Subjects studied elsewhere: Please note any additional subjects that you intend to undertake outside of the DECV, including subjects at other schools, VET subjects, or subjects studied at the Victorian School of Languages: --- Requested modifications to standard learning program: Please outline any requested modifications to the standard learning program outlined above. Your Learning Advisor will be in contact to discuss this prior to your enrolment. VSL Distance Education Courses Available for Students in Years F - 12 Note: Students who are unsure of the level at which they should enrol should ask for a placement test. Note: FL = First Language SL = Second Language SLA = Second Language Advanced Arabic: VCE Units 1 - 4 Chinese (SL): Year 7 - 10, Certificate in Applied Language, VCE Units 1 - 4 Chinese (FL): VCE Units 1 - 4 Chinese (SLA): VCE Units 3 - 4 Classical Greek: Accelerated 1 & 2, VCE Units 1 - 4 French: Year 6 - 10, Accelerated 1 & 2, Certificate in Applied Language, VCE Units 1 - 4 German: Year 6 - 10, Accelerated 1 & 2, VCE Units 1 - 4 Greek: Year 6 - 10, VCE Units 1 - 4 Hindi: VCE Units 1 - 4 Indonesian (SL): Year 6 - 10, Accelerated 1, VCE Units 1 - 4 Indonesian (FL): VCE Units 1 - 4 Italian: Year 7 - 10, Accelerated 1, VCE Units 1 - 4 Japanese (SL): Year 7 - 10, VCE Units 1 - 4 Japanese (FL): VCE Units 3 - 4 Latin: Year 7 - 10, Accelerated 1 & 2, VCE Units 1 - 4 Spanish: Year 6 - 10, Accelerated 1 & 2, VCE Units 1 - 4 Certificate in Applied Language is particularly suitable for VCAL students. Comments - Year 7 language courses assume little or no previous secondary language study. - Year 8 language courses are suitable for students who have completed approximately one year of secondary language study. - Year 9 language courses are suitable for students who have completed approximately two years of secondary language study. - Year 10 language courses are suitable for students who have completed approximately three years of secondary language study. - Accelerated Language Courses Level 1 involve a faster moving program for older beginners with little or no previous knowledge of the language. They lead to the Year 9 courses or to Accelerated Language Courses Level 2 and in some VCE languages. - Accelerated Language Courses Level 2 follow on from Accelerated Language Courses Level 1 to complete a two year pre-VCE program. - Certificate in Applied Language Courses assume little or no previous language study. - VCE Unit 1 is suitable for students who have completed approximately four years of secondary language study. VCE Unit 2 follows on from VCE Unit 1. - VCE Unit 3 follows on from VCE Unit 2. For credit at VCE level it must be combined with VCE Unit 4. - VCE Unit 4 follows on from VCE Unit 3. For credit at VCE level it must be combined with VCE Unit 3. Please note: The VSL also offers face-to-face classes in 40 languages in Years F-12 held out of school hours in 29 metropolitan and several country centres. For more information visit the VSL website: [www.vsl.vic.edu.au](http://www.vsl.vic.edu.au).
OTHER CF DIGESTIVE SYSTEM PROBLEMS Sometimes people with cystic fibrosis* (CF) can have several digestive system problems beyond malabsorption.* (See the CF FEP module Managing Nutrition and Digestive Problems for more information on malabsorption.) “Other CF Digestive System Problems” gives you an overview of different types of digestive problems. Your CF health care team can help you watch for other digestive or gastrointestinal* problems in your child. If your child develops one of these problems, the CF health care team will give you more detailed information. MECONIUM ILEUS Meconium* is the name for the first bowel movements* that a baby has after birth. A baby who has meconium ileus* has a blockage* in the intestines* from thick, abnormal meconium. The blockage prevents the baby from passing a stool. Meconium ileus is usually caused by malabsorption. Meconium ileus starts even before the baby is born. The small intestine* can rupture (also called perforation), which can be very serious. Not only is the small intestine blocked, but the colon or large intestine* also may be smaller than normal. Infants with meconium ileus will often need surgery soon after birth. CONSTIPATION Constipation* means a person does not have stools or bowel movements regularly. Most people have at least one stool a day. A person who is constipated may not have a bowel movement for several days. Abnormal stool or reduced motion of the muscles in the walls of the intestines can cause constipation. People can get constipated from not drinking enough fluids or not getting enough fiber in their diets. Children with CF may have constipation more often than other children. WATCH AND DISCOVER† Constipation Problems Watch for signs* and symptoms* of constipation in your child. - Fewer bowel movements than normal—usually less than once a day - Harder stools than usual - Stomach pain or cramps - Decreased amount of stool compared to that in the past - A feeling of fullness in abdomen* (or feeling full or not eating because of a distended or bloated belly) A person with constipation can have stools that look like diarrhea.* This is not really diarrhea but stools that are more liquid that are leaking around a stool blockage. THINK AND ACT† Managing Constipation Talk to your doctor if you think your child is constipated. Your doctor may recommend medicines that will help soften the stool and move it through the bowels* (intestines). Some people with CF need to use laxatives* or stool softeners regularly for constipation. Other ways to help prevent constipation are to have your child: - Drink enough fluids (at least 4 to 8 cups or 32 to 64 ounces per day). - Eat a high-fiber diet (your CF dietitian can give you tips on high-fiber foods, such as raw vegetables; fruit with skin, including pears and prunes; and whole wheat or high-fiber bread) - Use fiber such as psyllium (one brand name is Metamucil®). Talk with your child’s doctor or dietitian to see if psyllium might help. Drink at least 2 cups of water with the psyllium. - Get regular exercise. Exercise can help move stool through the intestines. (See “Exercise and CF” in Appendix 3.) DISTAL INTESTINAL OBSTRUCTION SYNDROME Distal intestinal obstruction syndrome (DIOS)* is a more severe form of constipation and is caused by a buildup of stool and mucus* in the intestines. Children with CF can have DIOS with malabsorption or with dehydration.* Sometimes DIOS can happen when children are ill with a respiratory infection* and are less active. Distal intestinal obstruction syndrome is also called meconium ileus equivalent* because it is similar to meconium ileus. Your child’s doctor will treat DIOS by giving your child laxatives and fluids to flush out the blockage. Some children need surgery if the intestine ruptures or cannot be cleared. You can help prevent † See the CF FEP module Becoming a CF Manager to learn more about Watch and Discover and Think and Act. See CF Words to Know Glossary. DIOS by managing malabsorption and taking proper steps to prevent constipation. Some children will benefit from taking laxatives or stool softeners daily. INTUSSUSCEPTION* Intussusception is another problem that can cause a blockage in the intestine. Intussusception is when one part of the intestine collapses or telescopes into another part. Most often the last part of the small intestine (the ileum) slides into the colon. Children with CF are at greater risk of intussusception than children without CF. Intussusception is not a common problem and is usually seen in very young children during their first few years of life. Intussusception can happen as a complication of a viral infection, malabsorption, or DIOS. Symptoms include severe cramps or abdominal pain (the pain may come and go), vomiting, and decreased number of stools. Blood may be seen in the stool. Intussusception is treated by giving a special enema done with X-ray imaging. If this does not relieve the blockage, the child may need surgery. GASTROESOPHAGEAL REFLUX (GER) Gastroesophageal reflux (also called GER or GE reflux) is a condition in which stomach acid and partially digested food can flow from the stomach back up into the esophagus (the tube from the throat to the stomach). You may also hear this called heartburn (or acid reflux). If acid reflux happens often it can cause problems. Many infants have GER, but can outgrow it. Gastroesophageal reflux can appear at any age. In CF, coughing can make reflux worse. Acid reflux happens more at night when a person is lying down or when a person’s stomach is full. WATCH AND DISCOVER Symptoms of GER Watch for these symptoms of GER: - Repeated vomiting or spitting up - Frequent burping - Stomachaches or heartburn - Decreased appetite or feeling full after eating only a small amount - Slow or no weight gain from not wanting to eat, or losing food and nutrients with vomiting Gastroesophageal reflux can also sometimes lead to coughing and wheezing. Some children may have only one symptom of GER; other children may have several. THINK AND ACT Managing GER If you think your child may have GER, talk with your doctor. With your doctor, you can decide if your child needs tests or medicine for GER. Several medicines are used to treat GER, including acid blockers to reduce stomach acid so it does not irritate the esophagus, and other medicines to help the stomach empty better. Besides giving medicines, you can help control reflux by having your child: - Eat smaller amounts more often—your CF dietician can help you plan small meals that are high in calories. - Avoid spicy foods and foods that have a lot of acid such as tomatoes, citrus fruits, or fruit juices. - Avoid carbonated drinks such as sodas, and drinks with caffeine (colas, tea, and coffee). - Do airway clearance before eating. - Avoid lying down right after eating a meal (eat the evening meal at least 3 hours before bedtime). - Play quietly for 30 minutes after eating. - Sleep with the head of his or her bed raised 6 to 8 inches, especially if problems with acid reflux continue at night. IMPAIRED GLUCOSE TOLERANCE AND CF-RELATED DIABETES (CFRD) Carbohydrates from the food we eat are broken down into glucose for the body to use as energy. Glucose enters the blood and the blood glucose (sugar) level goes up. A rise in blood glucose level sends a signal to the pancreas to release insulin into the blood. Insulin is a hormone that helps lower blood glucose levels by moving glucose into the cells to be changed into energy. Insulin also plays a role in how cells in the body use fat and protein. Sometimes a person with CF can have too little or no insulin to move glucose into the cells for energy. In fact, over time the pancreas of some people with CF may actually stop making insulin. Blood glucose problems in people with CF can vary. Many older children and adults with CF will have some problems with high blood glucose levels, and 15 percent or more will have more severe glucose problems. The problem may be worse when a person has an infection or is taking certain medicines, such as corticosteroids. Blood glucose problems can make it harder for people with CF to gain weight and also make them more prone to lung infections. See CF Words to Know Glossary. When a person has a mild problem with blood glucose levels, it is called impaired glucose tolerance. People with impaired glucose tolerance may have symptoms of high or low blood sugar levels at times or may have no obvious signs or symptoms. Impaired glucose tolerance can lead to a more serious problem called CF-related diabetes (CFRD). CFRD is treated with insulin. Watch and Discover Symptoms of Impaired Glucose Tolerance and CFRD - Weight loss or problems gaining weight - Tiredness - Unexplained worsening of lung function - Lung infections more often - Increased thirst and frequent drinking, which leads to the person urinating often The doctor uses blood tests to diagnose impaired glucose tolerance and CFRD. 1) Blood glucose level: The level of glucose can be measured in the blood. This often is done while fasting (having not eaten for at least 12 hours) in the early morning. If this level is too high, other blood tests need to be done. 2) Oral glucose tolerance test: Looks at how a person’s blood glucose level changes after drinking a certain amount of glucose. For an oral glucose tolerance test (OGTT for short), the person must fast (not eat for at least 12 hours) before the test. If blood glucose levels are too high, the body may have a problem handling glucose and it could mean the person has CFRD. There are recommended guidelines when annual glucose tolerance testing should start in children with CF. Think and Act Managing Impaired Glucose Tolerance and CFRD If you see symptoms that suggest your child may have blood glucose problems, act and talk with your CF health care team about having your child tested. If your child is diagnosed with either impaired glucose tolerance or CFRD, you will work closely with your CF health care team and/or a diabetes doctor (endocrinologist or pediatric endocrinologist) and learn how to manage it. This may include (1) learning how to check blood glucose levels at home, (2) closely watching how many carbohydrates your child eats, and (3) learning how to measure and give insulin. High Blood Glucose Caused by Medication or Infection People with CF can have problems with high blood glucose levels when they are sick or are taking certain medications. In these cases, the person will have to watch blood glucose levels and the CF health care team will decide if insulin is needed. If insulin is needed, it may only be for a short time. The CF health care team will help figure out what is causing the high blood glucose levels and what the best treatment will be to get levels back to normal. Liver and Gallbladder Disease Some people with CF can have problems with the liver or gallbladder. The liver is an organ that sits in the upper right side of the abdomen. The gallbladder is attached to the liver and helps store extra bile made by the liver. The liver is important in digestion. It stores nutrients for the body, helps break down medicines, and filters the blood. Liver disease in people with CF varies from one person to the next. Many people with CF can have some fat buildup in the liver. The liver can become larger and liver function test levels may go up. This happens most often to people who have malabsorption and poor nutrition. Another problem with the liver is a buildup of abnormal bile. In CF, bile can be too thick and does not flow well through the liver and gallbladder to the small intestine. Gallstones can develop in the gallbladder with the abnormal bile and may cause problems. The much more serious problem, however, is that while the bile builds up, the liver is damaged. This liver damage is called cirrhosis. Chronic infections (such as hepatitis C) or alcohol abuse can also cause cirrhosis. Like the lungs, the liver can continue to function even though it is damaged; however, over time it can fail. Cirrhosis in CF can also lead to high blood pressure in the veins connected to the liver (called portal hypertension), enlargement of the spleen, and blood clotting problems. Portal hypertension can cause veins around the esophagus to swell and weaken (called esophageal varices). These veins can bleed into the esophagus. Between 15 and 30 percent of people with CF can have cirrhosis. See CF Words to Know Glossary. WATCH AND DISCOVER CF-Related Liver Problems People with CF may not know their liver is having problems because they have little or no symptoms. Blood tests help detect liver problems. The CF Foundation recommends that people with CF have liver function tests once a year. The tests are the best way to find a problem early so treatment can be started before there are symptoms of liver damage. If liver disease or gallstones are suspected, the doctor will do other tests to find the liver problem. Signs and Symptoms of Liver Disease - Liver and spleen are larger than normal - Indigestion and stomachaches (from gallstones or fat malabsorption) - Easy bruising (from blood clotting problems) - Vomiting up blood (from esophageal varices) - Poor weight gain or weight loss THINK AND ACT: Managing CF-Related Liver Problems Although there is no cure for liver disease, treatments are important to help make sure a person with CF has good nutrition and his or her malabsorption is well managed. A medicine called ursodiol (or ursodeoxycholic acid, brand name Actigall®) helps with bile flow and may help with gallstones. If a person is having a lot of symptoms from gallstones, he or she may have the gallbladder removed during surgery. CF-RELATED BONE HEALTH Our bones depend on good nutrition to develop properly and remain strong. Vitamin D and calcium* are important to build and maintain bones. Regular exercise also helps bones stay healthy. In CF, sometimes the bones don’t have enough minerals* so they are less dense and can become weak and break more easily. Older women may have this condition called osteoporosis.* Osteoporosis is also seen most often in adults with CF and in people who have severe nutrition and/or lung problems. Good management of CF can help prevent bone health problems in the future. Some medicines can also affect the density and strength of a person’s bones. Talk with your CF health care team about medicines and supplements your child is taking. The CF health care team can help you watch for early signs of bone health problems or things that may put your child at risk for future bone problems. WATCH AND DISCOVER CF-Related Bone Health Problems Bone problems happen very slowly and often a person does not notice any symptoms (unless they are severe) until he or she has a break or fracture. Fractures hurt. If a person breaks a rib, it can be hard to breathe and cough, and airway clearance may be painful. Osteoporosis Tests Vitamin D and calcium blood levels are checked every year—and more often if a person has low levels or has problems with malnutrition.* The CF health care team may notice a change in bones on a chest X-ray, especially if the person’s bone density* is low. Bone density can be measured with equipment that takes a special kind of X-ray. The CF Foundation recommends all people with CF have a bone density study, such as a DEXA scan* (dual-energy X-ray absorptiometry) by 18 years old. Low bone density often means a person is at higher risk for broken bones. THINK AND ACT CF-Related Bone Health As your child grows, you can do several things to help keep his or her bones strong: - Make sure your child gets regular exercise that can help keep bones strong. Walking, running, and weight lifting are great exercises to help keep bones strong and healthy. Exercise early in life can build bone density that will help in the future. - Have your child eat a balanced diet, including enough vitamins* and minerals vital to good bone health. Alcohol and caffeine lower bone density and should be avoided. Have your child drink whole milk and other products that contain calcium, vitamin D, and calories. - Make sure your child avoids smoking and secondhand smoke exposure.* Tobacco smoke hurts the bones and lungs. - Help your child keep his or her lungs healthy and avoid lung infections. Healthy lungs will contribute to healthy bones as well. The CF health care team may prescribe calcium and/or vitamin D if your child’s levels are low. The team may also prescribe some medicines that can help strengthen bones. Decide with your CF doctor if medicine would be helpful for your child. To learn more, you can get a copy of “Bone Health and Cystic Fibrosis” (including a list of foods high in calcium) from the CF Foundation’s Web site: [http://www.cff.org](http://www.cff.org) or call 1-800-FIGHT-CF. *See CF Words to Know Glossary.
Fluorescence Lifetime Standards for Time and Frequency Domain Fluorescence Spectroscopy Noël Boens,† Wenwu Qin,† Nikola Basarić,†‡ Johan Hofkens,† and Marcel Ameloot§ Department of Chemistry, Katholieke Universiteit Leuven, Celestijnenlaan 200f – bus 02404, 3001 Heverlee, Belgium, and Universiteit Hasselt, BIOMED, Agoralaan, Building D, 3590 Diepenbeek, Belgium Jacques Pouget,‡ Jean-Pierre Lefèvre, and Bernard Valeur CNRS UMR 8531, Laboratoire de Chimie Générale, CNAM, 292 rue Saint-Martin, F-75141 Paris Cedex 03, and Laboratoire PPSM, ENS-Cachan, 61 avenue du Président Wilson, 94235 Cachan Cedex, France Enrico Gratton, Martin vandeVen,† and Norberto D. Silva, Jr. Laboratory for Fluorescence Dynamics, University of California, Biomedical Engineering Department, 3120 Natural Sciences II Building, Irvine, California 92697-2715 Yves Engelborghs, Katrien Willaert, and Alain Sillen Department of Chemistry, Katholieke Universiteit Leuven, Celestijnenlaan 200G, 3001 Heverlee, Belgium Garry Rumbles and David Phillips Department of Chemistry and Centre for Photomolecular Science, Imperial College, Exhibition Road, London SW7 2AY, United Kingdom Antonie J. W. G. Visser and Arie van Hoek Departments of Biochemistry and Biophysics, MicroSpectroscopy Centre, Wageningen University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands Joseph R. Lakowicz, Henryk Malak, and Ignacy Gryczynski Center for Fluorescence Spectroscopy, Department of Biological Chemistry, University of Maryland, 660 W. Redwood Street, Baltimore, Maryland 21201 Arthur G. Szabo and Don T. Krajcarski Faculty of Science, Wilfrid Laurier University, 75 University Avenue, Waterloo, Ontario N2L 3C5, Canada Naoto Tamai* and Atsushi Miura Department of Chemistry, School of Science, Kwansei Gakuin University, Uegahara1-1-155, Nishinomiya 662-8501, Japan A series of fluorophores with single-exponential fluorescence decays in liquid solution at 20 °C were measured independently by nine laboratories using single-photon timing and multifrequency phase and modulation fluorometry instruments with lasers as excitation source. The dyes that can serve as fluorescence lifetime standards for time-domain and frequency-domain measurements are all commercially available, are photostable under the conditions of the measurements, and are soluble in solvents of spectroscopic quality (methanol, cyclohexane, water). These lifetime standards are anthracene, 9-cyanoanthracene, 9,10-diphenylanthracene, N-methylcarbazole, coumarin 153, erythrosin B, N-acetyl-L-tryptophanamide, 1,4-bis(5-phenyloxazol-2-yl)benzene, 2,5-diphenyloxazole, rhodamine B, rubrene, N-(3-sulfolopropyl)acridinium, and 1,4-diphenylbenzene. At 20 °C, the fluorescence lifetimes vary from 89 ps to 31.2 ns, depending on fluorescent dye and solvent, which is a useful range for modern pico- and nanosecond time-domain or mega- to gigahertz frequency-domain instrumentation. The decay times are independent of the excitation and emission wavelengths. Depend- * To whom correspondence should be addressed. E-mail: email@example.com. Fax: +32-16-327 990. E-mail: firstname.lastname@example.org. Fax: + 32-11-268 599. † Katholieke Universiteit Leuven. ‡ Current affiliation: Division of Organic Chemistry and Biochemistry, Ruder Bosković Institute, Bijenička cesta 54, 10 000 Zagreb, Croatia. § Deceased. † Current affiliation: Hasselt University, Biomedical Research Institute, Agoralaan, Building D, 3590 Diepenbeek, Belgium. * Current address: Department of Chemistry, Kwansei Gakuin University, 2-1 Gakuen, Sanda 669-1337, Japan. ent on the structure of the dye and the solvent, the excitation wavelengths used range from 284 to 575 nm, the emission from 330 to 630 nm. These lifetime standards may be used to either calibrate or test the resolution of time- and frequency-domain instrumentation or as reference compounds to eliminate the color effect in photomultiplier tubes. Statistical analyses by means of two-sample charts indicate that there is no laboratory bias in the lifetime determinations. Moreover, statistical tests show that there is an excellent correlation between the lifetimes estimated by the time-domain and frequency-domain fluorometries. Comprehensive tables compiling the results for 20 (fluorescence lifetime standard/solvent) combinations are given. Fluorescence decay measurements are an exceptionally useful tool for investigating the dynamics of excited states in biology, chemistry, and physics. The two main methods for obtaining time-resolved fluorescence data are single-photon timing (also called time-correlated single-photon counting) and multifrequency phase-modulation fluorometry. Both techniques yield essentially the same information and differ mainly in how the time-resolved fluorescence data are obtained, i.e., time domain (TD) versus frequency domain (FD). For detailed information on the two methods (general principles, instrumentation, data analysis, and possible problems with data collection and analysis), we refer to several excellent books and reviews.$^1\text{–}$$^{15}$ Fluorescence lifetime standards are needed most in the areas of photophysics, photobiology, chemical sensing, physical chemistry, fluorescence lifetime imaging microscopy, flow cytometry, and single-molecule spectroscopy. Fluorophores with known lifetimes are necessary for testing the time-resolved instruments for systematic errors, for calibration of fluorescence lifetime instruments, and for use as reference compounds to avoid the wavelength-dependent time–response (the color effect) of photomultiplier tubes.$^{14,15}$ Although these wavelength-dependent effects are less pronounced with microchannel plate photodetectors,$^{16–20}$ it is still necessary to verify that such effects are not present or to correct for them by the use of fluorescence lifetime standards.$^{13}$ One of the earliest sources of single-exponential lifetime data is Birks’s Photophysics of Aromatic Molecules,$^{21}$ published in 1970. However, many of the lifetimes compiled in this pioneering work have since been shown to be seriously in error. At the time the book was written, relatively crude techniques for lifetime measurements were in operation. Another early standard work for lifetime data is Berlman’s Handbook of Fluorescence Spectra of Aromatic Molecules.$^{22}$ Because these lifetimes were obtained more than 40 years ago with an old-fashioned pulse sampling oscilloscope technique, it should not come as a surprise that many of them were found to be inaccurate. In 1974, Chen$^{23}$ proposed quinine and $\gamma$-pyrenebutyrate for use as lifetime standards. Later research$^{24}$ showed that quinine is unsuitable because it exhibits dual-exponential decay kinetics, whereas $\gamma$-pyrenebutyrate is easily photolyzed and sensitive to quenching by oxygen. A few years later, Grinvald$^{25}$ proposed $N$-acetyl-l-tryptophanamide, anthracene, and naphthalene as lifetime standards. While the first two may be useful standards, the long lifetime of naphthalene is very sensitive to oxygen quenching and makes naphthalene unsuitable as lifetime standard (see Lifetimes). The availability since 1978 of high repetition rate, mode-locked, sync-pumped (cavity-dumped) dye lasers$^{26,27}$ as picosecond excitation sources in TD and FD fluorometry allowed the very rapid collection of time-resolved fluorescence data and stimulated the search for more reliable fluorescence lifetime standards. Several groups contributing to the current study embarked separately on projects aimed at obtaining accurate nanosecond$^{11,28,29}$ and picosecond$^{11,20,30}$ lifetime values. Unfortunately, only a limited number of the same lifetime standards have been measured independently by different laboratories so that the reliability of the reported lifetime data is unclear. Moreover, no attempt has been made to compare the lifetime values or to evaluate which values are more reliable. Finally, until now, no systematic comparison has been carried out of the precision of the lifetime data obtained by the TD and FD methods. The conclusions of the earlier studies on fluorescence lifetime standards can be summarized as follows. There are only a few (fluorescent lifetime standard/solvent) combinations reported in --- (1) Demas, J. N. Excited-State Lifetime Measurements; Academic Press: New York, 1983. (2) Cundall, R. B.; Dale, R. E., Eds. Time-resolved Fluorescence Spectroscopy in Biochemistry and Biology; Plenum Press: New York, 1983. (3) O’Connor, D. V.; Phillips, D. Time-correlated Single Photon Counting; Academic Press: London, 1984. (4) van Hoek, A.; Visser, A. J. W. G. Anal. Instrum. 1985, 14, 359–378. (5) Janeson, D. M.; Gratton, E.; Hall, R. D. Appl. Spectrosc. Rev. 1984, 20, 53–106. (6) Eaton, D. F. Pure Appl. Chem. 1988, 60, 1107–1114. (7) Pouget, J.; Mugnier, J.; Valeur, B. J. Phys. E: Sci. Instrum. 1989, 22, 855–862. (8) Eaton, D. F. Pure Appl. Chem. 1990, 62, 1631–1648. (9) Boens, N. In Luminescence Techniques in Chemical and Biochemical Analysis; Baeyens, W. R. G., De Keukeleire, D., Korkidis, K., Eds.; Marcel Dekker: New York, 1991; pp 21–45. (10) Lakowicz, J. R., Ed. Topics in Fluorescence Spectroscopy, Vol. 1, Techniques; Plenum Press: New York, 1991. (11) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer-Verlag: New York, 2006. (12) Valeur, B. Molecular Fluorescence. Principles and Applications; Wiley-VCH: Weinheim, 2002. (13) vandeVen, M.; Ameloot, M.; Valeur, B.; Boens, N. J. Fluoresc. 2005, 15, 377–413. (14) Wahl, Ph.; Auchet, J. C.; Donzel, B. Rev. Sci. Instrum. 1974, 45, 28–32. (15) Rayner, D. M.; McKinnon, A. E.; Szabo, A. G.; Hackett, P. A. Can. J. Chem. 1976, 54, 3246–3259. (16) Murao, T.; Yamazaki, I.; Yoshihara, K. Appl. Opt. 1982, 21, 2297–2298. (17) Yamazaki, I.; Tamai, N.; Kume, H.; Tsuchiya, H.; Oba, K. Rev. Sci. Instrum. 1985, 56, 1187–1194. (18) Bebelis, D. Rev. Sci. Instrum. 1986, 57, 1116–1125. (19) Kume, H.; Koyama, K.; Nakatsugawa, K.; Suzuki, S.; Fatlowitz, D. Appl. Opt. 1988, 27, 1170–1178. (20) Boens, N.; Tamai, N.; Yamazaki, I.; Yamazaki, T. Photochem. Photobiol. 1990, 52, 911–917. (21) Birks, J. B. Photophysics of Aromatic Molecules; Wiley-Interscience: New York, 1970. (22) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic Molecules; Academic Press: London, 1965. (23) Chen, R. F. Anal. Biochem. 1974, 57, 593–604. (24) O’Connor, D. V.; Meech, S. R.; Phillips, D. Chem. Phys. Lett. 1982, 88, 22–26. (25) Grinvald, A. Anal. Biochem. 1976, 75, 260–280. (26) Spears, K. G.; Cramer, L. E.; Hoffland, L. D. Rev. Sci. Instrum. 1978, 49, 255–262. (27) Koester, V. J.; Dowben, R. M. Rev. Sci. Instrum. 1978, 49, 1186–1191. (28) Lampert, R. A.; Chewter, L. A.; Phillips, D.; O’Connor, D. V.; Roberts, A. J.; Meech, S. R. Anal. Chem. 1983, 55, 68–73. (29) Thompson, R. B.; Gratton, E. Anal. Chem. 1988, 60, 670–674. (30) Lakowicz, J. R.; Gryczynski, I.; Laczkó, G.; Gloyna, D. J. Fluorescence 1991, 1, 87–93. the literature that have been measured by several laboratories with sufficiently high precision (Tables S8 and S9, Supporting Information) and some of the proposed ones, like quinine bisulfate, are unsuitable because they do not display single-exponential decays.\textsuperscript{24} Furthermore, many of the proposed substances have lifetimes that are too long to provide a meaningful test for picosecond instrumentation. Additionally, long lifetime standards are difficult to use due to the effect of oxygen quenching. The use of collisional quenching to obtain different lifetimes\textsuperscript{23} is not advocated for lifetime standards, because of the risk of transient effects and nonexponential decays. Finally, the compounds cited in the literature do not cover fully the spectral region of interest (both in excitation and emission). To provide the research community with reliable fluorescence lifetime standards, an international, cooperative project, which involved nine research groups active in the field of time-resolved fluorescence, was initiated by the corresponding authors. Those nine laboratories from the United States, Canada, the U.K., Japan, and Continental Europe are identified in the text and the Supporting Information by a three-letter label based on the university or the associated university city (BAL = Baltimore, HAS = Hasselt, IRV = Irvine, LEU = Leuven, LON = London, NIS = Nishinomiya, PAR = Paris, WAG = Wageningen, WLU = Wilfrid Laurier University). For the venture to be successful, the following preconditions were set. Because a reliable lifetime value for a standard should be independent of the used measurement method, research groups using the TD and FD methods were invited to join the project. The combination of groups active in TD and FD allows one to verify the equivalence of the precision of both methods. Another prerequisite was the use of lasers as excitation sources. The majority of the instruments was equipped with microchannel plate photodetectors. This combination produces greater excitation light intensity and improved temporal resolution when compared to older setups. All phase and modulation measurements were done at multiple modulation frequencies. CRITERIA FOR THE CHOICE OF FLUORESCENCE LIFETIME STANDARDS AND SOLVENTS Although any compound with a single-exponential decay can theoretically serve as a lifetime standard, for the sake of convenience and standardization, the following criteria for the choice of possible lifetime standards in liquid solution were applied. (i) A first condition for a lifetime standard is that it should show single-exponential decay kinetics, independent of excitation and emission wavelength. (ii) The compound should be commercially available in sufficiently high purity (e.g., laser grade, scintillation grade, zone-refined, $\geq 99\%$, etc.) so that additional purification steps can be avoided. Our original aim was to evade dyes that have to be purified prior to the measurements, because the necessary skills/equipment to purify a compound to fluorometric grade might not always be on hand. Unfortunately, some of the commercial compounds, which were candidates for lifetime standards, showed dual-exponential fluorescence decay kinetics when used as received (see Table S4, Supporting Information). So, we had to relax this criterion for some compounds because the existing purity of the commercial products was not high enough. Therefore, some of the compounds were purified before the fluorescence lifetime measurements. (iii) From a practical point of view, an ideal fluorescence lifetime standard should have a (relatively) large Stokes shift (to ensure minimal spectral overlap of excitation and emission spectra) and a (relatively) large quantum yield. (iv) To cover the picosecond and nanosecond time scales, evidently a series of fluorescence standards with lifetimes matching that range should be on hand. (v) A variety of fluorescence standards should be available to cover different spectral regions. (vi) Insofar as possible, the standards should not pose health, safety, or environmental problems. (vii) A final criterion is chemical stability and photostability during the fluorescence measurements. The choice of solvents was guided by their commercial availability for fluorescence measurements. Water is an environment benign solvent. Ultrahigh quality water, delivered by a properly maintained Milli-Q system (Millipore),\textsuperscript{31} meets all the requirements of fluorescence spectroscopy. The life science and biotechnological communities prefer fluorescent standards in aqueous solution. Methanol, cyclohexane, and methylcyclohexane can be obtained in sufficiently high (spectroscopic) purity from several chemical suppliers. Good laboratory practices should be followed during storage, usage, and disposal of the lifetime standards and solvents. THEORY Kinetics. For a fluorophore that decays monoexponentially with lifetime $\tau$, the fluorescence $\delta$-response function $f(t)$ is $$f(t) = \alpha \exp(-t/\tau) \quad (1)$$ where $\alpha$ is the pre-exponential factor or amplitude. In single-photon timing experiments where the fluorophore is excited by an excitation pulse $u(t)$, the observed fluorescence decay, $y(t)$, is a convolution of $f(t)$ and $u(t)$: $$y(t) = \int_0^t f(t') u(t - t') \, dt' = \int_0^t f(t - t') u(t') \, dt' = f \otimes u \quad (2)$$ where $\otimes$ denotes the convolution operator and $t$ and $t'$ represent time. The instrument response function $u(t)$ is usually obtained by measuring the scattered excitation light. When $y(t)$ and $u(t)$ are experimentally known, the parameters $\alpha$ and $\tau$ of $f(t)$ can be estimated by a variety of techniques.\textsuperscript{32–36} In the frequency domain, the values of the phase shift, $\phi$, and the relative modulation, $m$, for a fluorophore with lifetime $\tau$, are given by $$\phi = \tan^{-1}(\omega \tau) \quad (3)$$ $$m = \frac{1}{\sqrt{1 + \omega^2 \tau^2}} \quad (4)$$ \textsuperscript{(31)} Preparation of ultra-pure water: http://www.millipore.com. \textsuperscript{(32)} O’Connor, D. V.; Ware, W. R.; Andre, J. C. \textit{J. Phys. Chem.} \textbf{1979}, \textit{83}, 1333–1343. \textsuperscript{(33)} Livesey, A. K.; Brochon, J. C. \textit{Biophys. J.} \textbf{1987}, \textit{52}, 693–705. \textsuperscript{(34)} Brochon, J. C. \textit{Methods Enzymol.} \textbf{1994}, \textit{240}, 262–311. \textsuperscript{(35)} Lakowicz, J. R.; Laczkó, G.; Cherek, H.; Gratton, E.; Limkeman, M. \textit{Biophys. J.} \textbf{1984}, \textit{46}, 463–477. \textsuperscript{(36)} Brochon, J. C.; Livesey, A. K.; Pouget, J.; Valeur, B. \textit{Chem. Phys. Lett.} \textbf{1990}, \textit{174}, 517–522. where $ \omega = 2\pi f $ is the angular frequency of the harmonically modulated excitation light and $ f $ the generator-set frequency (expressed in Hz). The lifetime can thus be independently determined from the phase shift (eq 3) and the relative modulation (eq 4). The lifetimes determined from phase and modulation data are equal regardless of the modulation frequency of the excitation light for a single-exponential decay while this is not the case for a multiexponential decay. To correct for the wavelength dependence of the instrument response function $ u(t) $ (sometimes referred to as color effect) in single-photon timing experiments, the use of the reference convolution method is recommended.$^{37–42}$ This method (also called delta function convolution method$^{39} $ or F/F deconvolution$^{39} $) requires a single-exponential reference, which has to be measured under identical instrumental settings as used for the sample. Under these conditions, the parameters of the modified sample decay, $ \tilde{f}(t) $, are obtained from the measured decays of the sample, $ y(t) $, and reference, $ d_r(t) $: \[ y(t) = \int_0^t \tilde{f}(t') d_r(t - t') \, dt' = \int_0^t \tilde{f}(t - t') d_r(t') \, dt' = \tilde{f} \otimes d_r \] (5) When $ f(t) $ is single-exponential, $ \tilde{f}(t) $ is given by eq 6, where $ \delta(t) $ denotes the Dirac $ \delta $-function and $ \tau_r $ the reference lifetime: \[ \tilde{f}(t) = \alpha [\delta(t) + (1/\tau_r - 1/\tau) \exp(-t/\tau)] \] (6) A variety of analysis methods$^{32} $ is accessible to estimate $ \alpha, \tau, $ and $ \tau_r $ of $ \tilde{f}(t) $ once $ y(t) $ and $ d_r(t) $ are known. To correct for the photomultiplier tube color effect in FD measurements, the phase shift and relative modulation are measured under identical instrumental settings for a reference fluorophore with lifetime $ \tau_r $ and the sample fluorophore. The values for the phase difference $ \Delta \phi $ and modulation ratio $ m/m_r $ between two fluorophores that decay monoexponentially are given by$^{29} $ \[ \Delta \phi = \tan^{-1}(\omega \tau) - \tan^{-1}(\omega \tau_r) \] (7) \[ \frac{m}{m_r} = \sqrt{\frac{1 + \omega^2 \tau_r^2}{1 + \omega^2 \tau^2}} \] (8) where $ m_r $ denotes the modulation of the reference. Single-Curve Fitting. (1) Data Analysis. In TD experiments, the nonlinear mathematical model is of the form $ y = f(t, \theta) $ where $ t $ (time) and $ y $ (decay data) are observable variables and $ \theta $ is the unknown parameter vector with $ p $ components. To estimate $ \theta $, one observes $ y $ at $ n $ values of $ t $. In FD measurements, the nonlinear model is of the form $ y = f(\omega, \theta) $ for phase and $ z = g(\omega, \theta) $ for modulation, where $ \omega $ (angular frequency of light modulation) is an observable variable. To estimate $ \theta $, one observes $ y $ and $ z $ at $ n $ values of $ \omega $. (Note that the number $ n $ of modulation frequencies $ \omega $ used in FD measurements is generally smaller than the number $ n $ of used values of $ t $ in the TD method.) Then $ \hat{\theta} $, the weighted least-squares estimate of $ \theta $, is determined by minimizing the reduced chi-square value $ \chi_\nu^2 $: \[ \chi_\nu^2 = \sum_i^n w_i (y_i^o - y_i^c)^2 / \nu = \sum_i^n R_i^2 / \nu \] (9) In eq 9, the index $ i $ sums over $ n $ data points, $ y_i^o $ and $ y_i^c $ denote respectively the observed (experimental) and calculated (fitted) values corresponding to the $ i $th data point, and $ w_i = 1/\sigma_i^2 $ denotes the associated statistical weight, where $ \sigma_i^2 $ is the variance of $ y_i^o $. $ R_i = (y_i^o - y_i^c)/\sigma_i $ stands for the weighted residual of the $ i $th data point; $ \nu = n - p $ is the number of degrees of freedom with $ p $ the number of adjustable fit parameters. (2) Statistical Criteria for the Quality of the Fit. A characteristic feature of single-photon timing experiments is the statistical nature of decay data accumulation. In the absence of systematic errors, the number of counts in each channel follows a Poisson distribution, so that in the Gaussian limit $ \sigma_i^2 = y_i^o $. For FD fluorometry, one has to determine the variances $ \sigma_\phi^2 $ and $ \sigma_m^2 $ of $ \phi $ and $ m $, respectively, by repeated measurements at each $ \omega $. Since the variance of each data point, measured by the single-photon timing technique (TD), is known, a rigorous statistical assessment of acceptability of fits is possible. Conversely, the variances of phase shift $ \phi $ and relative modulation $ m $ calculated from many FD measurements at a given modulation frequency $ \omega $ are often underestimated (because of systematic errors), as shown by the abnormally high $ \chi_\nu^2 $ values obtained when using these values. The comparison of the $ \chi_\nu^2 $ values for different models is a better test for model discrimination (e.g., dual-exponential decay versus single-exponential decay) than the absolute $ \chi_\nu^2 $ value. A rigorous examination of the deviations between the experimental and fitted data comprises both graphical techniques (weighted residuals, the autocorrelation function,$^{43} $ and the normal probability plot of the weighted residuals$^{44,45} $) and numerical statistical tests ($ \chi_\nu^2 $ and its standard normal deviate $ Z_{\chi_\nu^2} $, the Durbin–Watson test statistic,$^{46} $ and the ordinary runs test$^{45,47} $). Detailed descriptions of the various tests to assess the quality of the fit in single-curve analysis are given in the Supporting Information. Global Analysis. The advantages of the global (or simultaneous) analysis method$^{48–52} $ in comparison to single-curve analysis --- (37) Gauduchon, P.; Wahl, Ph. Biophys. Chem. 1978, 8, 87–104. (38) Libertini, L. J.; Small, E. W. Anal. Biochem. 1984, 138, 314–318. (39) Zuker, M.; Szabo, A. G.; Bramall, L.; Krajcarski, D. T.; Selingar, B. Rev. Sci. Instrum. 1985, 56, 14–22. (40) Kolber, Z. S.; Barkley, M. D. Anal. Biochem. 1986, 152, 6–21. (41) Vos, K.; van Hoek, A.; Visser, A. J. W. G. Eur. J. Biochem. 1987, 165, 55–63. (42) Boens, N.; Ameloot, M.; Yamazaki, I.; De Schryver, F. C. Chem. Phys. 1988, 121, 73–86. (43) Grinvald, A.; Steinberg, I. Z. Anal. Biochem.* 1974, 59, 583–598. (44) Chilko, D. M. Probability Plotting (SAS Technical Report A-106); SAS: Heidelberg, 1978. (45) Gunst, R. F.; Mason, R. L. Regression Analysis and its Application; Marcel Dekker: New York, 1980. (46) (a) Durbin, J.; Watson, G. S. Biometrika 1950, 37, 409–428. (b) Durbin, J.; Watson, G. S. Biometrika 1951, 38, 159–178. (c) Durbin, J.; Watson, G. S. Biometrika 1971, 58, 1–19. (47) Swed, F. S.; Eisenhart, C. Ann. Math. Stat. 1943, 14, 66–87. (48) Knutson, J. R.; Beecham, J. M.; Brand, L. Chem. Phys. Lett. 1983, 102, 501–507. (49) Frye, S. L.; Ko, J.; Halpern, A. M. Photochem. Photobiol. 1984, 40, 555–561. (50) Beecham, J. M.; Ameloot, M.; Brand, L. Chem. Phys. Lett. 1985, 120, 466–472. are improved model testing capability and accuracy of the estimated parameters.\textsuperscript{53} (1) Data Analysis. Instead of minimizing $\chi^2$ defined in eq 9, in global analysis one minimizes the \textit{global} $\chi^2$, $$\chi^2_{vg} = \left\{ \sum_l^q \sum_i w_i (y_i^0 - y_i^f)^2 / \nu_g \right\}$$ where the subscript g denotes global. The index $l$ sums over $q$ experiments, whereas the index $i$ sums over the appropriate number of data points for each individual experiment. $\nu_g$ represents the number of degrees of freedom for the entire multidimensional fluorescence decay surface. Equation 10 is formally equivalent to eq 9, but the summation is over the entire decay surface. (2) Statistical Criteria for the Quality of the Fit. The graphical methods include plots of surfaces (carpets) of the autocorrelation function values\textsuperscript{43} versus experiment number and of the weighted residuals versus time (or channel number, TD) or frequency $\omega$ (FD) versus experiment number. A good fit should produce carpets free of creases. The numerical statistical tests are the calculation of the \textit{global} $\chi^2$ (eq 10) and its corresponding $Z_{\chi^2_{vg}}$ $$Z_{\chi^2_{vg}} = \sqrt{\nu_g/2}(\chi^2_{vg} - 1)$$ Since $Z_{\chi^2_{vg}}$ is standard normally distributed, $Z_{\chi^2_{vg}}$ can be readily used to compare the goodness of fit of analyses with different $\nu_g$. The goodness of fit for the individual decay curves (TD) or phase and modulation data (FD) is examined by the Durbin–Watson statistic,\textsuperscript{46} the ordinary runs test,\textsuperscript{45,47} the local $\chi^2$ (eq 9) and its normal deviate $Z_{\chi^2}$. EXPERIMENTAL SECTION Materials. All fluorophores [anthracene, 9-cyanoanthracene, 9,10-diphenylanthracene (DPA), N-methylcarbazole, coumarin 153, erythrosin B, N-acetyl-l-tryptophanamide (NATA), 1,4-bis(5-phenyloxazol-2-yl)benzene (POPOP), 2,5-diphenyloxazole (PPO), rhodamine B, rubrene, N-(3-sulfolpropyl)acridinium (SPA), 1,4-diphenylbenzene ($\beta$-terphenyl)] were typically of the highest purity commercially obtainable (e.g., laser grade, 99%, scintillation grade, zone-refined, gold label, microscopic grade, etc.) and were procured from different commercial chemical suppliers. It is not the purpose of this report to distinguish “good” from “bad” suppliers of fluorescent dyes or to name the “best” buy for a specific dye. Indeed, the quality of a certain dye procured from a particular provider might change over time or from lot to lot. It also is possible that chemical suppliers will merge with other companies or cease to exist. \textit{Our aim is to determine if a certain fluorescent dye displays single-exponential decays and can be used as a robust lifetime standard.} The solid samples of the fluorescence lifetime standards are stable when stored according to the supplier’s instructions. The solvents (methanol, cyclohexane, methylcyclohexane) were of spectrophotometric grade, HPLC grade, fluorometric grade, high purity, etc., and were obtained from different chemical suppliers. The solvents showed negligible background fluorescence under the spectral conditions of the experiments. When water was used as solvent, it was of Milli-Q (Millipore)\textsuperscript{51} quality or doubly distilled. Some measurements were done in ACS certified aqueous buffer pH 7 in lieu of pure water. Instrumentation. Multifrequency phase and modulation measurements were performed using lasers as excitation source, either mode-locked lasers (IRV, LEU, BAL) or cw lasers associated with a Pockels cell (PAR), as previously described by the different research groups participating in the current study.\textsuperscript{7} Between 11 and 60 frequencies $f$ (25 is the average) were used in the measurements. The number of frequencies that is required to recover “correct” values for the decay parameters depends on the complexity of the model decay function $f(t)$. In principle, a few frequencies suffice for a single-exponential whereas at least 50 or more frequencies (producing $\geq 50$ $\phi$ and $\geq 50$ $m$ values) are necessary for a distribution of exponentials. Laser excitation was similarly used in the single-photon timing measurements, as described in the literature.\textsuperscript{54} The decay traces were collected in the channels (between 1/4K and 4K) of a (computer-integrated) multichannel analyzer. Since the obtained results are independent of the used FD and TD instrumentation (see Laboratory Bias and Comparison of Precision of TD and FD Methods), detailed instrumental descriptions will not be given here and can be found in the literature. The used excitation and emission wavelength ranges are compiled in Table 1. The excitation wavelength ranges might not be the optimal ones (see Figure S1, Supporting Information), because the best possible laser excitation wavelengths were not accessible on some laser setups. The absorbance of the fluorophores in all solutions at the excitation wavelength was less than 0.15, typically $\sim 0.05$. Magic angle (54°44′) detection\textsuperscript{55} was used to eliminate the effects of rotational diffusion on the intensity decays.\textsuperscript{11–13} If that preventative measure is not taken, the measured time-resolved fluorescence trace will be at least biexponential (eq 15 in Lifetimes; see the results of LON in Table S4, Supporting Information). All TD laboratories as well as LEU (FD) used a monochromator to select the emission wavelength. The FD laboratories IRV, PAR, and in some instances BAL used optical filters for this purpose. Dissolved oxygen was removed from all solutions by repetitive freeze–pump–thaw cycles or by purging the solutions with N$_2$ or Ar. The steady-state intensities of the samples were stable during the lifetime measurements. Excitation and emission spectra were recorded on a SPEX Fluorolog fluorometer and are corrected for fluctuations of the excitation source flux and wavelength dependence of the detection system. Excitation spectra closely matched absorption spectra, taken on a Perkin-Elmer Lambda 40 UV–visible spectrophotometer. All steady-state and time-resolved measurements were done at 20 °C. Figure S1 (Supporting Information) displays the normalized absorption and emission \textsuperscript{51} Löfroth, J.-E. \textit{Eur. Biophys. J.} \textbf{1985}, \textit{13}, 45–58. \textsuperscript{52} Beechem, J. M.; Gratton, E.; Ameloot, M.; Knutson, J. R.; Brand, L. In \textit{Topics in Fluorescence Spectroscopy}: Principles; Lakowicz, J. R., Ed.; Plenum Press: New York, 1991; Vol. 2, pp 241–305. \textsuperscript{53} Janssens, L. D.; Boens, N.; Ameloot, M.; De Schryver, F. C. \textit{J. Phys. Chem.} \textbf{1990}, \textit{94}, 3564–3576. \textsuperscript{54} Maus, M.; Rousseau, E.; Cotlet, M.; Schweitzer, G.; Hofkens, J.; Van der Auweraer, M.; De Schryver, F. C.; Krueger, A. \textit{Rev. Sci. Instrum.} \textbf{2001}, \textit{72}, 36–40. \textsuperscript{55} Spencer, R. D.; Weber, G. \textit{J. Chem. Phys.} \textbf{1970}, \textit{52}, 1654–1663. Table 1. Mean Lifetime Data $\bar{\tau}$ of the Fluorescent Lifetime Reference Dyes in Fluid Solution at 20 °C Measured in the Present Study \| compound$^a$ \| solvent \| lifetime $\bar{\tau} \pm s$ (ns)$^b$ \| $100 \, s/\bar{\tau}$ \| $\lambda_{\text{ex}}$ (nm) \| $\lambda_{\text{em}}$ (nm) \| $c$ \| $n^d$ \| \|--------------\|-------------\|--------------------------------------\|------------------------\|-----------------------------\|-----------------------------\|-----\|-------\| \| anthracene \| methanol \| $5.1 \pm 0.3$ \| $6.1$ \| $295–360$ \| $375–442$ \| 7 \| 7 \| \| \| cyclohexane \| $5.3 \pm 0.1$ \| $2.6$ \| $295–360$ \| $375–442$ \| 7 \| 7 \| \| 9-cyanoanthracene \| methanol \| $16 \pm 1$ \| $9.3$ \| $295–360$ \| $400–480$ \| 7 \| 7 \| \| \| cyclohexane \| $12.7 \pm 0.7$ \| $5.5$ \| $295–360$ \| $400–450$ \| 4 \| 4 \| \| DPA \| methanol \| $8.7 \pm 0.5$ \| $5.6$ \| $295–360$ \| $400–475$ \| 8 \| 8 \| \| \| cyclohexane \| $7.5 \pm 0.4$ \| $5.8$ \| $295–360$ \| $400–475$ \| 8 \| 7 \| \| $N$-methylcarbazole \| cyclohexane \| $14.1 \pm 0.9$ \| $6.2$ \| $290–325$ \| $350–400$ \| 6 \| 6 \| \| coumarin 153 \| methanol \| $4.3 \pm 0.2$ \| $4.5$ \| $295–442$ \| $495–550$ \| 5 \| 5 \| \| erythrosin B \| water \| $0.089 \pm 0.003$ \| $3.6$ \| $488–568$ \| $550–580$ \| 6 \| 6 \| \| \| methanol \| $0.47 \pm 0.02$ \| $4.0$ \| $488–568$ \| $550–590$ \| 6 \| 6 \| \| NATA \| water \| $3.1 \pm 0.1$ \| $3.6$ \| $295–309$ \| $330–410$ \| 7 \| 7 \| \| POPOP \| cyclohexane \| $1.12 \pm 0.04$ \| $3.6$ \| $295–360$ \| $380–450$ \| 8 \| 8 \| \| PPO \| methanol \| $1.65 \pm 0.05$ \| $2.7$ \| $295–330$ \| $340–400$ \| 8 \| 8 \| \| \| cyclohexane \| $1.36 \pm 0.04$ \| $2.6$ \| $290–325$ \| $360–450$ \| 8 \| 8 \| \| rhodamine B \| water \| $1.74 \pm 0.02$ \| $0.9$ \| $488–575$ \| $560–630$ \| 6 \| 5 \| \| \| methanol \| $2.5 \pm 0.1$ \| $4.0$ \| $295, 488–568$ \| $550–630$ \| 8 \| 8 \| \| rubrene \| methanol \| $9.9 \pm 0.3$ \| $3.2$ \| $300, 488, 514$ \| $550–610$ \| 5 \| 5 \| \| SPA \| water \| $31.2 \pm 0.4$ \| $1.4$ \| $300–330$ \| $466–520$ \| 5 \| 5 \| \| $p$-terphenyl\| methanol \| $1.17 \pm 0.08$ \| $6.5$ \| $284–315$ \| $330–380$ \| 7 \| 7 \| \| \| cyclohexane \| $0.98 \pm 0.03$ \| $3.3$ \| $290–315$ \| $330–390$ \| 7 \| 7 \| $a$ For abbreviations used, see text. All solutions are deoxygenated by repetitive freeze–pump–thaw cycles or by bubbling N$_2$ or Ar through the sample solutions. $^b$ Average lifetime $\bar{\tau}$. The quoted errors are sample standard deviations$^{64} s = \frac{1}{(n - 1)} \sum_{i=1}^{n} (\tau_i - \bar{\tau})^2]^{1/2}$. $^c$ Number of lifetime data measured. $^d$ Number of lifetime data used in the calculation of the mean lifetime $\bar{\tau}$ and its standard deviation$^{64} s$. The difference between columns $c$ and $d$ gives the number of outliers. Boldface numbers in column $d$ indicate that there is one outlier present. Spectra of all (fluorescence lifetime standard/solvent) combinations measured in this work. Additionally, this figure gives the chemical formula, the formula weight, the CAS number, and the fluorescence quantum yield of each of the fluorescent lifetime standards. Data Analysis. Each research group measured and analyzed its own time-resolved fluorescence data, using its individual data analysis software. To minimize any bias in the lifetime data, the fluorescence lifetime values obtained by the different laboratories were hidden from each other until the final version of this paper was distributed to the participating laboratories for final editing. Only the corresponding authors had prior access to all the data (for the statistical analysis). Since the used curve-fitting programs are not a critical factor, we shall not describe them here; details can be found in the literature.$^{41,56,57}$ LON analyzed the data as outlined in ref 3. The criteria to evaluate the quality of the fits are described in Single-Curve Fitting and Global Analysis and in the Supporting Information. The number and type of goodness-of-fit criteria varied among the contributing laboratories. Two FD laboratories (IRV, LEU) collected their phase shift and modulation data versus reference fluorophores, while four laboratories (BAL, LON, NIS, WLU) recorded the phase shift/modulation data and fluorescence decays versus a Ludox or glycogen scattering solution in water. The remaining three research groups (LEU/HAS, PAR, WAG) used both reference fluorophores and a scattering solution. The lifetimes of sample ($\tau$) and reference ($\tau_r$) were modifiable fit parameters. In most cases, the fluorophores used as references were also the fluorescence lifetime standards used in this study (Table 1). For example, erythrosin B in water as sample was measured versus erythrosin B in methanol as reference, and vice versa; PPO in cyclohexane as sample was measured versus $p$-terphenyl in cyclohexane, and vice versa; and so on. RESULTS Lifetimes. Table S4 (Supporting Information) shows all estimated lifetime values of 20 (fluorescent lifetime standard/solvent) combinations measured by the various laboratories. All time-resolved fluorescence data were analyzed by single-curve analysis. Three laboratories (LEU/HAS, LEU, LON) also used simultaneous (global) analysis of time-resolved fluorescence data measured at different emission wavelengths to verify the single-exponential character of the fluorescence decays. The values of the various criteria for assessing the quality of the fits indicated that all the decay data gave excellent single-exponential fits except the following. 9,10-Diphenylanthracene (97%), which was used as received from two different chemical suppliers, consistently yielded biexponential fits (eq 12). $$f(t) = \alpha_1 \exp(-t/\tau_1) + \alpha_2 \exp(-t/\tau_2) \quad (12)$$ It is only after recrystallization or sublimation that this compound displays single-exponential decay kinetics. One group (IRV) reported a biexponential decay for $N$-methylcarbazole (99%) in cyclohexane. The relative amplitude $\alpha_r [= \alpha_r / (\alpha_1 + \alpha_2)]$ of the 14.3-ns component was 0.902. Two biexponential fits were reported for coumarin 153 in methanol (out of five measurements). Apparently, laser grade quality from various suppliers leads to different (single-exponential versus dual-exponential) time-resolved Table 2. Comparison of FD and TD Data Precision of the Fluorescent Lifetime Reference Dyes in Fluid Solution at 20 °C Measured in the Present Study \| compound$^a$ \| solvent$^b$ \| $\bar{\tau}_{FD}$ (ns)$^c$ \| $n_{FD}$$^d$ \| $s_{FD}^{2e}$ \| $\bar{\tau}_{TD}$ (ns)$^c$ \| $n_{TD}$$^d$ \| $s_{TD}^{2e}$ \| $F^f$ \| $F^g$ \| $t^h$ \| \|----------------\|---------------\|-----------------\|--------\|-------------\|-----------------\|--------\|-------------\|------\|------\|------\| \| anthracene \| MeOH \| 5.00 \| 3 \| 0.191 \| 5.20 \| 4 \| 0.042 \| 4.54 \| 16.04 \| 0.819 \| \| \| CyH \| 5.32 \| 4 \| 0.025 \| 5.32 \| 3 \| 0.020 \| 1.22 \| 39.17 \| 0.004 \| \| 9-cyanoanthracene \| MeOH \| 15.29 \| 3 \| 2.967 \| 16.27 \| 4 \| 1.804 \| 1.64 \| 16.04 \| 0.850 \| \| \| CyH \| 12.39 \| 3 \| 0.294 \| 13.47 \| 1 \| \| \| \| \| \| DPA \| MeOH \| 8.71 \| 3 \| 0.541 \| 8.77 \| 5 \| 0.141 \| 3.83 \| 10.65 \| 0.152 \| \| \| CyH \| 7.17 \| 3 \| 0.232 \| 7.76 \| 4 \| 0.026 \| 8.86 \| 16.04 \| 2.347 \| \| $N$-methylcarbazole \| CyH \| 14.06 \| 3 \| 0.188 \| 14.15 \| 3 \| 1.691 \| 0.11 \| 39.00 \| 0.108 \| \| coumarin 153 \| MeOH \| 4.18 \| 2 \| 0.006 \| 4.33 \| 3 \| 0.056 \| 9.24 \| 779 \| 0.857 \| \| erythrosin B \| water \| 0.090 \| 2 \| $1 \times 10^{-5}$ \| 0.089 \| 4 \| $1 \times 10^{-5}$ \| 1.02 \| 17.44 \| 0.247 \| \| \| MeOH \| 0.45 \| 2 \| $2 \times 10^{-5}$ \| 0.48 \| 4 \| $3 \times 10^{-4}$ \| 18.16 \| 864 \| 1.772 \| \| NATA \| water \| 3.14 \| 3 \| 0.02 \| 3.01 \| 4 \| 0.003 \| 4.82 \| 16.04 \| 1.925 \| \| POPOP \| CyH \| 1.12 \| 4 \| 0.003 \| 1.12 \| 4 \| $7 \times 10^{-4}$ \| 4.22 \| 15.44 \| 0.016 \| \| PPO \| MeOH \| 1.63 \| 3 \| 0.003 \| 1.66 \| 5 \| 0.002 \| 1.80 \| 10.65 \| 0.946 \| \| \| CyH \| 1.35 \| 4 \| 0.002 \| 1.38 \| 4 \| $6 \times 10^{-4}$ \| 2.55 \| 15.44 \| 1.576 \| \| rhodamine B \| water \| 1.73 \| 1 \| \| 1.75 \| 4 \| 0.000 \| \| \| \| \| \| MeOH \| 2.48 \| 4 \| 0.010 \| 2.44 \| 4 \| 0.011 \| 1.14 \| 15.44 \| 0.614 \| \| rubrene \| MeOH \| 9.79 \| 2 \| 0.051 \| 9.97 \| 3 \| 0.156 \| 3.05 \| 779 \| 0.567 \| \| SPA \| Water \| 30.90 \| 2 \| 0.076 \| 31.37 \| 3 \| 0.185 \| 2.43 \| 779 \| 1.359 \| \| $\beta$-terphenyl \| MeOH \| 1.10 \| 2 \| $4 \times 10^{-5}$ \| 1.20 \| 5 \| 0.005 \| 122.91 \| 899 \| 1.913 \| \| \| CyH \| 0.96 \| 3 \| $9 \times 10^{-4}$ \| 1.00 \| 4 \| $1 \times 10^{-4}$ \| 7.80 \| 16.04 \| 2.971 \| $^a$ Abbreviations used; see text. $^b$ MeOH = methanol, CyH = cyclohexane. $^c$ Average lifetime $\bar{\tau}_{FD}$ (respectively $\bar{\tau}_{TD}$) determined from FD (respectively TD) measurements. $^d$ Number of FD (TD) lifetime data used in the calculation of $\bar{\tau}_{FD}$ ($\bar{\tau}_{TD}$) and its corresponding variance $s_{FD}^{2}$ ($s_{TD}^{2}$). $^e$ Variance for FD (respectively TD) measurements. $^f$ $F$-statistic calculated according to eq 16. $^g$ Tabled two-sided $F$-distribution value with $\alpha$ equal to 5%. This is equivalent to the cumulative function equal to 0.975. $^h$ Student’s $t$-statistic calculated according to eq 17. fluorescence. Five out of seven laboratories found clear monoexponential decay kinetics for NATA in water. The $\tau$-value (3.28 ns) determined by BAL is a borderline value (see Outlier Detection). In some cases, NATA (98% and reagent grade) must be recrystallized to yield monoexponential decays, whereas in other cases, NATA (98% and reagent grade), procured from a different supplier, gave excellent single-exponential fits. Three groups out of six found biexponential decays for rhodamine B in water or aqueous buffer. The $\tau$-value (2.8 ns) obtained by IRV is a clear outlier (see Outlier Detection). Rhodamine B in water readily forms aggregates (even at very low concentrations),$^{58–63}$ and therefore, one has to keep its concentration low enough for the time-resolved fluorescence to be single-exponential. For rhodamine B in methanol, excellent single-exponential fits were obtained by six out of eight groups. A single group (WAG) found a biexponential fit for rubrene in methanol. The relative amplitude $\alpha_r$ of the 9.58-ns component was 0.941. Two laboratories out of five reported biexponential fits for SPA in water. The relative amplitude $\alpha_r$ of the 31.09-ns component was 0.994 (PAR) and 0.917 for the 31.38-ns component (WAG). Apparently, SPA’s purity varies between different lots (all SPA came from the same supplier) and SPA has to be purified to ensure excellent single-exponential fits. At the start of the lifetime standards project, 6-methoxy-$N$-(3-sulfopropyl)quinolinium (SPQ), naphthalene, and L-tyrosine were also candidates as fluorescence lifetime standards. However, these compounds cannot be recommended as lifetime references for different reasons. The six research groups that measured the time-resolved fluorescence of SPQ in water all reported biexponential fits. Naphthalene is not included in Tables 1 and 2 because the two available, estimated $\tau$-values in methylcyclohexane vary too widely (95 and 113 ns). The value of 113 ns was obtained for a naphthalene solution degassed by four freeze—pump—thaw cycles and the 95-ns value for a solution deoxygenated by purging with $\text{N}_2$ gas. This may reflect the fact that naphthalene’s long lifetime is still quenched by the residual oxygen left by the less efficient deoxygenation method. Furthermore, since the only decay of naphthalene measured in cyclohexane yielded a dual-exponential fit, naphthalene cannot be recommended as a lifetime standard. Because we only have three reliable lifetime data for L-tyrosine in water, we cannot endorse this amino acid as a lifetime standard, although the existing experimental evidence points to a single-exponential fluorescence decay (average lifetime $\bar{\tau} = 3.1$ ns, sample standard deviation$^{64}$ $s = 0.4$ ns). Even if the dye is ultrapure, its measured fluorescence decay cannot be described by a single-exponential function if the effect of the rotational motion is not eliminated by measuring under the magic angle condition. Indeed, the fluorescence light will in general be polarized to some extent depending on the geometry of the experimental setup and the rotational motion of the emitting fluorophore. Consider the case of a macroscopically isotropic sample excited by light polarized along the perpendicular to the $^{58}$ Selwyn, J. E.; Steinfeld, J. I. J. Phys. Chem. 1972, 76, 762–774. $^{59}$ (a) López Arbeloa, I.; Ruiz Ojeda, P. Chem. Phys. Lett. 1981, 79, 347–350. (b) López Arbeloa, I.; Ruiz Ojeda, P. Chem. Phys. Lett. 1982, 87, 556–560. $^{60}$ López Arbeloa, F.; Ruiz Ojeda, P.; López Arbeloa, I. J. Lumin. 1989, 44, 105–112. $^{61}$ Illich, P.; Mishra, P. K.; Macura, S.; Burghardt, T. P. Spectrochim. Acta, Part A 1996, 52, 1323–1330. $^{62}$ Mchedlov-Petrossyan, N. O.; Kholin, Yu. V. Russ. J. Appl. Chem. 2004, 77, 414–422. $^{63}$ Ghasemi, J.; Niazi, A.; Kubista M. Spectrochim. Acta, Part A 2005, 62, 649–656. $^{64}$ According to ISO, Guide to the Expression of Uncertainty in Measurement, 1st ed. (International Organization for Standardization: Geneva, Switzerland, 1993) and Taylor, B. N. and Kuyatt, C. E. NIST Technical Note 1297 (Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results; U.S. Department of Commerce, NIST: Gaithersburg, MD, 1994), the estimated standard deviation $s_i$ evaluated by statistical methods should be termed standard uncertainty with suggested symbol $u_i$ (i.e., $u_i = s_i$) and is equal to the positive square root of the estimated variance. Whenever the term standard deviation ($s_i$) appears in the text, one can read standard uncertainty ($u_i$) to comply with the ISO and NIST guidelines. excitation–detection plane and where the fluorescence is detected at 90° with respect to the direction of the excitation beam. The polarized fluorescence decays with the emission polarizer set parallel ($i_{\|\|}$) and perpendicular ($i_{\perp}$) to the polarization of the excitation light can then be written as $$i_{\|\|}(t) = \frac{1}{3} f(t) [1 + 2r(t)] \quad (13a)$$ $$i_{\perp}(t) = \frac{1}{3} f(t) [1 - r(t)] \quad (13b)$$ where $r(t)$ stands for the time-resolved fluorescence anisotropy. For an excited species in a single type of isotropic environment, $r(t)$ is, in general, given by a linear combination of exponentially decaying functions: $$r(t) = \sum_j \beta_j \exp(-t/\phi_j) \quad (14)$$ Only when the rotational relaxation times $\phi_j$ are much shorter than the fluorescence lifetime $\tau$, the intensities $i_{\|\|}$ and $i_{\perp}$ become equal and proportional to $f(t)$. If this condition is not fulfilled, a polarizer oriented at 54°44′ with respect to the vertical has to be inserted in the emission path to eliminate the effect of rotational diffusion. If the emission polarizer is absent, the fluorescence $\delta$-response $i(t)$ can be written as $$i(t) = \alpha_1 \exp(-t/\tau) + \alpha_2 \exp\left[-t\left(\frac{1}{\tau} + \frac{1}{\phi}\right)\right] \quad (15)$$ for a single-exponential $r(t)$. For example, global biexponential analysis by LON of four decays collected at different emission wavelengths (420, 430, 440, 450 nm) for POPOP in cyclohexane ($\lambda_{\text{ex}} = 315$ nm) yielded a value of 1.11 ns for $\tau$ and 119 ps for the second decay time, $\tau \phi/(\tau + \phi)$. Similarly, biexponential decays were also obtained by LON for PPO in methanol and cyclohexane and for $p$-terphenyl in methanol and cyclohexane (Table S4, Supporting Information). Table 1 summarizes the calculated mean lifetimes $\bar{\tau}$ and the associated sample standard deviations $s$ for each compound in the used solvents, based on the data of Table S4 (Supporting Information). Comparison of columns c and d of Table 1 indicates that two data points were deleted from the calculations, because they are not consonant with the rest of the data. Such outliers can only be removed from the data set after close examination, as described in Outlier Detection and the Supporting Information. Figure 1 shows graphically the lifetimes $\tau$ estimated by the different laboratories for two illustrative examples (POPOP in cyclohexane and DPA in cyclohexane). Figure 2 displays a typical fluorescence decay of PPO in cyclohexane obtained by the single-photon timing (TD) technique. The best fit, the instrument response function $u(t)$, the weighted residuals $R_i$, and the autocorrelation function $C_j$ are also shown. Figure 3 shows the corresponding data for PPO in cyclohexane obtained by the phase-modulation (FD) technique. The best fits to the experimental phase shift $\phi$ and modulation data $m$, and the weighted residuals for both $(\phi, m)$ fits are also shown. To have an idea of the precision of the measured lifetime data, we calculated the relative standard deviation (the ratio of the sample standard deviation $s$ over the mean lifetime $\bar{\tau}$) expressed as a percentage, i.e., $100 s/\bar{\tau}$. These percentages range from 0.9% for rhodamine B in water to 9.3% for 9-cyanoanthracene in methanol (Table 1). Outlier Detection. A potential outlier is an observation that has an extremely large deviation: a peculiarity that does not fit in with the pattern of the rest of the data points. Removing an outlying observation from the data set should be done only after close inspection of the data. There are two methods for detecting outliers: graphical and statistical. For a discussion of the use of plotting techniques in outlier detection we refer to the literature. The simplest technique is by examining a scatter plot of the original data (lifetime $\tau_i$ versus laboratory $i$ as in Figures 1 and S2, Supporting Information) or, equivalently, of the (raw) residuals $r_i$ ($r_i = \tau_i - \bar{\tau}$, with $\bar{\tau}$ mean lifetime) versus laboratory $i$. Extreme points are often easily visible on such graphs. Visual inspection of all 20 plots of $\tau_i$ versus laboratory $i$ indicates that there are three possible outliers. Figure 1a (POPOP in cyclohexane) is indicative of the absence of outliers, while the data point from LEU (FD) for DPA in cyclohexane is a potential outlier (Figure Figure 2. Data obtained by the single-photon timing technique using a mode-locked Ti:sapphire laser pumped by the second harmonic (532 nm) of a YAG laser. The output of the Ti:sapphire laser was frequency tripled to obtain the UV excitation wavelength. Experimental fluorescence decay trace of PPO in degassed cyclohexane at 20 °C (excitation wavelength $\lambda_{\text{ex}} = 290$ nm, observation wavelength $\lambda_{\text{em}} = 380$ nm, number of channels used in the fitting 3620, channel width 1.85 ps). The instrument response function $u(t)$ measured with a Ludox scattering solution in water and the best monoexponential fit to the experimental decay data are also displayed. The plot of the weighted residuals $R_i$ versus time and the autocorrelation function $C_j$ are given in the lower panels. Results: $\tau = 1.357 \pm 0.002$ ns, $\chi^2 = 1.072$, $Z_p \chi^2 = 3.059$. The quoted error represents one standard deviation (standard uncertainty).$^{64}$ 1b). Examination of the other comparable plots reveals that the $\tau$-values determined by BAL for NATA in water (Figure S2b, Supporting Information) and by IRV for rhodamine B in water (Figure S2c, Supporting Information) can be labeled as potential extreme values and warrant closer analysis. Numerical outlier measures that assess the degree to which an observation in a small sample is a maverick point are complimentary to simple scatter plots, which are, however, somewhat subjective. The numerical tests$^{65,66}$ for detecting outlier observations in a small sample from a normal distribution are given in the Supporting Information. These numerical tests corroborate that two outliers can be found: one in the sample of 1,9-diphenylanthracene in cyclohexane and the other for rhodamine B in water. It is clear that these outlying observations are not used in the statistical analysis. Laboratory Bias. Possible occurrence of systematic errors in the participating laboratories can be demonstrated by using a two-sample chart.$^{67}$ The plot of the estimated lifetimes for sample A against the corresponding results for sample B for each laboratory should not show any discernible trend: one should obtain a random scatter of points about a line with zero slope. However, a positive trend in the data [i.e., low (respectively high) lifetime values for sample A corresponding to low (respectively high) lifetime values for sample B] is indicative of the presence of laboratory (or method) bias. One obvious condition to plot a two-sample chart is that the same laboratory has measured the lifetimes of both samples. Additionally, the chart should have enough data points to allow a possible trend in the data to be observed. As examples we have plotted in Figure 4 two-sample charts for the estimated lifetimes of anthracene in methanol ($x$-ordinate) versus those of POPOP in cyclohexane ($y$-ordinate) (Figure 4a) and of PPO in methanol ($x$-ordinate) versus DPA in methanol ($y$-ordinate) (Figure 4b). These examples (and many others not displayed here) show that systematic errors in the lifetime determinations are absent. Indeed, linear least-squares analysis indicates that the correlation between the lifetime data sets is insignificant. With 20 different samples, the theoretical number of possible two-sample charts is very large. Pairwise comparisons of all possible lifetime data sets are not recommended because the performed tests use each data set several times (i.e., the tests are mutually dependent). Because two-sample tests do not distinguish method (TD or FD) from laboratory bias, they cannot be used to compare different methods. --- $^{(65)}$ Grubbs, F. E. Technometrics 1969, 11, 1–21. $^{(66)}$ Dixon, W. J. Ann. Math. Stat. 1951, 22, 68–78. $^{(67)}$ Massart, D. L.; Vandeginste, B. G. M.; Deming, S. N.; Michotte, Y.; Kaufman, L. Chemometrics: A Textbook; Elsevier: Amsterdam, 1988. Comparison of Precision of TD and FD Methods. To know whether the two time-resolved fluorometries (i.e., TD and FD) yield the same or significantly different results, replicate analyses with each fluorometric technique were carried out by the participating laboratories. Ideally, the lifetimes obtained by both methods should be completely correlated, i.e., the correlation coefficient, $r$, should be equal to unity. However, because the correlation coefficient only gives a preliminary indication, it will not be discussed further. A $t$-test (assuming a normal distribution of errors) can be applied to investigate whether the differences between the mean lifetimes $\bar{\tau}_{\text{TD}}$ and $\bar{\tau}_{\text{FD}}$ obtained with the TD and FD methods are significant or not. The variances for the replicate analysis of each sample by the two methods can be compared using the $F$-test. These statistical tests are described in the next sections. Replicate analysis of a sample produces the estimated variances, $s_1^2$ and $s_2^2$, which are compared to test the significance of their difference. Let us assume that $n_1$ replicate measurements are done on a certain sample by using procedure 1 and $n_2$ replicate measurements by using procedure 2. The question is whether the variances of the two populations from which the samples are drawn are different: $\sigma_1^2 \neq \sigma_2^2$. The null hypothesis, $H_0$, is always a hypothesis of no difference. If the null hypothesis $H_0$, $\sigma_1^2 = \sigma_2^2$, is true, then the estimates $s_1^2$ and $s_2^2$ do not differ very much and their ratio should be close to unity. To test this condition, one calculates the test statistic $F$ as the ratio of the sample variances $s_1^2$ and $s_2^2$: $$F = s_1^2 / s_2^2$$ (16) By convention, one calculates the $F$-ratio by dividing the largest variance by the smallest. One then has to compare the calculated $F$-value with the theoretical, tabulated value $F_{(n_1 - 1), (n_2 - 1)}$ at the chosen significance level $\alpha$ (usually 0.05 or 0.01) for $(n_1 - 1)$ and $(n_2 - 1)$ degrees of freedom. $H_0$ is formulated as $\sigma_1^2 = \sigma_2^2$ and $H_1$ as $\sigma_1^2 \neq \sigma_2^2$. $H_0$ is accepted (and $H_1$ rejected) if the calculated $F$-value is smaller than the tabulated value. If the lifetime data are normally distributed and if the variances of the two populations from which the samples are drawn are equal, a $t$-test can be performed by calculating the statistic $$t = \frac{\bar{\tau}_1 - \bar{\tau}_2}{\sqrt{s^2 \left[ \frac{1}{n_1} + \frac{1}{n_2} \right]}}$$ (17) which is distributed as Student’s $t$ with $(n_1 + n_2 - 2)$ degrees of freedom. A composite variance, $s^2$ in eq 17, is calculated as $$s^2 = \frac{(n_1 - 1)s_1^2 + (n_2 - 1)s_2^2}{n_1 + n_2 - 2}$$ (18) The calculated $t$-value is compared with the tabled, theoretical value (Table S7, Supporting Information) at the chosen significance level $\alpha$ for $(n_1 + n_2 - 2)$ degrees of freedom. The null hypothesis $H_0$, $\mu_1 = \mu_2$, is accepted if the calculated $t$-value is smaller than the tabled value. To illustrate the test for the precision of the TD and FD methods, we used the measured lifetime data of PPO in cyclohexane (Table S4, Supporting Information, and Table 2): $\bar{\tau}_1 = 1.3825$, $s_1^2 = 0.000\,603\,7$, $n_1 = 4$ (TD); $\bar{\tau}_2 = 1.346$, $s_2^2 = 0.001\,542$, $n_2 = 4$ (FD). Since the sample sizes are small, two conditions have to be fulfilled for the validity of the $t$-test. Let us assume the condition of normality of the lifetime data is met. The condition of homogeneity of the variances can be checked with the $F$-test. The calculated $F$-value (2.554) is compared to the theoretical two-tailed $F_{0.05}^{0.95}$ value at $\alpha = 0.05$ and for (3, 3) degrees of freedom, i.e., 15.44. Because this calculated value is smaller than the tabulated value, $H_0$, $\sigma_1^2 = \sigma_2^2$, is accepted: no difference in variance can be shown. As both conditions are fulfilled, a $t$-test can be performed with $H_0$, $\mu_1 = \mu_2$, and $H_1$, $\mu_1 \neq \mu_2$, $s^2 = 0.001\,07$, $t = 1.576$. Because the theoretical values for a two-tailed $t$-test at $\alpha = 0.05$ and for 6 degrees of freedom is 2.447 (Table S7, Supporting Information), $H_0$ is accepted and one can conclude that at the chosen significance level no difference between $\bar{\tau}_1 (= \bar{\tau}_{\text{TD}})$ and $\bar{\tau}_2 (= \bar{\tau}_{\text{FD}})$ can be shown. Hence, the TD and FD procedures give comparable precision. One can repeat these calculations for the 19 remaining sets of lifetime data. When only one data point is on hand via the TD (9-cyanoanthracene in cyclohexane) or FD (rhodamine B in water) methodology, no values for $s^2$, $F$, and $t$ can be calculated. The results of the $F$- and $t$-tests for all investigated (fluorescent lifetime standard/solvent) combinations are compiled in Table 2. For each sample, the calculated $F$-values are smaller than the tabulated two-tailed $F$-value at $\alpha = 0.05$, showing no difference in variance between the two time-resolved fluorescence techniques. The $t$-tests indicate that no difference in means can be shown. It is only for $p$-terphenyl in cyclohexane that the calculated $t$-value (2.971) is somewhat larger than the tabulated $t$-test value at $\alpha = 0.05$ (2.571) for 5 degrees of freedom. However, at the $\alpha = 0.01$ significance level, the tabulated Student’s $t$-value (4.032) largely exceeds the calculated value. An easier way of assessing the comparability of the two methods is by least-squares fitting.$^{67}$ When the mean lifetimes $\bar{\tau}_{FD}$ obtained for the samples of Table 2 with the FD procedure are plotted against $\bar{\tau}_{TD}$ obtained with the TD methodology, a straight line should be found. Theoretically, this line should have a slope, $b$ (eq 19), of exactly unity and an intercept on the ordinate, $a$, of exactly zero. $$y = a + bx \quad (19)$$ By fitting eq 19 to the data, one obtains estimates $\hat{a}$ and $\hat{b}$ of $a$ and $b$, respectively, and by a goodness-of-fit test, one can find out if eq 19 really describes the experimental observations. A caret ($\hat{}$) above a symbol denotes an estimate of the quantity represented by that symbol. The presence of random errors in both methods yields a scatter of points around the least-squares line and a small deviation of the estimated slope and intercept from unity and zero, respectively. Since none of the two methods yields error-free lifetime data, one should use the orthogonal regression because it takes into account errors in $x$ and $y$. Indeed, the standard least-squares method’s implicit assumption that one variable ($y$) is subject to error and the other ($x$) is error-free produces biased estimates of the fit parameters when that assumption is violated.$^{68,69}$ Hence, when both $x$ and $y$ have measurement uncertainties, the standard least-squares procedure should not be used. Instead, a general least-squares method that accounts for measurement errors on $x$ and $y$ should be used to perform curve fits of $\bar{\tau}_{TD}$ versus $\bar{\tau}_{FD}$ (or vice versa).$^{69}$ The best orthogonal least-squares fit (eq 19) of $\bar{\tau}_{FD}$ ($y$, subject to error) versus $\bar{\tau}_{TD}$ ($x$, subject to error) obtained for 18 samples of Table 2 gives the following estimates: $\hat{a} = -0.001 \pm 0.005$ ($s_a$), $\hat{b} = 1.02 \pm 0.01$ ($s_b$) with $r = 0.978$. Because $n_{TD} = 1$ for 9-cyanoanthracene in cyclohexane and $n_{FD} = 1$ for rhodamine B in water (Table 2), these data were excluded from the least-squares regression. Figure 5 shows the results of the best straight-line fit of $\bar{\tau}_{FD}$ ($y$) versus $\bar{\tau}_{TD}$ ($x$) when both variables contain errors. The $r$-value reveals that the correlation between $\bar{\tau}_{TD}$ and $\bar{\tau}_{FD}$ is very good. To investigate whether $\hat{a}$ and $\hat{b}$ differ significantly from zero and unity, respectively, one must apply $t$-tests. To test whether the estimate $\hat{a}$ of the intercept is significantly different from $a = 0$, one calculates $t = \hat{a}/s_a$ and compares this value with the tabled Student’s $t$-distribution value $t_{n-2,\alpha}$ with $n - 2$ degrees of freedom at the significance level $\alpha$. The calculated $t$-value (0.2) is lower than the theoretical $t_{16,\alpha}$-value for $\alpha = 0.05$ (2.120) or $\alpha = 0.01$ (2.921) and 16 degrees of freedom. This means that $\hat{a}$ is not significantly different from 0, implying that there is no method bias. To test whether $\hat{b}$ is significantly different from $b = 1$, one compares the calculated $t$-value [$t = (\hat{b} - 1)(n - 2)^{1/2}/(1 - r^2)^{1/2}$ with $n = 18$] with the tabulated Student’s $t$-distribution value $t_{n-2,\alpha}$ with $n - 2$ degrees of freedom at the significance level $\alpha$. Also, the $t$-test shows that $\hat{b}$ is not significantly different from 1 (calculated $t = 0.539$). To conclude, the values of slope, intercept, and correlation coefficient estimated by the appropriate least-squares regression of $\bar{\tau}_{FD}$ and $\bar{\tau}_{TD}$ (both variables have uncertainties) demonstrate that the TD and FD procedures have very similar precision. DISCUSSION For lifetime determinations of the same compound carried out by several laboratories, each with their own personnel, lifetime instrumentation, data analysis software, lifetime standards, and solvents obtained from different suppliers, etc., one expects a normal distribution of errors broader than that when an experienced analyst carries out all determinations. The dispersion around the mean lifetime $\bar{\tau}$ can be considered a measure of reproducibility. To minimize systematic errors in the fluorescence lifetime determinations, all sources of variation must be taken into account and eliminated where possible. Therefore, to create experimental conditions as similar as possible, all laboratories participating in the cooperative fluorescence lifetime standards project had to comply with the following set of preconditions. (i) Since temperature may affect the fluorescence lifetime, the temperature was set fixed at 20 °C. (ii) To avoid the quenching effect of the ubiquitous quencher oxygen, all solutions had to be deoxygenated, either by purging the solutions with N$_2$ or Ar or, --- (68) Jefferys, W. H. Astron. J. 1980, 85, 177–181. (69) Lybanon, M. Am. J. Phys. 1984, 52, 22–26. preferably, by degassing the solutions by repetitive freeze–pump–thaw cycles prior to the measurements. (iii) The concentration of the solute in the solution had to be low enough to avoid systematic errors that depend on the concentration of the lifetime standard (such as reabsorption, aggregate formation, etc.). Therefore, the absorbance of the solutions was kept very low (typically $\sim 0.05$), so that the absorbed light flux was linearly proportional to the absorbance. (iv) Obviously, the impurity of both the lifetime standard and the solvent used to prepare the solutions can contribute to systematic errors. Hence, the use of standards and solvents of the highest possible purity commercially available was required. (v) To obtain reliable lifetime values, it is of utmost importance to collect high-quality experimental data, as well as to use sensitive criteria for judging the quality of the fit. High-quality data can be obtained only with state-of-the-art instrumentation that is free of systematic errors and that is expertly maintained and run. Systematic errors can be introduced into lifetime measurements by several factors [due to the fluorescent sample (purity, preparation, concentration, and geometry), the electronic and optical components of the time-resolved instrumentation, and data analysis]. Reference 13 gives an extensive list of possible error sources and their remedies in time-resolved fluorescence spectroscopy (TD and FD). All instruments used by the different groups had lasers as excitation source and possibly microchannel plate photomultipliers as photodetectors to ensure the highest temporal resolution. Actually, the use of lasers as excitation sources was chosen as an important criterion for participation in the project. Indeed, the use of a laser as excitation source was instrumental in demonstrating that quinine exhibits dual exponential fluorescence decay kinetics.\textsuperscript{24} Before 1982, the single-exponential lifetimes of quinine solutions were “known with better accuracy than those of any other solutions described in the literature”\textsuperscript{23} and there was “no other substance whose lifetime has been measured by so many laboratories with such good agreement as in the case of quinine”.\textsuperscript{23} Evidently, the single-exponential lifetime values measured for quinine sulfate by so many laboratories were not obtained by laser excitation. Since the signal-to-noise ratio in single-photon timing measurements is dependent on the number of counts, it is important to collect decay traces with high counts. Analogously, for the phase-modulation measurements, multiple modulation frequencies were used. (vi) Probably the surest way to minimize systematic errors in interlaboratory measurements is to select those qualified laboratories known to be particularly competent in the fields of pulse and phase-modulation fluorometry. The small relative standard deviations\textsuperscript{64} expressed as a percentage (Table 1) indicate that the interlaboratory uncertainty and the systematic error introduced by the use of the TD and FD methods are rather insignificant. This also is confirmed by the statistical analyses by means of two-sample charts. The unavoidable small artifacts of any instrumental setup and the difficulty of preparing identical solutions without a trace of impurity and quencher can account for the small deviations observed among the laboratories. Our lifetime values $\bar{\tau}$ (Table 1) agree very well with the following seven reported in the literature (Table S9, Supporting Information): anthracene in cyclohexane ($5.2 \pm 0.1$ ns, deoxygenated solutions), erythrosin B in water ($0.08 \pm 0.02$ ns, undegassed and deoxygenated solutions), NATA in water ($3.0 \pm 0.1$ ns, undegassed and deoxygenated solutions), POPOP in cyclohexane ($1.12 \pm 0.02$ ns, undegassed and deoxygenated solutions), PPO in cyclohexane ($1.39 \pm 0.03$ ns, deoxygenated solutions), and rhodamine B in water ($1.58 \pm 0.08$ ns, not specified whether solutions are deoxygenated) and in methanol ($2.5 \pm 0.3$ ns, not specified whether solutions are deoxygenated). The $\bar{\tau}$ and $s$ ($u$)\textsuperscript{64} values reported here and in Table S9 are calculated for minimally three measurements. When less data were available (Table S8, Supporting Information), no averages were calculated. The shorter lifetime ($4.02 \pm 0.07$ ns) obtained for undegassed solutions of anthracene in cyclohexane demonstrates that oxygen quenches the fluorescence by $\sim 20\%$ and confirms the need of deoxygenation to obtain consistent lifetime values. To assess the comparability of the TD and FD procedures, $F$-tests (eq 16), $t$-tests (eq 17), and linear least-squares fits (eq 19) of the mean lifetimes $\bar{\tau}_{\text{TD}}$ obtained for the samples of Table 2 with the TD procedure versus $\bar{\tau}_{\text{FD}}$ obtained with the FD procedure were performed. All these statistical tests indicate that both methods have comparable precision. Therefore, pulse (TD) and phase (FD) fluorometries are not only theoretically equivalent; they provide the same type of information [because the harmonic response is the Fourier transform of the $\delta$-response function $f(t)$] and do this with very similar precision. From the instrumental point of view, the state-of-the-art instruments used by the research groups in this project use both lasers and microchannel plate detectors. Because the time resolution is primarily limited by the response time of the detector, this parameter is the same for both techniques. Because the methodologies are different—indeed, they are relevant to the time domain and frequency domain—it is logical that one technique will be more appropriate than the other for obtaining certain information [time-resolved spectra and anisotropy, lifetime-based decomposition of spectra, fluorescence lifetime imaging microscopy (FLIM)]. The well-defined statistics in single-photon timing is an advantage for data analysis. Time-resolved fluorescence anisotropy measurements are more straightforward in TD fluorometry. Single-photon timing has an outstanding sensitivity, for very weak fluorescent samples requiring long acquisition times this is advantageous. For single-molecule lifetime determinations, single-photon timing is the only possibility. Conversely, the short acquisition time for $(\phi, m)$ measurements at a single frequency is a benefit in FLIM spectroscopy, provided the fluorescence intensity is high enough to get an analog signal whose zero crossing (for $\phi$ measurements) and amplitude (for $m$ measurements) can be measured with enough accuracy. For a comparison of the performance of the single-photon timing method and the frequency-domain method in two-photon FLIM, we refer to ref 70. The time of data collection depends on the complexity of the fluorescence $\delta$-response function $f(t)$. For complicated $f(t)$, the time of data collection is approximately the same for both techniques: in TD fluorometry, a large number of timed photon events is necessary, and in FD fluorometry, a large number of frequencies has to be selected. To summarize, TD and FD fluorometries each have their own advantages and drawbacks. They appear to be complementary methods rather than competitive ones.\textsuperscript{12} \begin{thebibliography}{10} \bibitem{70} Gratton, E.; Breusegem, S.; Sutin, J.; Ruan, Q.; Barry, N. \textit{J. Biomed. Opt.} \textbf{2003}, 8, 381–390. \end{thebibliography} Finally, traceability is the property of the result of a measurement (i.e., fluorescence lifetime) that can be traced back to the appropriate SI unit (time, in this instance) through an unbroken chain of measurements with properly evaluated uncertainties. The international standard for time and frequency metrology is the coordinated universal time scale (UTC). Technically, the traceability of fluorescence lifetime measurements depends upon the calibration of the time (TD) or frequency (FD) axis of the participating laboratories. The quoted standard uncertainties\textsuperscript{64} (Tables 1 and 2) imply that the contribution to uncertainty of the time (or frequency) axis calibration is marginal. ACKNOWLEDGMENT N.B. thanks the Fonds voor Wetenschappelijk Onderzoek—Vlaanderen (FWO) for grant G.0320.00. A.J.W.G.V. and A.v.H. thank The Netherlands Organisation for Scientific Research for investment grants to upgrade the time-resolved fluorescence equipment. B.V. is grateful to the Centre National de la Recherche Scientifique (France) for financial support. D.P. and G.R. thank EPSRC for equipment support through various grants. Y.E. thanks the FWO for project G.0092.01. SUPPORTING INFORMATION AVAILABLE Supporting Information Available: Absorption and fluorescence emission spectra, absorption and emission spectral maxima, molar absorption coefficients $\epsilon (\lambda_{\text{max}})$ at the maxima, fluorescence quantum yields, and literature data on all fluorescent dyes measured in the present study. Experimental conditions for the recording of the fluorescence emission spectra on the SPEX Fluorolog. Graphical and numerical criteria for assessing the quality of the fit. Numerical tests for detecting outliers. All lifetime data in liquid solution at 20 °C measured in the present study and corresponding lifetime data reported in the literature. Percentage points of $T_n$, percentage points of the distribution of $r_{10}$, and Student’s $t$-distribution. Fluorescence decays of several dyes measured by the TD and TD methods. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review November 16, 2006. Accepted December 18, 2006. AC062160K
The WACOM HAM Editor: Don Stark, N3HOW 724-228-4177 • email@example.com WACOM Officers President: Tom Fox, KA3NY 724-223-9666 Vice President: Joe Caldwell, N3XE 724-663-5708 firstname.lastname@example.org Treasurer: Ed Oelschlager N3ZNI 724-746-9235 • email@example.com Secretary: Jim Burtoft, KC3HW 724-228-0546 firstname.lastname@example.org Board of Directors Board President, Dave Wagner, KB3GOT 724-267-3400 email@example.com John Moninger, WA3VKC 724-228-5787 JAM @pulsenet.com Don Stark, N3HOW 724-228-4177 firstname.lastname@example.org Repeater Sam Mayberry, W3CYO 724-222-0367 email@example.com REMINDER— The next WACOM Board of Directors meeting is Thursday, January 30 at 7 PM at the WACOM Club room. All are welcome. JOIN WACOM ON THE AIR 2 meter net: Every Tuesday at 8:30 PM on the W3CYO repeaters 145.49 and 443.3 MHz 10 meter net: Every Tuesday at 9:00 PM at 28.340 MHz January 18, 2003 marks the 100th anniversary of Marconi’s first spark gap transmission between the United States and Europe. Well, my radio friends, it has been great fun. As I had previously announced, this is my last issue of the WACOM Ham, at least for now. Many of you know that Marjie and I have had a busy and difficult time balancing the many activities that we love with the situations with our home and family. Our lovely house was undermined last year by the coal company and was severely damaged. It appears that the only practical solution is to tear the house down and start over. Now we must find a place to move for temporary housing of 6 to 9 months. It is the same for everyone else on our street. My workshop tools have all been in a storage trailer for the past 9 months. Our well is gone. We have had a steady parade of experts through the house. We have lived with a lot of stress. I have not been able to spend much time on the radio. As bad as the last year has been, the next year will be difficult as well. I want to thank the many contributors to this publication. Many members have contributed photographs, information and articles. As radio operators, we usually communicate with the spoken word. It is great fun to see what impact written words and pictures can have. Ron Notarius, WN3VAW editor of the WASH Rag is going to take over as editor of the WACOM Ham. My great thanks to him for stepping in. You can contact him at firstname.lastname@example.org. TNX ES 73 DE DON N3HOW SK Dear Mr. Stark, I saw your column in The Wacom Ham, November 2002 edition, in which you mentioned acquiring the 5A5 portable. This being Christmas eve, I was thinking of the Christmas gifts I had received as a boy (I'm 68 now) and what were my favorites. The winner, hands-down, was the 5A5. There being no television at the time, I was a radio addict---there were many wonderful shows, shows in which the characters became "family" to the listener. Being able to listen to them anywhere, which the portable radio permitted, was nothing less than a miracle. I must have listened to my 5A5 literally thousands of hours. The batteries were expensive, so it was good that a power cord was coiled inside the battery compartment. There is a good color photo of the open radio at: www.geocities.com/old_radio_pictures/locate/radio-pictures/motorola-5a5.jpg. [editor's note—site not currently working] Seeing it was like seeing a picture of an old beloved friend, long-deceased. Very nostalgic. I envy you, having one. By the way, I was surfing the net, looking for pictures of the 5A5 when I ran across your article in the Wacom Ham. I'm not a ham myself, but my dad was (K5ARO) for many years. He died 7 years ago. John Nelms Dallas, Texas email@example.com ps: I am a Criminal District Judge in Dallas, just elected to a fourth and final term of four years. Stanley P. Cole, Jr, 60, died Sunday, December 8, 2002 in Washington Hospital. Our sympathy goes to his wife Susan, N3AOK, his sons Stanley P. Cole, III, NX3P and Jared Cole, and Stan’s family. Are you thinking of going to Dayton and can’t find a place to stay? Greg, KQ3DX, has arranged for suites at Dayton University. There would be 4 people per suite at $80 per night per suite. That would work out to only $20 per person per night. The suites have a living area, kitchen, refrigerator, etc. It sound like a good deal. Contact him if you are interested. Annual WACOM membership dues are only $15. A couple pays only $22.50. Send your check payable to WACOM to: Ed Oeschlager, N3ZNI, Washington Amateur Communications Inc. 1696 East Maiden St. Washington, PA 15301 Jim, KC3HW, worked NK3T/KC4 in Antarctica on Dec 5th at about 9 PM. He was 5 X 5. It turns out that the operator, Don Hess, grew up in Uniontown. How about you? Have you worked any rare DX lately? Let us at the WACOM Ham hear about it. Do you have any photographs of your station? We would love to share them with other WACOM members. Computers & Communications Portable radio Vertex The VX-210 Series, which meets or exceeds Mil Std 810 c/d/e standards for temperature, shock, vibration, solar radiation and rain, are pc-programmable and offer radio to radio cloning. The unit offers 500-megawatt audio output, and battery saver circuitry is included on both the transmit and the receive functions. Busy channel lock-out and busy tone lock-out are standard features, and the positive-lock channel selector features a “lift and turn” knob, virtually eliminating the possibility of accidental channel changes. What’s wrong with this ad? John Moninger, WA3VKC, submitted a copy of this ad of a communications transceiver that is used for fire and EMT crews. He suspects that with that level of audio power that the battery life might be just a little short. --- Upcoming Events January 2 - WACOM monthly meeting 7:30 PM South Strabane #1 VFD January 4 - Kid’s Day January 11 - WASH Two Meter Contest For more details see Ed Oelschlager, N3ZNI, 724-745-9235, firstname.lastname@example.org Go to www.WASHarc.org for entry forms and rules. January 11 - VE session sponsored by WPA Hilltoppers at University of Pittsburgh. Contact Dennis Tannar at 412-885-8073. January 11 - VE session sponsored by Breezeshooters at Perrysville Vol Fire Co. Contact Robert Benna at 412-366-0488. January 16 - VE session at McMillan Presbyterian Church sponsored by WASH. Contact Jacqueline Gosselin N3ZEL, 724-746-9235. January 19 - Western Pennsylvania Repeater Council Meeting South Strabane #1 VFD Fire Hall 1:30 PM. See Sam Mayberry for details. February 6, 7, 8, & 9 - Sportsman’s Show Special Event Station, W3C. Washington Mall. For more details see Ed Oelschlager, N3ZNI, 724-745-9235, email@example.com. February 22 Breezeshooters Ten Meter CW Groundwave Contest February 23 WASHfest 2003 Castle Shannon VFD Memorial Hall 8:00 AM—3 PM Contact Steve Lane, W3SRL 412-341-1043 --- Next WACOM Meeting: Thursday January 2, 2003 Monthly Meeting 7:30 PM South Strabane #1 VFD 1696 East Maiden St. Washington, PA 15301 “Holiday Radio Show and Tell” What new gear did you get for Christmas? March 2003 Technician Class - Technician Amateur Radio License Class. See Bob Ketzell, KB3IN, for more details or to offer assistance. May 16-18 2003 - Dayton Hamvention Dayton ARA www.hamvention.org/ October 5, 2003 - WACOM Hamfest 2003, Washington County Fairgrounds Opinions expressed in the WACOM HAM are those of the author and not necessarily those of Washington Amateur Communications, Inc. Copyright ©2002 by WACOM, Washington Amateur Communications Inc. Reproduction of material from The WACOM HAM is permitted if credit is attributed to the author and The WACOM HAM. The editor welcomes articles and timely information of interest to members and the general amateur community. Happy New Year’s Everybody, I hope that everybody got a new radio or other related Ham Radio gear for Christmas. In fact we welcome you to bring your new radio toys to the January meeting for a little “Show and Tell”. That way we all get to share in the fun. Speaking of fun, the Special Event station W3C is scheduled for February 6 - 9 at the Washington Mall. We need lots of members to come and help. There will be a sign-up list at the meeting or let Ed, N3ZNI, know when you can work the station. I am glad to hear that Carl, KD3KH, is home and is making good progress from his bypass surgery. He is not getting out much, but he can work the telephone or the radio if you want to give him a call. Bill Hill’s mother who is 96, recently had surgery for a broken hip. We wish her a speedy recovery. The Western Pennsylvania Repeater Council will be meeting on January 19th at the South Strabane #1 VFD at 1:30 PM. The meeting is open to all. Come out to see what factors affect our local repeaters. See Sam Mayberry, W3CYO, for more details. You might want to contact Bob Ketsell, KB3IN, about a TS 850 SAT with a built-in automatic antenna tuner that he has for sale. Santa brought him a new ICOM 746PRO! What a Christmas present! Who says there is no Santa Claus?! Chuck Shoemaker, formerly W3CCC, is now AB3AB. Nice call Chuck. Congratulations. See you at the meeting, 73 Tom Special event station KM1CC will be on the air January 11-19, 2003, to mark the 100th anniversary of Guglielmo Marconi’s inaugural wireless transmission between the US and Great Britain on January 18, 1903. Marconi used a 35 kW rotary spark transmitter and a massive antenna system to transmit a 54-word greeting from President Theodore Roosevelt to England’s King Edward VII. Details can be found on the Marconi Radio Club of Massachusetts W1AA website at http://personal.tmplp.com/k1vv/w1aa.
Twilight Gala and Auction January 30th 2010 The Capital Room of the Hartford Marriott To benefit the National Spinal Cord Injury Association, Connecticut Chapter TWILIGHT GALA COMMITTEE: Bill Mancini Gina Mancini Elaine Motta Kerry Sheltra Jim Quick Dave Morgana THE MISSION: The Mission of the Connecticut Chapter of the National Spinal Cord Injury Association is to support those with spinal related injuries or diseases and their families by being an advocate for their rights while serving as a resource to its members and the general public. We will provide individual support and furnish information to afford greater awareness of disabilities, spinal cord research, prevention and education. In addition, we will provide activities and events with the goal of empowering our members. The National Spinal Cord Injury Association, Connecticut Chapter NSCIA CT Chapter WE ARE: - We are individuals with spinal cord injuries and/or spinal cord injury related diseases. - We are family members and friends of persons with spinal cord injuries. - We are professionals in various related fields striving to improve the quality of life for persons with spinal cord injuries. Our purpose is to help persons with spinal cord injury and/or disease restructure and rebuild their lives. We offer the experiences and support of persons who successfully live with spinal cord injury. We can provide an individual with information regarding spinal cord injury. We want to help you learn to live with the extensive changes you will experience at home, at work, at school and in various relationships with others. The Connecticut Chapter of the National Spinal Cord Injury Association is part of a nationwide network of people committed to: Providing better services to individuals, families and communities to assist persons with spinal cord injuries. Prevention activities and campaigns to reduce the incidents that cause spinal cord injuries. Supporting research aimed at improving the quality of medical care and eventually, a cure for paralysis. The Chapter is committed to ensuring anyone with a spinal cord injury has the opportunity for a full, meaningful, independent and active life. GOALS: SUPPORT & SHARING We share our experiences and ideas with each other through community support groups. Through our Peer Assistance Program, spinal cord injured persons who have successfully coped with the results of their injuries assist newly injured individuals and their families adjust to changes in their lifestyle. RESOURCES We provide information about specific programs and services available to spinal cord injured persons. Use our Connecticut Resource Directory as a helpful resource. We also provide a quarterly membership newsletter, The Conn-Cord. We also have a website available to all interested parties, www.sciact.org. COMMUNITY AWARENESS & PREVENTION We share information on spinal cord injury and its impact on a person’s lifestyle. Our Stop Trauma Program explains the causes and consequences of spinal cord and acquired brain injury. This program is available to those interested on request to educators who want to bring the message of trauma prevention into the classroom. EDUCATION We sponsor and support informational seminars for persons with spinal cord injury and/or related diseases, health care professionals and the general public addressing issues of spinal cord injury. ADVOCACY & LEGISLATION We fight for the rights of present and future spinal cord injured persons and their families. We inform newly injured individuals of these rights. FOR MORE INFORMATION NSCIA CT Chapter Call 203-284-1045 Email firstname.lastname@example.org Website www.sciact.org EVENING EVENTS: 6:30 PM Registration 6:30-8:00 PM Cocktail Hour and Silent Auction 8:00-9:30 PM Welcome, Dinner, Live Auction & Awards Live Auction will be hosted by ‘Mark the Shark, 96.5 Radio Personality. Appreciation Awards Corporation Award – Gaylord Rehabilitation Hospital Individual Award – Kenneth ‘Obie’ Harrington-Howes 10:00 PM-Midnight Music, Cocktails and More. Corporation Award Gaylord Rehabilitation Hospital, Wallingford, CT Gaylord Hospital in Wallingford, Connecticut has supported the NSCIA CT Chapter since the inception of the chapter in 1978. The hospital has supported the organization by providing office and meeting space to conduct its daily operations. Gaylord provides acute rehab for individuals who have suffered with a spinal cord injury. The staff is very knowledgeable on all aspects of recovery, both emotional and physical, from a spinal cord injury. Individual Award Kenneth ‘Obie’ Harrington-Howes Obie has had a spinal cord injury for over 10 years. When he was injured his community came together and honored him by forming a foundation that has dedicated itself to improving the lives of individuals with spinal cord injuries. The Obie Harrington-Howes Foundation has given thousands of dollars in grant money to individuals with a spinal cord injury to purchase ramps, wheelchairs, home modifications, computers, vans and other assistive technology to lead an independent lifestyle. THE TWILIGHT GALA SPONSORS Table Sponsors: Gaylord Specialty Healthcare Rehabilitation \| Medically Complex \| Sleep Medicine Care beyond the ordinary. Fournier Motorcycles ATG rehab Life. In Motion. www.atgrehab.com Auction Table Sponsors: Mount Sinai Rehabilitation Hospital a SAINT FRANCIS Care Provider Beverage Sponsors: Hudson Home Health Care REHAB EQUIPMENT Eastern Rehabilitation Network Hartford Hospital Life can change in an instant. One moment I was standing at the sink. The next moment I fainted and fell into the wall, injuring my neck and spine. I woke up on the bathroom floor paralyzed from the neck down. No one thought I would move or walk again. After back surgery, I was transferred to Gaylord Hospital for physical and occupational therapy. My head and neck were locked in a brace and the only movement I had was in one thumb and a big toe. I felt hopeless. But the specialists at Gaylord treated me as if I were their only patient. They provided the encouragement and support that I needed to move toward recovery. I left Gaylord without a cane, walker or wheelchair! I am a walking miracle. To learn more or to see Richard’s Story, visit www.Gaylord.org. Gaylord Specialty Healthcare Rehabilitation \| Medically Complex \| Sleep Medicine Care beyond the ordinary The Obie Harrington-Howes Foundation is “dedicated to maintaining and improving quality of life by providing financial aid to individuals with spinal cord injuries.” A 501© (3) organization, it was originally established by members of the Darien community in response to Obie’s spinal-cord injury in a swimming accident on July 4, 1997, which left him paralyzed. The Foundation helps regional residents with spinal cord injuries and disease cover non-reimbursable medical expenses. Many people living with spinal-cord injuries reach a point where insurance is exhausted and state or federal funds are no longer available. We help bridge the gap between that point and the next level of independence. Since 1997 the Foundation has funded: - Wheelchairs, including sports wheelchairs - Assorted medical equipment - Computers and printers - Tuition assistance - Therapeutic leg braces - Hockey sleds and hockey equipment - Exercise equipment for rehabilitation - Beds - Ramps - Donations towards the purchase of vehicles which facilitate the return to work or school - Computerized speech augmentation devices - Respite care and home health care services - Minor home renovations - Door openers Congratulations NSCIA CT Chapter On Your 2010 First Annual Twilight Gala! We support and applaud your good work, and look forward to continued collaboration toward the goal of excellent care for all SCI patients and families. David A. Feingold, M.D. Matthew P. Raymond, D.O. Physical Medicine & Rehabilitation of Hartford, LLC Specialists in Spasticity Management With offices at: 35 Jolley Drive, Suite 101, Bloomfield, CT and 340 North Main Street, Southington, CT Phone (860) 243-3434 For information and appointments HARTFORD NEUROLOGY, LLC STEPHEN R. CONWAY, M.D. DAVID S. SILVERS, M.D. LAWRENCE S. BLUTH, M.D. ISAAC E. SILVERMAN, M.D. JOAO A. GOMES, M.D. 85 SEYMOUR STREET, SUITE 800 HARTFORD, CT 06106 6 NORTHWESTERN DRIVE, SUITE 302 BLOOMFIELD, CT 06002 2928 MAIN STREET, 1ST FLOOR GLASTONBURY, CT 06033 TELEPHONE (860) 522-4429 FAX (860) 249-6742 WWW.HARTFORDNEUROLOGY.COM Abner S. Gershon, M.D. Interventional Neuro Radiology Invision Medical, LLC 21 Arch Road, Suite B Avon, CT 06001 Office: (860) 673-1955 Cell: (860) 805-3624 Supporting NSCIA Are you looking for an IT partner who takes a business minded approach to your computer and computer network support? Who offers the Responsiveness, Budget Predictability and Peace of Mind you deserve? If so, then IT Direct is the company you’ve been looking for. Get started today by calling us at 860-656-9110 and request your complementary Problem Prevention Assessment. --- IT DIRECT “Getting You Connected” Phone: 860.656.9110 Web: www.gettingyouconnected.com --- CleanMaster, Inc. “Your Cleaning Source” Keith D. Todd, President FREE ESTIMATES P.O. Box 1019 Avon, CT 06001 phone: (860) 675-4004 cell: (860) 614-8726 email@example.com The Animal Hospital of Rocky Hill has been serving the community since 1978. Our veterinarians and highly skilled support staff are committed to providing high-quality medicine while maintaining a personal touch. Some of the services we provide are: - Laser Surgery - Dentistry - Senior Wellness - On-site Blood Analysis - Professional Grooming/Boarding - Ultrasonography - Digital X-ray - Prescription Diet Food The Animal Hospital of Rocky Hill is open Monday through Friday from 7:30 a.m. to 5:30 p.m. and Saturday from 8:00 a.m. to 1:00 p.m. For your convenience, there is an after-hours emergency hospital on the premises: www.animalemergencyct.net 860.563.4447. My best to the families and friends of NSCIA-CT. Wishing you a successful year in 2010. Paid for and approved by Dan Malloy for CT, Len Miller Treasurer David M. Barnett, D.M.D., F.A.G.D. Stuart M. Shoflick, D.M.D. Family Dentistry 2139 Silas Deane Highway Rocky Hill, CT 06067 Phone: (860) 529-0624 • Fax: (860) 721-0407 CLIFFORD G. BULLOCK PRESIDENT TEL (860) 870-7444 FAX (860) 870-1212 THE BULLOCK COMPANY BUILDERS AND REMODELERS ACCESSIBILITY CONTRACTORS www.bullockaccess.com 123 MIDDLE ROAD ELLINGTON, CT 06029 BENDER, ANDERSON & BARBA, P.C. 3308 Whitney Avenue Hamden, CT 06518 (203) 248-6440 “Your Time, Your Case, Your Firm, It’s All About YOU” 154 East Ave., 2nd Floor, Norwalk, CT 06851 Personal Injury Criminal Defense 203.642.3888 criminalmyFIRM.com injurymyFIRM.com “Ride-Away – America’s #1 provider of modified vehicles for people with disabilities. It’s no wonder why Ride-Away® customers remain so loyal — Superior Service! • 24/7 Roadside assistance • Full-service maintenance care • Operational maintenance check every 6 months • QAP Certified factory trained technicians • A reciprocal service network of over 180 dealers We Deliver Freedom. 104 Pakin Street East Hartford, CT 06108 “Ride-Away is a company that uses its success to build a better life for everyone we touch.” — Mark Lane Ride-Away President NEW BRITAIN MEDICAL SUPPLIES Specializing in Urology • Wheelchairs • Light Weight Ramps • Diabetic Supplies • Ostomy Supplies • Walkers • Wound Care • Hospital Beds • And Much More Over 20 Years Experience WE ACCEPT MEDICARE AND MEDICAID AND ALL OTHER INSURANCES 860-224-9017 www.NewBritainMedicalSupplies.com
Minicomputers move up with mixed memories MOS technology presents another American ideal to satisfy your revolutionary ideas... THE MCS 2050 RAM IS HERE! THE MCS 2050, a 256 BIT STATIC RANDOM ACCESS READ/WRITE MEMORY can do a lot for you, your customers, your nation... fast, without a big power grab! Consider the tenets to which its dedicated: Low Power Operation — Typically 200 Milliwatts, Static Circuitry — No External Clocks Required, Access Time—Typically 600 n sec, Standby Power Mode — Power Dissipation — 20 milliwatts, 200 n sec Input Address Skew, TTL/DTL Input and Output Compatibility, Complete Decoding Internal to Memory, Wire Or Capability, Supply Voltages —9V, −7V, +5V. THE MCS 2050, 256 BIT STATIC R.A.M. . . . FROM MOS TECHNOLOGY . . . IN VALLEY FORGE . . . WHERE IDEALS HAVE ALWAYS BEEN TRANSPLANTED INTO ACTION. We've Turned A Technology Into A Company EASTERN REGIONAL OFFICE—REGIONAL SALES DIRECTOR—Mr. William Whitehead, MOS TECHNOLOGY, INCORPORATED, 88 Sunnyside Blvd., Suite 307, Plainview, New York 11803 • Phone: (516) 824-5240 • ALABAMA, ARIZONA, GEORGIA, HAWAII, KENTUCKY, LOUISIANA, TENNESSEE, MISSISSIPPI—Parrrie Associates, P.O. Box 1424, Huntsville, Alabama 35807 • Phone: (205) 336-5656 • 2001 McCoy Road, P.O. Box 1320, Orlando, Florida 32801 • Phone: (305) 855-0848 • Box 558, Greensboro, North Carolina 27403 • Phone: (919) 273-3040 • NEW ENGLAND STATES—Victor Associates, 179 Union Ave., Framingham, Massachusetts 01701 • Phone: (617) 879-3370 • NEW JERSEY, NEW YORK (WESTCHESTER COUNTY, LONG ISLAND)—Falk Baker Associates, 383 Franklin Avenue, Nutley, New Jersey 07110 • Phone: (201) 667-4539 • PENNSYLVANIA, SOUTHERN NEW JERSEY—R. L. Smith, Box 338, King of Prussia, Pennsylvania 19406 • Phone: (215) 255-0000 • DELAWARE, MARYLAND, WASHINGTON, D.C., VIRGINIA, W. VIRGINIA—Bernard White & Company, Inc., 1 Church Lane, Baltimore, Maryland 21208 • Phone: (301) 484-4400. CENTRAL REGIONAL OFFICE—REGIONAL SALES DIRECTOR—Mr. Alan Mattal, MOS TECHNOLOGY, INCORPORATED, 10400 W. Higgins Rd., Suite 631, Rosemont, Illinois 60018 • Phone: (312) 298-2038 • ILLINOIS, INDIANA, WISCONSIN—Coombs Associates, Inc., 1001 E. Touhy, Des Plaines, Illinois 60018 • Phone: (312) 298-4830 • OHIO, KENTUCKY, WEST VIRGINIA—MOS Technology, Inc., P.O. Box 521, 123 Washington Street, Ohio City, Ohio 44115 • Phone: (216) 225-4562 • MICHIGAN—R. L. Messerly & Co., 18411 W. McNichols Road, Detroit, Michigan 48219 • Phone: (313) 535-6600 • MINNESOTA, NORTH DAKOTA, SOUTH DAKOTA—Met Ford Company, Inc., 7380 Bush Lake Road, Edina, Minnesota 55435 • Phone: (612) 941-7600 • MISSOURI, KANSAS, NEBRASKA, IOWA—Harlan J. Weisler & Assoc. Inc., 2050 Woodson Road, St. Louis, Missouri 63118 • Phone: (314) 428-3933 • TEXAS, OKLAHOMA, ARKANSAS, LOUISIANA—Norvell Associates, 10210 Monroe Drive, Dallas, Texas 75220 • Phone: (214) 357-6415. WESTERN REGIONAL OFFICE—REGIONAL SALES DIRECTOR—Jac Turin, MOS TECHNOLOGY, INCORPORATED, 21722 Durban Drive, Patio Bldg., Suite 221, Newport Beach, California 92660 • Phone: (714) 833-1600 • ARIZONA, NEW MEXICO—Ford Engineering Assoc., P.O. Box 100, Arcadia Station, Phoenix, Arizona 85001 • Phone: (602) 555-3193 • CALIFORNIA, NEVADA—Bertrand & Zooolian, Inc., 7340 Florence Avenue, Suite 205, Downey, California 90241 • Phone: (213) 927-4406 • HUNTSVILLE—John Fox Plaza, San Francisco, California 94102 • Phone: (415) 626-8576 • COLORADO, UTAH, WYOMING—R. G. Enterprises, 1107 South Pearl Street, Denver, Colorado 80210 • Phone: (303) 744-2468 • WASHINGTON, OREGON, IDAHO, MONTANA—J. A. Tudor & Assoc., Inc., 2605 Western Avenue, Seattle, Washington 98121 • Phone: (206) 682-7444. INTERNATIONAL SALES REPRESENTATIVES—ARGENTINA—T. R. C. Electronica, S.A.C.I.e.l., Cangallo 4450, Buenos Aires, Argentina • Phone: 88-4044/5/6, Mr. M. Lissner • AUSTRALIA, NEW ZEALAND—Impeckable Limited, Imperial House, 100 King Street, London W3 9LN, England • Phone: 01-992-5388 • FRANCE—Bureau de Liaison, 113 Rue de L'Universite, Paris • Phone: 551 00 00 • GERMANY, AUSTRIA, SWITZERLAND—International Micro Electronics, Inc., Arabellahaus 1838, Arabellastr. 5, 8000 Munich 81, Germany • Phone: 92-321 • INDIA—Electronic Enterprises, 46 Karami Building, New Charni Road, Bombay 4, India • Phone: 375375 • ISRAEL—Eastronics, Ltd., 75 Halfa Road, P.O. Box 21029, Tel-Aviv, Israel • Phone: 38359 • ITALY—Special-Ind. Corporation, Piazza Spoforno, 20159 Milano, Italy • Phone: 632-111 • JAPAN—Kokusai Kohan Company, Ltd., 1-7, Chome Kohijimachi, Chiyoda-Ku, Tokyo 102, Japan • Phone: 263-3211 • SWEDEN, DENMARK, FINLAND, NORWAY—G. F. Forsberg, Forsvarsgatan 50, P.O. Box 79, Farsta 1, Sweden • Phone: 647040 • SWITZERLAND—Ernst M. Egli, Ingenieur—Bureau Aeg, Witikonstrasse 295, Zurich (CH 8053), Switzerland • Phone: 53.38.11. The big buy in big-screen lab scopes 50 MHz $2200 100 MHz $2550 1 GHz $2550 You don't have to give up performance capabilities to save money on a big-screen scope; HP's 182A gives you both. For $2200, you can get a mainframe, a 50-MHz dual-channel amplifier, and a delayed-sweep time base. This combination gives you the biggest display area of any high-frequency scope (8-div x 10-div, 1.3 cm/div), 5 mV/div sensitivity, and 10 ns/div sweep time. And that's only the beginning. The 182A system isn't limited to 50 MHz in the plug-ins it can accept. Thanks to HP's pioneering advances in CRT technology, the 182A will take the entire family of 180 System plug-ins. For $2550 you can get a 100 MHz system (mainframe, dual-channel vertical amplifier, and non-delayed sweep; delayed sweep $450 extra). Also available at $2550 is HP's new 1 GHz sampling system (mainframe and plug-in) that's as easy to use as a real-time scope! And the 182A is the only large-screen lab scope that has these capabilities. So, if you're in the market for a high-frequency scope—get the 182A in the 50-MHz configuration, and protect yourself against having to buy a whole new system for 100-MHz capabilities in the future. It's like getting free "bandwidth insurance" with your mainframe! For further information on the 182A, contact your local HP field engineer, or write Hewlett-Packard, Palo Alto, California 94304. In Europe: 1217 Meyrin-Geneva, Switzerland. Scopes are changing. Are you? Electronics/October 11, 1971 Save $15 on our new communications kit and get something special in your system. The kit contains 12 Schottky diodes, 1 pin diodes, and one low-noise transistor. As well as application notes and data sheets. If you bought these components separately, it would cost you more than $34. But right now you can buy the whole kit and kaboodle for $19.40. We're sure that once you've tried them, you'll come back for more. This is the kind of performance that will bring you back: The HP 2800 Schottky diode is ideally suited for high level detector applications with a breakdown voltage of 70 volts. The HP 2835 Schottky diode with a breakdown voltage of 5 volts and a forward voltage drop of .33 volts at one mA makes an ideal VHF/UHF mixer. The HP 3080 pin diode as an AGC element guarantees low cross-modulation and intermodulation down to frequencies of 5 MHz. The HP 35821A transistor makes an ideal low noise RF amplifier with a noise figure of 3 dB at 1 GHz. So if you're looking for ways to improve the performance of your RF amplifiers, mixers, detectors or AGC's, tear out the coupon and a check and send them both in. You may also want one of our very popular Schottky diode kits. For $8.40, you get eight each of three different Schottky barrier diodes. Try them and see what a difference they make in your circuits. Clip this coupon and mail with check, money order or P.O. to: Hewlett-Packard Associates, 620 Page Mill Road, Palo Alto, California 94304. Your local HP sales office can also handle the order. $34.00 value for $19.40! (offer expires on January 31, 1972) - Send me one of your communications kits. (HP 5082-0051) - Here's $8.40 for a diode kit. (HP 5082-0050) (California residents add 5% sales tax) Name____________________________________ Company__________________________________ City/State/Zip_____________________________ Frequencies □ less than 10 MHz □ 10 to 250 MHz □ 250 to 500 MHz □ 500 to 1000 MHz □ greater than 1000 MHz HEWLETT PACKARD COMPONENTS 01109 Electronics/October 11, 1971 Electronics Review SOLID STATE: Current-switching scheme promises fast LSI logic, 29 MEMORIES: Sangamo division buys Sylvania's Soniscan, 29 LASERS: High-gain, flash-pumped laser gives coded pulse, 30 COMMERCIAL ELECTRONICS: Rotating-ring watch display may leapfrog others, 30 INDEX OF ACTIVITY: 31 COMPUTERS: Iliac 4 takes another step away from campus, 32; Univac aims 1616 at military, FAA, 32; Heat will limit specs, 34 COMPANIES: Solitron to head down C/MOS path, 34 EMPLOYMENT: Self-help groups see some daylight, 36 GOVERNMENT: Electronics figure heavily in negative trade balance, 36 PACKAGING: Brushes, tags bid for place in connector sun, 38 MILITARY ELECTRONICS: Displays, computers to teach dogfighting, 38 ADVANCED TECHNOLOGY: Diode key to first precise measure of speed of light, 40 FOR THE RECORD: 40 Technical articles COMPUTERS: Evolution breeds a minicomputer that can take on its big brothers, 62 (cover) COMPUTER-AIDED DESIGN: A convenient way to model the handy zener diode, 68 CIRCUIT DESIGN: Designer's casebook, 73 SOLID STATE: Impatt diodes and millimeter-wave applications grow up together, 78 COMMUNICATIONS: Digital data links deserve a bit error rate detector, 82 PACKAGING & PRODUCTION: Solving interconnection problems in big multilayer pc boards, 88 Probing the News COMPANIES: TI sees computers in its future, 93 SOLID STATE: Linear IC market in ferment, 97 INTERNATIONAL: Electronics to impact European automobiles, 99 AEROSPACE: Joint space talk outlook: clear skies, 103 New Products IN THE SPOTLIGHT: Linear 'LSI' shrinks scope for field work, 107 INSTRUMENTS: Light meter is highly stable, 109; Digital wattmeter reads to 20 kW at 20 kHz, 109 DATA HANDLING: Controller is multipurpose, 113; Portable cassette recorder designed for remote input, 113 PACKAGING AND PRODUCTION: Board tester isolates faults, 117; Logic aids in wire identification, 117 SEMICONDUCTORS: TI adds three Schottky TTLs, 123 MATERIALS: 126 Electronics International THE NETHERLANDS: Read-only memory cut out for the job, 131 WEST GERMANY: Fingering the display, 132; VW computerizes bugs, 132 FRANCE: Home-grown CAS system enters world competition, 133 JAPAN: Coding scheme opens way for semiconductor laser link, 134 GREAT BRITAIN: Car's image displaces radar in speed checking gear, 134 Departments Publisher's letter, 4 Readers comment, 6 40 years ago, 10 People, 14 Meetings, 21 Electronics Newsletter, 25 Washington Newsletter, 51 New Books, 126 New Literature, 128 Personal Business, PB1 International Newsletter, 29 Title R registered U.S. Patent Office. Copyright 1971 McGraw-Hill Inc. All rights reserved, including the right to reproduce the contents of this publication in whole or part Volume No. 44, Number 21 Our news staff, as we've pointed out here before, is strong on veteran reporters with newspaper, magazine, and wire service experience. But our eagerness to get the latest news into each issue of Electronics* is not limited to the news side of the magazine. Our technical section is a more difficult place to break an important story. But sometimes an important development warrants extra efforts that enable us to be first and to describe in depth the technical details of a technological advance. In this issue, for example, the significant evolutionary step in the development of minicomputers embodied in the new Digital Equipment Corp. 11/45 computer, with its mixed semiconductor memory (see page 62), impelled us to cut some corners on deadlines and feature it on the cover. Impressed as he was with DEC's work, Dick Gundlach, our associate editor who shepherded the article into the magazine, was struck by the improbable environment that spawned the development. "The same old brick factory that made blankets for the Union troops in the Civil War, is now turning out state-of-the-art computer gear," says Gundlach. It seems that, like other one-industry New England towns after the Second World War, Maynard, Mass., had a huge empty mill on its hands when the textile company moved out. When our computer editor, Wally Riley, first visited DEC a half dozen years ago, it was just one of many tenants rattling around in the old mill. Now, when Riley and Gundlach visited DEC to work on the article, the company occupied the whole place—and many newer glass-and-steel buildings around New England. There may come a day when an American or a Russian astronaut will owe the very air he breathes to his "enemy." The work now going on to coordinate U.S. and Soviet spacecraft design to make space rescues possible (see Probing the news, page 103) has the solid ring of history in the making. And in a way, it represents an historic turnaround in the political relations between the two countries, points out Ray Connolly, our Washington bureau manager, who wrote the story. The politics of U.S.-Russian confrontation has taken on a whole new tone. It used to be that the leaders of the U.S. space effort justified their big budgets by warning that the Russians might beat us to the punch. But now, with the U.S. successful in its race to first set foot on the moon, the competition has quieted down. Indeed, the most powerful argument now used by Washington politicians is the "peace ploy." But there are still elements of the fundamental competition between the U.S. and Russia. "Some suspicious people equate the Russian's strong cooperation in this project—and it is very strong—with a desire to know a lot more about our electronics expertise," says Connolly. "There appear to be some serious shortcomings in the Russian's communications systems. The fact that 10 of the 15-man Russian delegation are electronics experts is no accident." This $180 camera gives you a picture 15 seconds after you pull the trigger. The CR-9 is a new camera specifically designed for oscilloscope trace recording. It costs only $179.95.* And it weighs less than 24 ounces (700 grams). So, unlike permanently-mounted cameras, it can easily be used with more than one oscilloscope in your lab. The Polaroid CR-9 Oscilloscope Camera is designed to take any one of 8 interchangeable, light-tight hoods. These fit almost all oscilloscopes with 6 x 8, 6 x 10, and 8 x 10 cm graticles. (You specify the hood you need when you buy the camera; additional ones cost a little extra.) You don't need to focus, because the hood holds the camera at exactly the right distance from the CRT display. The camera gives you a picture that's virtually distortion-free. All you have to do is slip in an 8-exposure pack of Polaroid 3000-speed, self-developing film, and you're ready to shoot. The CR-9 is a real problem solver for scientists and engineers. Because one picture is worth a thousand waves. For additional information write: Polaroid Corporation, Dept. 35-218, 549 Technology Square, Cambridge, Mass. 02139. (In Canada: 350 Carlingview Drive, Rexdale, Ontario.) The new Polaroid CR-9 Oscilloscope Camera * Suggested list price. Polaroid® Looking for Bit Errors? Actual Eye Pattern 190 Here they are! Kingsley Roby of our engineering staff uses the DCS Model 4660 Link-BERC (Link-Bit Error Rate Calculator) to check out our own 10 megabit PCM equipment. Why not let him help you check your T-1, T-2, standard IRIG or special digital transmission links? With selectable bit-blanking, clock-phasing, bit error rate intervals and standard internal calibration... analyzing your link will be a lot easier and less time-consuming than complicated computer or pulse generator / scope lash-ups. Also remember your printer (or one we will supply) can be connected to Link-BERC to provide a hard copy record of system bit error rates. Call us or circle the reader service number for a free application brochure and a demonstration at your facility. Model 4660 Link-BERC DATA-CONTROL SYSTEMS, INC. Commerce Drive, Danbury, Connecticut 06810 203 743-9241 Readers comment Jeweled movements To the Editor: The Washington Newsletter of Aug. 16 erroneously reported that Hamilton Watch Co. is "the last fully integrated watchmaker in the U.S. able to turn out precision timing devices vital to this country's industrial base." In fact, Bulova is the only movement producer left among the 60 American companies that had produced jeweled movements domestically during the past century. Elgin ceased domestic manufacture of jeweled movements in 1967 and Hamilton ended domestic movement manufacturing in 1969-70. The Defense Department acquisition of Hamilton's facilities suggests there is a renewed appreciation of the unique defense capabilities of the quality watch industry. But Bulova—not Hamilton—is the sole domestic producer of precision jeweled movements. William Gowen Bulova Watch Co. New York City - Mr. Gowen is correct in noting that Bulova is the only domestic integrated watchmaker still producing precision jeweled movements for watches. However, both Bulova and Hamilton make such movements for applications other than watches. EEs in Britain To the Editor: Those taking part in your ballot on a union for EEs [Aug. 2, p. 50; Sept. 21, p. 72] will be glad to know that we are in the process of solving the same problem in England. We have formed, with the full approval of the engineering institutions, our own union, the United Kingdom Association of Professional Engineers. Confined to degreed engineers, it will be equivalent to the British Medical Association. It has the powers of intervention between employer and employee and is technically a trade union; but through its special contacts and lobbying powers it achieves its ends without disruption. We are pressing for a common code of professional conduct properly drawn up and enforced. We have also published a guide to salaCoup de RAM. When the 1101 random access memory was introduced, we knew it was a winner. So, we second sourced it. Likewise with the 11011, 1101A and 1101A1. Now we've designed our very own 256 word x 1 bit RAM, the MM1101A2. Basically, our new MM1101A2 gives you twice the speed of the 1101A1 (namely, 500nS Max.) with no increase in power. (A design feat we're frankly quite proud of.) Otherwise, our new MM1101A2 has all the other features you've grown to know in the 1101 series. Including the use of silicon gate technology. So, whether you need a good second source for the 1101, 11011, 1101A and 1101A1, or a good first source for our fast new MM1101A2, just drop us a line. Or give us a call. Or TWX us. Or cable if necessary. But don't just sit there. National Semiconductor Corporation, 2900 Semiconductor Drive, Santa Clara, California 95051. Phone (408) 732-5000. TWX: (910) 339-9240. Cable: NATSEMICON. Complete HF Receive Antenna capability in a 50 meter circle How Hermes did away with vast rhombic or log-periodic antenna farms. Shoed away by a shrewd array: Take 1 meter diameter loops, 4 meters apart, and get an omni directional broad-band receiving array. Covers 2 - 32 MHz. Optimum beam characteristics for both long and short range communications. Rosette configuration of linear arrays gives a number of overlapping high gain beams - all available simultaneously. * Using less than one hundredth of traditional real estate. Aperiodic Loop Systems are shrewd enough for restricted space, quick set up, roof mountable, or just below ground level. And portable version now available. From polar icecap to sweltering tropics. Installed in more than 52 locations for more than ten governments and more than 32 of their agencies. Give up the antenna farm. ASK US Hermes Electronics Limited Suite 315 2020 F Street N. W. WASHINGTON D.C., 20006 Telephone 202 296-2978 TWX 710-822-1106 Readers comment ries and conditions of service. We intervene when an employer deals with an individual unfairly. (We are acting tough with an employer who has dismissed 38 of his professional engineers for refusing to join a blue-collar union with which he signed an agreement.) No union can enforce guaranteed job security, but when there have to be redundancies, we advise the employer on how to handle the problem and see that our members get the best possible terms and their full rights under the law. We then redeploy them with the aid of our appointments service; we offer career advice when required; we shall one day be able to offer career planning as a package service. We are lobbying for portable pensions, mid-career training, and a promotion ladder on the technical side. Ukape emphasizes differentials, independence, loyalty to the firm, freedom of choice, and advancement by talent and not by age. Our member retains his professional integrity and outlook, remaining completely separate from the blue-collar union system, and yet strong enough to withstand its less desirable manifestations. He can get on with his engineering job, and leave the sordid details to Ukape, which sees that he gets a fair, not extortionate, reward. There is consequently no prejudice in management against Ukape. For Ukape, there are no "two sides to industry"; just one profession of engineering. R. L. Clarke United Kingdom Association of Professional Engineers London, England Who's who To the Editor: The article on Hall-effect ICs [Aug. 2, p. 46] attributes the discovery of the effect to E.F. Hall of Ohio State University. In fact, credit should go to E.H. Hall, fellow of Johns Hopkins University (E.H. Hall, "On a New Action of the Magnet on Electric Currents," American Journal of Mathematics, 2, 287, 1879). Jerome H. Perlstein Johns Hopkins University Baltimore, Md. GE WILL BEND FOR YOU! Now you can have performance, stability and a choice of six different lead configurations with GE's POWER-GLAS C122 silicone encapsulated SCRs. General Electric makes your mounting procedures simpler by factory forming the C122 round leads to match six standard configurations. The C122 also features a tab mounting hole that permits torque limit free mounting, thus eliminating possible pellet damage associated with center-mounting-hole packages. The 8 ampere C122 is available in 50 to 500 volt types. These features, plus the stability achieved by POWER-GLAS passivation, make the C122 the best value in plastic packaged SCRs. Standard 200V versions in 1000 lot quantities cost 85¢ each. GE offers the industry's broadest line of SCRs, Triacs and Triggers designed for all applications. Contact your local authorized GE distributor for complete information on GE's POWER-GLAS SCR and Triac products. GE'S POWER-GLAS MEANS IMPROVED PERFORMANCE AND RELIABILITY FOR YOU. GENERAL ELECTRIC \| BASIC TYPES \| TO-66 EQUIVALENT \| PRINTED CIRCUIT BOARD TYPES \| \|-------------\|------------------\|----------------------------\| \| \| \| \| Electronics/October 11, 1971 Over the years, helping companies get started in Georgia, we've learned a few things about finding workers. And we've put what we've learned into a computerized manpower matchmaker that's second to none. You tell us your workforce requirements. We'll tell you which Georgia communities can fill them. The information is accurate and updated quarterly for over 400 Georgia communities. Give us a call at 404-656-3599. Or send us the coupon. And we'll start talking to the matchmaker. In Georgia the odds of finding the workforce you need are 400 to 1. In your favor. --- Georgia Department of Industry & Trade Industry Division, Dept. EL-7, P.O. Box 38097 Atlanta, Georgia 30334 Please tell me more about your manpower matchmaker. And how Georgia will train my workforce at no cost to me. Name __________________________ Title __________________________ Company _______________________________________________________ Address ________________________________________________________ City _______________ State ______ Zip ____________ --- 40 years ago From the pages of Electronics, October 1931 Out of each depression, economists tell us, some new invention or industry has always arisen to lead the world into new business activity. If such is to be the way out of the present depression, then certainly on the horizon of 1931 there is no more promising prospect to perform such service, than the electronic tube with all its manifold applications. But always over the brightest picture of electronic possibilities, hangs the spectre of patent infringement. If the owners of electronic patents could see the advantage of cross-licensing and exchanging patent privileges under a voluntary pooling plan that would stimulate both manufacture and profits, how far would the development of this art be pushed ahead? In schoolrooms during the winter months, natural illumination goes through wide changes, imposing burdens on little eyes. But with a photocell electric lights will be automatically switched on each time the natural light falls below a predetermined value. Since there are 400,000 schoolrooms having electric lights, the market for photocell control looks bright. A device which amplifies an electric current ten quadrillion times was exhibited by E. S. Darlington of the vacuum-tube department of the General Electric Company at the radio and electric show of the Electric League of Washington, D.C., in September. The device is a low-grid-current tube, which in conjunction with a thyratron tube is capable of utilizing $0.000000000000000001$ (10^{-22}) ampere to control 0.1 ampere—or 100 milliamperes—directly. To demonstrate the remarkable sensitivity of the combination, Mr. Darlington utilized the relatively small current generated by rubbing an amber rod with a piece of paper, to turn on and off a 10-watt incandescent lamp, with the amber rod at distances varying from 5 to 15 feet. A current of $10^{-17}$ ampere thus directly controlled the 0.1 ampere used by the lamp. WE JUST SHIPPED OUR 100,000,000th SNAP ACTION SWITCH AND NOBODY NOTICED SIGH Sad but true. We were too busy working on our second hundred million snap action switches to bother much celebrating our first hundred million. And busy is the key word. Today, for instance, we'll ship about 70,000 snap action switches. And another 70,000 tomorrow. It goes on at this pace every working day. That's 17 million dependable Cherry switches a year to actuate 17 million dependable products. All because product designers like Cherry's long-life coil-spring mechanism. And because production people find high-overtravel Cherry switches easier—and faster—to install. See for yourself why Cherry snap-action switches are so popular. Send for the complete Cherry switch catalog. CHERRY ELECTRICAL PRODUCTS CORP. 3608 Old Deerfield Road—Highland Park, Illinois 60035 Makers of patented Leverwheel/Thumbwheel Switches, Matrix Selector Switches, Snap-Action Switches and Keyboards. The GR Systems Family Systems capability from General Radio Resistance Trimming is only the beginning... complete functional trimming is now a reality Resistance trimming at speeds to 10,000 trims per hour and accuracies to ±0.1% is only the beginning with Micronetic's new Model 80 Laser Trim System. Now you can select the specific functional trimming capabilities you need to trim your fully assembled hybrid circuits. Functional trimming is fast and economical Trimming assembled hybrid circuits to a functional specification can compensate for active device variations through the adjustment of only a few resistors. Throughput increases, yield improves, and the need for a separate final test operation is eliminated. In addition, because the Model 80 can functionally trim to compensate for wide fluctuations in active device parameters, specifications on these active devices can be relaxed for substantial added cost savings. General Radio instrumentation makes it possible Micronetic Systems has joined forces with General Radio Company to provide a complete functional trimming system. The Model 80 measurement system is built by General Radio in modular plug-in form. The basic module is the Resistance Measurement Unit, standard on the Model 80. Separate plug-in modules to measure ac and dc voltage and current, frequency, and capacitance are available as options to provide complete functional trimming capabilities. Functional software included Micronetic's exclusive Resistor Trimming Language (RTL) software package allows complete flexibility by controlling all parameters of the laser trim system with simple English language statements. Instructions to control all the GR plug-in modules are included in the RTL software supplied with every Model 80. Lowest capital investment The standard system, complete with the basic resistance measurement module and RTL software, costs only $59,000, net FOB Watertown, Mass. Modular expansion capability means you pay only for the capability you need. General Radio service The complete Model 80 System carries a full one-year warranty. Service, if required, is performed by General Radio's worldwide service organization. Write or call Micronetic Systems or your local GR sales engineer for information on how you can save money with the Model 80. Resistance trimming is only the beginning... People "I knew the silicon-on-sapphire technology was ready, but the company wasn't making any effort to shift into the manufacturing," says Joseph R. Burns, describing the situation during the last of the 11 years he spent at RCA's David Sarnoff Research Laboratories in Princeton, N.J. Burns' reaction was typical—he quit to set up his own company. But his timing, he's the first to admit, was terrible. It was September 1970, the stock market was down, and "electronics" was a dirty word among Wall Street investors. But Burns, holder of a Ph.D. in electrical engineering and a leading contributor to RCA's SOS expertise, when he was in charge of semiconductor device applications there. [Electronics, July 20, 1970, p. 82], figured he needed $1 million to get his company started. He got it and now as president of Inselek, the 34-year-old, 6'5" Burns, a forward on Princeton's basketball team for three years, is ready to make some big noises in his old college town. Inselek, billing itself as "the silicon-on-sapphire components company", is the electronics industries' first commercial supplier of SOS wafers—offered to semiconductor houses like Fairchild Semiconductor and Texas Instruments—and it will soon introduce both memory and linear SOS devices for circuit designers. Burns predicts great things for the fledgling technology. The inherently great isolation afforded by the sapphire insulator will lead to very high-speed MOS products having very high packing densities and dissipating low powers, he points out. And they will be extremely competitive with bipolar devices. By way of example he cites two devices he helped develop while at RCA; a 50-bit dynamic shift register operating at a spectacularly high 90 megahertz (for the Avionics Laboratory at Wright-Patterson AFB, Ohio) and a 256-bit, fully-decoded random access memory with 35-nanosecond access time (for NASA). Inselek will introduce in the next few weeks a 64-bit static RAM, compatible with transistor-transistor logic and with less than 40-nanosecond access time. And early next year, the company will have a 256-bit unit with less than 75-nanosecond access time. Both units will dissipate about 0.1 milliwatt per bit compared with the 2 to 5 mW/bit dissipated in the same speed range by bipolar units, Burns asserts. He is also planning large-capacity RAMs—1,024- and 2,048-bit devices for mid-1972. In the linear area, Inselek's first offering will be a quad-transistor configuration on a chip—four MOS devices to serve as building blocks for linear ICs. Overall, Burns' feeling is that Inselek is "just barely scratching the surface of the technology, without really pushing the devices' true capabilities." He predicts at least a factor of four improvement in both speed and packing density in the next 18 to 24 months. The communications and electronics industries don't seem to be having any problems with President Nixon's new economic policies [Electronics, Aug. 30, p. 21]—or at least that's the view the Federal Communications Commission (FCC) gets, concedes Executive Director John M. Torbet. "Of course we don't try to make economic policy here," explains Torbet, recently charged by FCC Chairman Dean Burch to field questions on the President's new economic stabilization plan, "but we have to interpret it for the industry in the industry's terms." To date, all the questions Cut costs... increase production... improve quality... just 90 minutes after your new Fluke ATE* comes through the door. Terminal/10, Fluke's new series of automatic test, control and data acquisition equipment comes to you after a three year development program and more than a year of on-line testing. Here's the ATE you can put right to work verifying and calibrating lab or production multimeters or ac and dc sources. Check out analog and combination analog-digital subassemblies. Collect, manipulate, reduce, store and report high accuracy data. Calibrate oscilloscopes, high frequency signal generators, meters and counters. Do it with ease and assurance because each of these ATE systems has been specifically designed with the user in mind by one of the world's outstanding precision electronic instrument firms. The Fluke Terminal/10 uses a non-dedicated computer so you won't have to alter it in any way or take it away from other tasks. In many instances you can plug right into the computer you already have. Up to 16 Terminal/10 systems can be served by a single computer. We use BASIC, a popular language compatible with a wide range of system requirements and standard industry-wide time-sharing packages. Other languages are also available. Instruments and processes can be added to the system at any time without changes in CPU core size requirements. All systems come with self-check routines to give you a fast way to verify performance and establish system pre-cal, post-cal and so on. Systems are manually programmable from the Fluke interface processor and CRT terminal. The CRT display interfaces with the operator to give him rapid instruction or provide a means of program development. For proportional systems control, an auxiliary keyboard is provided. The EIA compatible interface gives you a broad range of CPU configurations using an initial computer utility meeting both cost and need parameters. At the same time, you can economically expand the system to meet new requirements. The new Fluke Terminal/10 is on the air now and a touring demo unit will be in your area soon. To arrange a demonstration or get complete information call your local Fluke sales engineer or contact us directly. Automatic Test Equipment Fluke, Box 7428, Seattle, Washington 98133. Phone: (206) 774-2211. TWX: 910-449-2850/ In Europe, address Fluke Nederland (N.V.), P.O. Box 5053, Tilburg, Holland. Phone: (04250) 70130. Telex: 884-52237/ In the U.K., address Fluke International Corp., Garnett Close, Watford, WD2, 4TT. Phone: Watford, 33066. Telex: 934583. When you need something else to make it work... Say, as an engineer, you design a new product. Then you discover you need something else to make it work. What could be more frustrating? Especially if you don’t know where to find that something else. Stop biting your nails. We’ll help you find it. It’s probably in the Northern Plains Industrial Catalog. This complete, indexed, cross-referenced, multi-everything compilation service contains valuable information on something else. It tells who makes them, and where they are made. Clip the coupon and mail it. The service is free. Now you know. Aren’t you glad you asked? Iowa, Kansas, Minnesota, Nebraska, South Dakota, Upper Michigan and Wisconsin. INDUSTRIAL CATALOG Area Development Dept. NORTHERN NATURAL GAS COMPANY 2223 Dodge Street • Omaha, Nebraska 68102 Dear Sir: Can you put me on the track of a possible source of ____________________________ ______________________________________________________ □ Drawing enclosed NAME __________________________________________ TITLE _______________ COMPANY __________________________________________ ADDRESS __________________________________________ CITY __________________________ STATE _______ ZIP _______ People Torbet: We don’t make economic policy he’s forwarded to the Cost of Living Council have been concerned with the complex formula used to calculate television advertising rates—dependent upon changes in audience, season, and time of day or day of week. Torbet took the administrative helm of the FCC earlier this year after wearing the uniforms of the four services—most recently as deputy chief of staff of the U.S. Air Force Academy, Colorado Springs. The lean, gray Torbet, with degrees from Michigan State University and the Air Force Institute of Technology, is responsible for the day-to-day operation of the FCC. Activities on a recent day ranged from requesting a budget increase from the White House to ordering removal of Gay Lib’s lavender graffiti from the stone front of the Federal Communications Commission’s Washington headquarters. Torbet also has had to solve the FCC’s own problems resulting from Nixon’s new economic policies. “For example, a personnel realignment we didn’t foresee,” he admits. “Although we haven’t heard the details on the 5% Federal job cutback from the White House, when we do, we’re ready to accomplish it through existing vacancies and attrition. “In addition, we have kept the President’s views on economic stabilization in mind in writing the fiscal ’73 budget request,” he continues. Torbet last week asked the Office of Management and Budget for only a modest increase over last year’s $31 million. “We’re anxious to move into new programs,” he says, “but with the new economic policies, we will not achieve all of those we want.” in Ion Implantation there is only one manufacturer with proven performance: Accelerators, Inc. 40 Implanters already in the Field processing thousands of wafers per week after week after week... There are lots of people talking about Implanters these days, but not too many delivering them. Maybe that's because it's a little more difficult to build and deliver working systems than it is to talk about them. Or, perhaps success in this business takes a greater commitment than most are willing, or able, to make. After all, Implanters are Accelerators and it takes a lot of time and effort to put together an Accelerator company. We've done it. That's our name, and that's our business—our only business. And today, 5 years after starting the Implanter business, it's still our performance vs. the good intentions of others. If the industry's rapid swing towards Ion Implantation has got you thinking about a facility of your own, why not give us a call? We'll recommend a system, from our wide line of research and/or production Implanters, that best suits your needs. We can usually deliver in 60 days, and while you're waiting, we'll get you started by Implanting your wafers at our own facility. Accelerators, Inc. The full-time Ion Implantation Company For information on our Implanters and/or Implantation Service, write or call Accelerators, Inc./212 Industrial Blvd./Austin, Texas/78745/512-444-3639 Miniature LOW CURRENT—HIGH VOLTAGE MULTIPLIERS and POWER SUPPLIES HIGH VOLTAGE CAPACITORS Protected by "JET SEAL" CUSTOM DESIGNED HIGH VOLTAGE CAPACITORS MICA CAPACITORS and UHF TUBE SOCKET SYSTEMS CERAMIC GLASS AIR PLASTIC VARIABLE TRIMMER CAPACITORS Miniature CERAMIC SMI FILTERS World's Broadest Line Designing with Quality ERIE Electronic Components makes life just a little simpler... Try to beat this winning combination. A full line of electronic components produced to quality and performance standards that are right on target with your design and equipment needs. ERIE has long been known as the quality name in electronic components. And today we're the IN place to go for top-of-the-line components. ERIE plays a vital role in military, industrial, commercial, home entertainment and communications systems. You name it...ERIE's there. Why don't you join the swing to ERIE? Write TODAY for literature describing any component family illustrated. And if you would like to see your ERIE man, just jot a note. We'll do the rest. Who knows, ERIE components may even make your life just a little simpler. ERIE TECHNOLOGICAL PRODUCTS, INC. Erie, Pennsylvania 16512 Circle 19 on reader service card Try our straightforward method of reducing fixed resistor costs. You can't blame engineers or purchasing agents for trying to save every last penny on resistors these days. But lowest price doesn't necessarily mean lowest cost. For example, most manufacturer's color bands won't stand up to the cleaning methods used to remove excess flux. Or they darken and become illegible from the heat produced in normal usage. This can mean costly identification errors on your production line. The unnecessary expense of rework. Our solution? A-B quality. Bright, crisp identification of Allen-Bradley's specially formulated paints. Baked on to stay on. Designed to resist aging. Discover the other ways to save money. Ask your nearest A-B distributor for our free booklet "7 ways to tell the difference in fixed resistors." Or write Allen-Bradley Electronics Division, 1201 South Second Street, Milwaukee, Wisconsin 53204. Export: Bloomfield, New Jersey 07003. Canada: Galt, Ontario. United Kingdom: Bletchley, Bucks. Meetings Switching & Automata Theory: IEEE, Michigan State University, East Lansing, Mich., Oct. 13-15. Holm Seminar on Electric Contact Phenomena: IEEE, Illinois Institute of Technology, Drake Hotel, Chicago, Oct. 13-15. 1971 Region 8 Convention-Eurocon: IEEE, Le Palais de Beaulieu, Lausanne, Switzerland, Oct. 13-15. Fall Electronics Conference: IEEE, Pick Congress Hotel, Chicago, Oct. 18-20. Annual Electronic Connector Symposium: IEEE, Cherry Hill Inn, Cherry Hill, N.J., Oct. 20-21. Electronic & Aerospace Systems Convention: IEEE, Sheraton Park Hotel, Washington, Oct. 25-27. 1971 Joint Conference on Major Systems: IEEE, Disneyland Hotel, Anaheim, Calif., Oct. 25-29. Int'l Electron Devices Meeting: IEEE, Hilton Hotel, Washington D.C., Oct. 11-13. Northeast Electronics Research & Engineering Meeting (NEREM): IEEE, Sheraton Boston Hotel, War Mem. Aud., Boston, Nov. 3-5. Nuclear Science Symposium: IEEE, Sheraton Palace Hotel, San Francisco, Calif., Nov. 3-5. CALL FOR PAPERS International Conference on Computer Communication (ICCC '72): IEEE, Washington D.C., Oct. 24-26, 1972. Deadline for submission of papers is March 6, 1972 to Dr. Stanley Winkler, gen. prog. chairman, IBM Corp., 18100 Frederick Pike, Gaithersburg, Md. 20760 International Conference on Magnetics: IEE, Kyoto, Japan, April 10-13, 1972. Deadline for digests is November 20, 1971 to Professor Eiichi Goto, c/o INTERMAG '72 Secretariat, KDD Research & Development Laboratory, 1-23 Nakameguro 2-chome, Meguro-ku, Tokyo, Japan. A Nikkei/McGraw-Hill Publication. One of our products doesn’t fly, float, submerge, or blast off. It still carries a heavy load. Make a guess. How many phone calls do you make and get during an average day? The number you chose, we’ll bet, is too low. And it isn’t that you talk more than you think you do. It’s that everyone is talking a lot more. In fact, thirty-five million telephone conversations may be going on simultaneously in the eastern United States alone. Telecommunications is a big business and getting bigger. Which presents us both a problem and an opportunity: how to switch around this heavy load. Quickly. Reliably. And at a reasonable cost. Through our Stromberg-Carlson subsidiary, we design and make all sorts of telephone equipment for independent telephone companies, as well as complete private systems for large businesses. So Stromberg-Carlson was well equipped to find a better way to handle the increasing volume of telephone calls. The most common way was to switch calls mechanically. (Sometimes, when you dial a number, you can hear a series of clicking noises. That’s the mechanical switch physically moving your call to its destination.) Talk may still be cheap. But it’s a big business, and getting bigger all the time. While mechanical switching works, it's still mechanical, subject to maintenance, to a set amount of time to perform its task, to a small number of other functions that it can handle. (It can't handle push-button dialing, for instance.) Our answer was to create an entirely new type of switching system: an electronically controlled switch. We call it the Crossreed® Electronic Switching System. It is a major improvement over electromechanical switching. It can set up many more calls per minute. Switching each of them within a fraction of a second. Innovations like Crossreed switching aren't anything new to our Stromberg-Carlson subsidiary. But even when we aren't first with something, we look for a unique application. Our Vistaphone® telephone is a good example. The possible uses for it are endless. But our first Vistaphone system is being used experimentally by the National Technical Institute for the Deaf, in Rochester, New York. This is the wonderful part about it: for the first time, our Vistaphone system makes it possible for direct communications between deaf persons via the telephone. That's for today. In the near future, telephone communications will be changing dramatically. For one thing, by the end of this decade, computers will talk to other computers more than people will talk to people. For another thing, in the years ahead, newspaper facsimiles will be carried over phone wires. Children will learn at home by phone. A simple call to your home will open and close windows or turn your oven on or off. Endless possibilities. Many of the possibilities will become realities. Because we are able to use our technology to create one thing—like the Crossreed switching system or adapting Vistaphone systems for the deaf—and then find further uses for the same technology. This is an ability we apply to all of our markets. The ability to do things no one's been able to do before. It took 15 years for the computer to get from here... to here. GCA helped it get there. A decade and a half ago the miniature computer was unknown. Integrated circuits that could handle multiple computer functions on a single silicon chip did not exist. And one of the main reasons why miniaturization was impossible was the inability to make precise reductions of circuit artwork...the micromages that were the first photomasks. Then GCA's David W. Mann Company introduced the first commercially available Photorepeater®. It combined special optics with precision step-and-repeat motions into a macroimaging system which fast became the industry standard. As IC technology improved, GCA photomasking systems provided the reliable, precise means necessary to meet manufacturing demands for more circuit functions on smaller chips. The rest is design and packaging history. Through the Mann product line, GCA provides total photomasking systems which handle smaller and smaller circuit geometries and enable computer people to build more computer power into less space. If you could use a fast-turnaround, high yield photomask production capability backed by the most experience in the industry, turn to GCA. We've made the progress. We have the systems. And the information. Write GCA Corporation, Burlington Road, Bedford, Mass. 01730. Circle 24 on reader service card GCA CORPORATION The Mann 1795 Photorepeater Mexico staring hard at U.S.-owned 'garage operations' Growth of U.S.-owned electronics operations in Mexico has accelerated faster than most people—and the Federal Government—realized. The total is now an astonishing 180, sources at the Department of Commerce estimate. This hasn’t gone unnoticed by the Mexican government. There are now indications, based on industrial intelligence from Washington, that Mexico is planning to cash in on this growth by requiring owners to capitalize their operations to increase tax revenues. The extent of such operations was realized only recently by Commerce when operations owned by “U.S. individuals” not previously tabulated were added to those held by U.S. corporations. The planned move by the Mexican government is not expected to make Mexico less attractive as an offshore assembly area to the U.S. semiconductor industry and some systems houses. It would take something drastic to slow this trend, sources say, something that isn’t expected to happen. The Mexican government, they say, has taken no action against a large number of “garage operations” packed with “wall-to-wall workers and, in some cases, no toilets,” but is waiting for the industry to flourish—obtaining large orders and making delivery commitments—before it cracks down. First balloon goes up on Nixon R&D plan for 1972 The Nixon administration’s 1972 plan to dramatically raise America’s declining investment in industrial R&D [Electronics, Sept. 27, p. 33] will emphasize a more liberal tax structure for rapid depreciation and writeoffs of new R&D equipment and facilities, rather than direct Federal investment. This is the conclusion drawn by some electronics industries leaders after hearing former Treasury official Murray L. Weidenbaum address the Electronics Industries Association’s Fall Conference. Though economist Weidenbaum has returned to the faculty of Washington University, St. Louis, his recent role as Treasury’s assistant secretary for economic policy made him privy to the Nixon “game plans,” and some listeners sensed a trial balloon in Weidenbaum’s comments. Weidenbaum, who first gained prominence in the mid-60s as a critic of the defense industry’s profit structure, reflects Nixon Administration concern that “since 1964, this nation has been investing an ever smaller share of its national resources in R&D,” and believes it can best be turned around by private sector financing encouraged by a high and rising level of economic activity; tax advantages such as those employed by Australia, Britain, Canada, Norway, Spain, and other West European countries; a broad approach to civilian markets with steady funding not limited to single economic sectors or mechanism such as existing institutes or universities; and more research on the R&D process itself. Pad relocation goes into multiplier Possibly the most sophisticated bipolar logic device available anywhere is being offered by the Microelectronic Products division of Hughes Aircraft. It’s an eight-bit multiplier fabricated on a full 1.5-in.-diameter wafer using three layers of metal interconnects and a technique called pad relocation—a technique that Hughes officials believe is an improvement over Texas Instruments’ discretionary wiring [Electronics, Oct. 13, 1969, p. 44]. The TTL multiplier is the first of a family to come from Hughes, and consists of 52 full adders plus 96 additional gates for the equivalent of 616 gates. Developers say it can form 8 million products a second, multiplying two eight-bit numbers plus sign to get a 16-bit result plus sign each time it forms a product. It's aimed initially at military computer applications, such as airborne computers, for fast-Fourier transforms or digital filtering, but Hughes hopes to expand on the concept to provide other devices for the commercial computer market. Under-$100 calculator set for January* The first under-$100 electronic calculator is on the way—and it's all-U.S. made. Ragen Precision Industries of North Arlington, N.J., will ship in early January the shirt-pocket-sized device. Designed with two C/MOS logic chips and an eight-digit liquid crystal display, it's a four-function device with floating decimal point. Ragen builds both electronics and display. Ragen has an order from Alexander's, a New York Department store chain, for 20,000 calculators with an option for 20,000 more. Measuring 2⅝ inches by seven-eighths of an inch by 3½ inches, it will operate for more than a year on a 9-volt dry cell battery. The chips drive the liquid crystal displays directly. Coast Guard search for Loran resumes The Coast Guard is trying once again to acquire a solid state Loran C transmitter. Originally christened Transloc, for transportable Loran C, and contracted to Sylvania, the program ran into technical and funding problems and was dropped. The new contract is with Megapulse Inc. of Waltham, Mass., and is for about $940,000 all told. The old Sylvania contract was about $6 million. The cost differential is largely due to a proprietary approach taken by Megapulse to generation of ultrahigh-power pulses, and to a relaxation of the Coast Guard's pressure for a transportable system. Due to be delivered before October 1972, the megapulse transmitter would be a preproduction prototype capable of delivering more than 200 kilowatts effective radiated power—and, when paralleled with others like it, come up with the megawatt or more the Coast Guard requires. TI to sell TV tuning diodes Texas Instruments shortly will enter the market for tuning diodes for VHF and UHF television, a market it says can approach 1.5 billion units by 1980. TI products are electrically similar to parts Siemens has been making in Germany for a few years, but TI uses a standard small-signal glass diode package rather than the plastic used by Siemens and Motorola, the only other domestic producer. This means TI can use its automated diode production line, which is becoming available as ICs take over computer diode applications. Electronic Arrays sells calculator firm International Calculating Machines, the calculator manufacturing subsidiary established late last year by Electronic Arrays, has been sold to Lago-Calc Inc. An outright sale was one of three options Electronic Arrays president Mois Gerson suggested recently [Electronics, Sept. 13, p. 38]. Lago-Calc is headed by Max Lagomarsino, son of the founder of Totalia Calculator Co. of Italy, who has sold calculators as a distributor in this country for years. Lago-Calc will sublease the ICM facility, use ICM designs, and retain the top management. LED displays and Litronix have hit the big time. Our displays are in volume production. They are low cost and they are totally reliable. But don’t just take our word for it—try them. Compare the long run advantages in design, reliability, availability and cost. Take a look at a few of the LED stars now playing in full production at Litronix. Outshines MAN-1 at lower cost. Introducing the Data Lit 10A, MAN-1 economy model. Has an amazing brightness of 500 foot Lamberts at 10 mA with 0.27 inch character. High quality, high volume new star with low price tag, $6.75.* New 4 digit array. Small, bright display with .125 inch character heights and packaged for assembly economy. The Data Lit 34 is designed for low power and high brightness needs. It has four times the light emitting area as the MAN-3A. You can end stack it for 8, 16, etc., etc. digit displays. $3.80 per digit. Power Saver, Money Saver too. The Data Lit 8 display, with .24 inch character height, consumes half the power of the MAN-1 type display, and half the voltage drop. It’s just as bright. Mounts on 0.3 inch centers. Priced at $6.25. World Famous “Skinny DIR.” Perfect display for small calculator and portable instruments. Has .120 inch characters and comes in a very thin DIP package. Data Lit 10, priced at $3.80. Largest LED Display in the Known World. The Data Lit 6 is a 0.6 inch display with black substrate for fantastic distance viewing. Available in vertical mount that plugs into a standard .156 PC connector. Price: $14.00. Other Wonderful LED products. We also have a full chorus of second source and proprietary LED displays, discrete visible and infrared emitters, opto isolators and many other good things. Write for a copy of our product guide and a personally autographed photo of Carlos, producer of the show. LED’s will save you money in the long run. * All prices in 1,000 piece quantities --- The bright guys litronix Litronix, Inc. • 19000 Homestead Road • Cupertino, California 95014 • (408) 257-7910 • TWX: 910-338-0022 Electronics/October 11, 1971 ION IMPLANTATION... Your key to MOS performance — today! Our revolutionary production process, ion implantation, has quietly come of age. Today it's the most uncomplicated method for achieving such sophisticated benefits as low threshold voltages for TTL compatibility and depletion mode devices for state-of-the-art performance. Want proof? Then take a look at some examples of the advantages offered by MOSTEK's ion-implanted product line: (1) Threshold Adjustment. Take our MK 1007 P 4x80-bit dynamic shift register, for instance. Ion implantation makes possible a low threshold device for full TTL compatibility without the need for interface components or high voltage, high power clocks. Our clock is single phase, low capacitance and TTL compatible. Low power over a wide supply range for memories such as our MK 4007 P RAM is illustrated by the constant current drain versus supply voltage shown in Figure 1. Power remains low with the depletion loads compared with the enhancement loads of the competitive 256-bit circuit. Single 5V supply operation. MOSTEK's MK 5002 P 4-digit counter/display circuit for instrumentation purposes can operate from a single --5V logic supply, drawing less than 25 mW, or from any supply from 4.5 to 20V! Ion implantation is the only processing tool capable of delivering this performance from otherwise conventional, cost-effective P-channel LSI! Find out how easy it is to use the products and processes of the leader in MOS—MOSTEK. Call our home office at (214) 242-1494 and ask for marketing or applications assistance. Or contact your nearest MOSTEK sales office listed below or your local Sprague office or distributor. Ask for our new catalog! MOSTEK CORPORATION An affiliate of Sprague Electric Company 1400 Upfield Drive Carrollton, Texas 75006 Sales Offices: 12870 Panama Street, Los Angeles, Calif. 90066; (213) 391-2746 515 S. West Street, Jackson, Michigan 49203; (517) 787-0508 60 Turner Street, Waltham, Mass. 02154; (617) 899-9107 Circle 28 on reader service card Current switch scheme promises fast LSI logic By cutting propagation delay and power hunger, HP configuration can yield larger, faster logic chips Logic designers working with LSI circuits have a problem: their rulebook hampers both circuit size and performance because the logic configurations they are working with weren't intended for LSI circuits. DTL, TTL, and ECL configurations originally were designed as gate functions and not for highly integrated functions. The power/delay product of these circuit configurations is about 80 picojoules; a TTL gate typically has an 8-nanosecond propagation delay and requires about 10 milliwatts, and an ECL gate has a 0.9-nanosecond propagation delay and draws approximately 90 milliwatts. In terms of LSI, this means circuit size is limited by allowable propagation delay and package power limitations. Other roads. There are other logic configurations, however, and an engineer at the Solid State Laboratory of Hewlett Packard Labs, Palo Alto, Calif., is working with one that cuts this power/delay product by a factor of four. According to John E. Price, a member of the technical staff at Hewlett Packard labs, a power-delay product of only 5 pJ can be obtained with a form of current switching logic. Price is working with two versions of such a gate: one employs positive feedback and the other a reference signal—the base of the transistor that's tied to the $V_{\text{out}}$ output is connected to a 200-millivolt reference source instead of being connected simply to $V_{\text{out}}$. Price says that with the current switch configuration, "I've built gates that have a propagation delay of about 0.4 nanosecond and dissipate 10 mW, and gates that have delays of 2 nanoseconds and dissipate 2.5 mW. The basic figure of merit remains around 5 pJ." HP's interest in the high-speed circuit could be for new calculator circuits, high-speed counters, or high-frequency oscilloscopes, but Price won't comment on this. Price's work will be described at the International Electron Devices meeting in Washington. Both circuits' configurations have been built with both conventional phosphorous-doped and arsenic-doped emitter structures [Electronics, Sept. 13, p. 36]. The arsenic emitter devices offer a propagation delay of from 0.4 to 0.6 nanosecond with power dissipation from 6 to 10 mW while the conventional phosphorous emitter structures have delays of from 0.4 to 2 nanoseconds with dissipation amounting to from 2.5 to 18 mW. Price says that besides being fast, the current-switching configurations are easy to work with. A fanout of 5, for example, is easy to obtain, and noise margin is over 100 millivolts with an output swing of from 0 to 200 mV, which Price says "is quite good." Supply voltage is 2.5 to 3 V and there are no speed-limiting resistors in the signal path. As an example of how an LSI or MSI circuit could be improved with the current-switch configuration, Price points to the TTL type 7483 MSI four-bit full adder. This circuit requires 500 mW and has a propagation delay of from 32 to 48 nanoseconds. A similar function implemented with current switching logic requires only 150 mW and a propagation delay that adds up to only 6 nanoseconds. Memories Sangamo division buys Soniscan from Sylvania Soniscan, the magnetoacoustic bulk memory that promised multimegabit storage at 0.1 to 0.2 cents a bit [Electronics, July 6, 1970, p. 49], has been sold by GTE Sylvania to a division of the Sangamo Electric Co. The purchaser was Microsonics Inc. of Weymouth, Mass; the deal was consumated Sept. 27. The dealings were quite complex; Logical. The difference between reference-voltage circuit (top) and positive-feedback circuit is shown in color. Reference-voltage is faster, hotter, costlier. Electronics review negotiations went on for almost six months. A possible factor in the length of the negotiations were Sylvania's simultaneous investigations of "several joint venture and technology exchange offers," according to a company spokesman. Financial details have not been disclosed, although both firms appear to have established solid patent positions for themselves. And Sylvania, through its royalties, can hope to share in what many observers feel will be a lucrative piece of the mass memory market. But the question remains: why did Sylvania share when it could have had it all? Stability woes. Although Sylvania claimed greater success with ferro-acoustic techniques than other firms that had tried similar memory technologies, Sylvania was believed to have had problems with the long-term stability of Soniscan-stored data and with addressing—its engineering teams simply did not have the training and experience to solve them. This may have been a key reason for Sylvania's sale to Microsonics. Microsonics, according to its general manager, T.J. Geoghegan, has the required skills. "Our business for years has been delay lines for storage, display refreshment, and other applications. And we have the experience with bulk and surface-wave propagation, transducers, and production techniques which we feel can solve Soniscan's problems." Navy waits. Geoghegan says that the firm plans to seek out commercial markets. But the military may be on the preferred customer list. The Naval Air Systems Command still is interested in Soniscan for its Advanced Airborne Digital Computer System—and though neither the Navy nor Microsonics will comment, the Navy's own comparative tests during 1970 pegged Soniscan—with its low power needs, high packing density (5,000 bits per cubic inch), 75-nanosecond read-write time, and its low price—as the best among available and future bulk memory technologies. The Navy has been waiting like a cat for the mouse ever since, and so even though Geoghegan won't say anything about the Navy's potential as a customer, it's safe to say that the Navy brass in the Pentagon has his phone number. Lasers High-gain, flash-pumped laser gives coded pulse If lasers are to become realistic communications tools, it will be necessary to obtain efficiently a coded-pulse output from a high-power, flash-pumped laser. Two experimenters at Texas Instruments' Central Research Lab think they've found the key with an acousto-optic modulation technique that permits not only switching of high-power flash lasers, which isn't new, but variable output pulses. The approach taken by Warner C. Scott and M. DeWit was to use an acousto-optic beam deflector—a quartz resonator whose transmissive or reflective properties can be controlled by exciting it with radio frequencies—arranged so that the undeflected portion of the laser beam provides an output pulse, while the deflected portion re-enters the laser's feedback loop. The deflection of the acousto-optic device depends on the power applied to it, hence the threshold population inversion can be electronically controlled by varying the rf power; an additional feature is that very low rf power is required, typically only a watt, compared to the higher levels necessary with conventional Q-switching. Using the quartz resonator as a switch is not a new approach, but it's done in a different way: a Bell Labs technique uses the switch to vary the loss of a continuously pumped laser resonator from high (no lasing) to low, with consequent lasing and output. But it is impossible to obtain variable pulses in this manner from a flash-pumped laser, since the energy output is a function of the population inversion within the laser at the time of the flash; this means that threshold inversion is generally predetermined by mirror transmissions. Thus, variable pulses could be obtained only with a deflecting mirror, unsuitable for high frequencies and not amenable to wide control. In experiments, Scott and DeWit switched a krypton-pumped neodymium-YAG laser, providing both single and multiple pulses, using a 100-megahertz acousto-optic deflector. Single 70-millijoule output pulses, 30 nanoseconds wide (3½ to 4 megawatts peak) have been obtained, with even higher power obtained with double pulses. According to Scott, energy and times were completely stable and the pulses could be varied in energy from zero to the maximum available. Scott will not comment on possible applications at TI, but TI's equipment group is very active in military laser-controlled bombs and surveillance systems using infrared detectors. Possible applications could lie in this area, or in communications. Commercial electronics Rotating-ring watch display may leapfrog LEDs, crystals Low-cost complementary MOS digital countdown circuits have made electronic wristwatches a reality, but since most of these timepieces use gear-driven hands to display the time, they still have numerous moving parts. Light-emitting diodes also have been used to display the time, and liquid crystals hold promise, too. But to display digits constantly, both require more power than the 1.5-volt batteries in these watches can provide. And liquid crystals have limited lifetimes, and are temperature-sensitive. A Southern California inventor has come up with a new kind of display system for electronic watches that might offer an attractive alternative to gear-driven hands, LEDs, and liquid crystals. What's more, the rotating-ring display system, as it's called, is compatible with the 1.2-to-1.5-v batteries now used in electronic watches, and can have all the drive circuitry on one CMOS Electronics Index of Activity Oct. 11, 1971 While the index of total industry output for August climbed 4.5 points from July's figure, it was still 1.8% below the August 1970 total. Consumer was the largest contributor to the late summer rise as it chalked up an 11.4% increase. Experts are crediting President Nixon's new economic program for the generally improved air of consumer confidence. Industrial-commercial also made a slight contribution with a 0.2-point increase. The only August loser was defense, down 0.3% from the previous month and 10% from its year-ago level. Indexes chart pace of production volume for total industry and each segment. The base period, equal to 100, is the average of 1965 monthly output for each of the three parts of the industry. Index numbers are expressed as a percentage of the base period. Data is seasonally adjusted. Revised. Segment of Industry \| Aug. '71 \| July '71 \| Aug. '70 ---\|---\|---\|--- Consumer electronics \| 113.4 \| 101.8 \| 86.2 Defense electronics \| 112.4 \| 112.7 \| 126.5 Industrial-commercial electronics \| 116.3 \| 116.1 \| 117.0 Total industry \| 116.1 \| 111.6 \| 118.2 Ring movement, which corresponds to hand movement in conventional watches, is triggered by a pulse from the C/MOS drive circuit. This compensates the magnetic field of the fully aligned phase and one of the phases that are one-third aligned. Then the magnetic field from the permanent magnet in the remaining nonaligned phase reduces magnetic reluctance by aligning the magnetic areas of the ring with the stator poles. This moves the ring by two-thirds of a division. Direction of ring movement is governed by which of the partly overlapping phases is compensated. The single-ring prototype steps in less than 50 milliseconds with pulses that deliver about 100 ampereturns. Magnetic flux should accelerate the ring with about 20 g's, resulting in stepping speeds of 5 milliseconds. Dill calculates the power consumption of the entire system at 5 to 6 microwatts; this could extend the life of a 1.2-v battery to two or three years instead of the usual year. The rings may be driven separately. Electronics review Time turns. Rotating-ring display system for electronic watch can use single C/MOS chip. or be electronically coupled, but they're always completely sealed and are the only moving parts in the watch. Dill's next step is to build a better-looking model than his original. Then he hopes to license the rotating ring display system to a major watch manufacturer. Computers Illiac 4 to go operational by spring of '72 Spring of '72 is the magic date now for the giant Illiac 4 computer. That's when the unique system is scheduled to be turned on operationally. An object of intense interest by the data processing community because of its 64 processors, each with its own memory, the huge parallel system is also becoming associated less and less with the University of Illinois—its designer. First major break in the university's involvement was the decision to locate the system at NASA's Ames Research Center in California. The building to house Illiac 4 is under construction now, and officials say it will be ready in time for the spring date. The move to the Government facility was made primarily to safeguard the machine from possible student rioters. Hardware is being built and put together at the Paoli, Pa., plant of Burroughs, the builder, where debugging is currently under way. Not all the processors are working in the system yet—there are some input/output problems—but Illiac 4 will be ready for its spring timetable, officials maintain. The university still has contract responsibility for hardware, but transfer of diagnostics and maintenance from the school is the latest step in the gradual phaseout of the school's involvement. NASA has awarded a contract to Automation Technology Inc., Champaign, Ill., to continue development of off-line diagnostics, create a maintenance team to monitor the system on-line, and formulate and build a tester for the printed circuit cards. Under the new contract, some of the people who were working on the project at the school have transferred to ATI; and four Burroughs engineers will form the nucleus of the maintenance team. All eventually will be transferred to Ames. Even the existence of ATI is, in a sense, part of this phasing out: its president, Arthur B. Carroll, was second in command of the Illiac project for several years. Univac aims 1616 at military, FAA markets With an order for its new 1616 computer in hand, Univac has formally unveiled the breadbox-sized processor. The announcement comes as no surprise [Electronics, July 5, p. 18]. The sale of the first machine to ITT Giffillan for evaluation in its Navy AN/SPS-48 long-range air-search radar confirmed reports of Univac's market plans for the system. The three-dimensional SPS-48 shipboard system for attack carriers, cruisers, and frigates will feed into the Naval Tactical Data System and provide information for carrier aircraft interceptor control, and ship missile launching. Though Univac says the 1616 model to be delivered in December for the SPS-48 will have 32,768 words of memory, it declined to disclose price or application in the ITT system. However, the company said typical applications cover "air traffic control; converting radar data and flight facts into visual displays; preprocessors for the front end of centralized shipboard systems; sonar and beacon signal processing, transferring data in real time; and shipboard control systems such as radar, electronic countermeasures and navigation." Beyond military sales, Univac clearly is shooting for Federal Aviation Administration business, specifically the expansion anticipated for the Automated Radar Terminal System on which it's already prime contractor. And the Sperry Rand computer THE GROWING TEKTRONIX 7000-Series Oscilloscope Systems From 500 MHz at 10 mV or: 1 GHz @ <4 V/cm direct access plug-in For those who have requirements in between 500 MHz and 50 MHz there's the 150-MHz 7704 (R7704 rackmount), 90-MHz 7504 (7514 storage), all with 4 plug-in compartments; 90-MHz 7503 with 3 plug-in compartments. Exclusive to the 7000 Series is: CRT READOUT— Deflection factors and sweep speeds, the DMM and counter outputs, invert and uncalibrated symbols, etc., are automatically displayed on the CRT—where you look for information. CRT Readout can be ordered initially or as a conversion kit that is easily installed. In each case the cost is only $400. And it is available in all scopes except the 7403N. MULTIPLE-PLUG-IN MAINFRAMES Three or four-plug-in mainframes allow up to twenty combinations of vertical and horizontal operating modes. You can now use plug-ins with widely different features...simultaneously. If you wish, start with only one horizontal and one vertical plug-in and add more as your measurement requirements change. PLUG-IN VERSATILITY Plug-ins are available to make virtually any measurement desired. Examples are: - single trace - dual trace - combinations for multiple trace - 10 μV/div differential - 1 mA/div current amplifier - differential comparator - Sampling to 14 GHz - 45-ps risetime TDR - 1.8 GHz spectrum analyzer - curve tracer - single time bases - dual time bases with calibrated mixed sweep - 500-ps dual time base with triggering to 600 MHz - 525-MHz direct counter - digital multimeter. RUGGEDIZED MILITARY VERSION The AN/USM-281C consisting of mainframe, vertical plug-ins, dual time base and accessory cover is the militarized version of the 7403N Oscilloscope System. to 50 MHz at 5 mV dual-trace, 6½-inch CRT and delaying sweep for only $2200 7000-Series Scopes, complete with plug-ins, start as low as $1670. Call your nearby TEKTRONIX Field Engineer today for a demonstration of the scopes that make more measurements easier and quicker or write: Tektronix, Inc., P.O. Box 500, Beaverton, Oregon 97005. U.S. Sales Prices FOB Beaverton, Oregon Tektronix lease and rental programs are available in the U.S. division declares "it is not a minicomputer" in terms of performance, though its size and price suggest it. The 1616 base price is $16,000 for a 16-bit machine with 4,096 words of core memory that can be expanded to 65,536 words. Cycle time is 650 nanoseconds for the 2-foot-square box that stands 14.25 inches tall. The company says that "against computers in the same price range, the 1616 requires less memory for handling data, less instructions to solve a given problem, and has faster execution time." Univac says also that the industrial version of its 1616 with commercial-grade memory has about 6,000 hours mean time between failure for the total system. --- Heat will limit specs, says IBM researcher Between them, semiconductor and computer makers create the impression that computers continually will become faster, more powerful, and smaller, whether in single large mainframes or in nets of minicomputers. But IBM research scientist Robert W. Keyes figures that the time may be near when IC computer performance specifications and cost reductions are going to be limited simply by heat. "If you look at the indicators like throughput, speed, declining cost, and so on, they all project toward infinity. Circuits per computer appear to rise asymptotically, speed is rising about linearly as the years pass. But from the rise of the alloy diffused transistor to the IC, speed-power products have remained within a fairly well-defined ballpark," he says. And after a mathematical investigation of the facts of life within a computer, Keyes predicts that the clamps will come not because of gate propagation delay or device size limits, as some have predicted, or because packing density will level off ("Indeed, using variations of electron-beam lithography, it should be able to make ICs with active elements measured in molecules), but just because chips will overheat. And so long as computers are tied to semiconductor technology as we know it, there doesn't seem to be any way out of the bind. Speed—a factor of packing density, device size, and power—is expected by Keyes to level off at 10 times today's available speeds. Since the industry now is using circuits with gate propagation delays in the nanosecond region, things should top out at about a few tenths of a nanosecond. "But high packing densities combined with small device areas (junctions) are going to make the heat transfer problem ever tougher," he says. "Even using liquid cooling, it is hard to dissipate much more than 10 watts per square centimeter today. A hundred watts per square centimeter, not an unreasonable figure for some large-scale integrated devices in the minds of designers, if not on their drawing boards, is going to be just about impossible." This will limit propagation time and thus limit speed because LSI devices will have to be designed less densely than electron-beam lithography or other techniques might allow. And while they have not made a vigorous study like Keyes, the economics of the situation is becoming apparent to computer makers. A spokesman for one quality minicomputer producer notes that "for the next 10 years at least, it's software where the breakthroughs will come." Some of Keyes' colleagues disagree. One of them, Gerald A. Maley of IBM, would add new functions to existing chip designs and put more thought into architecture to improve machine performance. --- Companies Solitron to head down C/MOS path A new marketing strategy is beginning to emerge at the San Diego plant of Solitron Devices Inc. now that Walter Greenwood, the ex-Hughes Aircraft Co. executive, has had time to assess and influence it in his new job as executive vice president [Electronics, June 7, p. 14]. Perhaps surprisingly for a company that showed a loss in its 1970 net, the strategy includes ambitious plans to go up against the formidable competition already in the rapidly growing market of complementary MOS. From Japan, Solitron also has a foot in the calculator camp, with a contract from Matsushita Communications in Japan for 15,000 sets of five chips each. A sizable follow-on order is being negotiated with Matsushita, says Peter Gopal, manager of MOS custom systems. The calculator chips are p-channel MOS devices. Both the C/MOS and calculator capabilities were acquired from outside Solitron. The C/MOS expertise came with the hiring of three former employees of Hughes Aircraft's Microelectronics division. Carroll Perkins was most recently MOS marketing manager at the Hughes division; he's now Solitron's MOS product manager, with responsibility for both C/MOS and p-channel devices, excluding calculator devices. The C/MOS effort will emulate the RCA CD4000A series. Solitron's CM4000 series, like RCA's, will consist of 41 basic devices, but whether or not all 41 are produced will depend on demand. The company currently has four products available in sample quantities of a few hundred and is generating artwork on new types at a rate of two a week. "We have generated artwork on about 15 types and will start shipping about five types during October," says Perkins. He has his sights aimed high. "By year end we should have 25, and by the end of the first quarter of 1972 we should have caught up with RCA," he says. Solitron is following the RCA lead because the RCA series has been well promoted and well established as standard devices. However, Solitron regards the 4000 series mainly as a springboard for developing custom-made devices. Perkins calculates the 1971 marBell & Howell & Heads You Win "Wow! We've done it! A super headed recorder/reproducer like nothing nobody's ever seen!" This from one of our brighter back room boys as he charged into our office. "Take a look guys. 2MHz performance at 60 ips. Realize what that means? Perfect recording with half the tape. And at about $200 a reel, that's a cash register full of saved bucks." "Not bad." "And it's so good we've given the record head a whompin' 5,000 hour warranty." "Well, I think we got something to work with. Anything else?" "Yeah, those are both industry firsts from Bell & Howell." "And?" "Well, it has the usual 14 channels. 15/16 to 240 ips and stuff like that, in the conventional format." "Ummm, who uses the thing?" "R&D types mostly. It's about as sophisticated as you can get. They use it for aircraft tests. Cars. To pull down whole libraries of data from field recorders. That sort of thing." "Sounds like a heckuva machine." "System." "Okay, system. So what do you call the super headed system?" "The VR3700B." "Wow. That really tells it like it is. Okay, we'll do an ad. But first I'll flip you for coffee. Call it." "Heads." "You win." "Hey, not a bad headline for the..." "Are you kidding? Beat it. Sheez, can you imagine." Read about it. Bell & Howell, CEC/Instruments Division, 360 Sierra Madre Villa, Pasadena, California 91109. CEC/INSTRUMENTS DIVISION BELL & HOWELL © Copyright 1970 Bell & Howell Circle 35 on reader service card Electronics review Market for C/MOS at slightly more than $5 million, increasing to $15 million in 1972, $80 million in 1974, and $170 million by 1975. While counter-type circuits now account for more than 80% of sales, he expects that random access memories will be the biggest single factor within about three years. Employment Self-help groups see some daylight Are engineers becoming employable again? Maybe so. Certainly some of their self-help group efforts appear to be bearing fruit after nearly a year [Electronics, Jan. 18, p. 34]. The depressed Boston area spawned many such groups, and now one of them has decided to close—despite the recent RCA computer division shutdown—because 80% of its participants have been hired. Called the Sudbury Experiment, the group was small, too small for projecting national trends. But now other groups are reporting a gradual improvement in hiring. And in every case, the jobs are in engineering. Interestingly, the key men in some of these efforts are quickest to land jobs themselves. Once the most vocal, the Economic Action Group Inc., Needham, Mass., continues to function, though more quietly—its founder, Gerald Wallick, has moved after finding a new job in New Hampshire. Paul Jackman, head of the Topsfield Group, now wears two hats as head of the group and consultant to the Massachusetts Science and Technology Foundation. The largest effort in Massachusetts has been the state-sponsored Route 128 Center for Professionals, a clearing house for jobless technologists and potential employers. In August the center held a special seminar for about 20 jobless engineers. At the same time, the center kept records on 50 other engineers. Like the IEEE-AIAA sponsored "Tiger Breeding" sessions [Electronics, Jan. 18, p. 35], the seminar was aimed at improving the job-landing skills of the participants, and it seems to be successful. During a 30-day period beginning halfway through the seminar, the participants had about four interviews each, versus about 2.5 per man in the control group. "Apparently engineers need nudging," says a spokesman. Compared to the control group, the trained 20 are scheduling more interviews, going to more job placement agencies, carrying in more resumes personally, and generally being more aggressive than their counterparts. And the payoff is in jobs; of the 20 seminar-trained engineers three soon found work compared to one of the 50-man nonseminar group. Government Electronics figures heavily in negative trade balance Disclosure of August's trade figures—the numbers that triggered President Nixon's 10% import surcharge—show that electronics imports, though still on the plus side, made a significant contribution to that month's negative balance of $259.7 million. When added to earlier figures, the August total showed that U.S. imports overall exceeded exports in the first eight months of 1971 by $936.1 million—a sharp contrast from the $2.23 billion surplus in the same 1970 period. A precise breakout of figures on electronic products and components exports and imports in August will not be available for several weeks, but increases occurred in all electronics categories—manufactured goods and electrical machinery, apparatus and appliances. Commerce also came up with another set of supporting figures: for the first time the U.S. imported more than half of Japan's electronics exports in 1970. American purchases of $1.2 billion in Japanese electronics last year were up 14% from 1969. Consumer products accounted for 84.7% of the total, and showed an 11% increase. Since consumer products have long been Japan's bread-and-butter business in the U.S. market, the Commerce figures are less startling than the agency's disclosure that the U.S. more than doubled its imports of Japanese computers in 1970. Their value: $78 million vs $36.5 million the year before. Component shipments to the U.S., at $67.9 million, represented 5.6% of total electronics imports, although this figure slipped $6.6 million from the 1969 peak. All other equipment areas showed gains, however, rising to $116 million, or 0.7% of the total. Included in this last category are telephone and telegraph equipment—particularly switchboards and exchanges—where sales to the U.S. totaled $15.7 million, more than triple the 1969 level for such sales. One intriguing aspect of the U.S. data is the notation that Japan—which has mounted a sharp export drive in all world markets for every kind of telecommunications equipment—also is expected to import more of this hardware in 1971 as its own domestic requirements increase. For example, its imports of communications and electronics this year are expected to rise 20% to 30% from the 1970 total of $590 million. Leading this increase will be computers, integrated circuits, and test and measuring instruments. Nevertheless, Japan's Communications Industry Association expects its members' exports throughout the world this year will jump another 19%. The biggest boost, 30%, is expected in wire telephone and telegraph equipment, with respective increases of 17%, 16%, and 15% for microwave hardware, carrier equipment, and broadcast systems. Packaging Brushes, tags bid for place in connector sun "Brush" and "laundry tag" may conjure visions of a dry-cleaning establishment, but if the Bendix Electrical components division in Sid"ESP rectifiers. They're new from Unitrode. "Typical recovery time is 15 nanoseconds in any circuit. And that's the fastest available in a power rectifier. Forward voltage drop is as low as .8V at 20A. That's efficiency. And, as you can see, they're in extremely compact packages having continuous ratings of 2.5A to 20A in 50V, 75V, 100V and 125V types. And with many thousands of hours of life tests behind them, you can be sure of a state-of-the-art product with demonstrated reliability. Unitrode has the statistical test data to prove it. "With ESP rectifiers, you can design high current, high frequency power supplies to 500 KHz with very low diode losses and at very low cost. They're available off-the-shelf in three series. UES 101-4 for 2.5A, UES 201-4 for 6A, and UES 301-4 for 20A applications. Now you know what ESP stands for." "You mean Efficiency, Speed and Power." "That's very perceptive of you." For fast action, call Sales Engineering collect at (617) 926-0404, Unitrode Corporation, Dept. 11 C, 580 Pleasant Street, Watertown, Mass. 02127. Electronics review If Bendix, N.Y., is successful, they may become familiar terms in the connector business. Bendix has been researching two new types of contacts—the brush contact, a bundle of small wires that makes contact with another bundle by simply inter-leaving the wire ends, and the laundry tag, a contact intended for edge connections to printed circuit boards by gripping plated-through holes with a pair of jaws that closes through the hole when the board is inserted. The brush contacts, now being sold in prototype quantities, are envisioned in several applications, including almost any situation now using a pin-and-socket arrangement, according to Donald E. Michel, Bendix chief engineer for microelectronics. Michel says that the contact insertion force for a brush is about 1¼ ounces, much less than the typical 6-ounce insertion force for pin-and-socket contacts. When putting together matching halves of a 100-contact connector, Michel says, "you might need a sledge-hammer if you used pin and sockets" but it could be managed easily by hand with the brushes. Michel also points out that since brush contacts are hermaphroditic, it might be possible to build connectors that interconnect flat-pack hybrid circuits by stacking them up and do away with printed circuit board interconnections entirely. The laundry tag contacts are a little further away from regular use—Michel estimates samples will be available by early 1972—but he does feel that they can compete as a low-cost connection medium with any of the present pc board connector contacts, such as the tuning fork and the pin and socket. The laundry-tag grippers are shaped as chamfered jaws on the ends of a tuning fork. When the pc board is inserted, the insertion force can be transmitted to the jaws to cause them to close and lock through the plated-through hole, putting pressure on the inside hole surface to make electrical contact. Several mechanisms have been used to provide the closure force—rollers in the connector that ride up a ramp to a detent position to close the jaws, a set of polypropylene hinges where insertion force is used to move one hinge member relative to the other and close the jaws, and a simple cam mechanism that might serve several connectors that could be actuated with a screwdriver. The push contact is made by cutting chamfered ends on phosphor bronze wires each about 6 mils in diameter and then crimping the 14 wires in a holder. The contacts, when engaged with another bundle of brushes, are designed for a total engagement of 0.125 inch, resulting in 5 milliohms contact resistance. A lower engagement gives inconsistent contact resistance; a greater figure tends to accentuate wire bending as the two bundles are engaged. Test results show that up to 500 matings produced little change in contact resistance. Military electronics Displays, computers to teach dogfighting Close-range aerial combat is one of the most difficult skills a fighter pilot must master. But the most effective way to learn it is to get involved in the real thing—roughly analogous to learning to swim by being thrown in a river. The U.S. Navy hopes to do the next best thing to boost the dog-fighting skills of student pilots with an electronic system being developed by Cubic Corp., San Diego, Calif. Called the Air Combat Maneuvering Range, it will enable pilots to practice dogfights in the air while a ground instructor monitors the action in real time on a computer-driven video display and gives instructions by radio. Two important benefits of the system are that it will get the instructor out of the air and on the ground where he can better observe the action, and help teach pilots to identify the "missile envelope," the effective area in which he should fire to hit the target. This will be accomplished by transmitting an audio tone automatically when a pilot is in the right position. The system will consist of four major elements: a computation and control subsystem by System Development Corp.; a tracking instrument subsystem—a high-speed, phase-locked-loop ranging system operating at L band; display and debriefing subsystem by Adage Corp.; and an airborne instrumentation subsystem by Lear Siegler. The latter is an instrument pod 9 feet long and 5 inches in diameter that will be carried in place of the usual Sidewinder missile. The pod will house a distance-measuring-equipment transponder, attitude-heading reference unit, angle-of-attack sensor, weapons-bus monitor (for weapons status information), and a data encoder. The computation and control subsystem will consist of three Xerox Data Systems Sigma 5 computers. One of these will compute aircraft positions and one will simulate the firing of missiles when a pilot pushes a button, displaying the imaginary trajectories on the display. The third computer will be the system's overall controller. At the Yuma, Ariz., training range, the tracking instrumentation subsystem will consist of a manned master station and six unmanned remote transponder sites. On command from the computer, the master may instruct a particular remote unit to interrogate a given aircraft by its identification code. The aircraft pod's response will be picked up by all six stations and relayed to the master station, which will compute the slant ranges of all six, using a Kalman filter program, and transmit the data to the central computer. Are nervous resistors giving your products the jitters? Stabilize with Metal Glaze™ Stability. The magic word in electronic components. And in resistors, the word for stability is Metal Glaze. And the word is out, because millions are already in field use. Metal Glaze, noted for its mechanical and electrical ruggedness, is also an extremely stable resistive system. Operated at rated power, and in many cases at twice rated power, Metal Glaze resistors go coolly along, resisting change for thousands of hours. They are truly precision resistors, and we have reams of test printouts to prove our case. The glass-hard Metal Glaze film is permanently fused to a solid ceramic core. The lead wire terminations are soldered to the core in a patented process. This makes the resistive element and the terminations one continuous component ready for molding in an environmentally protective envelope. Competitively priced in power ratings from 1/8 watt to 2 watts, Metal Glaze resistors are available to 1% tolerance, and 100 ppm temperature coefficient. TRW Metal Glaze resistors are available in quantity from your local TRW industrial distributor. Or contact TRW Electronic Components, IRC Boone Division, Boone, North Carolina 28607 Phone (704) 264-8861. Electronics review Under contract to the Naval Air Systems Command, Cubic is scheduled to have the first unit operating at the range at Yuma by November 1972. Two more ranges are planned, including one at Cherry Point, N.C. Total value of the contract, including four tracking stations for aircraft and missile testing, is expected to be about $47 million. Advanced technology Diode key to first precise measure of speed of light MIT experimenters led by Ali Javan, father of the gas laser, may have come up with the most precise method yet of measuring the speed of light. They have developed a diode that's small enough to measure infrared wavelengths. The development is a result of work going on in the Physics department's laser group since 1968 on direct measurement of the frequency of light waves. Their method involves focusing laser beams on a diode and measuring a resultant harmonic beat on an oscilloscope. But this necessitated the diode development because ordinary silicon diodes have tungsten leads that are 1 mil or more in diameter, too large to resonate beyond the submillimeter range. But a diode that works with light waves must have an "antenna" smaller than one wavelength of light—or smaller than any antenna built before. The structure of the MIT diode is conventional, and the only differences lie in size and composition. The tungsten wire is about 2.5 microns in diameter and $ \frac{1}{8} $ inch long, and is electrochemically etched to a point about 1,000 angstroms in diameter. This wire is mounted at the end of a coaxial cable and makes a mechanical contact with a nickel post. David R. Sokoloff, a member of the group, says "We suspect that an oxide layer forms on the metal post so it's really a metal-oxide-metal diode. Electron tunneling causes rectification, and some theoretical studies have been done on the response, but no one really knows what is going on." Focusing the light beams from two-near-infrared gas lasers on a diode's whiskers sets up electrical oscillations at their infrared frequencies. Simultaneously, a microwave beam from a klystron is directed at the diode, which also responds electrically at that frequency. The frequencies of the laser and microwave beams set up a harmonic beat note at a low frequency. The tungsten transmits the optical frequency alternating currents to a thin film of oxide on the diode post, where electron tunneling mixes the currents and converts them to a current that carries the beat to a scope for measurement. The frequencies of the two laser beams are then compared through harmonic analysis and measurement of the beat note. In an experiment using four lasers, the laser frequencies were mixed in the diode and tuned to a time signal sent from the National Bureau of Standards, making it possible to compare the various laser frequencies in several steps. With this method, a carbon monoxide laser frequency was found to be precisely 58,024,341 million cycles per second. With silicon diodes the highest frequency measured was about a twentieth as high. Sokoloff says the group is now experimenting with a method of vacuum-coating thin filaments on plastic to make still smaller whiskers capable of measuring frequencies into the visual range. "Up until two years ago people thought this couldn't be done," observes Sokoloff, adding that now "we can make measurements that have never been made before." For instance, the speed of light is the product of its frequency times wavelength. Wavelength is normally measured by interferometer or grating, and frequency is calculated using the experimentally known value of the speed of light. But with the ability to directly measure frequency in the infrared and optical ranges, it will be possible to determine the true speed of light more accurately. It is also possible that the standards of length and time will be made interdependent through a defined value of the speed of light. One possible use for time and distance measurement by lasers is laser ranging of the moon. For the record Sale. United Aircraft Corp.'s Hamilton Standard Medical Electronics division, Windsor Locks, Conn., reportedly has been sold to General Medical Corp., Richmond, Va., at an undisclosed price. A spokesman for General Medical says the sale is not final, and the company will not discuss details until it is. Hamilton Standard has been in the medical electronics business for five years, spinning-off UA-developed space technology for use in artificial kidney machines, and blood circulation and respiratory apparatus. Goldmark's goal. Having turned down $750,000 a year at CBS to sit back and think, Peter C. Goldmark, retiring head of CBS Laboratories, Stamford, Conn., has instead set up a new company to solve communications industry problems. The firm, to be located in Stamford, will be a systems engineering venture to bring state-of-the-art CATV and video cassette technology to bare communication needs of urban centers. His long-range objective is to piece together the many forms of communications, including satellite, for the cities of the future. Fast miniprinter. A serial printer up to 10 times faster than available competitors has been unveiled by This new IEC 10 MHz Function Generator has a couple of things going for it! (like sweep & phase lock) And that's not all. Sweep and phase-lock are just two of the numerous features that distinguish this new IEC 0.0005 Hz to 10 MHz all-purpose test instrument. For example, trigger/gate, and calibrated output attenuation. The F55. The Great One. Top of the Series 50 line that includes four other brand new function generators. What do they have in common besides superb performance? 10 MHz capability, fixed and variable offset, variable width pulses as well as fixed width pulses and ramps, simpler man/machine interface and much, much easier maintainability. That's what. Plus the special features of each model, such as the sweep and phase-lock blend of the F55. Depending on requirements, there's a particular Series 50 model that's exactly right for your needs. And priced accordingly, from $595 up. Like a look? Call, TWX, write or wire our John Norburg at Interstate today. Ask for a demonstration. Get a new IEC Function Generator going. For you. Three new GE SSL's to put more energy in your work SSL-54 SSL-55B SSL-55C We've just come up with three lens-end infrared solid state lamps with more useful energy than before. SSL-54 has more power in the narrow 20° cone than its sister lamps, SSL-4 and SSL-34. And you still get the same fast rise time. SSL-55B and SSL-55C, in a 20° cone, generate almost twice the mW per dollar than any other GE SSL. So you can use less expensive detectors than before. Or place them farther away than before. Typical ratings for the SSL-54, SSL-55B and the SSL-55C are 1.0mW, 4.8mW and 6.0mW, respectively. They're just right for "mark sense" applications, detection systems, and computer-related uses like BOT and EOT sensing. 1000 lamp prices: SSL-54 — $1.21, SSL-55B — $2.26, SSL-55C — $2.52 ea. Complete technical information on the new SSL's — previously called light emitting diodes — is free. Just write. and seven other ways to make your job a little easier. NEW NEON GLOW LAMP #3AG-F. New circuit component lamp with tinned leads, and a special silicone Dri-Film™ coating that increases leakage resistance to 1,000 megohms minimum. Use with MOSFET, matrix or time delays. NEW RED NUMERIC DISPLAYS SSL-140 SSL-190. Red, easy-to-read seven segment solid state readouts with character heights of .140" and .190". Wide segments for each viewing. All products shown actual size. * Trademark of General Electric Company. For General Electric's 2-part SSL manual on theory, characteristics and applications — 106 pages in all — send $1.00 for each set to: General Electric Company, Miniature Lamp Department, #382, Nela Park, Cleveland, Ohio 44112. CIRCLE NO. 230 SSL; 231 RED NUMERIC; 232 WEDGE BASE; 233 NEON GLOW Electronics review Printer Technology Inc., Woburn, Mass. Priced at $1,700 in OEM lots, the Printec-100 is aimed squarely at the minicomputer market, now using output devices like IBM's Selectric (at 10 to 15 characters a second and about $3,000), Teletypes, and more costly line printers. The new device's claims to fame are full characters, not dot matrices, a 1,200-baud phone line communications capability (said to cut communications costs), and its pricing. Manpower. Electronics and aerospace engineers show the highest unemployment rates—5.3% each—in a new National Science Foundation survey of engineering society members. EEs with computer and mathematics specializations also suffered unemployment rates higher than the 3% average for all engineers, while electrical and communications engineers were better off, with rates of 2.2% and 2.9%, respectively. The survey, conducted by the Engineers Joint Council this summer and published in late September by the NSF, showed the highest unemployment figures for young EEs [Electronics, Sept. 27, p. 27]—5.5% for those under 24—followed by the 60- to 64-year age group with 4.2% and the 55- to 59-year-old group at 4.1%. Those engineers without an academic degree were the hardest hit with 4.4% unemployed, followed by those with master's 3.2%; bachelor's, 2.8%; and doctorates, 1.9%. Defense-aerospace unemployment was the highest, with engineers previously working in positions directly related to aerospace reporting a rate of 6.3%; defense, 4.8%; and transportation, 3.3%. The lowest rate, 1.3%, was reported by engineers previously employed in public works. Boost. By adding an optical cavity or waveguide region beside the p-n light-generating junction within a GaAs semiconductor laser, RCA Laboratories in Princeton, N.J., says it has doubled the output efficiency at peak powers. Varying cavity size affects output power, with units producing anywhere from 1 watt peak to 20 watts. SIGNED by Bourns; the quality leader. SEALED against solvents and processes normally used in circuit-board cleaning. DELIVERED off-the-shelf from your local Bourns distributor, or factory direct... A-BUCK-A-WATT Get 1 watt for the same price you formerly paid for a comparable ½ watt wirewound potentiometer. SPECIFY Bourns new Model 3345...the latest addition to Bourns growing family of LOW-COST TRIMPOT® POTENTIOMETERS MORE POWER PLUS... • SMALL SIZE: .450" diameter x .23". • LOW T.C.: 70 PPM/°C over temperature range of -55 to +125°C. • STANDARD RESISTANCE RANGE: 10 to 50K ohms • BOTH TOP and SIDE ADJUST Configurations. For complete details on the Model 3345, contact your local Bourns Distributor, Representative, Sales Office, or Bourns Trimpot Products Sales Department. 5000-9999 place price BOURNS INC., TRIMPOT PRODUCTS DIVISION • 1200 COLUMBIA AVE., RIVERSIDE, CALIF. 92507 Will The Designer Who Ordered Plastic Silicon Power In More Than One Package Size, Power Rating and Lead Form, Please Drop It Right Into It’s Ready! Thermopad plastic power — now available in just about any size, power rating and lead configuration you require . . . for immediate drop-in into any metal-device socket you might have: TO-66, TO-5 or PCB, for flat or flag-mounting, with or without heat sinks . . . in high voltage, complementary EpiBase® and Darlington technology! And you’ve got over 120 types to do it with: 33 registered and 90 house-numbered units that ensure your getting exactly what you need . . . eliminating over or under-design . . . without costing more than you want to pay! You have your choice of 40 “small” case 77 types offering the most compact design possibilities . . . available in either E-C-B or B-C-E pin styles . . . readily replacing TO-66 units and furnished in 10 different lead forms. Use the newest, “medium-sized,” case 199 to directly replace TO-66s and other plastic types with unformed leads where direct, metal-to-metal mounting screw attachment is desired. Like the case 77 and case 90, the 199 provides the most intimate heatsink-to-header contact seal in the industry and offers superior hermetic case integrity in high humidity environments. Power dissipation is also greater because of its 60% bigger heatsink area. “Large” in every way, the case 90’s minimum, .032” thermal path length provides essentially the same $\theta_{jc}$ as a copper TO-3 in spite of the fact it has only one-fourth the heatsink surface area! You also get Darlington/complementary capability and up-to-100 W performance! Get the whole story on Motorola plastic silicon power — write Box 20912, Phoenix, AZ 85036 and receive data sheets plus a copy of the new LEADFORMS brochure . . . it details dimensions, styles and applications of Motorola’s unique capability to fit plastic power devices to any of your sockets. It’s right now! His Sockets. 3 Different Sizes, 19 Different Lead Forms To Choose From! MOTOROLA POWER In The First Place, It's Easy. Circle 45 on reader service card This ROM doesn’t need masks. Why wait around and pay more for one that does? The masked ROM has had it. The future of the masked memory is being cut short by a 2nd-generation ROM of a different color: Intersil's 256-bit (32 x 8) 50-ns programable IM5600. It's a fully decoded T²L memory, housed in a 16-pin ceramic or plastic DIP or flatpack. It comes in full military or commercial ranges, with a choice of open-collector or tri-state outputs. The difference is that every IM5600 is delivered with logic "zeros" in all 256 locations. When you need a particular ROM, you yourself can program—permanently and very simply—the logic "ones" wherever your truth table dictates. And by every standard, the completed ROM measures up to and beats what's previously been available. \| \| BIPOLAR MASKED ROM \| INTERSIL IM5600 ROM \| \|----------------------\|--------------------\|---------------------\| \| Price \| Down to 1¢ per bit \| Below 1¢ per bit \| \| Mask charge \| $500 typical \| Zero \| \| Inventory costs \| High \| Minimum \| \| Availability \| In quantity \| In quantity \| \| Delivery \| 5 to 12 weeks \| Immediate \| \| Reliability \| Excellent \| Excellent \| Get a programer free. But here's the biggest bargain. Get a free programer box with your first order of IM5600s totaling 1,000 pieces or more. This suitcase-mounted unit automatically tests each ROM, programs it and then verifies the accuracy of the completed memory. All at a rate of better than two ROMs per minute. Reliability is a quarter-billion bit hours. And do they last? You bet. A constantly monitored study has logged well over 1,000,000 device hours of high temperature operating and storage life testing without a single failure. Send for full details covered in IM5600 Reliability Report. By the carload, quick as a wink. The IM5600 is in volume production right now. Ask for what you need and get it when you need it. It can be computer-programed at a millisecond a bit, or one girl can program a thousand memories a day using an Intersil Portable Programer Box. Order them pre-programed from your Intersil distributor and you still get off-the-shelf delivery. Or buy quantities of blank ROMs, mount them on your own PC cards and stock them. When you need it, program a complete card at a time, plug it in and go! And Intersil guarantees 100% programming yield. From Intersil, the one to remember when it comes to memories. 10900 North Tantau Ave., Cupertino, CA 95014. Get 'em here. Intersil stocking distributors. Schweber Electronics; Century Electronics; Semiconductor Specialists; DeMambro Electronics; R. V. Weatherford Co. Intersil area sales offices. Los Angeles (213) 370-5766; Metropolitan New York (201) 567-5585; Minneapolis (612) 925-1844; San Francisco Bay Area (408) 257-5450; Singapore, Telex RS 21607. Overseas representatives. Clichy, France: Tranchant Electronique. Amsterdam, Holland: Klaasing Electronics. Tokyo, Japan: Internix. Zurich, Switzerland: Laser & Electronic Equipment. London, U.K.: Tranchant Electronique. Munich, West Germany: Spezial Electronics. U.S. sales reps in all major cities. Boeing showed the way home. When the Apollo 15 astronauts went for a drive in their Lunar Roving Vehicle, they could depend on a reliable navigation system to get them "home" to the lunar Module again. Distance and heading back to the lunar Module were displayed continuously on the LRV console, derived from the output of a unique Signal Processing Unit designed and built by Boeing. This compact digital/analog computer had to meet some of the toughest specs ever for electronic hardware. Such as gravity forces of $12\frac{1}{2}$ Gs during launch. Temperatures from minus 250 to plus 250 degrees. And hard vacuum. What's more, the SPU, along with the rest of the navigation system and the complete LRV, was designed, built, tested and delivered to NASA by Boeing in the shortest time ever for Apollo-qualified hardware. It's a typical example of Boeing's total capability for on-time delivery of highly reliable special-purpose computers. Other Boeing computers now in development include a controller for the USAF Airborne Warning and Control System, a central computer for Mariner/Venus Mercury '73, and a low-cost attitude computer for spacecraft. Early development work is under way on advanced configurations for an ultra-reliable Grand Tour computer. When you need a special-purpose computer, come to Boeing. We can do the exact job you need. At very competitive prices. Write to Mr. Herbert S. Broadwell, Boeing Electronic Products, P.O. Box 3999, Seattle, Washington 98124. Or call him at 206-773-6116. BOEING ELECTRONIC PRODUCTS Electronics/October 11, 1971 Introducing the little counter that can. It can become four different systems. It can go anywhere you do. It can protect you against obsolescence. It can make buying and maintaining a counter less expensive than ever before. Meet the Hewlett-Packard 5300, the snap-together counter that's not much bigger than the palm of your hand. It has six digit accuracy, solid state display and autoranging. It'll make period, frequency, time interval and ratio measurements, operate on its optional snap-on battery pack and drive a printer. Rugged dust-proof aluminum case resists almost any bumps it might get in the field. Prices start at only $520 for one of the most amazing counters you've ever owned. Start with the basic mainframe ($395). Then snap on any of the following modules (more on the way) to make just the counter you need, and avoid obsolescence, too: 10 MHz frequency module. Model 5301A, $125. 50 MHz all-purpose module includes period, time interval. Model 5302A, $250. 500 MHz module with both 50Ω and 1 MR inputs. Model 5303A, $750. 100 ns time interval module with: unique "time holdoff" feature, dc coupling, slope and trigger level controls, and period and frequency measurements to 10 MHz. All the functions you'd pay $1,200 for in a universal counter. Model 5304A, $300. Rechargeable battery pack module works with any of the other modules for cord-free operation. Model 5310A, $175. The 5300 is one system you have to use to appreciate. If you've ever needed to accurately measure frequency or time interval, you owe it to yourself to call your nearby HP field engineer for further information. Or write Hewlett-Packard, Palo Alto, California 94304; Europe: 1217 Meyrin-Geneva, Switzerland. Counters that promise a lot and deliver it all. HEWLETT PACKARD ELECTRONIC COUNTERS Circle 49 on reader service card LINEARS FROM RAYTHEON SEMICONDUCTOR. OP AMPS 702 Wideband Amplifier 709 General Purpose 725 Precision 741 General Purpose 747 Dual 741 748 Uncompensated 741 4709 Dual 709 101 General Purpose 107 Compensated 101 108 Precision* 4558 Dual 741 (Available in mini-DIP) 4531 High Slew Rate 4131 Precision 4132 Micropower INTERFACE ELEMENTS 710 High Speed Comparator 711 Dual High Speed Comparator 9622 Dual-Line Receiver 7520 Series Core Memory Sense Amplifiers 4431 Dual-Line Receiver 1414 Dual 710 with strobes 1488 Quad Line Driver / EIA Compatible* 1489 Quad Line Receiver / EIA Compatible* VOLTAGE REGULATORS 723 Precision Positive / Negative 109 + 5V Voltage Regulator* RAYTHEON LINEARS ARE AVAILABLE IN ALL POPULAR PACKAGES: Plastics — 8-pin mini DIP, 14 and 16-pin DIPs T Pak — TO-3; 8 and 10-pin TO-5 Ceramic — 14 and 16-pin DIPs; 10 and 14-pin flat packs When you need a linear IC of any kind, give us a call. We've got what you want. We're stronger than ever in both proprietary and off-the-shelf linear circuits. And that's not our only strength. Raytheon Semiconductor has changed. Our new technical team is deeply involved in development programs to provide new design tools. New digital analog interfacing that allows active switching and amplifying. Greater integration of external parts in voltage regulators. Super Beta process for precision op amps. If you haven't been keeping up with us lately you've got a surprise coming. Write for more LIC information and see. Raytheon Semiconductor, 350 Ellis Street, Mountain View, California 94040. (415) 968-9211. RIP THIS PAGE OUT AND TACK IT ON YOUR WALL. YOU CAN USE IT. Available fourth quarter 1971. Circle 50 on reader service card Beyond the World Wide Military Command and Control System's (WWMCCS) 13 computers, scheduled for contract on Oct. 15, military sources predict awards for the purchase, lease, or expansion of 140 more systems during the next six months. Biggest single piece of hardware in the Army's 74 computer actions is the Persincom 70X, a large personnel management system using 126 remote terminals in the Washington area to replace three CDC 3800s, one Burroughs B3500 and six IBM machines. The Army also plans two computer moves for automated telecommunications centers at Bayonne, N.J., and Oakland, with options for more. With most of the acquisitions by all the services falling in the category of standard business computers, IBM stands tallest in the competition. Nevertheless, Univac and Burroughs are competing strongly for many of the big machine orders, with Honeywell, although not a military favorite, also pushing. Though the Navy plans 37 computer acquisitions and the Air Force another 18 in addition to WWMCCS, the second biggest dollar buy contemplated probably will be the Defense Supply Agency's award for 11 processors to be used in its Samms program for supply and material management. Big Navy programs include: 11 small shipboard computers to be acquired by NavShips, six large-scale machines for Naval Ocean Surveillance Centers, and one large-scale system to replace two CDC 3800s at the Naval Research Laboratory. Potentially big is the sale of a single machine for the first phase of the Automatic Data Distribution System that will interface with the Naval Tactical Data System. Other Navy EDP efforts in the next six months cover one general-purpose machine for the Underwater Systems Center, six small ones for hospitals, and seven numerical control systems for rework centers. The Air Force is still after six large systems for regional use in its delayed Advanced Logistics System effort, another to replace an IBM 7040 at the Human Resources Laboratory and 10 small systems for as many of its hospitals. The National Academy of Engineering has asked the National Institutes of Health to expand and clarify its role in biomedical engineering. To help do this, NAE wants NIH to create an overview body of industrial, medical, engineering, and social experts. In a report to be published late this month, the academy also recommends that NIH expand its in-house engineering competence, support goal-oriented research, help establish university programs for design and product-oriented biomedical engineers, and support engineering internships at the institutes. Dr. Herbert H. Rosenberg, director of NIH's resources analysis office, says he expects the hierarchy at the Department of Health, Education, and Welfare to respond favorably, but cautiously, to the report, which reflects an Arthur D. Little survey of 50 biomedical engineering firms. Biomedical engineers should be applying their skills to health service rather than to basic research, researchers at George Washington University medical school have decided in a study funded by the EngiWashington Newsletter neering Foundation. That school's James C. Aller estimates that there may be about 3,000 biomedical engineers currently employed in what is now called a "clinical" engineering capacity, but the health care industry could use up to 50,000 profitably—decreasing the cost of health care by reducing equipment costs. "The research and manufacturing segments of the biomedical instrumentation industry are reaching maturity, so the thrust of biomedical engineering curricula should be redirected towards keeping equipment in service," Aller says. "Engineers should be more concerned with the selection and upkeep than with the design of new instrumentation." LEAA leans to computers for voice recognition Future LEAA research in speaker recognition may be limited to computer techniques, implies a Law Enforcement Assistance Administration study to be published later this year. "This method of speaker recognition may prove to be the most promising," concludes the report, written for LEAA by the Michigan state police. The potential for legal acceptance of speaker verification by panels of listeners, and by observers examining voice prints, the report explains, is diminished by the fact that those methods are not completely objective. LEAA, which has put $500,000 into speaker verification so far, has subcontracted Stanford Research Institute for specification of criteria, and perhaps hardware, for computer recognition techniques. About $25,000 of the SRI award is going to Texas Instrument to maintain a liaison with a manufacturer. Jobs in aerospace engineering seen stabilizing by June Though Federal officials contend "the worst is behind us" when they speak of the economic slump and its related job losses, there is fresh evidence that aerospace and its related electronics will be among the last industries to stabilize. Overall jobs are expected to continue to decline through June 1972, says the Aerospace Industries Association, to a level of approximately 894,000. Of these jobs, 147,000 will be engineers and scientists—a drop of 37.5% from the 1967 peak of 235,000. A breakout of engineering jobs by product groups shows there were 83,000 engineering and scientific jobs in the aircraft industry in March, for example, but this slipped to 77,000 by June and will drop to 75,000 in December, before levelling off at 69,000 next June. Comparable openings in missile and space work were 51,000 in March and 49,000 in June, will decline another 1,000 by December and stabilize at 47,000 by June. Other aerospace engineering jobs—ground electronics, for example—stood at 32,000 last March, dropped to 31,000 by June and are expected to slip another 1,000 by year end before bottoming out at about 30,000 by the middle of next year. Addenda The Navy committed itself at the Sept. 30 deadline to buy 48 more F-14 fighters from Grumman Aerospace for $806 million when Senate floor opposition to eliminate the buy crumbled, as anticipated [Electronics, Aug. 16, p. 48]. . . . But the Navy apparently lost on its extremely-low-frequency Sanguine system for shore-to-submarine communications [Electronics, July 5, p. 35] with 44-42 Senate passage of an amendment to cut its $5.57 million appropriation by $2 million. The Navy may recoup, however, on a vote to reconsider if the 13 senators absent can be rounded up. Our new DIP gives you just what you need and no more. First we put four fixed resistors and a trimmer in one module. We called it our TRN (Trimming Resistive Network) package. Great. Now you can get the same module, with the same dimensions for automatic insertion and the same excellent characteristics, in a DIP trimmer. And just a trimmer. Not just a 3/4-inch commercial trimmer turned on its side, but one that has been developed from the ground up to meet all dimensions in accordance with EIA Microelectronic Outline DIP Family with 0.300-inch-row spacing. Resistance range of our new DIP is from 10 ohms to 1 megohm. The resistance tolerance is ±10%. And like the TRN, the DIP offers excellent TC of 100 ppm standard with 50 ppm available. The operating temperature range of this DIP is −55°C to +125°C. Power rating is 3/4-watt at 40°C. So if you want the whole works—trimmer and from 1 to 4 fixed resistors—in one package, get our TRN. But if you want a trimmer and no more, then our new DIP will give you just what you need. Both are now available through Amphenol's distributor network. For more information write Amphenol Controls Division, Bunker Ramo Corporation, 120 South Main Street, Janesville, Wisconsin 53545. ASK FOR NEW EARNINGS REPORTS IBM · ABC · XEROX THIS WEEK ONLY PORTFOLIO ANALYSIS $3.95 10 BUY SELL DECISIONS BEST BUYS TEXACO 35 5/8 AT&T 112 1/4 RCA 38 3/8 NO STAMPS WITH STOCK PURCHASES TODAY'S FEATURES AMPEX 18 3/4 BUDD CO. 13 5/8 GEN. FOS. 82 1/2 LOEWS TH. 18 7/8 LITTON IND. 27 7/8 MCI 16 1/4 NORTH AM. 57 1/2 PAN AM. 52 3/4 SCH. 11 1/8 LOW PRICE SPECIALS KIN ARMY BLUEBELL GABRIEL By 1979, you’ll shop for stocks in your supermarket. The stock exchange is going electronic. Millions of dollars are now being invested in computers, terminals and electronic components to break today’s paper bottleneck. Already, Wall Street is the biggest single market for read-out devices. By the end of this decade, the securities industry may well be one of the biggest single markets for all electronics. You’ll buy and sell stocks through terminals located not only in brokerage houses, but in hotels, restaurants, private offices and even supermarkets. So, if the price of sirloin is up, you may still save the day with a great buy on IBM common. Just slip your identification card into the terminal and buy or sell any stock you want. Each transaction will be verified instantly, with your account statement printed out on the spot. But helping you increase your portfolio is only one way electronics will make life easier in the future. The fact is, products of electronics technology will be doing more for us tomorrow than electricity does for us today. In business, transportation, communications, health care; even housework. Who are the master minds masterminding these changes? Our readers. Among them design, quality, and production engineers in components, computers and computer peripheral equipment. Every two weeks, Electronics presents them with a complete up-to-the-minute picture of the state of the technology. Plus all the fast-changing developments in their particular fields of interest. Industry-wide and world-wide. If you want to be a part of the future, speak to the men who are working on it today. Electronics, a McGraw-Hill market-directed publication. Our readers are changing the world. Do you face a make or buy decision on power supplies? USE LAMBDA POWER HYBRID VOLTAGE REGULATORS—AND BUILD YOUR OWN. ($16 in quantities of 1000) FROM MELVILLE, N.Y. 1-DAY DELIVERY FROM CHICAGO, ILL. 1-DAY DELIVERY FROM N. HOLLYWOOD, CAL. 1-DAY DELIVERY FROM MONTREAL, QUE. 1-DAY DELIVERY The world's first Power Hybrid Voltage Regulator, the biggest advance in power supply design since the silicon power transistor, is now available for 1-day shipment from four Lambda distributing points. 22 models provide up to 28 VDC, up to 5 A output, 85 watts dissipation, 0.2% line or load regulation . . . @ $16 in quantities of 1000. If you build your own power supplies, the Power Hybrid Voltage Regulator will save you money at every step from design through production. And now Lambda will save you time as well. Call or write for Brochure L-10. MELVILLE, NEW YORK 11746 515 Broad Hollow Road Tel. 516-694-4200 ARLINGTON HEIGHTS, (CHICAGO) ILL. 60005 2420 East Oakton St., Unit Q Tel. 312-593-2550 NORTH HOLLYWOOD, CALIF. 91605 13425 Wyandotte Street Tel. 213-877-0041 MONTREAL, QUEBEC 100C Hymus Blvd., Pointe-Claire, Quebec-730 Tel. 514-697-6520 WHETHER YOU MAKE OR BUY... LAMBDA ELECTRONICS CORP. A Venco Company What's smaller than a breadbox, has two channels and built-in preamps, and writes well under pressure? Answer: the Brush 220. The Brush 220. A rugged, two-channel recorder that's little to the tune of 25 pounds, accurate to 99½% linearity. Just plug it in. And then let Brush performance take over. It starts with our pressurized ink system. A system that forces ink into the paper, instead of just laying it on. So you get traces that are clear, clean, and crisp. And a disposable ink cartridge holds a year's supply of ink. Built-in preamps give you a measurement range of 1 millivolt per chart division to 500 volts full scale. And a choice of 20 signal conditioners give you an even wider range. Four chart speeds are push-button controlled. And two event markers are standard. The Brush 220. It just may solve your recording riddles. To see the Brush 220 in action, call your Brush Sales Engineer. Or write for more information to Gould Inc., Instrument Systems Division, 3631 Perkins Avenue, Cleveland, Ohio 44114 or Rue Van Boeckel 38 Brussels 1140, Belgium. The next generation is here now The SABRE IV is more than a laboratory recorder/reproducer. It's a complete magnetic tape system that is setting NEW STANDARDS in the accurate recording and recovery of data...whether low or high frequency, analog or digital. A SABRE IV in your laboratory means data reduction performance unmatched by other recorder/reproducers. The SABRE IV allows you to reduce data up to 2.0 MHz in the direct mode and 500 KHz in the FM mode with unequalled reproduction quality. With eight transport speeds the SABRE IV allows time base expansion and contraction ratios up to 128:1. The Total Bandwidth Convertibility of the SABRE IV provides more capability than previously found in two or more recorder/reproducers. If you would like to know more about the most advanced magnetic tape systems available today...including the best portable system available today...write or call: SANGAMO ELECTRIC COMPANY The Innovators in Tape Instrumentation Data Systems P.O. Box 3347 Springfield, Illinois 62708 (217) 544-6411/ Telex: 406-421 PLENCO - FLAME-RETARDANT - TRACK-RESISTANT - X-RAY SHIELDING PHENOLIC, MELAMINE-PHENOLIC AND ALKYD MOLDING COMPOUNDS Thermosets formulated to provide an extra factor of product safety. With greater range in choice of properties. Improved arc-resistance, excellent resistance to tracking under the influence of conductive contaminants, X-radiation shielding, and humidity-resistant characteristics are among the qualities offered by Plenco in this series of compounds. All are recognized under the Component Recognition Program of Underwriters' Laboratories, and were developed to help answer current concern for increased safety in electronic and electrical products. Where design and reliability requirements are being upgraded, Plenco gives the designer a wide range of properties and cost. Permits him to optimize the best approach to his problem. FOR COMPLETE INFORMATION on these or any Plenco molding compound, write for brochure, data sheets or property charts. Wide-ranging selectivity and experienced assistance is of course offered you whatever your molding problem or process. Just part of our regular service, and yours whenever you wish it. See us at the Show—Booth 1132 PLASTICS ENGINEERING COMPANY 1607 Gee.e Ave., Sheboygan, Wis. 53081 Through Plenco research . . . a wide range of ready-made or custom-formulated phenolic, melamine, epoxy and alkyd thermoset molding compounds, and industrial resins. Elco’s Series 6064 discrete card edge connectors look like . . . and work like . . . other discrete card edge connectors. With one major exception. Ours can save you up to 20% of your total package cost. Because we’ll supply our Series 6064 connectors mounted on our Variframe™ back panel which we’ll custom design and build to your requirements. With connectors wired (up to three levels of wire wrapping) to your specs. And with input/output connectors and voltage and/or ground buses. To give you the best performing package at lowest cost. Fast. Then again, if you’re in a hurry for a look-alike that is, call your local Elco distributor. He has our Series 6064 connector in three sizes: 36/72, 40/80 and 50/100 contacts. With bifurcated bellows contacts on .125” grid. All with .025” square tails. And dialyl phthalate insulators that accept .062” printed circuit cards. And he’s got them all on his shelf. Plus other card edge connectors at .100”, .150”, .156” and .200” spacings. Ready to go. Availability. And economy. These are what distinguish our look-alikes from theirs. Just two more services in keeping with CONNECTRONICS, Elco’s Total Connector Capability. For full details and specifications on our Series 6064 connectors, contact: Elco, Willow Grove Division, Willow Grove, Pa. 19090 (215) 659-7000 Elco, Pacific Division, 2200 Park Place, El Segundo, Calif. 90245 (213) 675-3311 Operations in USA, Australia, Belgium, Canada, Denmark, England, France, Germany, Israel and Japan. Sales offices throughout the world. Copyright © 1971 Elco Corp. All rights reserved. Minicomputer evolves with maxi capabilities: p. 61 (cover) What is a minicomputer? If your answer is a small computer that performs simple functions economically where speed and capability aren't important, you're not with it, says a trio of authors from Digital Equipment Corp. The latest addition to the PDP-11 family, with its Schottky TTL/MSI logic, semiconductor mixed memories, and floating-point units, has evolved into the general-purpose sector formerly the exclusive domain of much larger machines; yet it maintains the cost-effectiveness that originally made minicomputers feasible. The cover: The front panel of the new PDP-11/45 gives a hint of its impressive capabilities. The address control board, with its Schottky TTL/MSI circuits and its bipolar read-only memories for microprogramming functions, tells even more. Computer modeling works for zener diodes: p. 67 Despite the widespread use of the zener diode in a host of circuit designs, only a few computer models are available for it, and these don't adequately deal with its thermal effects. The latest of the models, however, not only can handle the thermal considerations, says author Emanuel Schnall, but uses easy-to-evaluate circuit elements and can adapt to linear and nonlinear analysis programs. Millimeter-wave applications and Impatt diodes—the perfect match: p. 78 The proliferation of millimeter-wave communications, alarm, landing, and radar braking systems is greatly increasing the attractiveness of silicon Impatt diodes as a solid state source, notes author N.B. Kramer. The diodes give high power with high efficiency, are compact, and offer impressive flexibility in a number of important applications. Bit error rate detector—handy tool for digital data links: p. 82 Digital data transmission system designers have a problem. As their systems proliferate, there's an increasing need for bit error correction, but the usual approach is through software—frequently expensive and complicated. The solution, says author Kingsley P. Roby, is to implement bit error rate detection in hardware: the technique is simple and inexpensive, and uses the same principles as in analog networks. Taking on interconnection problems in big multilayer pc boards: p. 88 In very large, very dense systems, designers can't afford to ignore the considerations involved in making thousands of reliable interconnections, says author P.O. Benham. One such challenge was a 3-foot-diameter, nine-layer pc board for a phased-array antenna system that needed 23,000 plated-through holes. It was conquered with a numerically controlled drilling operating and a plastic-encased preform solder system. And in the next issue... How electronics companies are automating to survive hard times... shortcuts for Chebyshev and Butterworth filter analysis... how gallium phosphide and gallium-arsenide-phosphide stack up... timing for semiconductor memories. Evolution breeds a minicomputer that can take on its big brothers Fast Schottky TTL/MSI, semiconductor memories, multiple unified buses, microprogramming, and floating-point units bring minicomputer up to large-machine capabilities without sacrificing cost-effectiveness by Richard J. Clayton, Bruce A. Delagi, and Rony Elia-Shaoul, Digital Equipment Corp., Maynard, Mass. By incorporating the newest technological developments and building on changing user requirements, the minicomputer is evolving along lines that are putting less emphasis on the "mini" and more on the "computer." Now, instead of remaining the limited-performance, stripped-down version of the bigger machines that met a need for economical computing power where speed and capacity weren't essential, the minicomputer is fast reaching the capabilities of the larger computers. At the same time, the cost-effective design philosophy that established the minicomputer concept is being retained. The latest step in this evolution is the PDP-11/45, the newest addition to the PDP-11 family. The 11/45 is a 16-bit, general-purpose computer that refines minicomputer organization and technology to solve problems normally associated with larger computers. Multiple unified buses, mixed memory systems (bipolar, MOS, and core), an exceptionally powerful instruction set, microprogramming, hardware floating-point units, memory segmentation, Schottky logic gates, and multilayer boards are some of the techniques that are helping the PDP-11 family to redefine the boundaries of minicomputers. To understand minicomputer evolution consider the basic structure of the first minicomputer, as shown in Fig. 1 (a). In such early machines, separate buses to and from the central processor were provided for input/output and for memory. In most cases, the input/output bus operated synchronously (clocked), could handle only those peripherals within a narrow range of response times, and was relatively slow. The memory bus was fast and generally asynchronous. With this configuration the limit on memory usually was no more than eight modules of 4 kilowords each. Also, bus length was held to about 10 feet and terminations had to be tuned rather carefully. Speed ranges were quite restricted, and mixed memory add-on capability was not available. Minicomputer design made a significant advance when the concept of unified busing was introduced in the PDP-11 architecture [Electronics, Dec. 21, 1970, p. 47]. The Unibus is essentially a fast asynchronous system for data transfer between computer system elements. Its features are overlap of bus transfer, selection of the next system element to be granted use of the bus (bus arbitration), fast asynchronous operation, and improved drive capability over previous minicomputer memory buses—twice as many loads can be driven over five times the distance. Overlap of bus transfers and bus arbitration increases bus channel capacity. Multichannel direct memory access and all interelement communications are obtained in this way with the involvement of no additional cost or time penalty. Because the Unibus operates asynchronously, a user can mix memories of different speeds, and bus buffers (bidirectional repeaters) can be added to expand load capability and line lengths as required. System communication on the Unibus, as used in the 11/20, the first general-purpose member of the PDP-11 line, is illustrated in Fig. 1. Even though the Unibus was successful in increasing system generality, the advent of high-speed semiconductor memories, with access times surpassing the maximum speed of the bus, required a new approach to memory organization. Basically, the new approach, shown in Fig. 1 (c), is to separate the memory system into two parts: one for the slower core, bulk memory devices, and conventional peripherals; and the other for the higher-speed semiconductor memory units. To accomplish this, a very fast, short, dedicated bus is added to the system. This bus connects a dual-port semiconductor memory system to the central processor. The second semiconductor memory control port is a standard Unibus interface (Unibus 2 in the figure). The dual-port architecture brings to a minicomputer a degree of flexibility and system performance improvement previously only seen in much larger systems. Because communications between the central processor and solid state memory need involve only one of the two semiconductor memory controls, the second control is free to handle external Unibus data requests up to the full Unibus rate of 40 million bits per second. In more advanced configurations, Unibus 2 need not be directly connected to Unibus 1, but may be used as a second bus operating simultaneously and independently of Unibus 1. Thus, Unibus 2 can be a very high-speed swapping channel while keeping Unibus 1 to handle interrupts and other external requests. One of the more interesting applications, shown in Fig. 1 (d), of this second Unibus is in cross-connecting two processors so that both operate in an environment 1. Comes the evolution. Early minicomputers (a) had separate I/O and memory buses, usually less than 10 feet long. Unified bidirectional bus concept (b) could handle more loads on lengths up to 50 ft. New approach in the 11/45 (c), required for fast IC memories, uses several buses: a Unibus for slower-speed memories; another for faster ones; and a dedicated unit to handle transactions between CPU and fast semiconductor storage. Dual-processor 11/45 system (d) allows both to share data with virtually no conflict or data time penalty. that offers virtually no conflict or data time penalty in accessing a common data base. The high-speed semiconductor memory system as shown in Fig. 2 comprises matrix and control modules. The matrix module contains the actual memory elements. The two semiconductor memories currently available for the PDP-11/45 are MOS units with 4,096 words of 16 bits and bipolar stores with 1,024 words of 16 bits. A control module provides timing, central processor interface, and Unibus interface circuitry for up to four matrix modules. MOS and bipolar matrix modules require different control modules. The 11/45 contains physical provisions for up to two controllers, each associated with four matrix modules. Thus, a maximum of 32k words of MOS, or 8k of bipolar, or 16k of MOS combined with 4k of bipolar, is possible. This organization is suited to accommodate any future higher-density semiconductor memory elements. The MOS memory used right now is the 1103-1, a 1,024-bit chip first delivered by Intel. This choice was based on cost and performance of production-proven devices. The MOS memory system level access time is 350 nanoseconds, and unlike core memory systems offering comparable access times, is much less dependent on addressing patterns. While the MOS memories offer a substantial speed improvement, the PDP-11/45 was designed for still faster operation and also offers a lower-density, higher-power, and higher-cost bipolar memory system for high-speed applications. On a system basis, access time here is 295 ns, again largely independent of addressing pattern. The primary memory component is a 256 bit, TTL-compatible bipolar chip available from Intersil and Texas Instruments. Because semiconductor memories are inherently volatile, several methods are available to the user to handle temporary power outages. Of these, the most common is to keep basic process-state information in core memory, which usually is available on all large systems, while dedicating semiconductor memory to program storage. Since programs are generally not self-modifying, they can be kept on disk or tape storage and then replaced automatically when power is restored. Also, a range of power-retention options is available to keep the system up until backup primary power can be restored. Without semiconductor storage the 11/45 is about twice as fast and costs, depending on memory and instruction execution mixes, about 1.5 times as much as the 11/20; with semiconductor storage, the system can be as much as seven times faster, but only twice as expensive. All of the PDP-11/45 memory systems offer optional byte parity. This increases the word length of the matrix to 18 bits and provides additional reliability. The PDP-11 family instruction set also represents an evolutionary step in minicomputer architecture. The worth of an instruction set is evaluated on the basis of three criteria: how easily coding to solve problems may be accomplished by the user; once coded, how fast the program runs; and how much storage is required to contain this code and the data on which it operates. In the PDP-11 family, focus is on the data types, the data structures, the operations, and the interrupt sys2. New approaches.* The CPU's memory management unit accesses the fast semiconductor stores through the dedicated bus, and provides address mapping and memory protection. Optional floating-point processor operates asynchronously to implement floating-point arithmetic. tems that form the basis of the solutions of problems. The entire range of data types (bit, byte, word, multiple word, signed and unsigned integer, and floating-point variables) can be effectively handled by the PDP-11 family instruction set. Byte data, for example, is commonly used in both operating systems and customer environments. Because a programmer working with the 11 family has many byte instructions available, he can handle words or bytes as the program demands without loss of coding, storage efficiency, or speed of operation. In addition, he can conveniently deal with signed and unsigned data. For example, it's frequently necessary to add, subtract, or compare signed operands, (data) or unsigned operands, (address), and detect overflows. Most machines can handle only one, not both. In earlier machines, for instance, unsigned arithmetic comparisons were done with hardware; the signed arithmetic comparisons were handled by software, which took several instructions to accomplish. This not only prolonged computer time, but greatly increased the probability of programmer error. In the 11/45, a complete set of conditional branches and condition codes provide the hardware to handle both signed and unsigned operands. In virtually all applications, data elements are organized into structures. A key feature of the 11 family is its ability to directly support a wide variety of data structures by any of 12 distinct addressing modes. This support lets the programmer choose one that's optimal to his specific problem. Even as recently as 1968, minicomputers had very primitive interrupt structures, and no way of controlling priorities. Thus, when a request for access to the bus was received from a teleprinter, disks, or tapes, the system had little flexibility to determine which request was more important and what time slot to allot to it. Further, all the devices on the line had to be polled to determine which had made the request. Moreover, there was no software interrupt system to book requests that could be serviced at a later time when processor priority level dropped. When the PDP-11/20 was introduced, four priority levels for interrupts were provided. And in the 11/45, the number of priority levels has been increased to eight, four for hardware or software interrupts and four for software interrupts only. Also, polling was eliminated—the interrupting unit provides a unique interrupt vector that identifies the requesting device. Another feature of the 11/45 is a second set of general registers. With it, the processor status word can be used to specify which set of registers is assigned to a particular device interrupt-service routine. Thus some of the general registers can be dedicated to servicing very fast devices only, reducing interrupt service times by as much as 50%. The classic approach to CPU design is to implement major state registers, instruction decoders, and other functions using networks of combinational logic in discrete elements. Although usable in a system with the extensive range of features of the 11/45, these networks are expensive, physically big, and difficult to debug. However, in the microprocessor design approach for the 11/45, bipolar read-only memories are used to interpret instructions and operate the control points both within the computer's data path and the external environment. Because microprocessor implementations directly realize instruction flows, the design goes much faster. Debugging of the initial breadboard proceeds rapidly because it's easy to change the control memory's Why Schottky TTL won High-speed logic was essential in achieving the 295-nanosecond operating speed of the PDP-11/45. The choice fell between the well-established emitter-coupler logic and a newer family recently introduced by Texas Instruments—Schottky TTL. Small component size is a key design element in fast systems because of the need for short signal paths. Availability of a wide range of MSI/TTL circuits (both Schottky and standard) was a key point in the choice of Schottky TTL over ECL. In many cases non-Schottky MSI gates operate faster than a logically equivalent collection of individual Schottky gates, so both types of TTL can be used wherever either type is most cost-effective. Schottky TTL utilizes the same thresholds, logic levels, and power supplies as conventional TTL, so that a single power supply, regulated to a tolerance of ± 10%, can be used. No level shifters or logic circuits are required to interface with TTL, DTL, and low-threshold MOS circuitry, thereby reducing system costs and eliminating associated system propagation delays. A system can be graded according to speed requirements by carefully mixing standard TTL, MOS, high-speed TTL, and Schottky TTL circuits to provide optimum cost/performance benefits to the equipment users. TI's series 54S/74S Schottky TTL has impressive specs: power dissipation of 20 milliwatts per gate and a typical propagation delay of 3 ns, similar to the speeds of ECL. This performance is achieved by using Schottky-barrier diode clamps across the collector and base on all normally saturating transistors. In Schottky TTL, the Schottky-barrier diode is employed as a Baker clamp across the base-collector junction of an npn transistor (Fig. a). Since the diode has a lower forward-voltage drop than the base-collector junction, base drive current is diverted through the diode when the transistor is turned on, preventing the transistor from reaching saturation due to elimination of excess charge storage in the base region. As a result, turn-off times are dramatically reduced. For example, the basic 54S/74S gate circuit, (Fig b), achieves 3-ns propagation delay while maintaining rise and fall times in the 2-3-ns range to minimize system noise considerations. All input transistors in the basic Schottky TTL gate schematic are clamped by Schottky diodes instead of pn diodes as in conventional TTL. The lower forward voltage drop of the Schottky units gives greater protection against both negative voltage transients and positive line reflections possible in high-speed digital systems. And since the reverse input characteristic of a Schottky TTL gate more closely approximates an ideal termination, resistive terminals usually are not required for lines of impedances between 50 and 150 ohms. In addition to the Schottky diode clamps, 54S/74S employs other basic design changes to reduce internal circuit propagation delays. Transistor geometries are approximately one-half the area of conventional TTL transistors, resulting in reduced delays due to minimize parasitic capacitive effects. Although Schottky diodes and advanced design geometries are used in Schottky TTL, basic processing steps are very similar to those employed in conventional TTL. Over 2 million device hours have been logged under high-temperature accelerated stress tests to Mil-Std-883 conditions. Failure rates that have been verified at TI have proven to be as good as those logged in for standard TTL devices. Use of Schottky barrier clamps and elimination of gold-doped slice processing in 54S/74S circuits results in stability of ac delays and dc thresholds over operating temperature ranges. Stability of dc thresholds (and thus, noise margins) is important where devices of varying complexities drive each other—when driving an MSI function with a simple gate, for example. Other circuit design techniques minimize switching overlap of the totem-pole output transistor, thus reducing current spikes—about 20% of those in 54H/74H high-speed TTL. Thus, fewer decoupling capacitors need be used with Schottky TTL. And the lower dynamic power dissipation permits increased complexity and use of smaller power supplies—Schottky TTL circuits of up to 80-gate complexity are now possible. The SN74S181 arithmetic logic unit used in the PDP 11/45 computer is a 75-gate chip that can perform 16 arithmetic or logic functions. Even when used with a standard TTL carry/look-ahead package, the SN74S181 provides a 16-bit addition time of 25 ns against 40 ns when using an equivalent conventional TTL circuit. In general, MSI-level Schottky TTL greatly decreases system interconnections (and thus, wiring delays) and reduces on-chip delays. The SN74S153 dual four-input multiplexer achieves a propagation delay of 5.8 ns through two logic levels, or 2.9 ns per level. THE SN74S153 was also used in the PDP 11/45 and was instrumental in achieving the processor speed goals. However, with basic Schottky TTL gates, delays of 4 to 5 ns per logic levels are typical; on-chip MSI delays usually remain between 2.5 and 3.0 ns per logic level. contents. But most importantly, the size and cost of control logic are significantly reduced. Both the central processor and the floating-point processor contain microprocessors with bipolar ROMs. These microprocessors, shown in Fig. 3, handle state sequencing and data-path and bus-control supervision. These microprocessors comprise the microprogram address register, control memory, control memory output buffer, and microprogram address control unit. The microprogram address register is made of D-type flip-flops holding the next address of the control memory to be read (similar to major-state shift registers of conventional machines). The control memory is a $256 \times 64$ masked ROM array. Its output is fed to the control memory output buffer, a 48-bit D-type flip-flop register, and also to the microprogram address control. The control memory output buffer drives multiplexer selectors and provides enabling levels for timing pulses throughout the machine. The control memory output also supplies its own next address and selects the external signals and internal sense points which modify that address. The optional floating-point instructions represent another area where microprogramming techniques and fast, high-density ROMs now permit the user to achieve substantial system improvements at low cost. Floating-point capability in a computer system relieves the user of keeping track of the decimal point which is cumbersome in applications where the dynamic range of the variables is large. It also allows the programmer to use higher-level languages, such as Fortran and Basic, providing faster problem-solving. Using these types of languages, at least sevenfold increase in programmer efficiency can be achieved over assembly language techniques. The floating multiply time is 8 microseconds for 32-bit numbers and 15µs for 64-bit numbers. As a result, more than 20 times as many calculations can be performed with the 11/45 floating-point processor than with programed operations. Multilayer boards, too. Because of the high speed achieved in the PDP 11/45's central processor, both crosstalk and the propagation characteristics of the signals become an important consideration. With Schottky TTL, typical rise times of from 1.5 to 3 nanoseconds made the use of multilayer boards for all high-speed elements the logical choice. Although multilayer technology isn't new, the PDP 11/45 represents one of its first high-volume, low-cost computer applications. However, an extra bonus resulted. Not only did this technique improve the electrical environment, but it resulted in a 20% increase in packaging density. This increased density, in turn, shortened interconnection distances, thus reducing system noise and lowering drive requirements of the individual circuits. These boards are 8.5 inches by 15 inches, and have four layers and 216 edge contacts along one side. One of the two inner layers is used for a ground plane; the other serves to carry power lines. All the logic signals are kept on the two outer layers. The components are placed on one of the outer layers; the other layer is wave-soldered. A standard grid pattern, which can house a maximum of 144 ICs or equivalent components, is used on most boards. This allows the two inner layers to be identical for all of the multilayer boards used in the computer. That way manufacturing costs and production lead time are reduced considerably. --- 3. Microprocessor control. Microprocessor controller supervises state sequencing and data-path bus control. It's analogous to instruction decode and major-state control of conventional machines. The computing power of the 11/45 system is well suited to time-sharing applications. The key to a time-sharing system is management of system resources—perhaps the most important of which is memory. The 11/45 segmentation unit provides a mechanism by which a system supervisory program can control allocation of active storage to each user. This is realized by segmenting each user's program and assigning such segments to specific elements of the storage system. In this way, for example, frequently executed portions of a program can be assigned to the high-speed bipolar memory, while large data arrays are located in a much slower memory (such as core). These procedures implement a software cache system. What's ahead for the 11/45 architecture? One development will be extension of the semiconductor memory system as soon as more cost-effective, higher density, faster components are available. While continued higher performance systems will be implemented, the major direction of future evolution will be toward multiple-processor configurations. A convenient way to model the handy zener diode Unlike earlier computer models, this set copes adequately with the zener diode's thermal effects, uses easy-to-evaluate circuit elements, and is well suited to available linear and nonlinear analysis programs. by Emanuel Schnall, Design Automation Inc., Lexington, Mass. The zener diode is an important and frequently used circuit component, but only a few computer models of it are available for computer-aided design. None of them fully account for thermal effects, and they are often difficult to adapt for circuit analysis programs. Moreover, a special computer program is frequently needed to determine model parameter values. A new set of linear and nonlinear zener models solves these problems. Besides providing for all thermal effects, they can be adapted to all popular linear and nonlinear analysis programs. Their parameters also may be found directly from a data sheet, by graphical analysis, or by measurement. A linear zener model can be regarded as a simplified portion of a nonlinear model. Any single linear representation cannot be used for all zener operating regions and then, of course, it is necessary to modify it to account for the nonlinear aspects of diode performance. In any case, it's easier to start by developing linear models, since this will establish certain circuit elements that can also be used later in a nonlinear model. Zener impedance, for instance, is common to both types of models, as are thermal compensation components, a constant back-voltage, and transition and package capacitances. Once these elements are described by parameters that can be evaluated directly, then the problems of fitting the model to existing computer programs can be tackled. 1. Zener data. Tabulation of typical parameter values for small-signal zener model lists 27 representative diodes. The data can be used to approximate circuit parameters for both linear and nonlinear zener models. Diodes are arranged by power rating and then by voltage rating. further modeling is necessary for the zener. In developing a linear model for the reverse breakdown region, the first thing to consider is zener small-signal or dynamic impedance, $Z_z$. In the breakdown region, $Z_z$ is generally measured with an alternating test current that is a tenth of the dc zener current, $I_Z$. Actual curves of $Z_z$ versus $I_Z$ for a number of zener diodes indicate that $Z_z$ can be expressed as: $$Z_z = R_b + R_d (I_Z) = R_b + R_{dT} \left( \frac{I_{ZT}}{I_Z} \right)^{N_z}$$ \hspace{1cm} (1) where $R_b$ represents contact, lead, and semiconductor bulk resistances; $R_d$ the small-signal, $I_Z$-dependent, dynamic resistance; and $N_z$ is an exponent that permits a best fit to available data. Subscript T refers to a specific test condition, which usually corresponds to a quarter of the power rating. Linear model. Equation 1 provides an accurate fit for $Z_z$ within about 25% for the published impedance curves of 26 of the 27 representative zener diodes listed in Fig. 1. The exception is the type 1N4552 diode, which is modeled with only 65% accuracy. At the other extreme, the type 1N5226 diode has an impedance curve that is accurate within about 1% over three decades of current. As can be seen from the table, $N_z$ ranges from only 0.62 to 1.37, while diode ratings vary from 0.5 watt to 50 w and 3.3 to 110 volts. Typical $R_b$ and $R_{dT}$ values also vary considerably as do zener voltage and power ratings. One reason is the drop in dynamic resistance $R_{dT}$ at about a 6-v breakdown rating, corresponding to the end of the transition from a field-emission (low-voltage) mode to an avalanche-breakdown (high-voltage) mode. By way of illustration, Fig. 2(a) shows a typical dynamic impedance curve between the zener's knee test condition (the point at which avalanche breakdown begins) and the primary test point, while Fig. 2(b) plots the impedance curve using Eq. 1. There are many zener diodes, however, for which impedance curves are not published, but maximum $Z_z$ values at two currents are provided. In many instances, $R_b$ can be neglected or set equal to some typical value. Then, $N_z$ is approximated by: $$N_{zA} = \ln(Z_{zK}/Z_{zT})/\ln(I_{zT}/I_{zK})$$ \hspace{1cm} (2) where the overscored quantities are maximum values, and subscript K indicates the zener knee test point at a current ranging from 0.0033 to 0.33 of $I_{zT}$. Figure 2(c) plots Eq. 2; Fig. 1 lists the values of $N_{zA}$ that approximate $N_z$ within +47% and -26%. Thermal effects. Zener current can vary enough to cause significant junction heating or cooling and so affect voltage. This voltage change can be modeled by adding a thermal-effect resistance, $R_\theta$. $R_\theta$ represents the linear effect of zener current on junction temperature ($T_j$) and breakdown voltage ($V_z$): $$R_\theta = \frac{\partial V_z}{\partial T_j} \frac{\partial T_j}{\partial P} \frac{\partial P}{\partial I_Z} = a_v \theta_{jA} V_z^2$$ \hspace{1cm} (3) where $a_v$ is the fractional temperature coefficient (in percent per C) of breakdown voltage $\theta_{jA}$ the thermal resistance (in °C per watt) from junction to ambient temperature, and P the power dissipation ($P = V_z I_Z$). Maximum $\theta_{jA}$ (or $\bar{\theta}_{jA}$) is equal to the reciprocal of the diode power derating factor. The zener thermal voltage effect has a dynamic lag due to the thermal time constant of the junction. Since only one mode of cooling is usually significant, a simplified single-node thermal model can be used. And the dynamic lag is realized by shunting $R_a$ with a thermal-effect capacitance, $C_o$; $$C_o = \tau_{JA}/R_a = C_{JA}/a_v V_z^2$$ \hspace{1cm} (4) where $C_{JA}$ is the thermal capacitance (in watt-seconds per °C), and $\tau_{JA}$ the thermal time constant ($\tau_{JA} = \theta_{JA} C_{JA}$). When the zener is cooled primarily by convection, the single-node approximation lumps the thermal capacitance of the case and the diode chip together. This produces thermal time constants of about 0.5 second to 50 seconds for axial-lead-mounted diodes, varying with lead length and case design. Alternatively, if cooling is mainly by conduction, the junction-to-case thermal resistance predominates, and the thermal capacitance of only the diode chip applies. Thermal time constants are only about 0.05 to 0.5 second for diodes that are bolted and heat-sunk. Finishing the model. Now that the essential circuit elements for the reverse-breakdown zener are defined, a linear model can be synthesized for this most frequently used mode of zener operation. Combining all elements (bulk resistance, dynamic resistance, and thermal-effect resistance and capacitance) with a battery that represents breakdown back-voltage results in the linear model of Fig. 3(a) for zener reverse-breakdown operation. The fact that it's valid only when current flows in the direction indicated implies that the nominal breakdown voltage, $V_z$, is greater than the breakdown back voltage, $V_b$: $$V_b = V_{ZT} - I_{ZT}(R_b + R_o + R_{dT})$$ \hspace{1cm} (5) Voltage $V_b$ is determined by a specific test voltage and current. In contrast to its performance in the reverse-breakdown mode, the zener behaves similar to an ordinary diode when reverse-biased. For this reason, the elements used for its reverse-biased linear model are the same as those for an ordinary diode. However, the parameters defining these elements can be so selected as to simplify the job of relating the description to the real device. The linear model of Fig. 3(b) synthesizes zener leakage and capacitive currents in the reverse-bias region, below breakdown. Resistor $R_c$ represents the junction ohmic leakage resistance at a given temperature, $C_p$ the capacitance of the diode package, and $C_t$ the voltage-dependent transition capacitance. Transition or depletion-layer capacitance $C_t$ varies with junction voltage, $v$: $$C_t(v) = C_0/(1 - v/V_c)^{N_e} = D_j (V_c - v)^{N_e}$$ \hspace{1cm} (6) where $V_c$ is the diffusion potential (typically 0.7 V for silicon, 0.3 V for germanium), and $N_e$ represents a grading constant, which ranges from 0.33 for a graded junction to 0.5 for an abrupt junction. Constant $C_0$ represents the capacitance at $v = 0$, while constant $D$ is the capacitance at $v = V_c - 1$. Eq. 6 applies for reverse junction voltages smaller than $V_{zK}$, at which $C_t$ rapidly drops towards zero. There are several methods for evaluating parameter values. Even low-accuracy techniques are helpful since they aid in determining expected-value ranges when deciding if and how laboratory testing should be used. 3. Linear models. Circuit (a) synthesizes the zener in reverse breakdown. It consists of $I_z$-dependent dynamic resistance $R_o$, bulk resistance $R_b$, breakdown back-voltage $V_b$, and thermal-effect components $R_a$ and $C_o$. Reverse-biased model (b) includes leakage resistance $R_l$, package capacitance $C_p$, and voltage-dependent transition capacitance $C_t$. Ordinary unit models forward-biased zener. The typical or approximate values, indicated for most model parameters, may be accurate only within 20%, but yield an overall model whose accuracy can be within 5%. In estimating resistance $R_b$, either the values in Fig. 1 or a multi-point data curve can be used. A log-log plot of zener impedance versus $I_z$, like the one shown in Fig. 2(b), yields an asymptote that will fix a value for $R_b$. This same curve also yields exponent $N_e$, though it can also be approximated from the two-point data of Fig. 2(c) or by using Eq. 2. A simple and often acceptable method is to set $N_e$ equal to unity. Then, once $Z_i$, $I_z$, $R_b$, and $N_e$ are known, $R_d$ can be found from Eq. 1. Thermal-effect resistance $R_o$ is computed with Eq. 3, after $a_v$ and $\tau_{JA}$ have been determined from test or design data. Alternatively, thermal resistance $\tau_{JA}$ can be assigned some typical value like 0.7 divided by the power derating factor. Or again, $a_v$ and $\theta_{JA}$ can be estimated from data for diodes with a similar package type, zener voltage, and zener current. Next, leakage resistance $R_l$ can be obtained in two ways—from the average slope of a curve for $v$ versus reverse current ($I_r$), or by dividing a reverse-bias test voltage by the appropriate leakage current. Back-voltage $V_b$ can then be computed directly from Eq. 5. Transition capacitance $C_t$ is measured at various test voltages. Next, exponent $N_e$ can be found from a log-log plot of $C_t$ versus $v$, or assumed to be 0.33 for diffused-junction diodes (0.5 for alloy-junction diodes). Diffusion potential $V_c$ can be solved graphically as the intersection of horizontal and diagonal asymptotes on a log-log plot of $C_t$ versus $-v$, or it may be assumed to be 0.7 V (for silicon at 25 °C). Once $C_t$ is known at some voltage, the value of $D$ can be computed with Eq. 6. Otherwise, approximations derived from handbook data can be used: 5,200/$V_{ZT}^{n_g}$ picofarads for ¼-w diodes, 24,000/$V_{ZT}^{n_g}$ pF for ¾- to 10-w diodes, and 58,000/$V_{ZT}^{n_g}$ pF for 50-w diodes. What's been done for the zener? Since its introduction in the late 1950s, just five computer models have been developed for the zener diode through 1970. The first and still most widely used linear model is a battery and a series resistance. Accuracy was improved somewhat by Chandler$^1$ in 1960. He added a second series resistance to represent the zener self-heating effect. Chandler's addition was extended by Mollinga$^2$ in 1963. A third independent investigation by Rigg$^3$ in 1962 led to another refinement of the linear zener model. Rigg introduced a junction transition capacitance shunted by a junction resistance to represent zener impedance below the diode's knee region. There are only two nonlinear zener models, and both are difficult to program. Daniel's$^4$ model (in 1967) includes nonlinear current sources that realize dynamic, bulk, and leakage resistances, and transition and diffusion capacitances, but does not satisfactorily represent zener self-heating effects. Todd's$^5$ model (in 1970) employs a temperature-dependent, constant-current source for leakage. REFERENCES: 1. J.A. Chandler, "The Characteristics and Applications of Zener (Voltage Reference) Diodes," Electronic Engineering, Vol. 32, February, 1960. 2. T. Mollinga, "Effect of Temperature and Current on Zener Breakdown," Electronics Technology, Vol. 72, October, 1963. 3. B. Rigg, "The Characteristics and Applications of Zener Diodes," Electronic Engineering, Vol. 34, November, 1962. 4. M.F. Daniel, "Development of Mathematical Models of Semiconductor Devices for Computer-Aided Circuit Analysis," IEEE Proc., Vol. 55, April, 1967. 5. C.D. Todd, "Zener and Avalanche Diodes," Wiley, N.Y., 1970. The value of package capacitance $C_p$ can be obtained from a bottom asymptote to the log-log plot of $C_t$ versus $-v$. It can also be measured by using an internally open-circuited packaged zener. Thermal-effect capacitance $C_a$ is computed from Eq. 4, after the thermal time constant $\tau_{1A}$ has been obtained from test data. (It is also possible to use a typical value estimated from a similar package for $\tau_{1A}$.) Analysis problems. Now that linear models for synthesizing the zener in either its reverse-breakdown or reverse-bias region have been described and the ways to evaluate their parameters discussed, it's possible to consider how to use the models in current computer programs. Unfortunately, there are potential problems in any adaptation of a model to a computer program, but fortunately, they are not insoluble. In general, most linear circuit analysis programs, such as ECAP (electronic circuit analysis program) and equivalent time-shared software packages, allow the models of Fig. 3 to be represented by fixed-value resistive and capacitive elements. Naturally, the loss in accuracy that results from using fixed values for $R_d(I_z)$ and $C_t(v)$ depends on the magnitude and importance of the variations omitted. That loss can be minimized by ECAP's transient analysis capability if $R_d$ and $C_t$ are expressed as piecewise-linear variables. A few numerical considerations should be observed for computer analysis. Thermal-effect resistance $R_a$ can have negative values at zener voltages below about 6 v, where temperature coefficient $a_v$ can become negative. This leads to convergence problems in dc analysis if circuit resistance external to the zener is unusually low—for instance, if it is comparable to dynamic resistance $R_d$. The problem can be avoided by reducing the magnitude of $R_a$ to obtain a positive net series resistance for $R_d + R_a + R_n$ in the model of Fig. 3(a). The extremely short time constant of the path through $R_n$ and the two capacitors in Fig. 3(b) could make the time taken by transient analysis excessive. Very short integration time steps and, therefore, long computation times are needed to prevent a numerical error buildup. The difficulty is usually circumvented by placing package capacitance $C_p$ in parallel with transition capacitance $C_t$, since $C_t$ is normally higher in value. The two can then act as a single capacitor. Using linear models for a zener diode, however, will not yield the most accurate results possible because such models are an approximation of the diode's nonlinear performance. But linear models do provide a basis for developing a computer-adaptable nonlinear model—and a major problem with existing zener models is the difficulty in adapting them for computer use. A linear small-signal model is unable to deal with fractional current variations that are large enough to cause significant changes in zener diode voltage, $R_d(I_z)$, or $C_t(v)$. It also cannot handle transitions from one zener operating region to another. These functions require some kind of nonlinearity. An appropriate form for the nonlinearity can be derived from a consideration of the voltage regulation of the linear zener diode model of Fig. 3(a). From linear to nonlinear. Zener voltage regulation is found by integrating dynamic impedance $Z_z$ in Eq. 1 with respect to current. When $N_z \neq 1$, the calculation yields: \[ V_z - V_{ZT} = (R_b + R_d)(I_z - I_{ZT}) \\ + \frac{R_{dT}I_{ZT}}{1 - N_z} \left[ \left( \frac{I_z}{I_{ZT}} \right)^{1-N_z} - 1 \right] \] If $N_z = 1$, then: \[ V_z - V_{ZT} = (R_b + R_d)(I_z - I_{ZT}) + R_{dT}I_{ZT} \ln(I_z/I_{ZT}) \] Equation 7 could cause computational difficulties if $N_z$ is close to unity, because it approaches an indeterminate numerical form. Moreover, the equation is difficult to rewrite into the explicit voltage-dependent current form needed for compatibility with presently available nonlinear circuit analysis programs. It is convenient, therefore, to let $N_z$ be unity and use Eq. 8, which is determinate for all positive values of $I_z$, and which results in an explicit voltage-dependent current source. Both Eq. 7 and Eq. 8 have a voltage regulation coefficient, $R_{dT}I_{ZT}$, associated with the nonlinear logarithmic current term. In Eq. 8, this coefficient gives the effective nonlinear regulation for a current change of 2.718:1. Figure 1 includes typical regulation coefficient values, which range from 0.01 v (for a 5.1-v breakdown rating) to 0.61 v (for a 110-v breakdown rating). Nonlinear model. A basic nonlinear zener diode model that covers all three regions of operation is shown in Fig. 4. The circuit employs an Ebers-Moll model for the forward diode, which consists of $R_b$, $R_v$, $C_t(i_v)$, $C_t(v_v)$, and $i_v(v_v)$. The model also has an exponential current source, $i_z(v_z)$, which represents zener voltage regulation. The dependent Ebers-Moll diode current source, $i_1(v_1)$, can be expressed as: $$i_1(v_1) = I_{s1} \left[ \exp \left( qv_1 / MKT \right) - 1 \right]$$ (9) where $I_{s1}$ is saturation current, $M$ junction emission constant, $k$ Boltzmann's constant, $T$ junction temperature (in K), and $q$ the charge on an electron. $M$ usually ranges between 1 (ideal) and 2, while $KT/q$ is always 25.7 millivolts at 25 C. Diffusion capacitance $C_{d1}$ is: $$C_{d1}(t_1) = q[i_1(v_1) + I_{s1}] / 2\pi F M K T$$ (10) where $F$ is the intrinsic diode cutoff frequency. Back-voltage $V_B$ is defined to obtain zero terminal current at zero terminal voltage, allowing the model to be used for both high and low operating currents. If $v_1 = 0$ at the zener test condition ($V_Z$, $I_{ZT}$) and $i_1(v_1)$ approaches $-I_{s1}$, then $V_B$ can be given by: $$V_B = V_{ZT} - R_u I_{ZT} - R_u I_B$$ (11) where: $$I_B = I_{ZT} - I_{s1} - (V_{ZT} - I_{ZT} R_u) / R_c$$ (12) The value for $V_B$ lies between breakdown back-voltage $V_B$ and zener test voltage $V_{ZT}$. The logarithmic component of zener voltage in Eq. 8 is represented by current source $i_2(v_2)$: $$i_2(v_2) = I_{s2} \left[ \exp \left( v_2 / R_{dT} I_{ZT} \right) - e_n \right]$$ (13) where $I_{s2}$ is: $$I_{s2} = I_{s1} / (1 - e_n)$$ (14) and where $e_n$ is defined as: $$e_n = \exp(-V_B / R_{dT} I_{ZT})$$ (15) to obtain zero diode terminal current at zero terminal voltage when $v_2 = -V_B$ (for accurately modeling zener leakage current or a low operating current). Since regulation coefficient $R_{dT} I_{ZT}$ ranges from 0.01 to 0.61 V for the diodes listed in Fig. 1, the exponent in Eq. 15 varies from $-8.3$ to $-510$. This indicates that typical $e_n$ values are below 0.0003, making $e_n$ significant only at currents lower than those in the zener test current region. Except for $R_d(I_Z)$, all the elements in both linear models of Fig. 3 are explicitly called for in the nonlinear model of Fig. 4. Dynamic resistance $R_d$, however, is implicitly modeled through current source $i_2(v_2)$. Differentiating Eq. 13 with respect to $v_2$ and inverting the result yields: $$dv_2 / di_2 = R_{dT} I_{ZT} (I_{s2} + I_{s1} e_n)$$ This expression represents $R_d$ for $N_z = 1$, in accordance with Eq. 1. Nonlinear analysis. Parameter values for the nonlinear model are obtained in the same fashion as for the linear models. The methods already discussed for finding $R_u$, $R_c$, and $C_p$, and for solving the equations for $R_u$, $C_u$ and $C_p$ are identical. Values of saturation current $I_{s1}$, cutoff frequency $F$ and junction emission constant $M$ are determined by measuring diode leakage current, storage time and forward voltage, respectively. Regulation coefficient $R_{dT} I_{ZT}$ can be estimated from data like that of Fig. 1. The remaining parameters of diffusion capacitance $C_{d1}(t_1)$, back-voltage $V_B$, and current source $i_2(v_2)$ are computed using Eqs. 10 through 15, once all necessary variables are known. Again, certain computer numerical analysis precautions must be observed. Those previously mentioned for $R_u$ and $C_p$ also apply to the nonlinear model. Forward voltages approaching diffusion potential $V_c$ can lead to numerically excessive values for transition capacitance $C_t$ in Eq. 6. This may require a restriction that limits the magnitude of $C_t$ or a convenient larger value of $V_c$; the first modification is preferable. If NET-1 (network analysis program) is used, the magnitude of $C_t$ is automatically limited. But floating voltage source $V_B$ requires a series element, which should be a small resistor approximately equal to 0.001 $R_u$. The nonlinear model of Fig. 4 makes it possible to save programming time since the dependent exponential currents, $i_1(v_1)$ of Eq. 9 and $i_2(v_2)$ of Eq. 13, have the same general mathematical form: $$i = I [ \exp(a) - b ]$$ where $I = I_{s1}$, $a = qv_1 / MKT$, and $b = 1$ when $i = i_1(v_1)$, and where $I = I_{s2}$, $a = v_2 / R_{dT} I_{ZT}$, and $b = e_n$ when $i = i_2(v_2)$. On the other hand, the model may be difficult to represent with some circuit analysis programs. NET-1 and Circus (circuit simulator), for example, must include both exponentially dependent current sources as part of the diode model. Since the sources are connected to elements that cannot be eliminated from the network, an internal computation loop is formed with very short time constants. This, in turn, results in excessive computation time for transient analysis because extremely small integration time steps are required. Sceptre (system for circuit evaluation and prediction of transient radiation effects), however, provides ideal diode current sources that implement Eq. 9 and that can be used without an $R_u$ element. Then $C_t$ and $C_p$ shunt $i_1(v_1)$, thus avoiding computational delay errors in Sceptre's state-variable transient analysis, without introducing a short time constant. Dc analysis usually presents no problems; modeling compromises are generally necessary only for efficient transient analysis. In many applications, the forward-biased portion of the nonlinear zener model can be eliminated to ease the short-time constant problem for transient analysis with NET-1 and Circus. The variables $C_{d1}(v_1)$ and $i_1(v_1)$ are simply omitted from the model. A better way to maintain oscillator specifications. A lot of oscillators can’t meet their own specs once in operation. That’s why we came up with a better way. A die cast zinc enclosure to keep specs what they should be. No other oscillator offers our kind of enclosure. Or our kind of performance: Models in 1MHz, 3MHz, and 5MHz ranges. And when you write for our spec sheets, what you read is what you get. Write to Motorola Component Products Dept., 4545 W. Augusta Blvd., Chicago, Ill. 60651. One-of-eight decoders test and correct parity by Michael J. Gordon, Jr. Psynexus Systems, Wilmette, Ill. Only four one-of-eight decoder packages are all it takes to build an eight-bit parity checker that also can be used for parity correction. Either odd or even parity testing can be done. The basic circuit can be adapted to suit almost any logic family; complete data is given for TTL and MOS packages. When wired as indicated, the decoders produce a logic 0 output for correct parity, a logic 1 output for incorrect parity. Suppose the input word is 10111001, and it is to be checked for even parity. Since there is an odd number of 1s, a parity error should be indicated. The three least-significant bits become the $I_3$, $I_4$ and $I_5$ inputs. The next three bits, going from right to left, are inputs $I_6$, $I_7$ and $I_8$. The last two bits become $I_9$ and $I_{10}$. Because decoder $D_1$ receives only a single 1, its output is 1. The input to $D_2$ contains three 1s, causing another parity error. Decoder $D_3$ also produces a 1, since its input contains just one 1. Three 1s, therefore, are applied to $D_4$, resulting in a 1 output, indicating a parity error. If the input word has correct, even parity, the output will be a 0. Let the word be 10010110. Now $D_1$ produces a 0, $D_2$ a 1, and $D_3$ a 1. Since the input to $D_4$ contains an even number of 1s, the output is 0, indicating correct parity. Using the output as the parity bit corrects any parity error. If the circuit is wired for even parity, there will always be an even number of 1s. When wired for odd parity, the circuit will always supply an odd number of 1s. For example, if the seven-bit word 1101101 is checked for even parity, the circuit’s 1 output can become the eighth bit to correct the parity error. Of course, whenever word parity is correct, the circuit produces a 0, which does not change parity. The outputs of the decoders shown in the diagram are wire-ORed together. If the decoder package used does not have this feature, an OR gate must be added to its output. Testing, testing. Eight-bit parity checker generates a logic 0 for correct parity, a logic 1 for incorrect parity. Decoders $D_1$, $D_2$ and $D_3$ accept input data word and supply three-bit input for output decoder $D_4$. Circuit can test for either odd or even parity, depending on how it is wired. An OR gate must be added to output of each decoder if the type of logic employed cannot be wire-ORed. \| DECODER \| LOGIC \| PARITY \| PACKAGE PIN NUMBERS \| \|---------------\|-------\|--------\|---------------------\| \| MOTOROLA MC4038 P \| TTL \| ODD \| A B C D E F G \| \| \| \| EVEN \| 13 2 4 11 15 6 1 \| \| \| \| ODD \| 3 5 12 14 15 6 1 \| \| \| \| EVEN \| 2 5 12 13 15 9 7 \| \| MOTOROLA MC1150 L \| MOS \| ODD \| 4 3 11 14 15 9 7 \| \| \| \| EVEN \| \| Electronics/October 11, 1971 With feedback, isolation amp gives better-than-unity gain by Roland J. Turner RCA Corp., Missile and Surface Radar division Moorestown, N.J. An isolation amplifier with degenerative feedback provides gain over a large bandwidth, yet keeps output impedance low. An amplifier set up this way can be an excellent driver for long, low-impedance transmission lines or other highly capacitive loads. The circuit can even be used as a gate when its dual MOSFET is driven to cutoff. (Conventional isolation amplifiers, like emitter-followers, are usually unity-gain circuits.) Moreover, gain and output impedance in the improved amplifier are insensitive to variations in the MOSFET's transconductance ($g_m$) and the bipolar transistor's beta ($\beta$). Normally, $g_m$ drops by a factor of three as temperature rises from room conditions to 60°C, while $\beta$ increases with rising temperature. In this amplifier, the changes in $g_m$ and $\beta$ can offset each other. Gain is 17 decibels from several kilohertz to 20 megahertz; output impedance is less than 10 ohms. Furthermore, isolation greater than 50 dB can be achieved at frequencies as high as 6 MHz. The circuit's input can be operated either matched or unmatched. As can be seen from the diagram, MOSFET $Q_1$ has two control gates. When the bias ($V_{G_{12S}}$) of the second control gate ($G_2$) is decreased from 4 to -3 V dc, the forward gain through gate $G_1$ can drop by 50 dB, without adversely affecting its input characteristics. As $V_{G_{12S}}$ becomes negative, $Q_1$ is driven into pinchoff, reducing its drain current. Transconductance through the signal gate then drops from 15 millimhos to 50 micromhos. The voltage transfer function for the amplifier is: $$\frac{e_o}{e_i} = \frac{g_m[\beta R_1 + (\beta + 1)R_2]}{1 + g_m(\beta + 1)R_2}$$ which indicates that amplifier gain is directly proportional to $g_m$. If $g_m(\beta + 1)R_2$ is greater than 1, $e_o/e_i$ reduces to: $$\frac{e_o}{e_i} \cong \frac{1}{1 + 1/g_m(\beta + 1)R_2} + \frac{R_1}{R_2}$$ and circuit output impedance becomes: $$Z_o = \frac{1}{g_m(\beta + 1) + 1/R_2} + \frac{R_1}{\beta + 1}$$ which equals $R_1/\beta$ for frequencies to 10 MHz. As $Q_1$ approaches pinchoff, its drain current drops, thereby driving $Q_2$ towards cutoff. Since $Z_o$ increases as $\beta$ decreases, amplifier gain is effectively reduced by mismatch. In the limit, as both $g_m$ and $\beta$ approach 0: - $e_o/e_i$ approaches $1 + R_1/R_2$, and - $Z_o$ approaches $R_1 + R_2$. The input transmission-line transformer is similar to a Z-match type 50-200E. If the secondary of the transformer is terminated with 200 ohms, a low input VSWR can also be realized. Total parts cost is less than $5. Line driver. Isolation amplifier makes ideal line driver. It supplies more than unity gain while holding output impedance under 10 ohms through frequencies as high as 20 megahertz. Degenerative feedback through $Q_1$ makes circuit performance insensitive to device parameter changes. Bias of $Q_1$'s control gate $G_2$ effects gating action without harming input characteristics of signal control gate $G_1$. Diode and SCR protect multiple-voltage equipment by Peter T. Uhler Midwest City, Okla. Placing a diode between two different power supply lines permits one low-power SCR to control both supplies, thus protecting components from an overvoltage condition. Often, several supply voltages must be used to power a single piece of equipment, creating a hazard for those circuits that operate from a low voltage. If critical components fail, or if two or more supplies are accidentally shorted, these low-voltage circuits are usually destroyed. Multiple-voltage protection is quite important for equipment containing integrated circuits, since ICs generally operate from low-voltage supply lines. The circuit shown is intended primarily for overvoltage protection—overcurrent protection is secondary in this case. The voltage at point A or B determines when the SCR fires. Transistors $Q_1$ and $Q_2$ act as current-sensing elements. Resistor $R_1$ and diode $D_1$ trigger $Q_1$ at around 250 milliamperes, as do $R_2$ and $D_2$ for transistor $Q_2$. The SCR requires only about 5 microamperes to trigger it. In the event of a fault (when the output currents of $Q_1$ and $Q_2$ exceed 250 mA), the SCR fires, turning transistor $Q_3$ off. Diode $D_4$ then turns off transistor $Q_4$, thereby permitting two supplies to be controlled by one SCR. More than two supplies can be controlled by a simple extension of this basic technique. Diodes $D_1$, $D_2$, and $D_3$ assure that transistors $Q_1$ and $Q_2$ are turned off completely. Capacitor $C_1$ at point A provides a time delay of about 0.1 second so that initial load-current surges will not trigger the SCR. Should the 15-volt supply rise about 16.4 V, transistor $Q_5$ fires the SCR immediately since there is no capacitor at point B. Diode $D_7$ protects $Q_5$ if the 15-V supply becomes shorted. Premature triggering of the SCR due to its differential voltage ($dv/dt$) effect is prevented by resistor $R_3$. Switch $S_1$, which should be part of the power control switch, resets the SCR after the supplies are turned off. Approximate parts cost for the protection portion of the circuit shown is $10. Double protection. Overvoltage protection circuit monitors 15- and 30-volt supply lines at the same time. Because supplies are interconnected with diode $D_3$, only one SCR is needed. If output currents of $Q_1$ and $Q_2$ exceed 250 milliamperes, SCR fires, turning $Q_3$ off, while $D_4$ turns $Q_4$ off. $Q_5$ can also trigger SCR if 15-V supply reaches 16.4 V. Capacitor $C_1$ prevents transient current overloads from firing SCR. Dual op amp comparator controls ramp reference by Ronnie Jack McKinley Duncan, Oklahoma A control comparator can be used to establish the end points of a sawtooth generator's ramp output. That way, the sawtooth's end points don't drift with temperature, providing a stable output. The ramp, which travels from absolute zero potential to 5 volts, can be varied over a frequency range of 0.33 hertz to 1 kilohertz. Constant-current source $Q_1$ linearly charges capacitor $C_1$, whose voltage is buffered by unity-gain follower $A_1$. Comparator $A_2$ is formed by two operational amplifiers operating at open-loop gain; a flip-flop is realized from NOR gates $G_1$ and $G_2$. Diodes $D_1$ and $D_2$ prevent gate inputs from becoming more negative than -0.5 V, while the 4.7- and 3.3-kilohm voltage dividers stop the gate inputs from exceeding 5 V. The comparator sets up two threshold reference points—one at absolute zero potential, the other at 5 V. When the voltage across $C_1$ reaches 5 V, $A_2$ switches, causing the flip-flop to produce a negative pulse. This turns $Q_2$ off and $Q_3$ on so that $C_1$ discharges. Since $Q_3$'s emitter is biased at -2 V, the discharging capacitor tries to reach $V_{SAT}$ -2 V. However, when the voltage across $C_1$ becomes zero, the comparator switches, returning the flip-flop to its original state. Transistor $Q_2$ now conducts, $Q_3$ goes off, $C_1$ begins charging again from a stable zero point, and the cycle repeats. Because of the comparator reference, the baseline of the ramp cannot be shifted by any temperature variance in the $V_{SAT}$ of $Q_3$. Supplying temperature-stable output. Dual comparator in this sawtooth generator sets lower and upper end points of ramp. Constant current from $Q_1$ charges $C_1$ until the capacitor's voltage becomes 5 V. Comparator $A_2$ then switches, toggling flip-flop formed by $G_1$ and $G_2$. $Q_2$ turns off, $Q_3$ turns on, and $C_1$ discharges until its voltage is zero. Then $A_2$ switches again, triggering the flip-flop, which turns $Q_3$ on and $Q_3$ off. Philips offers a range of minicomputers, backed by the largest sales and service network of any electronics company in the world. OEM's will benefit from large volume Philips production, that gives prices as low as $1,800 for the P850, to quantity purchasers. The P850, our smallest mini, has a 16-bit word, 3.2 microsecond cycle time 512 word memory (expandable). Punched tape I/O, teleprinter, card reader, display unit and magnetic cassette tape units are all available. Transfer rates are up to 50,000 words per second. Special options include interrupts, power failure/automatic re-start and a real time clock. Software includes a mathematical library, relocatable loader and restricted assembler — impressive for such a small machine. If you'd like to know more about getting a little Philips into your system, get in touch with OEM Marketing in: Europe: N.V. Philips-Electrologica P.O.Box 553, Apeldoorn, Netherlands North America: North American Philips Corp. Dept. 007 100 East 42nd Street, New York NY 10017 Far East: Philips Industrial Development and Consultant Co. Ltd. Kokusai Building, (7th floor) 1-1 Marunouchi 3-chome, Chiyoda-Ku, Tokyo 100, Japan PHILIPS Philips - Sales and Service worldwide Impatt diodes and millimeter-wave applications grow up together As millimeter-wave communications, alarm, and landing systems proliferate, silicon Impatt diodes, with their impressive flexibility, power output, and efficiency, become more attractive as solid state sources. by N.B. Kramer, Hughes Research Laboratories, Torrance, Calif. The applications for systems using millimeter-wave frequencies are growing, and so is the attractiveness of silicon Impatt diodes as solid state sources. Recent development work has done much to enhance the advantages of the diodes. Among those advantages: Impatts give the highest millimeter-wave cw power of any solid state device—as high as 150 milliwatts at 100 gigahertz. Moreover, Impatts are many times more efficient than LSA diodes and are at least as efficient as the best Gunn devices. And Impatts are vastly smaller, lighter, and simpler than the old varactor harmonic generators. Flexibility is another important plus for Impatts. They can be operated efficiently as stable amplifiers and already are providing serious competition to traveling wave tubes at the higher millimeter-wave frequencies. Small, simple, wideband Impatt sweep generators are becoming standard throughout the millimeter transmission band. Parametric amplifier pumps using stabilized Impatt oscillators are being designed into a number of new systems and appear to provide paramp noise figures as good as conventional pump sources. Millimeter-wave communications systems, such as Bell Laboratories' trial transmission unit, will use silicon Impatts for transmitters and local oscillators. And other potential uses include small doppler radar for intrusion alarms, aircraft landing systems, and collision avoidance networks. Many new applications will be explored as experience with Impatt devices increases. A silicon Impatt's electrical performance is impressive. At millimeter wave frequencies, Impatt diodes operate at less than 30 volts bias with bias currents in the 100–300-milliampere range. Conventional single-drift structures regularly give power output levels in the 100-milliwatt range at efficiencies of a few percent. And recently developed configurations, such as the double-drift diode, promise significant increases: power output approaching the 1-watt level with 10% efficiency will be possible under operating conditions. One problem with Impatts has been the high noise encountered in avalanche devices. But in many applications, such as power amplifiers, noise is not important; in others, such as paramp pumps, problems due to noise can be eliminated by using a stabilizing cavity. In fact, extensive investigations have shown that Impatt sources make attractive pumps for low-noise, wideband paramps for application in radar and communications systems. State-of-the-art Impatts. The first practical Impatts were silicon pn junction diodes for computer switching and these form the basis of the present cw Impatt structures. Operating in X band, the early devices were planar or mesa diodes with junction areas of about $10^{-4}$ 1. Old ways. Showing the way to today's Impatts were these old mesa diode structures. Used primarily for computer switching, these devices relied on close bond between surface adjacent to junction and copper heat sink to provide a good heat transfer. 2. New ways. Modern Impatt design eliminates excess high resistivity material by forming epi-layer; Punch-through occurs only at avalanche. Less material also reduces rf series resistance and lowers the overall thermal resistance to the back contact. 3. Lab pack. This typical stand-off package exhibits low loss because stand-off volume can be made small. It can also be tailored for low parasitics but the drawback is that it cannot be sealed completely immediately around the actual diode area. 4. Tough contender. More suitable for industrial applications, this hermetically sealed package is good at the lower millimeter-wave frequencies; however, it has higher rf loss and less flexible parasitic adjustment than does the stand-off package. cm$^2$, compared to the $10^{-5}$ cm$^2$ areas attainable now for mm wave devices. In addition to a geometry change, the new devices have been greatly refined to minimize thermal resistance, resulting in higher power capability. Careful attention also has been given to precise control of the doping profile to minimize rf series resistance. Then scaling of the basic diode design to operate at higher-frequency ranges is easily accomplished and performance agrees with simple lower-frequency models. The great improvement in Impatt performance levels over the last five years has been closely related to reduction in diode volume. Early cw diodes were upside-down-diffused mesa structures that relied on the intimate bond between the surface adjacent to the pn junction and a copper heat sink for good heat transfer (Fig. 1). A modern diode is represented in Fig. 2. Excess high-resistivity material is eliminated by tailoring the epitaxial layer so that the depletion layer just punches through to the substrate at avalanche breakdown. Most of the excess substrate material also has been removed to reduce rf series resistance and lower the thermal resistance to the back contact. New silicon processing techniques, as well as careful application of some old techniques, work together to yield the high performance of Impatt devices: epitaxial layers grown at low temperatures (below 1,000 C) have given the best results. Non-abrupt doping gradient at the nn' interface, caused by outdiffusion during conventional high-temperature epitaxial growth, resulted in lower efficiency. More recently, ion implantation has been utilized successfully for junction formation. The high temperatures associated with conventional diffusion processes are not required in the ion implantation process; therefore, abrupt doping transitions at the p'n and nn' junctions can be maintained. Also, the extremely shallow junctions that can be made with ion implantation result in minimum thermal impedance. Complementing ion-implantation methods, a low-temperature diffusion process has been developed that provides equally good junctions. Wafer thinning and special bonding techniques also have contributed significantly to improved performance at millimeter wavelengths. Ion implantation and low-temperature diffusion will play an increasingly important role in silicon Impatt diode fabrication. In addition to the usefulness of implantation in forming highly controllable shallow p' regions, it is an indispensable tool in forming the new p'pnn' (double-drift) Impatt structure for millimeter-wave operation. The double-drift diode is more efficient because it utilizes both carrier species (electrons and holes) created in the avalanche process. Moreover, the diode's active region is thicker and junction area can be increased while maintaining sufficiently high diode impedance levels. The combination of increased efficiency and larger active volumes result in significantly improved power output capability (it's theoretically quadrupled). Ion implantation is used to build this new type of diode structure by employing the range-energy relationship to tailor the diode's p' region. Packaging of diodes for millimeter-wave applications has always been a problem. Reduction of package parasitics is a must because of the high frequencies and low diode impedance levels. Furthermore, millimeter waveguide dimensions are very small, putting size and tolerance constraints on diode packages. Rf loss increases with increasing frequency, so only low-loss materials can be used in package design. The additional requirement of minimum thermal impedance for power devices like Impatts makes package design critically important for efficient rf performance. A package that is convenient for laboratory experiments is shown in Fig. 3. A small metalized quartz standoff is mounted near the diode and a ribbon bond is made from the standoff to the back of the diode. The standoff's small volume gives this type of package its low loss. And parasitics can be easily tailored—ribbon inductance and shunt capacitance can be varied by changing standoff dimensions and position. However, this lab package is completely open, preImpatt power Avalanche diodes operating in the Impatt (impact avalanche transit time) mode develop their considerable power output when carriers at a pn junction build up due to avalanche and then move across the structure's drift region. A simple three-layer (single drift) pnn device is shown in Fig. a. To function as an amplifier or oscillator, the device must deliver power to its output circuit—it must have a negative-resistance characteristic. In the Impatt mode, negative resistance develops as the result of a phase delay of drifting carriers. This is due to both avalanche generation and transit time delay. In avalanche, internal secondary emission occurs at a pn junction that has been reverse-biased into breakdown. At high fields (several hundred kilovolts per centimeter), carriers will acquire enough energy to knock valence electrons into the conduction band, producing hole-electron pairs. These new carriers cause further generation, until a critical field is reached and avalanche occurs. Under steady state conditions the maximum field across the junction will be limited to the avalanche or critical field. But under transient conditions, if the field is moved rapidly from below critical level to above it, and then below it again, the current will still be increasing when the field has passed its maximum; in effect, a phase delay will be introduced. This is the delay caused by avalanche buildup. The current-voltage phase shift can be as great as 90° under small-signal conditions. This, combined with transit time delay (between 90° and 270°) allows the diode to exhibit the required negative resistance over a microwave frequency band. However, a double-drift-space (2d) Impatt diode gets its greater power from the additional drift region implanted onto the simpler structure. It has four layers (Fig. b) instead of the usual three, and this double-drift region, together with the resulting bigger diodes, is responsible for the power increase over the single-drift structures. Thus, in a double-drift diode, (a p'pnn' structure for example), in addition to the drifting electrons on a p'nn' diode, holes drift across the added p region in phase with the electrons, resulting in greater power outputs. —Laurence Altman 5. In tune. Typical millimeter-waveguide cavity structure has considerable mass for heat sinking but mass can be reduced if external sinking is available. Mechanical tuning in the structure is accomplished through a sliding-short technique. 6. Run for the money. Impatt packages such as this one are becoming strong competitors to klystron oscillators. This unit, cheaper than comparable klystrons, gives continuous tuning over a 30 to 40 gigahertz range with a minimum output of 50 milliwatts. cluding hermetic sealing immediately around the diode. In addition, the small standoff makes the package delicate; its use is justified only in systems where rf power output and efficiency are of paramount importance. A rugged, hermetically sealable package is shown in Fig. 4. This type can be very useful at the lower millimeter-wave frequencies even though rf loss is higher than in the lab version and tailoring of the parasitics less flexible. The package also can be made with a ceramic body for added strength. Due to the high power densities (2 x 10^3 w/cm^2) required for efficient Impatt operation at millimeter wavelengths (corresponding to junction temperatures of approximately 200°), some attention must be given to 7. Widely used. Paramp pumps made with Impatts are being adapted for radar systems. These pumps are smaller, lighter, and less complex than varactor chains they replace. 8. And back at the lab. Millimeter-wave sweep generator has Impatt oscillators for laboratory testing. Bias-current tuned, the head is fixed by flexible cable to the power supply. Diode life. Certainly, silicon Impatts will not be attractive as millimeter-wave power sources if the junction temperature must be limited to values equivalent to conventional transistor and diode ratings. Though extensive life data is not available, experience in limited life testing and in operational systems has revealed no fundamental failure mechanisms that would inherently limit longevity at these power densities and junction temperatures. Care must be taken in diode design and fabrication to avoid metalization or mechanical weaknesses that might be aggravated by high-temperature operation. For optimum efficiency, cavity design must be closely related to diode design. Waveguide circuitry is used almost exclusively at millimeter wavelengths and waveguide cavities for Impatt diodes generally have proven highly versatile and flexible. This has resulted in similarity of physical size and configuration among millimeter-wave Impatt diode mounts. A schematic of a typical cavity with bias connector and tuning elements is shown in Fig. 5. At high frequencies and power outputs the cavity mass must be large to assure adequate heat sinking. Where external heat sinking, such as a cold plate, is available, cavity volumes can be reduced significantly. Applications. With the appropriate cavity design, mechanically tunable cw oscillators using Impatt diodes will be replacing klystrons at millimeter-wave frequencies. An oscillator capable of continuous tuning over a 10-GHz band (30 to 40 GHz) with 50 mw minimum output is shown in Fig. 6. This unit is cheaper than a klystron of comparable performance, requires a simpler power supply, and is easier to tune. The oscillator also has an electronic tuning sensitivity of 2 MHz/ma, making it useful for phase-lock jobs. A millimeter-wave Impatt oscillator used with a stabilizing cavity and an appropriate leveling circuit makes an efficient solid state pump source for low-noise parametric amplifiers. A block diagram of such a paramp pump, being designed into several wideband radar and communications systems, is shown in Fig. 7. A complete Impatt pump is much smaller, lighter, and less complex than a varactor multiplier chain providing equivalent paramp performance. Stable circulator-coupled reflection amplifiers employing Impatts have been successfully demonstrated at millimeter-wave frequencies. At V band, small-signal gain-bandwidth products of 15 GHz have been measured with power outputs exceeding 100 mw obtainable at 8-db power gain. Comparable performance has also been observed in the 30–40-GHz range. This type of amplifier will be useful for low-level transmitters and for TWT preamplifiers. Injection-locked oscillators for power amplification of wideband fm and phase-modulated millimeter-wave signals also have been demonstrated. Impatt oscillators have a good locking figure of merit (product of fractional locking bandwidth and voltage gain) and can be used to efficiently amplify angle-modulated signals. The advantage here is that unwanted amplitude modulation is stripped from the signal due to the limiting characteristic of the locked oscillator. A new millimeter-wave sweep generator for laboratory instrumentation has been developed using a bias-current tuned Impatt oscillator; a photograph of the complete unit is shown in Fig. 8. The rf head is connected by flexible cable to the power supply, permitting the sweeper to be conveniently connected at any location in a test system. The Impatt sweeper is more compact and less complex than a comparable backward-wave oscillator sweeper. Sweep bandwidths up to 15 GHz can be obtained with power output of greater than 1 mw over the band. The power output variations with frequency are smooth when suitable isolation or padding is used. The wideband bias-current tuning capability of this oscillator should make it attractive for other electronic tuning applications. Doppler radar is another type of millimeter-wave system that should find wide use with the availability of simple solid state sources. Demonstration units have been built in the 50-GHz range that employ a single-drift Impatt diode for the transmitter/local oscillator in a homodyne configuration. This type of radar can be small, inexpensive, and portable. Range over 100 yards can be easily obtained in an inexpensive unit. With a good-quality mixer and a high-power Impatt, range could be extended to 1,000 yards or more. Foreseeable applications include such products as intrusion alarms, and such automotive tasks as braking radars and speedometers. Digital data links deserve a bit error rate detector Hardware is replacing software for the job of picking up errors in digital data transmission; the equipment is based on the same mathematical principles as are used in monitoring analog networks. by Kingsley P. Roby, Data-Control Systems Inc., Danbury, Conn. Since the early days of digital data transmission and processing, there have been software programs for bit error correction. But as the use of data links spreads, the need arises for simple, inexpensive hardware to monitor digital equipment. Recently the surge in digital transmission has made the need more urgent. The tool that has been developed for this job is the bit error rate detector. The technique is basically the same as that used for the evaluation of analog data networks, where the input characteristics of frequency versus phase/amplitude are compared with those of the output, except that for digital networks an equivalent digital mathematical model is used. With digital systems, however, distortion of the signal is not necessarily cumulative. If signal distortion exceeds the system design threshold, the information is totally lost; if it is less than the threshold, the information can be totally recovered. A bit error rate detector simply compares a pattern that has passed through a digital device with an identical pattern generated by a reference code generator, as the block diagram in Fig. 1(a) shows. The comparison is made on a bit-for-bit basis. If the two states disagree, an error is counted, and displayed as a function of time or as bit errors per n bits transmitted. The detector is as versatile as it is easy to use. It can check performance in any digital line—telephone lines, magnetic tape recorders, data modems, or pulse code modulated telemetry links. About a dozen different versions of the device are being manufactured, and some can be used to test at rates up to 400 megahertz. In almost all cases, the equipment to be tested must be taken out of operation. The detector generates a known digital pattern, and inputs it to the digital device. Because an independent code is generated at the receiver, the delay through the system does not have to be compensated for. But it is necessary to synchronize the reference code generator to the incoming data. When a link does not transmit its own clock, or where the received data includes much jitter or noise, a bit synchronizer is needed. This accepts the data, and recreates a clean data stream and an accompanying clock by using the data transitions (ones and zeros) to synchronize a phase-locked, voltage-controlled oscillator with the incoming data. The oscillator then functions as the clock. In the absence of transitions, the VCO will "coast," or mark time, and still be synchronous at the end of the coasting period. To check the bit synchronizer itself, jitter is simulated by modulating a VCC that acts as a bit rate oscillator and controls jitter amplitude and frequency. Noise is summed with the code generator output, and the composite signal (jitter plus noise) is applied to the bit synchronizer input, as in Fig. 1(b). The synchronizer's performance is verified by comparing its bit error rate with curves of bit error rate versus signal/noise ratio. There are three basic types of digital communication links that can be tested: simplex links, where data flows in one direction only; duplex links, where there is simultaneous two-way flow; and half-duplex links, where data flows in one direction some of the time and in the opposite direction the rest of the time. The basic configurations are shown in Fig. 2. Simplex testing requires two bit error rate detectors, one at the data source to act as the transmitter and one at the data sink to act as the receiver. Duplex testing also usually requires two error rate detectors, the pattern generator output of one unit being sent to the error rate detector of the other unit. 1. What BERDs are and do. Bit error rate detectors are used to test digital links by comparing a known pattern transmitted from a data source with an identical pattern simultaneously generated at the data output (a). By adding a noise generator and a jitter oscillator, a BERD also may test the ability of the bit synchronizer to recreate clean data stream and the accompanying clock (b). The arrangement for testing a half-duplex loop is similar to duplex, except that the link can only be used in one direction at a time. In this case, each error rate detector is used as a pattern generator when testing in one direction, and as an error rate counter when testing in the other direction. In doing this, the pattern generator is independent of the error rate detector, since data may be sent at different rates in either direction. It is possible to measure error rate on a duplex or bidirectional simplex communication link with one detector, but only when the data link can be looped back to the source. In this case, if errors are incurred randomly throughout the link, the actual bit error rate is half that recorded. When no clock is available, it may be necessary to include a bit synchronizer at the receiving end to simulate the clock function. If there is no jitter or frequency variation, the clock used in the pattern generator to the bit error rate counter can be bypassed. This method is useful particularly with hardline links which can be looped back to the source. The type and length of digital pattern determines the extent that the bit error rate measurement includes a sufficient variety of pattern combinations. A simple pattern—a series of marks, a series of spaces, or alternating marks and spaces—suffices for a primitive test setup. This type of pattern is easy to generate and to synchronize, but provides only limited information about system performance, and is effective only for low-speed data systems. Another type of pattern generator, intended for systems with a fixed word length, is a fixed-length, switchable, repeating word generated by a presettable shift register, as shown in Fig. 3(a). This is useful where a data sink, such as a teletypewriter, can interpret patterns. The switched shift registers may be replaced by a programmed sequence which would step through each combination of $ n $ bits. When an 8-bit pattern is used and placed within proper framing, the pattern generator provides signaling to the teletypewriter, which then types out a particular sequence of symbols, such as "The quick brown fox, etc." Bit error rate is then determined by the number of incorrect symbols. The most frequently used pattern generator is the pseudo-random noise, or shift register, generator, shown in Fig. 3(b). It combines the output of some of the shift register stages to produce a repeating pattern of a particular length. One of the primary advantages of this arrangement is that the output is a long, random sequence which contains all the combinations of 1s and 0s characteristic of normal data transmission. A fairly continuous spectrum from dc to the bit rate is obtained by using a maximum pattern length, which is usually $ 2^n - 1 $ bits long. The maximum number of 1s in a row is $ n $ and the maximum number of 0s is $ n - 1 $. (Observing the code $ n $ bits at a time, that is, bit 1 through $ n $, then 2 through $ n + 1 $, will reveal all possible combinations of logical 1s and 0s in $ n $ bits.) The longer and more complex the pattern, the greater the spectral density, or number of pattern combinations, that can be analyzed. There are several standard interfaces used to adapt bit error rate detectors to various digital systems. Usually, the test equipment interface is determined by some common logic family, such as TTL, where 0 to 0.8 volt indicates one state, and 2 to 5.5 v indicates the other. But bipolar and neutral transmission are still used in some teleprinter systems. The neutral mode uses a pulse current for a 1 and the absence of current for a 0. The bipolar mode is similar, except that the polarity of the current is used to distinguish 1s from 0s. The EIA standard interface is designated RS-232. The transmission is basically a polar signal ±5 to 15 v in amplitude where a 1 is negative and 0 positive. An interface similar to RS-232 used by the military is Mil Std 188. The logic levels are polar ±6 v with the positive voltage indicating a 1. In addition to interfacing voltage level, code formats must also be considered. The seven Inter-Range Instrumentation Group (IRIG) codes are the standard of the aerospace industry (Fig. 4). The RZ (return-to-zero) code is used when a dc response is not required. Where the absolute polarity of the logic does not have to be maintained, the mark and space codes have the advantage. This is because a logical 1 (in the case of a mark code) is indicated by the change of state and not the absolute level as in the case of NRZ-L (non-return-to-zero level) code. A biphasic code requires a higher frequency response for a given data rate, but due to its high transition density provides improved performance in the absence of excessive jitter. For this reason, it is often used with tape recorders. Delay modulation, or DM, was developed to --- 2. Digital link trio. In simplex transmission, data flows in one direction (a); in duplex and half-duplex, two BERDS are needed because data flows in both directions (b); however, one BERD may be used (c) if the transmission line can be looped back to the source. 3. Generating patterns. Test patterns for a BERD can be pre-set into a shift register (a) so that it repeats each n bits. The output of the bit rate oscillator provides the clock for the shift register. To generate a variety of more complicated patterns, the feedback points can be changed and more flip-flops may be added to the shift register (b). provide maximum transition density with minimum frequency requirement. DM has a transition in the middle of the bit period for a 1 and a transition at the beginning of a bit period for a 0, except that when a 0 follows a 1 there is no transition. DM can provide packing densities of up to 20 kilobits per inch on magnetic tape. Several highly specialized codes have also been developed. One is the two-tone coding used over telephone lines for long-distance calls: the flurry of beeping heard after dialing is a series of tone pairs representing numbers which transfer billing information into the telephone company office. A different technique is used by the telephone company to transmit 1.5-MHz synchronous data over pairs of wires. It's done in the T-1 PCM transmission system with a bipolar format in which there's a pulse for each 1, and no pulse for a 0. Each successive pulse is of opposite polarity. Every 6,000 feet, a repeater detects, resynchronizes, and reamplifies the data to its nominal value. This system is transformer-coupled and balanced, and has a high common-mode rejection ratio that allows it to perform well in a multiconductor cable with other noise-producing signals. The average signal current approaches zero because each pulse equals the previous pulse in amplitude and is opposite to it in polarity. This means that the system requires no dc response. With a repeatered line, signals can be transmitted over at least 25 miles without degradation. After the interface requirements have been met, the digital information may be considered in terms of the logic family in the bit error rate detectors. To measure the bit errors in the incoming data stream, a synchronous clock is needed to feed the data into the sink portion of the bit error rate detector. It is also necessary to synchronize the reference pattern with the incoming pattern. The synchronous clock can be provided in a number of ways. Usually, the input data is transferred into a flip-flop register bit by bit. If the data is entered in NRZ-L form, then an optimum clock transition occurs in the middle of the bit period. However, because the data exists at a 1 or a 0 level for the entire bit period, the input clock and data may be at any phase relationship so long as they are synchronous, that is, with one clock pulse occurring within the period of each data bit. To make bit-for-bit comparisons, the input data must be synchronized with the reference pattern generator. There are several ways of doing this, depending on the type of pattern generator used, length of code, and number of bits. The most obvious method is to observe the incoming data, recognize a particular part of the pattern, and simultaneously preset the reference generator to that state. A faster method is to preset the reference pattern generator to the state of the incoming data. To synchronize input data with the reference pattern at high bit rates, the clock to the reference generator can be deleted once each pattern cycle as long as the pattern comparator exhibits a high error rate. Eventually, the two patterns will fall into synchronization, and no more clock pulses will be deleted. The problem here is time. It takes an n-bit code approximately $m2^n/2$ bits to acquire sync, where m is the number of bits used to determine an out-of-sync condition. When using a shift register generator, the method of synchronization is simple and efficient, as Fig. 5(b) shows. The state of a pattern generated by a maximum-length n-bit shift register counter is determined by n bits of the code. To acquire sync, at least n bits of incoming data will be gated into the reference shift register. After the register is filled, the feedback loop is reconnected, and incoming data is applied to the comparator. All bits used for synchronization must be error-free. If not, the shift register will be faked into a different part of the pattern. For this reason, if the error rate is very high, several attempts at synchronization may be required. Because of errors and jitter, the reference code generator may jump out of synchronization with the input data. If this happens, it is convenient to have a circuit to detect the out-of-sync patterns and automatically resynchronize them. Loss of sync is evidenced by a 50% error rate. Consequently, the obvious way to uncover out-of-sync conditions is to detect error rates of 50% over a given interval. However, the random errors can approximate loss of sync, and, as the error rate increases, there is a greater probability of mistakenly attributing an error burst to loss of sync. By observing the error rate over fairly long periods, 4. Codes. The top seven codes are aerospace standard. DM is for high-density tape recording; T-1 for telephone PCM systems. The chances of such misinterpretation are reduced, except that if the patterns are in fact out of sync, a large number of errors are incurred while the decision is being made, and this results in a distorted bit error reading. The problem boils down to a tradeoff between accuracy in determining out-of-sync conditions and the number of bits required to do this job. A practical solution is to have a monitor of resynchronization pulses, so that the operator can interpret bit error rate measurements according to the incidence of resync pulses. When bursts of errors cause a false resynchronization pulse and the error rate is high, the attempt to resynchronize usually fails. The net result is that the automatic resync circuits, instead of keeping the reference pattern in sync, are knocking it out of sync, and causing an abnormally high error-rate reading. One solution to this problem is a switch that disables the automatic resync whenever error rate reaches a certain threshold. But if this is done, a manual resync switch must be included in the design, to allow the operator to resynchronize whenever the 50% error rate indicates an out-of-synchronization condition. To see how all these elements combine to form a practical bit error tester, take the example of the DCS Model 4660 Bit Error Rate Counter. This unit was developed for aerospace and wideband synchronous data communications. It uses a 2,047-bit pseudo-random noise code at data rates up to 10 Mb/s. A sync pulse is provided for both the source pattern and the reference pattern, and is used to synchronize an oscilloscope with the source pattern. To modify the source and reference pattern, a control is included that provides a period of no transitions in the code called “blanking.” This blanking control forces the pattern output to the zero state for the first 32, 64, 128, or 256 bits after the sync pulse in each pattern cycle. Basically, this provides an extremely low-frequency component in the pattern to test the “coasting” ability of bit synchronizers. Because the control is separate for source pattern and reference pattern, it can also be used to introduce errors at a known rate; that is, by selecting “0” blanking in the source pattern and “32” blanking in the reference pattern, the source 5. Synchronizing methods. If the clock code format is in NRZ-L form, the optimum transition occurs in the middle of the bit period (a). To synchronize input data with the reference pattern, the shift register generator must be filled with return data (b). pattern will have, for instance, 23 logical 1s that the reference pattern isn't expecting. This provides a calibrated error rate of $23/2047$, or $115/10^3$ on the three digit display. To test a two-way data loop, the bit error rate counter is connected to the data source and sink. A source code clock synchronizes the returning data. (If the counter were used to test a radio transponder link or one with severe data degradation or frequency shift, a bit synchronizer at the data sink would have been necessary.) After the switches have been adjusted for the proper code format and data inversion, the bit error rate counter acquires sync either automatically or manually, and bit error rate is shown as a digital display. If it indicates an excessively high error rate or an out-of-sync condition, the return data is examined to determine that it really exists and isn't a random result, and that it meets the minimum operating criteria for the 4660 input limiter or for the bit synchronize input, if used. The first step in characterizing system performance is to observe the input data on an oscilloscope synchronized to the data transitions and study the "eye" pattern. Some of these are shown in Fig. 6. By superimposing many data transitions, the full range of effects of jitter and noise can be seen at a glance. The measure of data ambiguity is related to the amount the "eye" of Fig. 6(b) is closed. This indicates the difficulty encountered by the input limiter in making the logical 1/0 decision. If the scope external sync is connected to the source pattern sync on the link counter, the pattern can be seen bit for bit. With a delay sweep scope it is possible to examine the entire 2,047-bit pattern. The next step is to examine the errors. These appear as pulses each time a disagreement occurs between the two patterns. By observing error pulses and data simultaneously on a dual beam oscilloscope, with source pattern sync connected to the scope external sync, it is possible to determine the errors characteristic of the transmission system. Random errors are caused by noise and appear on the scope display as scattered pulses. Characteristic errors are those that occur consistently at a particular part of the pattern. Both random and characteristic errors are shown in the scope trace in Fig. 6(c). Errors that occur after a period of no transitions indicate poor low-frequency response, while those that occur only during a peak transition—101010—indicate poor high-frequency response. Another method of ferreting out the cause of errors is to synchronize the scope to the error output; this means the first bit on the scope is always the erroneous one. If the errors are predominantly 1s they may indicate induced spikes, if predominantly 0s they indicate signal dropouts. The scope trace may be used to synchronize the errors with other equipment, as well. For example, if errors were being introduced by a nearby printer, the print command pulse could be synchronized on the scope trace by the error pulse from the link counter. Another characteristic which can be measured using this configuration is transmission delay. This is done by making source pattern sync serve as the oscilloscope external sync and observing the reference pattern sync on the oscilloscope display. A calibrated sweep shows the delay as the interval between the start of the sweep (source sync) and the occurrence of the pulse (reference sync). For accurate measurement, it's necessary to subtract from this the delay occurring in the bit error rate counter by connecting the pattern output directly to the bit synchronizer input. If the total delay is greater than 2,047 bits, it is also necessary to estimate overall delay and add an integral number of 2,047-bit delays to the final figure. In bit error rate testing, it is important to remember that this is basically a digital distortion measurement. Just as the phase error in a distortion meter is nulled out to take a reading, errors introduced by the test equipment must be removed. To make these measurements, test points on the bit error rate detector itself are usually provided. Test points for the model 4660 are shown in Fig. 7. Ideally, the test fixture should simulate exactly the equipment receiving the data. Since the information is digital, this is usually not a question of impedance and levels but rather of timing and inversion. Can you imagine what the signal source of the future will be like? Will it have excellent frequency accuracy and stability? Better amplitude characteristics? Systems capability with inexpensive interfacing and software you can handle yourself? And how about a reasonable price tag? Well, the new Hewlett Packard 3320A and 3320B Frequency Synthesizers give you all this—plus a lot more! Both instruments give you frequency accuracy and stability measured in a few parts per million per year. That's a thousand times better than any RC oscillator. They both give you the signal purity of a good RC oscillator—a new feature for a frequency synthesizer. The 3320B offers you amplitude accuracy, resolution and frequency response measured in a few hundredths dB over a 100-dB attenuation range. So it is both a frequency standard and a very precise level generator. The 3320A and 3320B have optional remote control and the 3320B can even be controlled directly from most tape readers or card readers (like the new HP 3260A Marked Card Programmer). And to top it off, the 3320 has the widest frequency range of any test oscillator, programmable oscillator or low cost frequency synthesizer on today's market, 0.01 Hz to 13 MHz. How about the price tag? The 3320A, priced at $1900, and the 3320B at $2400 give you two great buys in signal sources—both today and in the future. We call the 3320 "the new price/performance benchmark for precision signal sources." Electronics magazine called it a "pacesetter." You'll call it "a steal." For further information on the 3320A/B, contact your local HP field engineer. Or write Hewlett-Packard, Palo Alto, California 94304. In Europe: 1217 Meyrin-Geneva, Switzerland. ...at $1900, It's a Steal! Solving interconnection problems in big multilayer pc boards The task was formidable: interconnecting 23,000 plated-through holes in a 3-ft, nine-layer phased-array antenna board; the solution: numerically controlled drilling and a plastic-encased preform solder system. by Philip O. Benham, Raytheon Co., Bedford, Mass As systems grow larger and more complex, formerly routine considerations such as interconnections for printed circuit boards are becoming increasingly important and potentially troublesome for designers. A very large multilayer board presents soldering, drilling, and interconnection problems that become a critical part of the overall design. These problems came to a head in a multilayer board designed for use in a phased-array antenna system, and their solution provides a model for other high-density pc board applications. The antenna, developed under an Air Force contract, required a nine-layer pc board about 3 feet in diameter with about 23,000 plated-through holes, each of which required a soldered contact for a receptacle that holds a phase shifter. The antenna, a space-fed reflective array of ferrite phase shifters, is illuminated with rf energy by a single four-square horn (a cluster of four horns that defines the four quadrants). The reflector is an array of 3,850 high-permeability Hipernom metal tubes welded together (Fig. 1); each tube is fitted around a ferrite phase shifter (Fig. 2). Each phase shifter plugs into a six-pin connector mounted on the 34-inch-diameter multilayer circuit board, which carries the logic and control currents (Fig. 3). Before deciding on a single, large multilayer board, a set of smaller boards was considered. But the problems of interconnecting the smaller boards with wires having controlled capacitance and inductance were too formidable; hence the single-board approach. However, the single-board method presented four challenges: generating accurate artwork; finding a laminating press large enough to handle the 34-in. board; assuring good laminating over the entire board; and accurately drilling the 23,000 holes. A board fabricator was found whose presses were large enough and whose laminating techniques were adequate to assure quality work, but the artwork and 1. Down the tubes. Array of high-permeability Hipernom tubes fits over phase-shifter array in space-fed system. 2. Inside look. Mockup shows how Hipernom tubes enclose ferrite phase shifters with their coils and drive networks. drilling were more difficult. To attain the needed accuracy in the artwork, it was necessary to apply a technique commonly used in integrated circuit mask-making. And accurate drilling of the holes in the boards required the use of a numerically controlled drilling machine. With such a large board, it's not practical to generate oversize artwork and then reduce it with a camera, as is common practice with smaller boards; accurate 1-to-1 artwork must be produced. But since the pattern is composed mainly of six-dot groupings corresponding to the receptacle terminals, a master six-dot pattern was produced at 10-to-1 scale (a larger scale, say 100 to 1, would not have been justified for the accuracy needed). Then it was reduced and reproduced with a step-and-repeat process to form the artwork master for the complete board. When the antenna is finally assembled, each of the Hipernom tubes must line up accurately with its own ferrite phase shifter. To assure alignment, the tubular array was made first and measurements were taken directly off the array and transformed into coordinate information for a numerically controlled drilling machine. Each multilayer board is thus drilled to match one particular array of Hipernom tubes, and a separate drill tape is made for each board. Once the board is made and drilled, the receptacle contacts must be attached. Hand soldering is insufficiently reliable as well as time-consuming and costly. Flow-soldering methods also are not practical because of the great size of the board—large copper areas would cause uneven heat distribution, leading to warpage. The choice fell to a system based on prepackaged, measured amounts of solder in irradiated heat-shrinkable plastic. The plastic recovers from its initial shape to a flat sheet when heated above a predetermined temperature. It's a specially compounded polymer composed of long, chain-like molecules. To achieve its heat-shrinkable properties, the polymer is bombarded with ionizing radiation from a high-energy electron beam to form cross links between the molecules. This molecular change creates a new material that no longer melts and flows at the low temperature it would have prior to cross linking. The new melting point is higher than the temperatures needed to melt the preform's solder. The cross-linked polymer can be expanded, molded, or shaped when heated above its melting point before irradiation. If the irradiated material with its cross-linked molecules is cooled while being held in a particular shape, it will retain that shape until it's heated above the original pre-irradiation melting point again. The material then returns to the shape it held prior to cross-linking. Thus, it has an "elastic memory," and it is this action that forces the molten solder into the plated-through holes. A commercial version of the soldering preform, called Dimple-strip (made by Raychem Corp., Menlo Park, Calif.), was redesigned to make the size and spacing of the solder preforms fit this job. Most connector pins are essentially flush with one side of the board, as shown in Fig. 5(a), and thus require only a solid solder preform. However, some pins also pass through the board and extend outward to connect to flat cable. To handle these pins, a solder ring and a plastic dimple with a perforation was developed so that the pin could pass through but still make an effective seal to the plastic, thus preventing solder from escaping. To form the Dimple-strip, cavities or dimples are shaped while the material is heated above its original melting point, then solder forms are inserted in the dimples. The solder used is 50-50 indium/tin with a melting 5. Shrink and press. Three stages (a) of heating for flush-pin connections show original solder sphere in its dimple, molten solder moving up plated-through hole as dimple shrinks, and final state, with solder fillet on both top and bottom. Similar action occurs for the through-pin connection, except that a solder ring and a punctured dimple are used. Normal elasticity of the material creates a seal around pin. 6. Seal it. The Dimple-strip is sealed to the multilayer printed circuit board to prevent solder leakage, with pc board hold-down plate, base plate bolted together. Before soldering, all pins, which have been plated with tin, are cleaned and coated, and the board is coated with liquid flux, such as the type 611 made by Alpha Metals or Kester type 1544. This assures good solder coverage of the holes. The hold-down plate then is loaded with the flush-pin Dimple-strips, omitting areas where through-pins occur. The through-pin Dimple-strips then are placed in position on the hold-down plate. The board, with flush pins inserted, is placed over the Dimple-strips on the hold-down plate. After it's registered to the circuit board with special pins, the through-pins are inserted. The connector backup plate then is placed over the connectors and the whole assembly is bolted together. The assembly next goes to a preheated base plate, with a preheated top plate clamped in place. Preheating is needed to bring the mass of metal in the clamping plates to near the soldering temperature before the assembly goes in the oven; without preheating, the solder might melt in the oven before the copper of the plated through holes is at the proper temperature, resulting in poor solder joints. The assembly stays in the 300° oven for 5 minutes under a vacuum of 25 inches of mercury. Thus, more than 23,000 connections are made in just a few minutes without temperature variations that might cause warpage in wave-soldering. After removal from the oven, the board is inspected, and a small soldering iron is used to reflow solder around unsound joints. Yields generally have ranged near 97%, with the remaining 3% easily touched up. Your water—expensive Conditioned water—cheap How good should good water be in order to operate efficiently in the electronics business? We know what it means to you when a semiconductor is spotted with impurities. We understand why you must have a reliable source of water that meets your exact megohm specifications. Meeting these needs—constantly, dependably, automatically, economically, is our business. Our product and service equipment lines are complete and efficient. For example, a Culligan Aqua-Cleer reverse osmosis system, followed by our deionizer service units, combine to provide extremely high quality water at exceptionally low cost. And there are no acid or alkali caused waste treatment problems. In addition to conditioning process and production water, Culligan capability includes equipment for softening, filtration, boiler water, cooling towers, and waste water treatment. Call your local Culligan Man today for a consultation. He will evaluate your water needs as to flow, quality, application, and treatment equipment requirements. Culligan® THE WORLD-WIDE WATER CONDITIONING PEOPLE WHO SERVE YOU BETTER LOCALLY Franchised Dealers in the U.S. and 85 countries. Culligan International Headquarters, Northbrook, Illinois 60062. Franchises available. Expanded Versatility for Tektronix Automated Measurement Systems Testing versatility in TEKTRONIX Systems is a product of proven hardware performance, a high degree of programmability and the ability to communicate with computers. The R1340 Data Coupler, with its options, is the deceptively plain-looking box that brings the computer and systems components together for exceptional versatility. Perhaps the most exciting function the R1340 can perform is enabling the computer and the system to digitize waveforms for direct computer processing. In waveform digitizing, a programmable data acquisition instrument responds to commands selecting test points, sensitivity to voltage or current and a measurement time slot. The outputs of the data acquisition system are then made available to the computer for digitizing through interfaces in the R1340. A computer may be used to select testing programs stored within the system. The R1340 Data Coupler has 32 bit input and output buffers available to allow systems components such as DVM's, test fixtures, etc. to communicate with the computer. Data logging runs, with or without direct computer involvement, is another function that is made possible by the R1340. TEKTRONIX Systems test both digital and linear devices. DC, functional and dynamic testing can be done with just one system. Can you afford to select a test system without talking to your TEKTRONIX Field Engineer? TI sees computers in its future Though avoiding confrontation with IBM, Dallas firm seeks 'significant' share of market for big scientific machines, minis, and process control by Paul Franson, Dallas bureau manager While RCA Corp. is folding its computer operations, Texas Instruments Inc. is slowly unfolding its plans for expanded activity in that very area. However, the Dallas-based organization, unlike RCA, will avoid competing with IBM and will concentrate on minicomputers, process control systems, and giant scientific machines, typified by its Advanced Scientific Computer (ASC). TI won't release any details on its ASC or say what its specific plans are; when asked what part of the market they expect to corner, company officials smile and reply, "a significant share." The answers to questions like "how much", "how big", and "how soon?" are still locked up in TI's executive row. But there's no question that TI wants to significantly expand its computer sales. Right now, the company is selling minicomputer systems, manufacturing almost in secret two ASCs, and diversifying its line of peripherals. TI firmly denies it is getting into computers because mainframe makers are manufacturing more and more of their own semiconductors. Although IBM contributed greatly to TI's early semiconductor growth, and has always been TI's major customer, the machines from Poughkeepsie, N.Y., are using more and more in-house ICs. And TI accordingly has reduced its dependence on sales to IBM. But other computer makers are not expected to follow IBM's example in developing in-house components, according to Grant A. Dove, TI's vice president, corporate development. "In fact," he says, "some computer makers seem to be getting out of semiconductor production." Overall growth is providing the impetus for TI's push into computers. Last year's sales were $826 million, and by 1980, the company says it will nearly quadruple that figure to $3 billion. That kind of growth probably cannot come from semiconductors, airborne radar, oil exploration, and the numerous smaller business areas that TI is active in. The company is highly integrated vertically, producing and selling materials and equipment, and coordinating these activities from a sophisticated technological and management base. A. Ray McCord, group vice president, equipment, says, "Our major goal is solving the customer's problem. If that requires a computer, we'd like to sell a TI computer. But we want to keep the problem-solving vector far in front of the black-box vector. We have only so many resources and feel we can get the best return this way." TI has plenty of experience in automating its own crystal-growing, transistor, and IC production and testing. And recently, the firm has been increasing its emphasis on outside marketing of computers and customer service, according to McCord. Sales offices will be increased from 19 to 57 by the end of next year, he says. More coming. TI's Silent terminal (below) checks solid state radar. Multiple-head disk (right) is one of four units used in TI's ASC. the business that led to TI's entry into computer development and manufacture. As early as 1958, Geophysical Services Inc., TI's subsidiary, developed the TIAC 827 for handling the immense quantities of seismic data involved in locating oil deposits. Several machines were purchased by major oil companies. And in 1967, GSI introduced the 870A, which is still being built and sold for seismic work. But the big machine for seismology will be TI's ASC. TI once considered bidding on another giant computer—the University of Illinois' Illiac 4—but didn't because it looked like a money-loser, according to insiders. Burroughs Corp. got the contract. Now, with TI trying to get into the big computer market, James Marley, marketing manager at Burroughs, says TI won't be a serious competitor for some time because even with its technical capability, the Texas firm doesn't have Burroughs' experience in software and maintenance. Meanwhile, TI's competition in maxi and minicomputers is maintaining a wait-and-see attitude. "We don't know enough about TI's Advanced Scientific Computer to know whether or not we should be worried," says James Thornton, vice president of Control Data Corp., Minneapolis, Minn. Thornton doesn't expect to learn much about the ASC until either TI gets a contract and has to reveal its designs or the company changes its traditional tight-lipped policy. At the Foxboro Co., Foxboro, Mass., Bruce H. Baldridge, manager of corporate marketing and product planning, recalls discussions with TI about a year and a half ago regarding a possible OEM agreement for a TI process-control computer: "At that time," says Baldridge, "they had no software, no marketing group to sell to end users, no systems engineering, no systems analysis. We looked upon them as a possible computer supplier." But TI has now established application engineering and software groups, and has beefed up its marketing. If TI starts selling process control systems to customers who will do the systems application, analysis, and other work, says Baldridge, "we wouldn't likely shrug them off." But if TI decides to stay in the semiconductor process control area, Baldridge says, "we may not bump into them at all. Process control is a big field." While TI's ASC grew out of the requirements of its geophysical operations, the TI minicomputers were motivated by factory automation and, to some degree, military activity. One of TI's early minis was the 856/7, developed to replace an SDS 910 that, in turn, replaced paper tape in programming the TI continuous automatic test (CAT) system for transistors. These 5-microsecond, 24-bit computers were patterned after the 910, and a smaller, more practical version, the 853, was used for the CAT and other test systems. A recent venture was a joint development by TI and Applied Kinetics Inc., a subsidiary of Alloy Metal Products, Inc., of Davenport, Iowa, of an advanced process-control computer system for melting applications in the metals industry. The AKI series 70 system includes a TI 980 minicomputer; the system is installed by AKI and serviced by TI. Peripherals. In peripheral equipment, TI's Silent terminal uses a thermal printhead and is made in portable versions, and as replacements for the IBM 1050 and 2741 terminals and teleprinters. The company's line of tape transports includes plug-in replacements for the IBM 360 series 2400 (TI has a $1.8 million contract from the General Services Administration) for relatively low-cost OEM transports and transports and controllers for the IBM System 370. TI's focus is ruggedness, mechanical simplicity, and electronic sophistication. McCord says TI needs a broad line in peripherals, and is working to develop more equipment. Moreover, though TI hasn't announced any definite intentions, it's hinting strongly that some of the work on the big scientific computer could lead to sales of the high-speed disk recorder for the ASC. Right for the times! Here's why these new counters—offered in four frequency ranges of 50, 200, 512 MHz and 3 GHz—are right for today's requirements: New 6150 series expandable universal counter-timers. S-D developed this line to handle almost every counter/timer requirement—bench and systems. Four basic models cover frequency ranges of 50, 200, 512 MHz, and fully automatic 3 GHz. Buy for current requirements and upgrade frequency range at any time. These instruments were designed from scratch for programmability: single line or binary, total control including attenuators, and analog or digital trigger level control. 6150 options and features: Choice of 100 nsec or 10 nsec TIM resolution—choice of five oscillators—four types of BCD output—up to 9-digit readout — versatile remote programming — 3½" height in full rack width—10mV rms resolution to 0.1 Hz. Expandability. The expandable counter concept satisfies many needs, present and future. Why? Because it's so simple and economical to upgrade the frequency range of your counter and add options. BCD output, additional readout digits, and a 200 MHz frequency range (in place of 50) are added by inserting new plug-in PC cards right inside your lab in minutes. Go to 512 MHz or 3 GHz, higher stability oscillators, 10 nsec TIM resolution (on 6150 universal series), and remote programming—all are offered as expandable option kits installed by your local S-D service center. \| Frequency \| 50 MHz \| 200 MHz \| 512 MHz \| 3 GHz \| \|-----------\|--------\|---------\|---------\|-------\| \| Universal Model \| 6150 \| 6151 \| 6152 \| 6153 \| \| Counters Price \| $1195 \| $1495 \| $2150 \| $2995 \| Systron Donner Concord Instruments Division, 888 Galindo Street, Concord, CA 94520. Phone: (415) 682-6161 World Radio History Circle 95 on reader service card Capacitor Problems That Require A Lot Of Self-Control...Chemically Speaking Problem 1: How to make sure the silver paste composition used for electrodes provides the best results for each electrical parameter in a given capacitor design? Problem 2: How to improve the recognized moisture reliability of our dipped mica capacitors without adversely affecting life reliability? Problem 3: How to upgrade the reliability of molded mica capacitors to equal that of dipped mica capacitors so designers can take advantage of body uniformity and axial lead design? Solution: Chemical self-control! To do this we operate our own chemical manufacturing plant where we formulate silver pastes, phenolic dipping compounds, and epoxy molding compounds — all under strict controls. Result: Dipped mica capacitors and molded mica capacitors of equally high reliability that operate up to 150°C. Send for technical literature and always insist on El-Menco brand capacitors . . . your assurance of better quality and reliability through control. THE ELECTRO MOTIVE MFG. CO., INC. WILLIMANTIC, CONNECTICUT 06226 Dipped Mica • Molded Mica • Silvered Mica Films • Mica Trimmers & Padders Mylar-Paper Dipped • Paper Dipped • Mylar Dipped • Tubular Paper West Coast Manufacturers contact: COLLINS & HYDE CO., 900 N. San Antonio Rd., Los Altos, California 94022 5300 Whittier Blvd., Los Angeles, California 90022 ALSO SOLD NATIONALLY THROUGH ELECTRONIC PARTS DISTRIBUTORS Linear IC market in ferment Reshuffle caused by recession and saturation of op amp sector pushes designers into consumer, computer interface, other special-purpose ICs by Lawrence Curran, Los Angeles bureau manager There's a shakeup going on in linear integrated circuits. The day of the dominance of the general-purpose operational amplifier is fading and semiconductor companies in the linear business are concentrating on the more-specialized devices. The shakeup has been triggered by a number of factors that have combined to slow the brisk new product introduction pace of the years from 1967 through mid-1970. They include an apparent near-saturation of the op amp market, a recession that redistributed linear designers and process engineers, and a recognition that the growth markets are now in consumer and computer-interface circuits. The action has already started. There's a big surge in demand for consumer circuits—mainly devices that go into sockets in television and stereo sets—and the market for computer-interface ICs is blossoming. Not that linear suppliers are abandoning op amps. The LM 118 introduced in August by National Semiconductor Corp. has most people, including at least one competitor, convinced it's the newest industry standard. More than one firm will become alternate sources for the unit, including Motorola's Semiconductor Products division. National calls the LM 118 an easy-to-use device with a very high slew rate—the minimum rate is 50 volts per microsecond with zero offset—and the unit is internally compensated. High-slew-rate op amps have typically required external components for compensation. Few, if any, other truly new op amps are being introduced in the industry, as suppliers think the op amp market is saturated. So linear IC houses are going through a transition period as the emphasis shifts to consumer circuits and a variety of interface devices for the industrial market, such as line drivers and receivers, sense amplifiers, digital-to-analog and analog-to-digital converters, and comparator circuits for level sensing. There's also some effort to develop new voltage regulators and custom circuits for the automotive market. But the transition hasn't been smooth so far, and it doesn't look like it will settle down for awhile. Most linear suppliers were caught with big inventories in mid-1970 when the recession began to hit hard. This led to price cutting, which is still going on, even though business is picking up at most firms. Just how hard this price cutting has hit makers is reflected by first-half linear sales. Units were up to 26.5 million compared with 24 million in the first half of 1970, but dollar sales dropped from $44.6 million in 1970 to $40.3 million this year. This dropped the average selling price for a linear IC from $1.86 to $1.52. The price cutting is evidence of the linear market's "immaturity," says Steve Thompson, linear IC marketing manager at Motorola, who says the pricing trends have become "almost irrational" at some companies. If it continues he says, it will lead to the cutthroat pricing that hit bipolar TTL circuits. But he doesn't think the downward trend can continue for too long: "enough people are aware of the situation so that economic conditions may not force the same degree of price cutting that happened when inventories were high." Thompson looks for 1971 linear IC sales to be in the area of $86 million, up $6 million over last year. For 1972, he sees sales hitting $98 million. Slowing new products. Playing a big role in the linear business shakeup are key personnel shifts among the companies. As the recession set in, many companies were forced to let some of their linear designers go. Other key linear people left voluntarily to accept better job offers. As a result, new linear products have been slow in coming on the market. For example, late last year Fairchild Semiconductor officials were hoping to introduce some 40 new linears in 1971. It now looks like only about a third of these circuits will reach the market by year end. The reason is personnel cutbacks and "some premature announcements," says Anthony Livingston, Fairchild's product marketing manager for linear ICs. No more super op amps? Nearly all linear IC makers feel the evolution of the general-purpose op amp has reached the end of the string. "No more super op amps are coming," says Fairchild's Livingston, "although you'll see specials like the 776 programmable op amp." The industry has been through four generations of op amps now, "and it's difficult to design the fifth-generation device, though the LM 118 looks like it could be it," says Thompson. Agreeing with Thompson on the op amp's future is Robert Dobkin, National's director of advanced circuit development. Linears grew initially because of the popularity of the Fairchild μA709, but he says that they may have reached a practical limit with the LM 118 in terms of ease of use and wide application. "I don't see much in the way of innovation in op amps," he observes, "but I do see new device types—circuits that people don't normally think of as being linears. National is working on a new type of voltage regulator and a high-speed comparator circuit and expects to introduce 40 linears in 1972. While Sprague Electric Co. isn't talking about op amps as a growth business either, it is moving strongly into linear ICs. The company is beefing up its consumer-oriented linear IC design and development efforts with the emphasis on TV and automotive circuits. It expects to double the number of designers for these circuits to about 20 by the end of the year. "We're so enthused about projections in these fields that we are more or less downplaying digital circuit development to concentrate more fully on the linears," a Sprague official notes. Texas Instruments, which hasn't been recognized as a strong factor in the linear market, is coming on strong now, particularly in consumer parts and computer-interface circuits. And even though Lawrence J. Housey, product manager for linear circuits in Dallas, says the industry can still expect evolutionary improvements in op amps he concedes that "the tradeoffs are getting harder to make." Signetics Corp. is about the only major semiconductor manufacturer hinting at big new things in the op amp arena. Alan Gregory, director of linear products, says that while the company initially "chased the market" with its versions of Fairchild and National op amps. "We will not have any new op amps that employ present technology for the time being. To make a significant improvement in op amps will require a technological breakthrough—and we're working on it." Signetics appears to have a leg up on its linear competitors with its phase-locked loop devices. These give the company an edge in linear circuits for the communications and computer peripheral market because they're being designed into modems, printers, and receivers. National Semiconductor has already become a second source for the phase-locked loop units, with Fairchild and Advanced Micro Devices to follow soon. Consumer: the action. The potentially lucrative high-volume consumer circuit business is the market that has the linear houses really turning on. Motorola's Thompson says the consumer portion of the market grew by 120% in dollar sales from 1970 to 1971 and the same kind of growth is predicted for 1972 over 1971. Reducing it to dollars, National's Dobkin pegs the consumer linear market, excluding automotive parts, at $10 million this year, doubling to $20 million in 1972. National has established a new group to concentrate on consumer circuits, from which products should be flowing within the next few months, Dobkin says. Fairchild's Livingston says his company has developed or is developing a color processing IC for television sets, a circuit that contains the complete audio section of a television set including the 5-watt output amplifier, and a general-purpose monolithic circuit that can control all the timing functions in an appliance such as a washer or dryer. And Motorola is moving into volume production with a phase-locked loop stereo demultiplexer that's scheduled for formal introduction in November. Most of the major linear suppliers are casting a wary eye toward Dallas as TI readies its consumer product push. Motorola's Thompson characterizes TI as being "very aggressive on price, and they're second-sourcing everyone in the industry in the consumer area." TI's Housey is reluctant to talk about TI's specific plans because so much of its work is custom and is being done for a limited number of customers. He will say, though, that the firm has 40 to 50 consumer linears in development. Computer market: good. Linear IC houses are also watching growth curves in computer-interface circuitry, which are not expected to be as steep as those in the consumer product arena, but offer great potential, nevertheless. National's Dobkin says the 1971 industry figure for sense amps, line drivers, and receivers will be about $20 million, growing to about $30 million next year. TI's Housey feels his firm is especially strong in line drivers and receivers, memory drivers, and sense amplifiers. TI is heavily involved in custom work, also, including what Housey calls "linear LSI" for multiple memory interfaces. Fairchild recently introduced a precision comparator circuit and a high-speed comparator as the vanguard of a new series of interface circuits. Motorola has developed an ac-coupled sense amp for plated wire memories, which is described as a comparator type of amplifier that will probably bow later in October. Motorola also has a six-bit monolithic digital-to-analog converter circuit ready for the market; it will sell for about $5 in large quantities. This compares with $10–30 for the nearest monolithic six-bit unit and $70–80 for an eight-bit monolithic d-a converter on the market. Motorola also has coming an MOS clock driver aimed at doing a job done only by hybrid circuits today—driving the heavy capacitive loads of MOS devices and handling swings in voltage from –30 to +5 volts. What it all boils down to is that high-volume, general-purpose linear IC makers are bumping into the limits of optimization and innovation. The tradeoffs are forcing them to the specialized and custom devices. How it changes the job of the circuit maker is best summed up by Signetics' Gregory: "Rather than telling a customer what we have and what he should use, we will go to bed with him and build what he wants, based on the technology we have." --- Probing the news International Electronics to impact European automobiles Fuel injection, electronic ignition, antiskid braking will proliferate as pollution, safety issues catch on To many car buffs, U.S. automobiles are not even in the same league as European models. Yet visitors to the Paris Auto Show this month will not find much difference between new European offerings and 1972 U.S. cars in the way of electronic accessories. Like U.S. car makers [Electronics, May 10, p. 105], European manufacturers are slowly adding nonentertainment electronics to their products, with the big expansion expected in three or four years. And despite their current lead in developing car electronics, Europe may eventually lag the U.S., since the legislative outlook in Europe is not nearly as clear as in the U.S. Voluntary adoption of electronics will rest heavily on substantial price cuts, asserts a spokesman for British Leyland Motors, makers of Austin and Morris cars. He feels electronically controlled alternators and fans are good bets, but believes antiskid brakes and electronic fuel injection and ignition won't be able to compete with conventional approaches without tough laws to back them up. In Paris, Daimler-Benz will unveil its new 35-slc sedan with the electronically controlled fuel injector developed several years ago by Robert Bosch, GmbH. Mercedes models will include electronic windshield and headlight wiper-washer combinations. The only electronic gear on a French car will be the Citroen DS-21's Bosch fuel injector and electronic tachometer, neither new this year. The Aston Martin and Jaguar V-12, however, will have the new Joseph Lucas electronic ignition. Also not being shown are the antiskid systems developed by Daimler-Teldix, Bosch, Bendix, and Peugeot. The legislative situation is most advanced in West Germany, making that country the leader in automotive electronic systems. By 1975, 80% to 90% of all cars in West Germany are expected to have electronic fuel injection and in the next four years the car electronics market will quadruple from its present $10 million in total sales to $40 million, according to Eckhard Zeiffer, marketing manager at Texas Instruments' Freising, Bavaria, facility. At TI's Freising plant, circuits are being developed for electronic direction indicators, speedometers, tachometers, fuel injection, antiskid units, and voltage regulators. TI is cooperating with nearly all major car equipment suppliers in Europe, including Bosch, Smiths Ltd., and Lucas in Britain, and with Italian firms. Siemens AG, Munich, and several other German firms are working on custom-tailored ICs. Intermetall GmbH, a subsidiary of the ITT Group, has already hit the market with an integrated circuit, the SAK 110, designed for tachometers; the circuit has a maximum frequency of 10 kilohertz and comes in an eight-pin mini-DIP package. Innovative. However, Bosch is far and away the most active automotive electronics supplier in Europe. Bosch, which recently opened a $10 million R&D facility near Stuttgart, makes an electronically controlled fuel injection system that's used on a dozen European cars, including the Volkswagen 411 and the Mercedes-Benz 280 SE. And now the Probing the news system is being redesigned to make all the circuitry integrated. Bosch has developed a four-wheel antiskid device, an electronically controlled transmission that shifts gears in 0.3 second, and an electronic control unit to evaluate inputs from the brakes, accelerator, steering wheel, and engine. Bosch's four-wheel antiskid systems are being tested by several automobile companies. The big obstacle to mass application of such systems, Bosch maintains, is price: the system that will go into the Mercedes 350 SL as optional equipment next year will run about $400, easily within range of buyers of expensive Mercedes models but too high for the low-priced market. Bosch feels that the price must be halved before large demand can be expected. Bosch is working on the electronically controlled transmission in conjunction with Zahnradfabrik Friedrichshafen (ZF). The heart of the system is a solid state control that functions without any interruption of tractive power during the gearshift operation. One version now being tested and developed specifically for trucks shifts gears up to 10 times faster than an experienced truck driver can. Electronic transmission units should hit the market in about two years initially as optional equipment on trucks and then for passenger cars. Future. Collision avoidance is another Bosch activity, but an economically viable system is still far off. Last year, the company made an extensive evaluation of systems proposed so far and ruled out radar-based or laser-based approaches as impractical. The company feels that an inductive loop-based system is technically better, although very expensive. The company also is making electronic windshield wiper control units; a windshield washer-wiper combination; wiper control systems for front and rear lights, and a headlight illumination regulator that helps prevent drivers in oncoming cars from being blinded. Reports from France say that all four of the country's car makers—Peugeot, Citroen, Renault, and Simca—have felt no competitive pressure from the U.S. car sales overseas to go electronic, although Peugeot is working on its own fuel injection system and Citroen's high-priced DS-21 model will include the Bosch fuel injector. In England, while Lucas has electronics systems for fuel injection, ignition, antiskid brakes, systems monitoring, and engine speed control, buyers are few. Steve Rayner, chief engineer of the company's electronic design group, says he wouldn't be surprised if the biggest orders for car electronics come from Continental Europe rather than Britain. Jaguar and Aston-Martin use the Lucas electronically switched breakerless ignition, but neither will talk about future plans. The two auto manufacturers in Sweden, Volvo and Saab, both use the Bosch electronic fuel injection on certain models; 30% of Saab cars are equipped with it. The only other electronic system in Saab's auto is the transistorized timing units in the car's gasoline-fired heater. The heater is made by Ebers Pacher of West Germany, which also builds the timer. Electronically controlled windshield wipers and turning signals look attractive, says Bertil Ylhage, head of Saab auto electrical systems, but the costs for such devices are still too high. The outlook for automotive electronics is less than certain in Italy, too. A planner from the Fiat electronics division in Milan says, it is safety that the Italian will pay for and it is probably in that area that electronics will gain entry into the car market." Tests. Toward that end, Fiat is testing an antiskid device for its model 130 luxury sedan. The company says the device meets U.S. standards that will go into effect in 1973. Fiat is also experimenting with a fuel injection system developed by its subsidiary, Magnetti Marelli in Milan. Marelli's biggest seller is electronic ignition used on prestige sports and racing cars. Marelli is also working on an electronic windshield washer, which will provide 12 to 40 strokes per minute and is slated to hit the market next year. Contributing to this article were Electronics staffers John Gosch in Frankfurt and Michael Payne in London, and McGraw-Hill World Newsman Michael Johnson in Paris, Andrew Heath in Milan, and Robert Skole in Stockholm. Sorensen’s new SRL power supply has a great thing going for it. Built-in overvoltage protection that can be quickly set and instantly checked even under full load. SRL is a new low voltage, regulated DC power supply for systems and laboratory applications. We designed it by asking power supply users what they liked and didn’t like, the features they wanted or needed. Features like exclusive front panel monitoring and adjustment of overvoltage setpoint without removing the load. SRL comes in 14 models: 4 voltage ranges to 60 VDC, 4 power levels to 2000 watts. It offers higher power density, low ripple and noise, fast response time over full load range, operation to 71°C, IC reliability. Check out SRL. Or any of the hundreds of other Sorensen power supplies with output voltages from 3 to 150,000 VDC, output currents from 1.5mA to 1000 A. They are all listed, with prices, in the Sorensen catalog. Write for your free copy to Raytheon Company, Sorensen Power Supplies, 676 Island Pond Road, Manchester, New Hampshire 03103. Tel: 603-668-1600. From the wonderful people who brought you the first DAC under $30 now... the barrier buster DAC-9-8B $9.95 the first DAC for only in singles SPECIFICATIONS: ELECTRICAL Digital Inputs Resolution 8 binary bits Coding straight binary, eight parallel lines Data inputs DTL or TTL compatible, positive logic Loading: one standard TTL load Iₜₘₐₓ = 1.6 ma @ Vₜₙ = 0.4V 5MHZ Update rate ANALOG OUTPUT Type of output current Output current ±2.6 ma Output voltage compliance 1.2v max. Output settling time 1 μsec to ± 0.4% of FS Resolution 10 bit Linearity ±½ LSB Temperature coefficient ± 100 ppm / °C of FS Input power +1.75 v ≤ 10 ma Operating temperature range 0 °C to + 70 °C Storage temperature range -45 °C to + 85 °C Size 2" x 4" FL x 0.4"H plug-in module Weight 2 oz. FEATURES: Fast settling time ... 1 μsec. Small size ... 0.8 cubic inches DTL/TTL compatible Ultra linear ... ± ½ LSB Low power consumption ... 15 milliwatts Complete ... simply apply DC power ORDERING INFORMATION: DAC-9-8B PRICE: $9.95 each Clip out and mail to VARADYNE SYSTEMS • 1020 Turnpike Street, Canton, Massachusetts 02021 Something old / something new / nothing borrowed Send today for your Free New 1971 Fall/Winter Catalog giving the complete details and specifications on 100 models, 30 of which are brand new. D/A's from $9.95. A/D's from $69.00 name ____________________________________________________________ company _________________________________________________________ street ___________________________________________________________ city ___________________________ state ________________ zip ________ VARADYNE SYSTEMS A DIVISION OF VARADYNE, INC. 1020 Turnpike Street Canton, Massachusetts 02021 or phone (617) 828-6395 Joint space talk outlook: clear skies Spotlight at second round in Moscow will be on guidance and communications group; though hardware is a long way off, both sides praise progress by Ray Connolly, Washington bureau manager Achieving compatibility among electronic systems, even on the smallest scale, is one of the most trying tasks engineers face. And they could be forgiven for thinking that anything on such a grand scale as compatible systems for U.S. and Russian manned space flight would be all but impossible. Yet the prospects for attaining this goal are good as engineering teams of the National Aeronautics and Space Administration and the Soviet Academy of Sciences make ready for a November meeting in Moscow. This will be their second in a series designed to make the civilian space hardware of the two superpowers compatible. How long the 15-man American team, headed by Glynn Lunney of NASA's Manned Spacecraft Center, will meet with the Soviets is not yet certain, but U.S. sources believe the stay will far outlast the initial two-day session in Houston last June since NASA wants its negotiators to complete agreement on a number of technical specifications. Details. While the June meeting produced agreements on more than 60 points in principle, technical details remain to be worked out on spacecraft docking and related communications, stabilization, and close-in navigation systems. As the minutes of that meeting note, "A great amount of work will be necessary, jointly and in each country." Before the two nations reach the point of undertaking a joint effort from scratch, they will begin by docking spacecraft in orbit, most likely an Apollo vehicle with a Salyut-class craft, both proven systems. The second phase will be docking of a Russian Salyut spaceship with America's Skylab scheduled for 1973 launch [Electronics, Sept. 13, p. 40]. Most significant for the long-term plans of the U.S. electronics community are the efforts of the second of three working groups, the one responsible for compatibility of radio guidance, optics, and other guidance and communications systems. During the June sessions, NASA's Donald C. Cheatham, chairman of the five-man U.S. team on Working Group 2, found Viktor Legostayev and his 10-man Russian team to be "most cooperative." Continuation of that relationship is seen as essential in achieving mutually acceptable operational standards in the 37 areas covered by that group. There are 15 standards for communications alone to be decided on. Rendezvous and docking methods are being explored by Working Group 1. This group also must establish requirements for automatic linking of electrical power and communications systems after docking. Working Group 2 has agreed that the docking procedure itself will be best achieved manually using onboard systems only. "For the present," said the group, "rendezvous guidance systems would not depend upon the accuracies of ground tracking systems." Both sides agreed to look further at laser and radio systems with passive reflectors for rendezvous. Papers proposing values for these and other still-undefined parameters are being exchanged by both countries as a basis for discussions in November, according to NASA. Agreement has been reached on some of the more obvious communications requirements. For example, Soviet and American spacecraft in Earth orbit will have simplex communications links at distances to at least 150 kilometers, voice communications, and omnidirectional antennas. A Russian proposal to use telegraphy to extend communications range and simplify possible language barriers still is Now you can make a sales trip to Japan for $400... and talk to 15,000 buyers in their native tongue! The vehicle? Nikkei Electronics — the Japanese version of Electronics Magazine published in Japanese for consumption in Japan only. Circulation - Nikkei Electronics will begin publication with a paid Japan circulation of 15,000 in April and with a guaranteed paid Japan circulation of 20,000 within the first year of publication. - Nikkei Electronics circulation is restricted to: engineers and management in companies manufacturing electronics products; users of electronics products; and managers of governmental and independent R&D; educational and electronics studies associations. - The subscriber universe for Nikkei Electronics is 120,000 electronics engineers plus 18,000 new engineers graduating annually. - Nikkei Electronics is the only Japanese Electronics magazine going ABC. All others are not audited. Probing the news being considered by the U.S. But both sides agree on the frequencies for spacecraft communication. Russian cosmonauts will use the uhf band between 120 and 140 megahertz, while the U.S. has tentatively agreed to 250-300 MHz. The type of modulation to be used by U.S. spacemen also is pending. Both countries will carry transceivers using hardware provided by each for its specific frequency. Both nations also decided that communication "would be useful" between spacecraft, astronauts outside the vehicles, and ground stations, although ground stations on each side will use only their own frequency. The U.S. said NASA is considering technical parameters for the transceivers proposed by the Soviets. Once technical requirements are drawn up at the November session, both countries will participate in design, development, manufacture, and testing of the new transceivers. More specs. Though both countries agree that a coherent transponder using an omnidirectional antenna (4-pi steradian) will be required on board passive spacecraft, at least five transponder specifications still must be worked out in Moscow. The list, which may be expanded later, includes receiver and transmitter frequencies, receiver sensitivity, transmitter power output, transponder phase delay, and bandwidths. In addition, planning of maneuvers leading to rendezvous will demand accurate data exchanges between Russian and U.S. ground stations on spacecraft state vectors. The Americans and Russians have agreed that their first considerations should not depend on ground tracking for rendezvous guidance. They also concurred that more long-range discussions will be needed to define space mission models, their probable ranges, and approximate accuracies applicable in each case. Though an equivalent set of nonelectronic hardware problems also faces the negotiators (space docking, vehicle atmosphere, and astronaut transfer systems are a few) engineers at NASA see none that cannot successfully be resolved. □ Advertising in Nikkei Electronics - If you sell, or want to sell your products in Japan, you can talk to 20,000 buying influences in their native tongue for just $400 a page. - If you are already advertising in any of the unaudited lesser media in Japan, here is your opportunity to consolidate your schedule in the one strong medium, make a dominant impression and save advertising dollars. If you are interested in advertising in Nikkei Electronics contact your local Electronics salesman or any of the Electronics sales offices in the U.S. and Western Europe. Editorial The editorial staff of Nikkei Electronics consists of ten full-time editors plus contributions from Electronics editors, the McGraw-Hill World News staff and the news services of NIHON KEIZAI SHIMBUN—the world's largest publisher of financial newspapers. Every two weeks, these editors will produce a news and technology Japanese language magazine for electronics engineers in Japan. When you come to us for connectors, you could have your whole job wrapped up. In the old days, we made connectors only to your specifications. We still do. But now we've gone a lot further. We added off-the-shelf connectors. And now we'll mount either type to your specifications. And we'll have them wrapped to your specifications. In short, we'll give you a connector or a complete wired backpanel assembly. Or anything else in between. You save time, money and aggravation, because you don't have to set up an in-house operation. And we can probably do it at lower cost because we make our own plastic bodies, form our own contacts, and assemble our own connectors. We even draw our own wire for wrapping. It used to be you could only come to us for connectors. Now you can throw the whole wrap at us. To find out what we really can do, call Russ Sanders at 814-723-2000 or write him at GTE Sylvania, Precision Materials Group, Parts Division, Warren, Pennsylvania 16365. It plots. And it prints. And it’s reasonable. And it’s only one machine. If you wanted someone to design a COM System that could plot, who would you call? CalComp. We’re the people who taught your computer to draw. And while working with our early plotters, we could see the need for (1) getting large quantities of information reduced onto something smaller; and (2) producing it a lot faster. Thus, Computer Output Microfilm. A COM System that could turn programmed information into bar charts, line graphs, or even a complete set of engineering drawings. And because our CalComp 1670 COM System could print as well as plot, you could add written information to charts, or transfer lengthy written records onto easy-to-store film. In fact, the equivalent of 100 pounds of information would now fit on 60 4 x 6 inch microfiche. Or on 16 mm., 35 mm., or 105 mm. film. Finally, we included still another feature. Simplicity. Once our system is programmed, it needs virtually no adjustment. It turns out plotted and printed information that’s consistent and accurate. And CalComp COM Systems are reasonable. In fact, we offer the best price/performance in the industry. And we support them with 52 sales and service offices around the world. See for yourself. Write to California Computer Products, Inc., Dept. EM-10-71 2411 West La Palma Avenue, Anaheim, California 92801. Or call (714) 821-2011. New products Linear 'LSI' shrinks scope for field work By Stephen Wm. Fields, San Francisco bureau manager Three ICs hold most of circuitry in battery-powered instrument; new gain cell controls sensitivity variation A new low—in size, that is—has been reached with the introduction of the model 211 portable oscilloscope from Tektronix. The 211 is a single-channel, laboratory-quality, 500-kilohertz, battery-operated scope that measures 3-by-5-by-9 inches and weighs about 2.5 pounds. The 211 is aimed at field maintenance applications where small size and low power drain are important. Vertical calibration is variable from 1 millivolt per division to 50 volts per division. Horizontal sweep rates are variable from 5 microseconds per division to 200 milliseconds per division, and a continuously variable sweep magnifier provides uncalibrated sweep rates to about 1 μs per division. CRT viewing area is 3 by 5 centimeters. One of the keys to the 211's small size is large scale integration. Most of the 211's circuitry is contained in three monolithic integrated circuits, all proprietary to Tektronix. According to David Allen, the designer of the 211, "two of the ICs contain 70% of the active circuitry—sort of like linear LSI." One IC contains the sweep circuit, the trigger circuit, and the horizontal amplifier, and is made up of both npn and pnp transistors. This device, says Allen, is manufactured with a compatible FET/bipolar process that is proprietary to Tektronix. Another IC contains both the vertical and horizontal output amplifiers. It provides the 4-milliamp vertical drive current and the 1-mA horizontal drive current required by the CRT. The third IC is a quad op amp in which two act as current sources and the other two are the vertical amplifier. According to Allen, the biggest problem in designing the ICs was in "squeezing in all of the components. When you are putting 104 transistors on a 65-by-65-mil chip, you run into signal path problems and parts placement problems. For example, during one of the early designs, we had the trigger interfering with the sweep circuit." As a result of this, about 90% of the silicon surface is used—there are even devices placed between the bonding pads. Power supply design was also an obstacle in reducing the scope's size. Allen says that the sensitivity of an oscilloscope is a function of the CRT intensity. In standard-size scopes, the power supplies are regulated twice—once at the input and once at the high-voltage output. But to save space in the 211, only input regulation (2 to 5%) is employed. To control the intensity/sensitivity variations, the 211 uses a gain cell in a feedback loop. This gain cell is a Tektronix development that's based on a circuit whose gain is dependent on the input current. In the 211, as the intensity control is turned up, the voltage goes down and the sensitivity goes up. "To correct this," says Allen, "we feed back high voltage through the gain cell." This approach is employed in both the horizontal and vertical circuits. The 211 will be available by the end of this year at a price of about $500. Tektronix Inc., Box 500, Beaverton, Ore. 97005 [338] On the spot. Portable scope weighing 2.5 pounds is designed for field maintenance jobs. Are quartz oscillators good enough for your communication system? Let's face it. Your system's performance is directly related to the quality of its frequency source. And that's why so many people are switching to atomic frequency standards. We say this although we make some of the best quartz oscillators on the market. But then we make the best alternative. The HP 5065A Rubidium Frequency Standard is 100 times more stable than quartz. For example, the long-term stability is $1 \times 10^{-11}$ per month, and the short-term stability is $5 \times 10^{-12}$ for one second averaging—the best guaranteed rubidium stability available. This means better accuracy over longer intervals between calibration and better signal to noise ratios. Then there's our 3-year Rubidium Frequency Vapor Reference warranty. This covers the operation of the entire module, not just the gas cell part which only goes wrong once in a blue moon anyway. So if you're working in color TV, navigation or communication systems, we can solve many of your worries. The 5065A, which costs $7500, has proved itself in the field for the past three years. Should your applications require even greater long-term stability, ask about our Cesium Beam Frequency Standard. Its long-term drift is virtually unmeasurable. In fact, we're the only company that can supply you with all three types of frequency source—quartz, rubidium and cesium. And your local HP field engineer is the one source for all your frequency needs. So give him a call. Or write us for complete information at Hewlett-Packard, Palo Alto, California 94304; Europe: 1217 Meyrin-Geneva, Switzerland. HEWLETT PACKARD FREQUENCY STANDARDS Circle 108 on reader service card Instruments Light meter is highly stable Compact, low-cost unit uses photometric filter and Schottky-barrier diode As the field of optoelectronics expands, the need for instruments to accurately measure light increases—especially for instruments that are highly stable, compact and portable. A Santa Monica, Calif., company, United Detector Technology, says it has such an instrument in its model 40A Opto-Meter. The unit solves the chief problems of those who have had to use bulky, difficult-to-calibrate instruments in this fledgling field, says Paul Wendland, president. His company has specialized in silicon photodiode detectors since starting in business five years ago, and such a detector is a key part of the model 40A. The unit can be hand-held and operates on power supplied by two internal 12-v mercury batteries. Bulkier instrumentation using photomultiplier tubes can't be built for less than $3,000, Wendland says, and would have to be calibrated daily against a National Bureau of Standards lamp. The model 40A is guaranteed not to drift more than 1% in six months in responsivity after its initial calibration against such a lamp standard by the manufacturer. The Opto-Meter sensor head incorporates a computer-selected photometric filter and a Schottky-barrier p-i-n photodiode combined with an operational amplifier. The photometric filter consists of four pieces of colored glass cemented together after being computer-selected to give the desired spectral response in the filter. The Schottky-barrier diode's response is nearly flat. Wendland says that at 4,500 angstroms, this means a response that's 80% of the peak response vs. about 1% for a photomultiplier tube. The model 40A measures both radiometric and photometric power and energy. The light power range is from 0.001 microwatt to 10 milliwatts in six ranges in the radiometric mode, and 0.001 foot candle to 10,000 ft. in the photometric mode. On the most sensitive scales, the Opto-Meter can resolve 10⁻⁴ watt and 10⁻³ ft. The instruments light energy range is from 0.001 microjoule to 1 millijoule in five ranges in the radiometric mode, and 0.001 lumen-candle-second to 1,000 ft-c-s in the photometric mode. The $495 price includes the sensor head and both the radiometric and photometric filters. Delivery is from stock. United Detector Technology, 1732 21st St., Santa Monica, Calif., 90404 [351] Digital wattmeter reads to 20 kW at 20 kHz Simplicity is the strong suit of the Model 235 wattmeter, made by Clarke-Hess Communications Research Corp. For one thing, it's simple to read—it has a digital display using cold-cathode tubes. But more important, it's easy to use. The unit has a wide frequency and power range. The 235 measures power from 2 watts to 20 kilowatts over a span of 20 hertz to 20 kilohertz. Such ranges usually require external circuitry before taking a reading. But with the 235, voltage and current leads from the load are attached to the instrument, a range is selected with a front-panel switch, and that's it. "There aren't 62 shunts or 62 dropping resistors to put in," points out company president Kenneth Clarke. The selectable ranges are delineated according to maximum voltage and current. The A setting permits a maximum rms input of 632 volts and 31.6 amperes. For B, it's 200 V and 10 A; for C, it's 63.2 V and 3.16 A; and for D, it's 20 V and 1 A. For all settings, the accuracy is ±1% of the full-scale reading for currents between 1 and 30 A. The 235 computes power with a solid state multiplier. At the instrument's input is a 30-A shunt in series with a variable-gain amplifier. It's this amplifier that's adjusted to select ranges. "We're still digging up applications," says Clarke of his instrument. Among those already found are measuring the output of choppers and silicon controlled rectifiers. Size of the 235 is 8½ by 10 by 11 inches, and weight is 13 pounds. It sells for $2,500. Clarke-Hess Communications Research Corp., 43 W. 16th St., New York, N.Y. 10011 [352] System quickly pinpoints communication-line faults Diagnostic system called Detect performs end-to-end testing of data networks. The system consists of a New products central station and individual units at remote locations in the network. Applications are in commercial end-user organizations and common carriers. It can be incorporated into any existing single- or multi-line network on a plug-in basis. The unit may be either desk- or wall-mounted. Data Products Corp., Telecommunications Div., 17 Amelia Pl., Stamford, Conn. 06904 [353] Oscilloscope has 5 mV/cm sensitivity at 60 MHz Model 1062 scope offers 60 MHz bandwidth at 50 mV/cm. A triggering system provides flat full-bandwidth sensitivity. Variable hold-off allows synchronous triggering on digital word lengths. Price for rack-mounted version R1062 is $2,045, and for 1063/R1063 without delayed sweep it is $1,845. Dumont Oscilloscope Laboratories Inc., 40 Fairfield Pl., West Caldwell, N.J. 07006 [356] Capacitance meter/ converter provides 4½-digit readout A capacitance meter/converter series 2450 measures high-Q capacitors with 0.2% of full scale accuracy. Single-capacitor or three-terminal differential measurements can be made in ranges of from 0 to 1,999.9 pF, to 0 to 1,999.9 pF. Drive signal is a 10 kilocycle square wave, and analog output is 1.9999 volts full scale. Price is $1,875. Spearhead Inc., 1401A Cedar Post Lane, Houston, Texas 77055 [354] Time interval meter gives resolution of 1 ms or 1 μs Time interval meter, model 355, has five-digit display, separate input channels and input signal conditioning. Resolution is 1 ms and 1 μs, switch selectable, and readout ranges from 1 μs through 99,999 s. An internal crystal-controlled oscillator determines accuracy. Price is $450 and an optional computer printer interface is available for $100. Eldorado Electrodata Corp., 601 Chalomar Rd., Concord, Calif. 94520 [357] Digital voltmeter offers resolution of 100 μV DigiTec model 266 digital voltmeter features 4½ digits, and an accuracy of 0.02% of reading. Indicators are LED displays, and front panel controls include self-check zero and calibration to assure maximum accuracy. Other features are guarded input, isolated BCD and system functions, 100 μV resolution, and voltage measurement to 1,000 V dc. Price is $525. Delivery time is up to eight weeks after receipt of order. United Systems Corp., 918 Woodley Rd., Dayton, Ohio 45403 [355] Non-reset counters designed for low-cost metering jobs Series 8Z9 counters go up to 99,999 and then repeat. Applications are in metering or timing of business machines, photo equipment, and packaging and production machinery. Unit runs at 500 counts per minute and has an operating life of five million counts at 10 unit counts per revolution. Price of the model 8Z9 is $4.50. Stock Drive Products, Div. of Designatronics Inc., 55 S. Denton Ave., New Hyde Park, N.Y. 11040 [360] Portable photometer aimed at process control Model 3100 photometer calibrates illumination and brightness for inspection of photocells and phototransistors, can also function as a process control instrument and as an edge sensor. Unit is available with four probes having a sensitivity to 0.0002 foot-candle. Price of the photometer is $295. Vactec Inc., 2423 Northline Ind. Blvd., Maryland Heights, Mo. 63043 [358] 7-channel monitor scope fits in 3½-inch-high rack Designed around linear and digital integrated circuits, monitor oscilloscope system includes seven independent oscilloscopes using a 1-in. by 3-in. CRT, packaged in a 3½-in.-high rack mount configuration. Features include no vertical drift from 0 to 55 C, bandwidth from dc to 10 MHz, and sensitivity from 50 mV/in. to 10 V/in. Automatic triggering is on a 5-mV signal. Primary applications include calibration, real time data monitoring, and data playback for wideband analog tape recorders in pcm, fm, direct, and pdm record modes. Price is $2,795. California Instruments Co., 5150 Convoy St., San Diego, Calif. 92111 [359] Up to 200 hits/minute... Up to 6 stacks of p.c. boards at a time... Up to 20" x 24" board size... Up to 12 different drill sizes/setup. ...The QUADRAMATIC MARK II is up to it! TOOLING UP for p.c. board drilling need not rely on your skill with a crystal ball. It's true that there is no predicting just what your circuit designers will need next year... even next week... but you can anticipate a need for fast drilling of clean holes by the thousands or millions. That's where the QUADRAMATIC MARK II comes in. It's a numerically controlled, granite framed, very accurate, very fast p.c. drilling machine that can simultaneously drill 0.0135" to 0.250" holes in four stacks of 20" X 12.45" circuit boards... or three stacks 20" X 16.60" ... or two stacks 20" X 24". And with the optional Excellon automatic drill changer, the MARK II can drill up to twelve different hole sizes per spindle with the same setup; or it can exchange cull drills for sharp ones automatically; or both! With four boards/stack, and four stacks, a $5/hour operator can deliver up to 330 holes for one penny of labor cost! That includes multiple hole sizes. And that means versatility plus economy... which is what the QUADRAMATIC MARK II is all about. Get the rest of the facts from this brochure from Excellon, the recognized leader in p.c. drilling. Where variety is the problem, versatility is the answer EXCELLON INDUSTRIES 23915 GARNIER / TORRANCE, CALIFORNIA 90505 Phone: (213) 325-8000 Telex: 674562 — Cable: EXCELLON Torrance ENGLAND • Excellon International • Arkwright 18, Astmorp Indus.; al Estate Runcorn • Cheshire, England • Phone: Runcorn 66545 • Telex 62749 FRANCE/BELGIUM • Electra Outillage • 31, Rue du Gouverneur-General Eboue • Issy-Les-Moulineaux (Seine) • Paris • Phone: 46-30 • Telex 20168 SCANDINAVIA • Edsby AB • Schenck • Malmenkalladatan • Stockholm 3, Sweden • Phone: 23 24 201 • Telex 17434 THE NETHERLANDS • Excellon Europa N.V. • Vijhuizerdijk 15b, Vijhuizen • Netherlands • Phone: 02504-156 • Telex: 178 • ITALY • Excellon Europa GmbH • Via S. Giovanni 30/38 • Villetta b. Frankfurt • Alle Strasse 37 • Switzerland • Phone: 061 93-85-7 • Telex: 41011 ITALY • France & Portugal • Via Mauro Macchi, 70-20124 Milano, Italy • Phone: 2042245 SPAIN • Antonio Casas, S.A. • Carretera Sta. Cruz de Calafell Km. 10 Aoañado de Correos (P.O. Box) No. 16 • San Baudilio de Llobregat • Barcelona, Spain • Phone: 204-2245 • Cable: TALVITKO AUSTRALASIA • Tungsten Carbide Tooling Pty Ltd. • 3 Expo Court Mount Waverley, Victoria 3149, Australia • Phone: 544-9700 FAR EAST • Nippon S.T. Johnson Sales Co., Ltd. • No. 3-5, Uchiranda-3 Chome Chiyoda-Ku • Tokyo, Japan • Phone: Tokyo (252) 1251-65 Telex: 2222307 • Yuan Kong Enterprises Ltd. • P. O. Box 22687 • Taipei, Taiwan • Phone: 338833 The China Engineers, Ltd. • Alexandra House Pm. 250 • Hong Kong • Phone: 230181 • Telex: HK-3297 How can Philips guarantee the performance of its RF power transistors for mobile transmitters? Yes...the word is guarantee. No more wondering "50 watt is 50 what?" With Philips RF power transistors for mobile transmitters, you know precisely what you're ordering. You get precisely what you order. If used as specified, it will not fail...or your money back! How is this possible? Not as difficult as it seems. It starts, of course, with a product of superior reliability. But the real secret is in the specifying. We provide you with a set of SOAR (Safe Operating Area) curves setting forth, in detail, at exactly what combination of VSWR, temperature and supply voltage variations the transistor will operate correctly. When we say "25 watt", we mean, for example, "25 watt performance, with a 70°C heat sink, withstanding a VSWR of 50, 20% overvoltage and an increase of 20% drive power". Rather than forcing you to build in your own safety margin...with the probability you will over-specify and over-spend to be on the safe side...we determine all the derating parameters for you. The specification is part of the product...and we stand behind it. The full range of Philips RF power transistors for mobile transmitting equipment...civil, military, aerospace or maritime...is shown in the accompanying table. Write for full specifications. \| Pout [Watt] \| SSB 30 MHz \| VHF 175 MHz FM \| UHF 470 MHz FM \| \|------------\|-------------\|----------------\|----------------\| \| 0.5 \| \| \| \| \| 1.0 \| 2N4427 \| \| \| \| 1.5 \| \| 2N3866 \| \| \| 2.0 \| \| \| \| \| 2.5 \| \| \| \| \| 3.0 \| 2N3924 \| \| \| \| 4.0 \| BES22A \| BLS23A \| \| \| 6.0 \| \| 2N3375 \| \| \| 7.0 \| \| \| \| \| 8.0 \| BLX13 \| BLY87A \| BLY91A \| \| 12.0 \| \| 2N3927 \| \| \| 13.0 \| \| \| \| \| 15.0 \| \| \| \| \| 20.0 \| \| \| \| \| 25.0 \| \| \| \| \| 40.0 \| \| \| \| \| 50.0 \| BLX14 \| BLY90 \| BLY94 \| \| 100.0 \| BLX15 \| \| \| N.V. Philips Gloeilampenfabrieken Electronic Components and Materials Division Eindhoven, The Netherlands. Distributed and sold in the U.S. by: Ampex Electronic Corporation Semiconductor and Microcircuit's division Providence Pike, SLATERSVILLE R.I. 028/6 In Canada: Philips Electron Devices a division of Philips Electronics Industries Ltd. 118, Vanderhoof Avenue TORONTO 17 - Ontario PHILIPS 112 Circle 112 on reader service card Electronics/October 11, 1971 Controller is multi-purpose Asynchronous serial model handles CRTs, printers, cassettes for Varian 620 One of the problems a minicomputer manufacturer faces is that of designing custom interfaces for the myriad peripheral devices offered with the machines. It can cost $5,000 to $10,000 to develop such a custom controller, and it's usually expensive to produce because it has to be a wire-wrapped rather than a printed, flow-soldered board. Engineers at Varian Data Machines have attacked this problem with a universal asynchronous serial controller that interfaces any Varian 620 minicomputer to any peripheral device using an asynchronous serial interface. Thus the range covered is from cathode ray tubes through serially connected printers to tape cassettes and teletypewriter terminals. James Orris, director of product management, says that 40% of his firm's business is in peripheral devices—and it's growing, so Varian wants to cut out that custom interface development cost and be able to mass-produce the controllers. The trend among peripheral equipment makers to furnish serial interfaces with their equipment has enabled Varian to come up with the model E-2184, a versatile serial controller that will be available on one flow-soldered circuit board to plug into any Varian 620 computer. There are three versions of the unit, ranging in data rate from 40 to 10,000 bits per second, and with five, six, seven, or eight bits with parity. Cable lengths from 20 feet to one mile are offered. A buffered CRT would require the 10,000 b/s rate, as would interfacing with another computer, Orris says; a peripheral device such as a card punch would require the 40 b/s rate. But Orris adds that printers are probably the biggest market for the serial controller. Varian usually includes two printers or two CRT units in its standard price list for the 620 computer line. The model E-2184 will allow the firm to offer a wider range with far less interface development cost. The controller board has rear edge connectors and requires only one input/output slot in the central processor mainframe or I/O expansion chassis. All versions are serial, direct-connect, character-buffered units capable of half or full duplex operation. Existing teletypewriter command set is utilized. Operation can be either under program control or in an interrupt mode using a priority interrupt module option. Another option permits automatic block transfers in conjunction with a buffer interface controller. A typical model E-2184 will sell for approximately $600. Delivery time for the controller is 30 days after receipt of order. Varian Data Machines, 2722 Michelson Dr., Irvine, Calif. 92664 [361] Portable cassette recorder designed for remote input In many business and record-keeping operations, data is entered during the day via inexpensive, slow-speed teleprinter, and at night the remote terminals are polled automatically at high speed by a central computer. This is the type of application that International Computer Products Inc. has in mind for its TermiCette 3300, a portable unit that is truly incremental in record and playback. It permits corrections and editing, even deletion of characters, at any location on the tape. The recorder uses a bit-mark sequencing technique that permits the recorder to sense the beginning and end of each character, and position the tape accurately for recording or corrections. It operates at 10, 15 or 30 characters per second for compatibility with data terminals, and can record at the 10 c/s of a conventional teleprinter, then transmit at higher speed over the telephone lines, or to a higher-speed printer or a terminal with cathode ray tube display. It can be used with wired or acoustically coupled terminals, and simply plugs into the terminal through the standard EIA RS-232-C interface. A current loop is available as an option. The standard TermiCette is controlled from the front panel switches, but a remote-control option permits control either from the keyboard of the terminal, or over a line from the central computer. Each 300-foot cassette can hold approximately 50,000 characters, and the error rate is less than one per million. Fast-forward operates at 400 characters per second, and fast-reverse at 100 c/s. A feature of the TermiCette is easy manual search for data. Pressing the standby button (or hitting the proper key in remote operation) puts a six-second space in the tape, and this can be located in fast-forward or reverse by watching an indicator that lights when data is being recorded or transmitted over the system. The 3300 is packaged in a carrying case with protective cover. It is 16$\frac{3}{4}$ inches deep, 10$\frac{1}{2}$ inches wide and 6$\frac{1}{2}$ inches high, and weighs 18 pounds. It requires 75 watts at 115 volts. Price is $1,650 in unit quantity, with discounts available for voOur Reconstituted Mica Capacitors Are Tested Harder You get higher capacitor reliability for less Because Custom Electronics, Inc., is the reconstituted mica capacitor maker who precisely grades the dielectric material before it reaches production, other manufacturers' final mistakes are never started at Custom. The result is fewer production rejects, and lower job quotes for Custom Electronics' customers. Add this grading system to Custom Electronics' rather fanatical Quality Control efforts throughout the production process, and it becomes clear why our client list reads like the who's who of high voltage electronics. If your high voltage capacitor needs are RELIABILITY and IMMEDIATE SERVICE write or call Custom Electronics, Inc., Browne St., Oneonta, N.Y. 13820. Phone 607-432-3880. TWX 510-241-8292 The QC Fanatics CUSTOM ELECTRONICS, Inc. New products lume users. Delivery time is 30 days after receipt of order. International Computer Products, Inc. P.O. Box 34484, Dallas, Texas 75234 [362] Tape perforator handles 120 characters per second Model P-1200 tape perforator operates at 120 characters per second, using punch pins that extend mechanism life by reducing the force necessary to perforate heavy-duty tape materials. Other features include synchronous operation, five-through eight-level operation, and a capstan drive that requires no adjustment. Price is $1,500, with quantity discounts available. Tally Corp., 8301 S. 180th St., Kent, Wash. 98031 [363] Desktop display terminals offer 1200- or 2400-baud rate Interactive display terminals, series 45, have detachable keyboards with 64 characters, 32 control codes, 17 functioning keys, and 12 editing keys. Models 80 and 84 are fully buffered, with 12-in. screens and switch-selectable data rates of 1,200 baud asynchronous or 2,400 baud synchronous. The model 80 displays 1,000 characters in 25 lines of 40 characters and the 84 displays 960 characters in a 12-by-80 format. Price of the desktop-size units is $3,430 to $4,650 depending on quantity. Leases are available. Photophysics Inc., 1601 Sterling Rd., Mountain View, Calif. 94040 [364] 80-column-card reader processes 500 per minute The capability of plugging into the IBM System/3 computer is offered by the model 8063 card reader designed for 80-column-card reading at 500 per minute. Life expectancy is more than 10 years or 20,000 hours, read error rate is less than one in $300 \times 10^6$ data bits, and each column of data is individually synchronized with a strobe signal. Price is less than $5,000, and leasing for a 36-month period costs about $200 per month. Bridge Data Products Inc., 738 S. 42nd St., Philadelphia, Pa. 19104 [365] Graphics display terminal aimed at business uses A business-oriented graphics display terminal, the model 4010, is compatible with more than 20 timesharing systems, more than 20 minicomputers, and with IBM 360/370 systems. A software package, Plot-10, is provided. The CRT terminal offers alphanumeric input to the computer by a teletypewriter, and graphic input by a dual thumbwheel arrangement on the keyboard. Price is $3,950 or $3,400 for 20 or more. Leasing costs $200 a month, includData coupler provides link to switched networks Data coupler, model EDC1001A, connects customer-provided automatic data equipment to the switched telecommunications network for data and voice communications. The unit has a voltage interface for use with terminal equipment that cannot tolerate contact bounce and electromagnetic radiation. The EDC1001A provides automatic linear control of signal levels above a specified threshold. Elgin Electronics Inc., Wainut St., Waterford, Pa. 16441 [367] Remote batch terminals feed into any computer Designed to communicate with any computer, the series 120 remote batch terminals are multiprogrammable. Models 123 and 124 use a 4,096-word core memory, a 300-card-per-minute reader, and a 135- or 600-line-per-minute printer. The model 125 offers a standard 4k memory, two magnetic tape cassettes, an IBM 735 Selectric typewriter, and either a synchronous or asynchronous controller. The 125 can also act as a stand-alone computer when it is not being used online. Eldorado Electrodata Corp., 601 Chalmar Rd., Concord, Calif. 94520 [368] Here they are: the best buys. Our new Series 72 single-turn $ \frac{3}{8}'' $ cermet trimmers at an eye-opening 49¢ each in the 1,000 quantity. They're equally well-priced in other quantities, too. One alone costs just 70¢. But there's also that dependable Helipot quality and performance, plus things you don't get elsewhere. Like our resistance and slider stability ... change is less than 0.5% in the first 1,000 operation hours, progressively less thereafter. Impressive, when the life expectancy exceeds five years' continuous duty under full-rated power. And the Series 72 sealing — compatible with wave soldering, low-viscosity potting, and board washing even with chlorinated solvents. Add the wide range of values, \( \pm 100 \text{ ppm/}^\circ\text{C tempco, and off-the-shelf availability from 17 nationwide stocking points and what have you got? The best reasons in the world for requesting complete specs and data on this breakthrough line of Helipot cermet trimmers. Do it now — it won't cost you a cent. Packaging and production Board tester isolates faults Computerized unit runs variety of checks on digital, analog and hybrid modules At first glance, test systems for LSI circuits and those for circuit boards and assemblies are similar. But there's an important difference: you throw away a bad IC, but generally fix a defective module. And the problem of isolating a specific bad part on a board can be even more difficult than finding out whether the complex logic board is bad. The equipment group of Texas Instruments, which markets transistors and IC testers developed for TI's components group, has now begun selling an assembly tester developed for its own use. The ATS-960 (it uses a TI 960 computer) performs functional, ac and dc parametric and special tests on analog, digital or hybrid modules. Results can be in go/no-go form or actual measurements, and the system can also be used for adjustments to bring the module into specification. For the latter, a CRT display can tell the operator "turn potentiometer R15 clockwise 15 turns, then counterclockwise until the pass light appears." The ATS-960 can be used with up to four independent test stations; each station can be running tests on different modules. The maximum number of pins per station is 256. Any commercially available or special programmable test instrument can be used. The system can even be bought without any test instruments, for users who already have their own or wish to make voltmeters, counters, function generators and pulse measurement systems. TI supplies special forcing/measuring power supplies rated at up to 100 volts or 1 ampere, a digital functional test unit, and analog and digital fault isolation units. Up to 16 low-frequency, eight high-frequency, and 32 sampling probes can be used. For fault isolation on analog boards, clip-on probes are employed with special diagnostic routines. For digital assemblies, two techniques are available: the conventional approach using computer-generated diagnostic routines is most appropriate for high-volume boards where the considerable cost of program generation can be amortized. An alternate technique developed by TI utilizes a universal program with a multi-pin IC probe. The ATS-960 test system is priced at $50,000 to $150,000. Texas Instruments Equipment Group, Box 1444, Houston, Texas 77001 [391] Computer logic helps solve wire identification problems Identifying a wire in a jungle of cables has always been a hair-pulling experience. Now the W.H. Brady Co., Milwaukee, Wis., promises to make the job easier with a system called the Pathfinder. The machine locates and identifies wires by using computer logic; coded pulses sent through the cables are picked up by a display and instantly translated into a digital number. Capabilities include locating random wires, specific wires among any number, and shorts or selected terminals on all types of connector assemblies. The Pathfinder eliminates the use of multicolored wire, as well as assembly errors, and it reduces inspection time. In addition, the unit will detect crossed wires, incorrect wires, high and low resistance shorts, and open circuits. The Pathfinder is capable of coding up to 999 wires and will identify pulses up to two miles away from the encoder unit, using a common-ground principle. The unit will also trace and identify telephone circuits, and complex missile and aircraft cabling. Users can design their own systems by selecting the necessary components to accomplish specific requirements in wire identification. Two basic versions of the Pathfinder are available. The model 99 is for assemblies, harnesses, or wiring systems requiring as many as 99 leads; and the model 999, a high-capacity unit, offers a minimum of 199 pulse signals, with capabilities up to 999. W.H. Brady Co., 727 W. Glendale Ave., Milwaukee, Wis. 53201 [392] Ultrasonic bonder can shift to thermocompression type Model NU-822 wire bonder is for hybrid circuit, semiconductor and IC production. It can be converted from cold ultrasonic ball bonding to thermocompression bonding by the operator. Work height range is ½ New products in., and a manual control permits multilevel bonding without search height adjustment. Mech-EI Industries Inc., 73 Pine St., Woburn, Mass. 01801 [395] DIP cartridge has its own heat sink and rfi shielding IC packaging system, designated Dipstik, eliminates soldering, pc boards, card files, extractors, extenders, and related hardware. Instead, a cartridge and receptacle will hold up to five dual in-line ICs, providing its own heat sink and emi/rfi shielding. Receptacle provides standard 0.025-in. wirewrap pins for interconnecting logic wiring, while the cartridge with feed-through terminals provides test or tie points. Price is about $1 per position. SAE Advanced Packaging, 2165 S. Grand Ave., Santa Ana, Calif. 92705 [393] Pc terminal strengthens DIP-to-board connections Printed circuit terminals, model 1938-8, are available in chain form or precut strips to fit specific IC requirements. The DIP connector has a grid contact arrangement, and holds typical on-center spacing of 0.100 inch between terminals of DIPS. Molex Inc., 5224 Katrine Ave., Downers Grove, Ill. 60515 [398] Matrix board simplifies data, control programing Matrix programing board is designed for computer, machine tool control, data handling, and processing applications. Program preselection is executed by connecting the X-Y coordinates by means of a shorting pin or diode plug. The board measures 20 by 30 by 4.5 mm, and has 4.5-mm spacing between centerlines of holes. A special tool is not required for this 4.5-mm configuration. Price is $66 in quantities of 50. Delivery is from stock. Interswitch, 770 Airport Blvd., Burlingame, Calif. 94010 [399] Desoldering tool uses jet of high-velocity air Desoldering tools, models GSS and ESS, are self-cleaning and are designed for one-hand operation. The work is done by a jet of high-velocity air, controlled by a finger valve, moving past a hollow heated tip. Air jet creates a suction which entrains solder particles and catches them in a receptacle. Price is $21.95. Hunter Associates, 792 Partridge Dr., Somerville, N.J. [394] Wide-cavity MOS packages accommodate larger chips MOS packages with 28- and 40-lead composite housings are available with 0.240-sq.-in. cavities, 0.040 in. deep with a 0.050-in.-wide gold-plated seal ring. The housings are designed with the leads brazed to the bottom of the package, allowing the cavity to be expanded without extensive modification. Price is from Avery uncomplicated new OEM recorder with just one thing going for it... You'll like what you see in our new approach to dedicated OEM strip chart recorders. First, we eliminated all those complicated moving parts from the writing mechanisms. No more pulleys, cables and slip clutches. Instead, there's just one simple moving part—the slider/pen assembly. That's because a linear servo motor keeps the pen going magnetically...and very reliably. When you see the HP Model 7123, you'll notice how the low power servo system makes the recorder smooth, precise and trouble-free. You could drive it off scale around the clock without noise or danger. Even with all that, you've got a lot more going for you with the 7123. Like a swing-out chart paper drive for quick reloading and reinking. The viewing writing area is slanted so you can make notes right at the disposable pen tip. And you can work without worrying about a lot of circuit adjustments. They're simply not needed anymore. Since it's an OEM machine from the ground up, the 7123 has options for everybody. Select any chart speed and voltage span in English or Metric scaling. In all, nearly 50 options will customize the recorder exactly to a specific application. You'll probably be most intrigued by an option we call electric writing. Normally, the ink system works like a cartridge fountain pen. But electric writing is designed for people who don't even want to mess around with that. A highly stable electro-sensitive paper that gives you a crisp, clear trace without ink. Available in full rack or half rack versions, the 3½ inch high 7123 makes totally unattended operation a reality. Simplicity, reliability, precision and even electric writing. With all that going for you, you can turn it on Friday and forget about your work all weekend. To see the uncomplicated new 7123 and its matching price and OEM discount schedule, call your nearest HP sales office. Or write, Hewlett-Packard, Palo Alto, CA 94304; Europe: 1217 Meyrin-Geneva, Switzerland. HEWLETT PACKARD GRAPHIC RECORDERS Electronics/October 11, 1971 Talk to Gudebrod about your tying operation this month... And about Lacing Tapes, harness rooms and systems. About temperature and vibration... speed and rejects! About Nylon, Dacron, Teflon, Nomex, Glass tapes and cords—treated and untreated... that meet or exceed military and industrial specifications, about cost comparisons with other methods... and all backed up with one hundred years of manufacturing knowledge. Dacron, Teflon, Nomex—Du Pont Registered Trade Marks Talk to Gudebrod this month! Write to this address for prompt return of our Product Data Catalog. Electronics Division Dept. 8310 Gudebrod Bros. Silk Co., Inc. 12 South 12th Street, Philadelphia, Pa. 19107 Circle 120 on reader service card New lab-quality instrumentation at famous HEATHKIT do-it-yourself savings! (A) New Heathkit solid-state digital multimeter. A true lab-grade digital multimeter at about half the cost of comparable DMMs! 5 overlapping DC voltage ranges, 100 uV to 1000 V; AC ranges, 100 uV to 100 V; 10 ranges for resistance, from 2 ohms to 2 amperes on AC or DC; 6 resistance ranges, 0 ohm to 20 megohms. Automatic polarity indicators. Automatic decimal point, over-range light. Precision DC calibrator furnished plus transfer method for AC calibration. Comes with two 4-digit readout tubes & "memory" circuit for stable, non-blinking indication. Assembles in about 10 hours. Kit IM-102, 9 lbs. ... $29.95* New Heathkit 175 MHz scaler & counter combo...for less than $300! The most versatile dual counter from 1 Hz to over 15 MHz. Hz/kHz ranges & over-range indicators let you make an 8-digit measurement down to the last Hz. 5-digit cold-cathode readouts, extremely low input impedance, less than 10 mV. 15 MHz; all solid-state with 26 ICs, 8 transistors. (C) Heathkit IB-102 scaler extender capable of counting pulses at a rate far below a 175 MHz counter. Compatible with virtually any counter. 10:1 & 100:1 scaling will give resolution down to 10 Hz...1:1 ratio provides high-through counting. Solid-state, fully regulated supplies. Carrying handle/tilt stand. Kit IB-101, 7 lbs. ... $199.95* Kit IB-102, 7 lbs. ... $99.95* (O) New Heathkit DC-15 Mhz dual trace solid-state oscilloscope...$399.95* 10:1 or 100:1 rock-solid traverses sweep full bandwidth in both automatic & normal modes; complete dual trace capability — Ch. 1, Ch. 2, chop, alternate; X-Y mode with 5% or less phase shift; 24 nsec rise time; 9-position 1, 2, 5 sequence vertical attenuator; 8 x 10 cm rectangular flat-face CRT with mu-metal shield. Kit OC-105, 15 lbs. ... $399.95* HEATH COMPANY, Dept. 520-28 Benton Harbor, Michigan 49022 ☐ Rush my new Heathkit Catalog ☐ Enclosed is $._______________________, plus shipping. Please send model(s)_____________________ Name ____________________________________ Address __________________________________ City __________________ State ______ Zip ______ Prices & specifications subject to change without notice. Mail order prices; F.O.B. factory. New products $1.07 to 97 cents for 28-lead packages, depending on quantity, and from $1.30 to $1.18 for 40-lead packages. They are available within six weeks of receipt of order. 3M Co., St. Paul, Minn. 55101 [396] Circuit card connector uses low-withdrawal-force contact Two-piece metal-to-metal pc card connector features a contact with a withdrawal force of one to six ounces. Contact is designed to provide low insertion and extraction forces, permitting mating of large numbers of contacts without mechanical jacks, screws, or cams. Called series 8219, connector is a high-density unit with contacts spaced on 0.050-in. centers. Connector sizes are 18, 30, 36, 42, 54 and 72 contacts. Elco Corp., Willow Grove, Pa. 19090 [400] DIP handler feeds and sorts up to 4,000 units per hour Integrated circuit handler, model IC 2500A, accepts dual in-line packages having from eight to 24 leads. It can handle up to 4,000 devices per hour in general-purpose applications of feeding and sorting DIPS, such as digital and linear integrated circuits, pulse transformers, and resistor networks. The unit accepts most commercially available shipping magazines and interfaces with all test systems using TTL or DTL levels. Test leads are brought out in short wire lengths to connector, resulting in low capacitance. Price is $6,500. Computest Corp., 3 Computer Dr., Cherry Hill, N.J. [397] INVITATION TO DEMONSTRATE AN AUTOMATIC VEHICLE IDENTIFICATION SYSTEM FOR BUSES All present or prospective suppliers of Automatic Vehicle Identification (AVI) systems are invited to demonstrate the performance of their equipment in a preliminary screening test any time during 1971. Successful systems will provide transponders to be mounted on buses, and capable of transmitting at least 9,999 individual codes as the vehicle is moving at least forty miles per hour, to an interrogator located on or near the particular lane in which the bus is traveling. Systems which demonstrate this capability in the approximately two-hour screening test involving temporary mounting of transponders on three buses and temporary installation of an interrogator and appear promising with respect to final installed costs and benefits, will be invited to participate in a full scale test. Lasting at least six months, the full test will involve approximately forty buses each equipped with at least one transponder from each participating supplier. At least one interrogator from each supplier will be installed. Signals from each bus will be transmitted to a computer at the Lincoln Tunnel, which will be checking performance of the AVI systems and interpreting the AVI information to determine the actual travel time of individual buses through the corridor. Transponders and interrogators are to be supplied and maintained without cost to the test. Costs for installing transponders and interrogators, transmitting signals, computer processing, evaluating and reporting the performance of the candidate AVI systems will be borne by the project. The United States Department of Transportation has authorized the test to evaluate the benefits and costs of an Automatic Vehicle Identification (AVI) system for buses using the I-495 corridor between the New Jersey Turnpike and the Port Authority Bus Terminal. Participating agencies include the Tri-State Regional Planning Commission, The Port of New York Authority, the New Jersey Department of Transportation, the New Jersey Turnpike Authority, and Transport of New Jersey (the major operator of interstate buses in the New Jersey-New York area). If the full test is successful, recommendations will be made to USDOT for equipping most, if not all, buses using the Port Authority Bus Terminal with an AVI system. Installation of a system of this magnitude is likely to set the pattern for more widespread introduction of AVI for such purposes as fleet control by truck and other operators, automatic toll collection, motor vehicle administration, etc. Interested suppliers may obtain a copy of "Feasibility Study-Automatic Vehicle Identification Systems" from the Clearinghouse for Federal Scientific and Technical Information, Springfield, Virginia 22151. The Reference No. is PB 185387, and the cost is $3. This study was performed for the United States Department of Transportation to determine the feasibility of adopting a national standard for an AVI system. Prospective suppliers desiring further information are invited to contact Mr. Robert S. Foote; Manager, Tunnels and Bridges Research Division; The Port of New York Authority; 111 Eighth Avenue, New York, New York 10011; telephone (212) 620-8144. Remember... MAIL MOVES THE COUNTRY but... ZIP CODE MOVES THE MAIL Now—complete coverage of Electronic Measurement—in 1 Vol. Electronic Measurement and Instrumentation* by Bernard M. Oliver and John M. Cage 720 pp. $29.50 This is the most complete current work on electronic measurement available. It reviews the scientific philosophy of measurement, discusses drawbacks and opportunities of network theory, and details the latest information on instruments and instrument systems . . . oscillography . . . optimization of digital techniques for signal analysis . . . engineering noise measurement . . . theoretical bases of network applications to measurement . . . and topics of similar engineering importance. Computer techniques for practicing engineers Application of Computers to Engineering Analysis by John R. Wolberg 288 pp. $14.50 This new book tells how to make effective use of the computer in engineering analysis and outlines the principles of developing mathematical models from physical statements to computer models. It contains useful numerical techniques, practical aids, methods for evaluating program quality, algorithms and adaptations of new methods, and case studies of real engineering problems. It offers a systematic approach to the kinds of computer analysis problems engineers regularly encounter and explains advanced concepts and techniques that can otherwise be picked up only through experience. A pocket-sized guide to 1500 essential engineering formulas Engineering Formulas by Kurt Giech 384 pp. $8.50 Now you can have at your fingertips the engineering formulas you use so often. This unique, reliable, practical pocket book brings together over 1500 formulas and 300 diagrams to simplify engineering calculations and is printed on one side of each page to provide room for your notes. It presents tables of measurement, areas, solid bodies, trigonometry, analytical geometry, differential and integral calculus and statistics and provides formulas of kinematics, dynamics, hydraulics, heat strength, machine parts, electrical engineering, optics, and chemistry. Now available in book form—ELECTRONICS' most significant articles on computer memory technology Electronics Computer Memory Technology Edited by Wallace B. Riley 263 pp. $14.50 Written by outstanding authorities, these articles explain how the various forms of computer memories work, how similar ones vary from each other, which systems may be used where and industries are using them and what systems are on the planning boards. Ferrite cores, thin films, MOS and other semi-conductor memories, read-only memories, electro-mechanical and acoustic memories, electro-optic and opto-electronic forms, sonic films, cryoelectrics, ferroelectrics, memory matrices, magnetic bubble memories, ways to cut costs, and making memories stand up under severe military applications—all these and many more areas are covered. Also of special interest to ELECTRONICS readers Organizing for Innovation—A Systems Approach to Technical Management—by J. A. Morton 192 pp. $11.50 A highly successful, internationally known engineer-manager shows how to motivate and integrate creative personnel in technological organizations. Intrinsic Safety—The Safe Use of Electronics in Hazardous Locations—by R. J. Redding, F.I.E.E. 192 pp. $12.50 The world's foremost authority on intrinsically safe design shows how safety is built into basic design without performance or convenience loss and within budget. --- Send me the book(s) checked below on 10 days' approval. At the end of that time I will remit price plus tax, postage and handling or return book(s) postpaid. This offer valid only in U.S. and subject to acceptance by McGraw-Hill. - [ ] Electronic Measurement and Instrumentation (478500) - [ ] Application of Computers to Engineering Analysis (713900) - [ ] Engineering Formulas (232024) - [ ] Electronics Computer Memory Technology (529159) - [ ] Organizing for Innovation (434204) - [ ] Intrinsic Safety (942242) Name ____________________________________________ Address __________________________________________ City ________________________ State ______________ Zip ___________ McGraw-Hill Book Co., Dept. 23-EL-1011 330 W. 42nd St., New York, N.Y. 10036 Circle 122 on reader service card TI adds three Schottky TTLs High-speed MSI units include arithmetic logic circuit with 15-nanosecond add time An easy way for designers to upgrade the speed of their conventional transistor-transistor-logic equipment is to use Schottky-clamped TTL, but not all popular functions are available in STTL. Texas Instruments, which produces the only generally available STTL line, has now expanded its family with high-speed versions of three popular TTL MSI circuits. The three provide speeds previously obtainable only with emitter-coupled logic, which can not be conveniently added to a TTL system. First in the trio is the SN54S/74S181 arithmetic logic unit, the Schottky version of the SN54/74181. This MSI function has proved a great attraction to users, for it can save many packages; Motorola has even taken the number (181) for an ECL part performing the same function even though it isn't a plug-in replacement. The S181 performs 16 binary arithmetic manipulations on two 4-bit words, including add, subtract, decrement, and direct transfer. Typical add/subtract time is only 15 nanoseconds (the conventional part requires 24). The 181 also performs 16 logic functions of two Boolean variables, including NAND, AND, NOR, OR, and exclusive-OR. Power dissipation of the 54S/74S181 is about 700 milliwatts, or 10 mW per gate, and it operates from the standard 5-v TTL supply. The plastic DIP, commercial range 74S181 is priced at $26 in 100-piece quantities, and delivery is 2 to 4 weeks. Military range units and ceramic DIP and flat packs are also available. The SN54S/74S153 is a dual 4-to-1 multiplexer, the Schottky version and plug-in replacement for the SN54/75153. It can be used to convert serial to parallel data or to randomly select data from a number of sources. Typical delay time from the data input is 6 ns, compared to 14 for the standard 153. Power dissipation is typically 225 mW. Price is $6.85 for the plastic version in quantities of 100. The third new Schottky part is a 110-MHz dual type D flip-flop, the SN54S/74S74, a plug-in replacement for the standard 7474 and high-speed 74H74 flip-flops. Up to now, TTL was limited to about 40 MHz input. Each half of the unit features a typical propagation delay of 7 ns. This part, unlike ECL versions, is a true dc-coupled flip-flop rather than a tracky ac type. Power dissipation is typically 75 mW per flip-flop. The least expensive version, the SN74S74 in plastic, costs $3.01 in quantities of 100. Texas Instruments Inc., Inquiry Answering Service, Box 5012, M/S 308, Dallas, Texas 75222 [411] Optical switch provides 50 ns square-wave rise time Optical switch, model os-391s-060 or -200, offers Schmitt trigger circuitry in a device using LEDs and phototransistors. This circuitry allows the switch to directly drive DTL and TTL levels. Typical square-wave rise time is 50 ns, and the trigger permits clean logic switching and wave shaping with no oscillations in response to slow mechanical action through the optical switch. Price is $12.90 in 100 lots. HEI Inc., Jonathon Industrial Center, Chaska, Minn. 55318 [417] Read-only memories are field-alterable Off-the-shelf read-only memory units can be supplied after customer specifies desired data contents in punched tape, card, or tabulated listing formats. Bit capacities range from 8 k to over 200 k per system, and a variety of basic word/bit configurations are available, many of them alterable to give additional formats by changing the input code and drive matrix circuitry. Price is 0.8 to 2 cents per bit depending on speed, capacity, and quantity. Datapac Inc., 3180 Redhill Ave., Costa Mesa, Calif. 92627 [419] LED display and decoder included in plug-in package Light-emitting diode displays called GaAsLites include a digital display-module, solid state indicator lamps, and a display bezel assembly. The display-module combines a seven-segment display, current limiting resistors, and a decoder/driver into one plug-in package. Price is $21 to $12 depending on quantity. Monsanto Co., 277 Park Ave., New York, N.Y. 10017 [413] Optical coupler offers high input-output isolation Optical coupler permits ±3,000 (or more) volts isolation between input and output sides and higher output-input current ratios. The unit can also serve as a high-speed solid state switch or relay, as an interface device between systems, or as a line driver-receiver element. It is housed in a DIP compatible with standard 14-pin IC sockets. M7 Inc., 210 Campus Dr., Arlington Heights, Ill. 60004 [416] Low-cost phototransistor aimed at card-reading jobs Model TIL78 phototransistor for card-reading and other applications features a light current of 7 milliamperes typically at 20 milliwatts per square centimeter. Dark current is 25 nanoamperes at 30 volts reverse voltage, and continuous power dissipation at 25°C is 50 milliwatts. The unit has an emitter-collector voltage and a collector-emitter voltage of 7 and 50 V respectively. Price is 50 cents in 1,000 piece orders. Texas Instruments Inc., P.O. Box 5-12, MS/308, Dallas, Texas 75222 [418] Buffer ICs drive clocks, lamps, lines, memories TTL Quad 2 NAND buffers drive transistor-transistor logic and DTL clock lines, lamps, medium length data lines, TTL gates, and high capacitive loads such as MOS memory arrays. Unit output is low only when all inputs are high, and propagation delay time is less than 15 ns. Version N7437A is 78 cents in 100 to 999 quantities, and the S5438F is $5.04 in the same quantities. Signetics Inc., 811 E. Argues Ave., Sunnyvale, Calif. 94086 [414] Quad-line receiver design gives high noise immunity Quad-line receiver, type HD-1489A, is designed to interface between single-ended transmission lines and data terminals. Typical hysteresis is 1.15 V, resulting in high noise immunity, and a response control pin in each of the four circuits permits the filtering of high-frequency noise or the paralleling of several receivers on a single line. Price is $4 in quantities of 100 to 999. Harris Semiconductor, Melbourne, Fla. [420] IC sockets accept TO cans with up to 12 leads A series of 12 integrated-circuit sockets, type 131-55, fit into TO packages with up to four leads on a 0.100-in. pin circle, and up to 12 leads on a 0.200-in. pin circle. The body material is made of glass-filled nylon to provide continuous operation from -55 to +125 C. Usage is in a wide variety of production applications, and three different styles are available. Barnes Div., Bunker Ramo Corp., 24 N. Lansdowne Ave., Lansdowne, Pa. 19050 [487] C/MOS digital ICs offered in chip form Unencapsulated chips, previously available in standard packages, are offered in a line of logic circuits selected for a wide variety of applications in hybrid design such as automotive circuits, industrial controls, appliances, clocks, medical electronics, computers, and communications equipment. Prices of the CD4000A series, for 100 to 900 lots, range from $1.98 to $8.95 depending on type of circuit. RCA Commercial Engineering, Harrison, N.J. 07029 [486] NEW! First comprehensive guide to every aspect of TTL devices and their practical applications DESIGNING WITH TTL INTEGRATED CIRCUITS Prepared by the IC Applications Staff of Texas Instruments Incorporated Edited by Robert L. Morris and John R. Miller This thoroughly comprehensive and practical volume—the first to explore the entire family of TTL integrated circuits—is a complete source book on the newest, most versatile, reliable, and economical innovation in systems technology. It covers not only design philosophy, economics, basic descriptions, and electrical performance of TTL devices; but many practical applications of the circuits in digital systems. It will save you time and work by providing many completed designs easily converted to your particular needs... reduce costs by suggesting the most cost-effective integrated circuits to use in your system... help your decision making by reviewing all modern logic types and giving you a comparison of the characteristics of sub-series within the TTL series. The first book devoted exclusively to the transistor-transistor logical family of integrated circuits, this volume will prove invaluable to electronics engineers, computer designers, and systems analysts and to non-engineer managers who want information on the best uses of TTL devices. 384 pages, 380 illustrations, $18.50 MAIL COUPON TODAY McGraw-Hill Book Company 330 West 42nd Street, New York, N.Y. 10036 Send me Texas Instruments' DESIGNING WITH TTL INTEGRATED CIRCUITS (637458) for 10 days' examination. At the end of that time I will remit $18.50 plus tax, postage, and handling charges. If I decide not to keep the book I will return it in the carton provided within 10 days. (Remit in full with coupon, plus tax, and McGraw-Hill pays all delivery costs.) Name ____________________________________________ Address __________________________________________ City __________________ State __________ Zip _______ Offer good in U.S. and Canada only. 23-6-7 20 YEARS OF RELIABILITY NON-LINEAR SYSTEMS, INC. DEL MAR, CALIFORNIA (714) 755-1134 / TWX 910-443-1132 Circle 125 on reader service card New products/Materials Acrylic optical components including prisms, beam splitters and light pipes can be custom-generated in a variety of sizes. Impact strength is said to be six to 17 times greater than glass, and other features include low cost, high visible light transmittance, and haze held to 2% or less. Applied Products Corp., 451 Caredean Rd., Horsham, Pa. [481] Solder process called Solderex is for through-hole rack plating of printed circuit boards. The tin/lead method does not emit hazardous or offensive fumes, and has high throwing power. It produces bright, readily-flowed 60/40 deposits. Sel-Rex Corp., 75 River Rd., Nutley, N.J. 07110 [482] Epoxy molding powder, Ecomold 1099, is formulated for transfer-molding applications where low density and/or dielectric constant are required. Typical applications include encapsulation of aerospace components, integrated circuits, coils, capacitors, and semiconductors. Price is from $6 to $8 per pound. Emerson & Cuming Inc., Canton, Mass. 02021 [483] Epoxy coating powder called Novaloy 6521 is for fast packaging of heat-sensitive electronic components. Processing temperatures may be as low as 120 C. and coating times as fast as one second are possible at higher temperatures. Edge coverage values range from 40% minimum to 65% or higher. Coating thicknesses from 0.010 to 0.035 in. can be obtained with one application. Rogers Corp., Rogers, Conn. 06263 [484] New books Thick-Film Microelectronics Fabrication, Design, and Applications Morton L. Topfer. Van Nostrand Reinhold Co., 210 pp., $10.95 No one becomes an expert by reading alone; but a book should provide a start in the right direction, some insight into potential trouble areas, and a good list of references to back up statements in the text. Morton Topfer's book fulfills all these criteria. The strongest sections are those covering technology, equipment, processing, assembly, and packaging techniques, which as an aggregate make up about half the text. Though each specific topic is necessarily brief (it's impossible to completely cover resistor pastes in only two pages) the central points are covered. In the equipment and processing sections, there's a good discussion of screen printing, including a useful table showing production rates. Assembly techniques are discussed, with special attention to flip-chip and beam-lead connections, while the packaging sector details all the various approaches, including plastic encapsulation. About the only point that's open to criticism is that the sections on applications seem to stray from the main point of the book—they get involved in details of circuit operation simply from a circuit standpoint without any real relation to the book's main topic, thick-film technology. Other sections, on system partitioning, reliability, and thick films in hermetic and nonhermetic applications, also contribute to the book's value. Several examples are included to show how the problems of hybrid design differ from those of discrete circuits. For example, it's pointed out that level shifting usually may be implemented with a capacitor, diode, and resistor clamp, but in the hybrid approach, the large capacitor presents a serious problem. The circuit therefore is redesigned to a resistor-transistor unit, which can be easily handled in hybrid form. Data on reliability also is included. CONSIDERING Frequency, Time, or Voltage Measurements? NEED Digital Systems or Time Code Instrumentation? CALL ELDORADO! ... Eldorado invented the 1 ns Time Interval Meter ... Eldorado invented direct gigahertz frequency measurements ... Eldorado pioneered low-cost IC counters ... Eldorado introduced the first U.S. made electronic calculator at under $200 ELDORADO ELDORADO ELECTRODATA CORPORATION 601 Chalomar Rd. • Concord, Calif. 94520 • (415) 686-4200 Circle 127 on reader service card Meet the Extremists: Measure from 1 Gamma to 50,000 Gauss! A complete line of precision magnetic measuring equipment. MODEL 505 GAUSSMETER • Lowest cost • Self-contained A completely self-contained instrument for magnetic field measurement. MODEL 750 GAUSSMETER • Wide range • Versatile Features wide measurement range, high accuracy and ease of operation. MODEL 3265 GAUSSMETER • Unexcelled readability • High precision Features absolute, differential and incremental measurement operations. Send for Free technical booklet on RFL Magnetics equipment. RFL Industries, Inc. Instrumentation Division • Boonton, N. J. We make 100+ different kinds of termination hardware but that's not the end We are in termination hardware because our customers asked us. They had some definite ideas about miniature posts, sockets, plugs or test clips. Binding post caps that don't melt at soldering temperatures, for example. Most of our termination products were developed for superior insulation, dielectric strength, contact resistance. (Transistor sockets with minimum insulation resistance of 500,000 megohms at 100 VDC.) But despite the length of our line, it's not the end. We will develop new hardware for new applications with the same commitment to quality in design, materials and workmanship that has fed the growing demand for our rotary and push button switch lines. Like to know more? Write or phone for our latest general engineering catalog. Grayhill, Inc., 523 Hillgrove Ave., La Grange, Ill. 60525, (312) 354-1040. Grayhill pioneers in miniaturization Circle 144 on reader service card Circle 145 on reader service card New Literature Relays. International Rectifier Corp., Crydom Div., 1521 Grand Ave., El Segundo Calif. 90245 has published a brochure describing a line of computer-compatible, solid state SPST ac relays. Specifications and a chart are included. Circle 421 on reader service card. Linear encoders. Sequential Information Systems Inc., 249 N. Saw Mill River Rd., Elmsford, N.Y. 10523. A two-page product sheet provides features, applications, and outline drawings of modular optical linear encoders. [422] Thin film etchants. Transene Co., Route 1, Rowley, Mass. 01969. Techniques for generating thin film microcircuits and data sheets on thin film etchants are included in a 15-page bulletin. [423] Audio signal delay. Gotham Audio Corp., 2 W. 46th St., New York, N.Y. 10036 has available a six-page brochure describing the Delta-Tau model 101 for digital processing of audio signals. [424] Flat load cells. Strainsert Co., Bryn Mawr, Pa. 19010. Brochures 365-2 and 365-2MP give dimensional data, specifications, and model designations for general-purpose, precision universal, and compression flat load cells. The four-page bulletins describe mechanical properties and include tabulations of deflection and spring rates. [425] Computer graphics system. Concord Control Inc., 1282 Soldiers Field Rd., Boston, Mass. 02135. Two brochures describe a graphic data processor which is a digitizer, plotter, and edit station; and two high-precision plotter systems, the MK 8 and the MK 10. [426] Panel meters. Digilin Inc., 1007 Airway, Glendale, Calif. 91201. A set of technical notes describes the uses of digital panel meters, with specific applications for the company's line of 2½-, 3½-, and 4-digit meters. Included is a discussion of noise rejection, active filtering techniques, and display of sensor outputs. [427] The custom touch: Suddenly a man’s car is his castle One prosperous New York motorist hardly flinches anymore when buzzers sound and lights flash on the dashboard as he turns the ignition key. The audio-visual display is merely a built-in warning system reminding him to fasten his seatbelt. Federal safety regulations say a new car (all built after Jan. 1) has to have one. But how about the refrigerator telephone and television set in the back seat? That was his own idea. Between government edicts on safety “extras” and his own desire for personalized motoring, the affluent American car buyer can—and increasingly does—keep a platoon of white-smocked technicians busy installing, adjusting and repairing a bevy of intricate hardware that has very little to do with moving the car down the road. For anywhere from $300 (for a 9-in. TV set) to $30,000 and up (for the accoutrements of a salon-on-wheels), a veritable arsenal of luxury equipment is being marketed for the motorist with a penchant for comfort, convenience, and individual expression on the road. Any car nowadays can be personalized, but usually the investment is warranted only on the costlier ones, such as the Continentals, Cadillacs, Thunderbirds, and the like. Only the owner’s imagination limits what can be done to them in the way of customization. Whole new interiors are being fashioned for some out of crushed velvet. Others have deep-pile carpeting, or one-way glass. Station wagons are being made out of Cadillacs and Lincolns, which do not ordinarily come from the factory in that form. The list of possible custom “extras” is just about endless: • A 2½ cu. ft. bar and refrigerator that starts at about $1,000. A complete service of goblets, decanters, ice bucket and walnut cabinet can run to $1,700. • A toaster and coffeemaker set with a 110-volt A.C. adapter, at least $700. • A rolling executive conference room installed in a limousine. It starts at $10,000, but a walnut cabinet-desk ensemble can be had for only $1,600. • High-intensity lights for reading. A small battery-powered unit may cost $100. An installation of several Tensor or quartz lamps can run to $1,300. • Personal television. A 9-in. black-and-white set may cost only $300, but a large color set can hit $1,700—and also present some tricky antenna problems. • Private communications. Telephones have to be purchased from the local telephone company, but a short-wave radio makes a good substitute, at between $300 to $1,500. A custom paint job can run $1,335 and up; consider, for instance, one in Morano Pearl, which gives something in the way of added value by seeming to change colors as the car travels down the highway. On the less frivolous side, executives and professional men claim that customizing to provide on-the-road conference space or such things as You can't replace a $30,000 house with $20,000 worth of insurance. If you haven't increased the value of your homeowners insurance recently, you may be seriously under-insured. Inflation has upped the value of the average house by 43% over the past ten years. See a State Farm agent about insuring your home and your possessions for what they're worth today with no worry about tomorrow... because a special inflation coverage feature automatically increases the amount your policy pays, as inflation continues to boost the cost of things. We're the world's largest home insurer. We protect you from loss by fire, tornado, burglary, vandalism and more. State Farm Fire and Casualty Company. Home Office: Bloomington, Illinois. For your home, your life, your health and your car—State Farm is all you need to know about insurance. State Farm inflation-coverage Homeowners Insurance Detroit's car builders will install some luxury items, such as plush carpeting and stereo tape sets (which start at $35), right at the factory. With the new 1972 cars, indeed, they are offering a host of comfort-and-convenience items. Air-conditioning (starting at $400) and "climate control" devices ($500 up) are increasingly common options. Rear window defoggers ($30) take some of the sting out of winter driving, and electric seat adjusters can be had for between $90 and $140. For driving comfort and safety, the manufacturers gladly come through with disc brakes, which last longer, stop more smoothly, and come as standard equipment on all Imperials, Cadillacs and Lincoln Continentals; they are at least a $75 option on most other cars. Anti-skid brakes are also now available on Thunderbirds, Mark IVs, Cadillacs and Imperials, for $220 and up. Automatic devices are also proliferating. For $50, a motorist can have his headlights rigged to dim automatically as other cars approach. With another $40 attachment, his headlights won't flick off as he leaves the car until he has time to get into the house. Anti-theft devices on the steering column and warning buzzers to alert drivers that they left the key in the ignition are standard equipment (by law last year), but the sporty Chevrolet Corvette is the only '72 car with an anti-theft audio alarm system as a standard item. One, of course, can be put on any car. Both domestic and foreign auto companies—and a host of custom shops—stand ready to add just about anything a motorist's heart desires. First step in getting a customized car is to get the car itself from a local dealer. Some money may be saved by dealing directly with a custom shop for the alterations, but letting the dealer make arrangements may save some headaches. It saves some fretting, too, when an owner realizes that his car is not going to be transformed into the chariot of his dreams overnight. Customizers say he should expect to allow as much as six weeks for a major job. Most automobile dealers can recommend a custom shop, but two prominent Midwest customizers who do work across the country are Custom-Craft, Inc., a division of American Sunroof Corp. (which builds all sun roofs for U.S. cars) and Moloney Standard Coach Builders of Chicago. The 1972 models: little change in a mark-time year The big-spurgle so traditional to Detroit's announcements of its new cars is oddly subdued this year. The reason is that Detroit has fewer really "new" cars to talk about than it has had in recent memory. Rising costs, a tight market, and unanswered questions about future federal safety standards (which could dictate major design changes next year) have made it so. There are, to be sure, enough cosmetic changes in grilles and exterior metal work to distinguish a '72 model from a '71. But the car buyer who feels he must look over everything new that Detroit has to offer will find he has an easy assignment this season. The Ford Motor Co. is the only domestic manufacturer offering a totally different '72 body. The all-new Thunderbird and luxurious Lincoln Continental Mark IV have it. So do the Ford Torino intermediate lines and the Mercury Montego. For the economy-minded, the modest appearance changes in most other lines offer a special break. Detroit insiders suggest that a thrifty buyer look over his dealer's '71 left-over cars before signing up for a '72—they are not only lower in list price than the new models, but the dealer also gets a 5% price rebate from the factory on the out-dated models and may be willing to shave the price even further. The new Mark IV is competing this year with Cadillac's unchanged Eldorado for buyers in the sporty $8,000 and up (far up) range. For luxury at a lower price (in the $5,500 to $8,000 category), shoppers should check the new Thunderbird, which Ford is pitting against the Oldsmobile Toronado and Buick Riviera, both hardly altered from last year. The all-new Torino and Montego are aimed at the buyer in the mid-$3,000-and-up category, whose alternate choices might be the Chevrolet Chevelle, Pontiac Le-Mans, Dodge Coronet and American Motors' Ambassador—all of which have only received a face-lift for '72. In a season of minimal change, there are still enough variations to invite some looking. Dodge's top-of-the-line Polaro and Monaco ($4,000 and up) are more different from each other than in several years. The Plymouth Fury ($3,000 and up) and Chrysler ($4,000 and up) also have some new styling worth noting. For the high-style-minded, Detroit observers of the new-model year suggest a close inspection of the Chevrolet Caprice. When it is loaded with extras, it takes on a remarkable Cadillac-like look—and it costs about 40% less, or near $5,000. The mutual fund derby is off and running again Small, fast-moving mutual funds are back in action, and once again there is a profit-performance contest to watch. But they aren't the go-go funds of a few seasons back. They are saner, more sober, and managed with more caution. Maybe, just maybe—for an investor—they are a place to lay new bets. "But before stepping in," says a top Wall Street pro, "a man will want to bone up. Among other things, this means studying the prospectus for a change." In 1967 and 1968, a youthful generation of mutual fund managers refined the art of stock-price appreciation and renamed it to suit the speculating public. They called it "performance" and this quickly became a watchword for smart innovation and unique ways of evaluating securities for a fund portfolio. The method—as Fred Carr, former head of Enterprise Fund once said—was to "conversationally develop the concept of the company" whose stock was being considered. "Fundamentals" such as price-earnings history and company rank and position in the industry often took a back seat to a company's real or imagined "concept"—or its uniqueness. The result: a host of fund investments in thinly-capitalized stocks, many of them over-the-counter. For a time, it worked like clockwork, and men such as Carr, Manhattan Fund's Jerry Tsai, Winfield Growth Fund's Dave Meid, and Security Equity Fund's Fred Alger became as familiar to Wall Street board rooms as a big-city commuter is to broken down trains and delayed buses. By discovering unexplored "situations" and searching the woodwork for obscure upbeat companies, these men of the go-go funds were called "gunslingers"—and they lived and breathed high appreciation. They made the conservative "Dow" and the Standard & Poor's average pale by comparison. In 1967, for instance, Winfield Growth doubled its assets, and Carr's speedy portfolio managers at Enterprise managed a 117.5% jump. Then the clock stopped. The gunslingers hadn't reckoned on a crashing bear market. So when the Dow went from its December 1968 high of 995 to a woeful May 1970 bottom of 631, the go-go funds fell in line—and fell apart. It was so bad that even the "little guy"—the naive small investor—figured that too many stocks were overpriced. Suddenly, "fundamentals" became vital again, and even the gunslingers were lining up and pledging allegiance to them. For nearly 18 months or so, it seemed that the "performance" craze was dead. But like the Phoenix rising from ashes, a new bull market began to bob and weave last November. With it, the go-go fund managers reappeared. True, the cast of characters has changed (Carr, for one, no longer runs Enterprise), and attitudes have been tempered by past disaster. But if the results in 1971 are an indication, a number of fund managers are once again taking up the basic pattern: fast in-and-out trading, and picking long-shots. In the first half of this year, the Dow went up a modest 6.23%, and the S&P 500, jumped 8.19%. At the same time, some 55 mutual funds turned in impressive gains in asset value of 25% or more, according to Arthur Lipper Corp. in a survey of the fund business. Topping the Lipper list: Channing Venture Fund, a tiny $1-million growth fund with gains of nearly 80%. Channing's style has been not unlike that of the small high-performance funds of years back: 40% of its assets riding on two companies—National Patent Development (soft contact lenses, a volatile stock), and Banister Continental (Canadian pipeline). Before Channing sold Banister convertible stock, its investment had doubled. Of course, it is probably much easier to manage $1-million than it is to oversee $100-million or more. Thus, most of the new-breed high-performance funds are small—most have assets somewhere under $20-million. At this size, they can locate enough small fast-moving "situations" to largely fill their portfolios, whereas $100-million funds are forced into more conservative investments at the expense of high performance. In any case, the manager of the small go-go hopeful must strive to discover the so-called hot stocks—and time their acquisition precisely. The investor must learn to live with the idea of taking maximum "reasonable" risk to get maximum gain—or else switch to bluer chips. Though they believe in their approach, the managers of the small, higher-risk funds are—today—aware that the golden gate can always swing shut. For one thing, the ordeal by fire that saw the Dow decline 50% has not been deftly skipped over by the fund men. They know the real score. High-risk situations and volatile stocks still have their old appeal; but for the most part, the go-go mentality that characterized the 1967-1968 bull market period has been diluted to the point where "fundamentals"—meaning sound analysis—are given at least reasonable attention. Moral: Review the fast appreciation funds—but do so cautiously, with a keen eye on both management and prospectus. Corporate bonds: booming yields for easy sleepers Yields in the corporate bond market have been so high lately that investors have been nailing down 8% returns by simply buying bonds of the Bell System. Some have gone even higher in profits by picking such companies as ITT and Chrysler. "But it's crucial to buy quality," says a top bond man on Wall Street. "If you're bond-minded, you want one that will let you sleep at night." But what about yields that are truly high, in the 11% to 15% range, for instance? If sleep is no object, an investor can find literally dozens of unusual situations—what might be called Bonds for Iron Stomachs. Hunting for high yields is like taking a tour of the disaster areas of American business. Aerospace, airlines, conglomerates, computer leasing—these industries are fraught with problems that range from the nightmarish to the merely hair-raising. Buying their bonds can be so perilous that ordinary brokers hardly know such securities exist. "Nobody in his right mind would touch those things," says one bond trader who sells strictly "investment grade" paper for large institutions. Yet some investors—or better, some gamblers—buy them, otherwise, their prices would sink to zero. But, cautions Russell Fraser, the vice-president who heads Standard & Poor's corporate bond rating operation: "You've got to be the kind of guy who takes a crack at new issues in stocks—a really speculative kind of animal." In general terms, the risks are twofold, and they are very real risks indeed. One is that the corporation's business may get so bad that it can no longer meet the interest payments. In that case, the bonds go into default, and the bondholder loses his yield. The other risk is that the company may simply go bankrupt, or go into reorganization to avoid bankruptcy, in which case the investor might very well lose all his principal, too. The other side of the coin, though, is that yields are fat, and if the company's fortunes improve, the bondholder can make substantial capital gains as well—gains that are taxed at only half the rate of interest payments. Capital appreciation might occur, too, if the operation is taken over by another company. Further, bondholders, as creditors, rank far higher than stockholders when it comes to gathering in the remains of a defunct business. If the company fails, there is often enough left over from auctions to pay bondholders 10¢ or 15¢, say, on the dollar. With this in mind, an adventurous soul might, for instance, rummage among the shards of James Ling's old empire. LTV Ling Altec has a sinking-fund debenture—a senior security—with a 6½% coupon due to mature in 1988. Originally issued with warrants, since detached and exercised, these debentures trade for around $480, far below their $1,000 par value. At that price the coupon equals a current yield of 14.06%. Savvy bond buyers also deal in yields to maturity, which are calculated according to a complex formula that takes into account both compound interest and the capital gain to be scored when (or if) the bonds are payed off at par when they mature. In practice, bonds' market prices usually rise in anticipation as the maturity date nears. In the case of Ling Altec's 6¾s of '88, the yield to maturity works out to nearly 15.5%. J&L Industries, which controls Jones & Laughlin Steel, has a 6¼ senior debenture due in 1994. Priced around $490, it has a current yield of 13.78% and a yield to maturity of 14.44%. Okonite, a cousin in the Ling empire, has a 6½ senior debenture of 1988, trading around $460, to yield 14.13% currently, 15.62% to maturity. Okonite was recently taken over by Ling's new vehicle, Omega-Alpha. The old holding company, Ling-Temco-Vought itself, has a bond outstanding, the 5s of '88, that trades around $445. Its current yield is 11.24%. Yield to maturity is 13.39%. S&P's ratings, which are familiarly triple-A or double-A on high quality bonds, descend from B down through triple-C and double-C to no rating at all on high-yield securities. Such a rating means pretty much "outright speculation," with increasing emphasis on "outright." LTV and Okonite have B ratings; J&L and Ling Altec, are rated triple-C. The nearer the maturity date, the higher the yield to maturity. The computer leasing company Levin Townsend, for instance, has a 7½ debenture due in 1979. Recently priced around $600, it boasts a current yield of 12.5%. The yield to maturity, though, is 17.2%. Leveraging, of course, can make the potential gain even greater. If he can arrange to buy on 25% margin, an investor need put only $150 to buy a $600 debenture, making the potential capital gain equal to 267%. He would have to pay 7% or 8% interest, naturally, on the $450 he borrowed from his broker, but that would be more than covered by the 12.5% current yield. If he is optimistic enough, he might even reason that, should worse come to worst, enough might be salvaged in bankruptcy proceedings to pay bondholders 15¢ on the dollar—and he would recover his $150. Anyone daring enough to plunge into deep-discount bonds would be wise to check how large the issue is outstanding. If the issue is less than $10-million, the market may be so thin that only a small amount of trading will produce sharp price swings—when the bondholder decides to sell his own bonds, for instance. That is a risk, for example, in First National Realty & Construction 6½s of '76, of which only $2.6-million are outstanding. Priced around $580, these yield 11.21% currently and 19.54% to maturity. Another Levin Townsend issue, the 9s of '74, have $5-million outstanding. They trade for around $770. The current yield: 11.69%. The yield to maturity: 19%. But the thinness of the market needs inspection. Beehive: NYSE's busy Bond Room National dilemma: How to cope with college drop-outs EMOTIONAL OUTBURSTS WIN LOW MARKS WITH CAMPUS COUNSELORS Steve’s parents started saving for his education when he was a toddler. All of a sudden, it seemed, Steve was in high school, earning good grades and participating in sports. To everyone’s pleasure, he was accepted by his first-choice college, an Eastern “name” school. Then Steve came home for Thanksgiving. He had a knapsack and a girl in tow, and announced that he was dropping out of school. For his parents, the news bore all the sting of the proverbial serpent’s tooth. The scene that followed is being replayed with increasing frequency in homes across the country. The college “drop-out” is not only a major heartache for an increasing number of American families, but he—or she—is also a troubling concern for college administrators as well. Of all the freshmen they see entering their ivied gates this fall, college officials are acutely aware that only 55% will stay around long enough to earn diplomas. The real puzzler is that reasons for campus failure these days are far more often of an emotional nature than academic, and thus harder to remedy. A study in 1965 by Edward A. Levenson, M.D., of the William Alanson White Institute in New York, draws a portrait of the typical drop-out which highlights the problem. The study indicates that he is generally above average in intelligence and creativity. His mind is bright, but mixed up. Socially he is likely to be a “loner.” When these psychological traits collide with the alien environment of the campus, the potential drop-out becomes an actual drop-out. Lately there has been a growing feeling among administrators that perhaps dropping out in many cases is not entirely the wrong move. Many of these students, they say, may actually be in the wrong school, with choice of colleges probably directed more toward prestige than toward curriculum. Once enrolled, they may find that their “education” is preparation for graduate school and practically nothing else. They may back down on the idea of “adjusting” to college life. Beyond this, there is the student who is suddenly confronted with more independence than he has ever known. Administrators realize that behind such pressures lies one that is even more difficult to handle—the American success syndrome. Success to many American parents is synonymous with a house in the suburbs and a college education for the children. They push for these goals, and consider anything short of them as failure. College deans hear constant reports of parents who resort even to bribery (a new car or trip to Europe) to bend their children to working for that all-important diploma. But Dr. Robert W. Pitcher, co-author of Why College Students Fail, finds that today’s teenagers simply are not motivated as strongly as their parents. “They are frequently over-protected by people who want them to have everything they didn’t have—such children have never really heard ‘no.’” Consequently, says Dr. Pitcher, the adolescent’s motivation and perseverance are slow to develop. Wise parents, say such experts, will start emotional preparation for college in the high school years, if not sooner. They will try to discover their child’s true ambitions, and provide realistic help in attaining his goals—not their own. Fathers have a special role to play, these authorities say. Typical “businessmen” fathers find it hard to encourage their children to make their own decisions. If the father is quite successful in his business or profession, he poses another stumbling block; a college-age son may feel he could never compete on his Dad’s level, and quits. “Too many boys come to my office still holding onto their father’s coattails,” says Eugene S. Wilson, former dean of admissions at Amherst College. “The boy, not Dad, should map out a college admissions program, for example. But a lot of fathers can’t let go.” Compound paternal pressure with maternal possessiveness, says Dr. George Hall, head of psychiatric services at New York University, and you have a child with a highly developed “separation anxiety.” In this state of helpless dependency, he is a prime candidate to flee school. Counselors advise parents to allow their youngsters to make some real blunders before they have to face college. One suggested trial is to let the youth manage his own checking account, and to learn to live with the consequences of his own bloopers. “Let him cover his own bounced checks,” Dr. Pitcher strongly advises. Finally, campus counselors implore parents to soft-pedal their emotions if and when a son or daughter does drop out. Irate fathers, full of blame for the college, are familiar to deans across the country, and mothers who suffer “nervous breakdowns” are more common than realized. All of which, the officials point out, does no good for anybody. Psychiatrists now suggest that for some young people dropping-out may even be beneficial—if the youth is put on his own to seek a job or serve in the military, and find his own way back to campus. “He’ll get back, if he has the basic stuff,” says Amherst’s Wilson. “Many do, and then college has some real meaning for them.” Besides Dr. Pitcher’s Why College Students Fail, other recent works that explore the drop-out problem include Anyone Can Go to College, by Herbert Livesey, Ph.D., director of admissions at NYU (Viking, $6.50), and Einstein’s 1971 College Entrance Guide (Grosset & Dunlap, $3.95). We know that when you get plenty of help in the planning stages, better workmanship and on-time delivery, you're not likely to try other sources. So we put more manpower in the field; operate three strategically located plants and provide the largest dealer network. That way, you can always count on timely personal help from a Lyon Dealer salesman, a factory representative, or both. You have a lot more going for you when you take full advantage of Lyon leadership. Proven quality. The best selection—more than 1600 stock items. The easiest ordering system. The best packaging. All of our products are unit packaged and shipped as shown in the catalog. And most orders are rolling in 24 to 72 hours. Next time, spare yourself: enjoy the kind of service that made us No. 1. Call your Lyon Dealer. We know what it takes to stay Number 1: more personal service LYON METAL PRODUCTS, Inc 75 Monroe Avenue, Aurora, Illinois 60507 Please send me a copy of catalog No. 100 Name_____________________________________ Firm______________________________________ Address___________________________________ City_________________State________Zip_____ Look for us in the Yellow Pages under LYON "STEEL SHELVING," "LOCKERS" or "SHOP EQUIPMENT" LYON METAL PRODUCTS, INC. General Offices 75 Monroe Ave., Aurora, Ill. 60507 Plants in Aurora, Ill., York, Pa., Los Angeles Dealers and Branches in All Principal Cities AMERICAN MOTORS INTRODUCES THE GUARANTEED CAR. If anything goes wrong with one of our 72's and it's our fault, we'll fix it free. Anything. AMERICAN MOTORS 1972 BUYER PROTECTION PLAN. For years, car buyers in this country have been saying that instead of shiny new chrome every year, what they really want is a good car they can count on. That's why American Motors and its dealers put together their 1972 Buyer Protection Plan. Here's how it works: A STRONG GUARANTEE IN PLAIN ENGLISH. When you buy a new 1972 car from an American Motors dealer, American Motors Corporation guarantees to you that, except for tires, it will pay for the repair or replacement of any part it supplies that is defective in material or workmanship. This guarantee is good for 12 months from the date the car is first used or 12,000 miles, whichever comes first. All we require is that the car be properly maintained and used for under normal use and service in the fifty United States or Canada and that guaranteed repairs or replacements be made by an American Motors dealer. This guarantee gives you 12-month or 12,000-mile coverage on a lot of things most car warranties don't. It covers air conditioning, battery, radio, wiper blades, front end alignment, light bulbs—literally everything we put on the car except tires. In other words, if something we did goes wrong with one of our '72's, you won't have to pay for the parts or the labor. We will. A MORE THOROUGHLY-CHECKED CAR. FROM THE FACTORY AND THE DEALER. Naturally, we wouldn't be backing our cars like this if we weren't more sure of them than we've ever been before. For 1972, we've put in more quality control steps, more tests and more people to see that every car leaving our factory is as perfect as man and machine can make it. At the dealer's, it's checked over again. Every car (not one of 50, one of 5, every one) is road-tested for starting, handling, braking and overall performance. When the serviceman completes the test, he signs his report and slides it over the sun visor, where you can see it. And, when a man signs his name to something, you can be sure he's done his job. And done it right. A LOANER CAR WHEN YOU NEED IT. FREE. Not only do we offer you a strong guarantee, we've set up a system to back it up without inconveniencing you. If you have to leave your '72 overnight for guarantee repairs, over 2,000 dealers will loan you one of their cars. A nice, clean, well-equipped car in good condition. Free. A DIRECT LINE TO DETROIT. TOLL-FREE. We're making some big promises, and we fully intend to keep them. But, just in case you have a problem, we have a way of handling it. When you buy a '72, you get the name and toll-free number of a person in Detroit. If you call, we promise you you'll get results. And fast. These aren't just fancy words. We mean it. Nobody in the business does as much for you after you buy a car, and it takes a lot of time, men and money to do all of these things. But we think it's worth it. Since we're giving American car-buyers exactly what they say they want, we should sell more cars than we've ever sold before. A pre-game treat: Drop that tailgate and pass the Klops GERMAN MEATBALLS AND MULLED WINE? IS THERE A TASTIER DEFROSTER? To paraphrase the poet, October's bright blue weather all too often is the time when the frost is on the bleachers and the spectators are turning blue. Even the excitement of football can be thin cover from the keening wind, unless, of course, proper provision has been made to insulate the inner man from all that hostile cold. One of the more hospitable ways to do it is with a tailgate lunch outside the stadium just before the game. This is no time for a dry ham sandwich and a hard-boiled egg. The station wagon tailgate should beckon with something sturdy and rib-sticking, and festive as well—perhaps a lusty hot soup plus charcuterie, or the makings of a Klops party. Klops, for the uninitiated, are German meatballs. To begin with the soup, the occasion calls for a good thick potage, one that can be made one day, heated up the next, and poured into thermos jugs for the trip to the stadium. One candidate is a split pea soup which the French call St. Germain. It takes the following ingredients for two quarts (six servings): 1 pkg. (2 cups) split peas, 1 ham bone (or ½ lb. ham in 2-in. cubes), ½ cup chopped onion, 1 cup chopped celery (with leaves), ½ cup chopped carrots, 1 bay leaf, 1 clove garlic (slivered), 1 tsp. sugar, ¼ tsp. thyme, 2 tbsps. butter, 2 tbsps. flour, salt and pepper to taste. Soak the peas overnight in enough water to cover. Save the liquid, and add enough water to it to make 12 cups. Put the bone (or ham) into the peas and water, and simmer three hours. Add the rest of the ingredients, and simmer another hour. Put through a sieve or strainer. Mix the butter and flour into a paste. Mix with a little of the soup, then add it to the pot. Bring to a boil once, then season and pour into a thermos. The consistency should be that of thick cream; if it's too thick, add a little milk, and if too thin, more flour and butter. At the game, serve in big mugs with a few croutons on top, and offer a platter of charcuterie—sliced ham, Swiss, tongue, cervelat or salami, and some French bread, buttered and cut horizontally, for make-it-yourself sandwiches. Bring along the mustard (try Dijon)—or some Major Gray's English chutney. For drinks, a hot rum toddy (in thermos) is in order for starters. To serve six, dissolve six lumps of sugar in a tumbler of hot water. Add a good shake of cinnamon, another of allspice, three or four small pieces of lemon rind, and one Good Old American Inflation Fighters Fight inflation four different ways with Wang calculators and computers. A Wang instrument will increase the productivity of anyone who works with numbers. Since a Wang calculator is a capital investment which raises productivity, you qualify for the expected 10% Job Development Tax Credit. Since every Wang calculator is made in the heart of New England, our prices will not be affected by the tax on imports. And since every Wang calculator is field expandable, your purchase will not become obsolete. Increased employee output, 10% tax advantage, no import duty and a guarantee against obsolescence — that's fighting inflation. And Wang can fight anywhere, in any department; because we have the widest line of calculators — from business machines to full computers. To find out how our army can fight inflation for you, call collect, Mr. Courtney at 617-851-7211. If you read the May 31 issue of TIME, you won’t want to miss the 1971 issue of Georgia’s Industrial Survey. If you caught the cover story of the May 31 issue of TIME, you read that “the Southern boom has urbanized and industrialized Georgia more quickly and completely than the rest of the Deep South. “Georgia leads the region’s indexes of growth and change.” Of course, you don’t have to take TIME’s word for what’s happening in Georgia. You can send for your free copy of our 1971 Industrial Survey and get it straight from the source. Georgia Department of Industry & Trade, Industry Division, Dept. MH-3, P.O. Box 38097, Atlanta, Georgia 30334. Georgia, the unspoiled. This unique book helps you manage your personal affairs with the same skill you bring to your business: Business Week’s Guide to Personal Business $9.95. “Many a man who should know better—whose business or professional income is, say, $25,000 to $50,000 or more—treats his own personal business like a third cousin who turns up in town in search of a loan. He brushes them off,” says Editor Joseph L. Wiltsee. If that describes your situation in any way, here’s help. This unique book, distilled from the popular “Personal Business” section of Business Week, offers you a wealth of new ideas on planning and managing your personal finances. Real estate, investments, insurance, school costs, taxes, estate planning—even advice on how to choose and evaluate an advisor! The money you can save—or make—from just one of the ideas in this 320-page guide should more than pay for its modest cost. Get it now from your favorite book store for $9.95. Or send your order (enclose payment please) to the address below. Note: a special deluxe edition, bound in buckram, gold stamped, slipcased is available at $2.50 more per copy, by mail only. Business Week Guide Book Service Office 330 Broadway Marion, Ohio 43302 It’s the season for apples—and cider, too Apples have been bred in just about as many varieties as their distant cousin, the multinonymous rose. But does an apple by any other name taste as sweet? Americans, according to a recent survey, couldn’t care less. Just three names—Delicious, McIntosh, and Winesap—top all others as the favorites. Delicious is semi-tart and particularly favored as a desert just as it comes from the tree. It has excellent keeping qualities, although it does turn mealy when over-ripe. The more globular McIntosh has dense meat that is crisp and tangy; its main drawback is that it bruises easily. Winesaps are particularly fine for pies and for making sweet cider. Cider is most easily made with a cider press. Anyone without one, however, is free to substitute time and muscle. One process is to run cut-up apples through a meat grinder, then mash them to a finer pulp with a mortar. The pulp is then placed in the center of a length of strong but loosely woven cloth. By twisting from both ends, a willing worker can hope to express about a gallon of juice from every 30 pounds of apples. A few ripe crabapples added to the mix lends zest. The cider one buys in markets and at roadside stands usually contains preservatives. Occasionally it can be found without these additives, and the temptation is to use it to make hard cider; indeed, some of it is sold that way, with or without the proper credentials for peddling alcoholic beverages, which it is, running to about 12 to 14 proof. But trying to ferment your own sweet cider can prove to be a disaster unless the proper equipment and strain of yeast are used. One source of information and supplies can be found at Semplex of U.S.A., Box 122276, Minneapolis, Minn. Just how “hard” pure sweet cider gets depends, in the end, on the degree of aging. For the impatient, there are some quick substitutes. One is concocted by shaking together 8 oz. of freshly expressed apple juice, 1½ oz. of vodka, a dash of bitters, the juice of half a lime, and a pinch of cinnamon. Take a Sony on the road. Wednesday. And your Girl Friday is minding the home office while you're tending to business in the hinterlands. What you need is a replacement. Well, we've got one. Hit the road with the new Sony TC-90 AC/DC portable Cassette-Corder®. It features Sony's built-in condenser microphone that picks up your voice with amazing clarity from any direction while leaving your hands free. Only $99.95 you'll find the TC-90 a snap to use. Cassettes snap in instantly and pop out at the touch of a button. Another good roadmate is Sony's TC-60 portable Cassette-Corder® for just $49.95. This versatile economizer is built with the same high standards of quality as our high-priced models and can operate with equal ease on internal batteries or house current. Get two for the road—or anywhere else—from Sony/Superscope. SONY SUPERSCOPE You never heard it so good® ©1971 Superscope, Inc., 8130 Vineland Ave., Sun Valley, Calif. 91352. Send for free catalog. Pennsylvania Grade motor oils: important when you powered along with 200.5 "cubes" (Vital as you drive 426) Big power. Power for sustained high-speed driving. And for all of today's power accessories. Big power is rough on ordinary motor oils. It causes them to break down. But don't worry. Pennsylvania Grade motor oils are anything but ordinary. They provide the greatest stability under high compression...the surest resistance to build-up of acid, sludge, and varnish. No wonder Pennsylvania Grade motor oils are the most asked-for brands in America. Our informative booklet describes the types of motor oils to use under varied service conditions. For your free copy, write Dept. PB110. PENNSYLVANIA GRADE CRUDE OIL ASSOCIATION P.O. Box 250, Oil City, Pennsylvania 16301 Desk set in classic jet crystal with clock - $80.00. Others, $3.95 to $175.00 SHEAFFER • the proud craftsmen W. A. SHEAFFER PEN COMPANY, A [extron] COMPANY, FORT MADISON, IOWA 52627 TRAVEL Kennedy Center: VIPs and tourists jamming the aisles IN WASHINGTON, THE ARTY FARE SPANS AFRO AND BOSTON POPS When the John F. Kennedy Center for the Performing Arts opened in Washington last month, it had an immediate and paradoxical effect on night life in the capital. It made it harder to get tickets for plays and concerts—but more fun for the man who lands a pair on the aisle. Music-loving tourists as well as locals and all stripe of VIPs—who formally picked up tickets to most Washington shows at the last minute—are learning to plan in advance. Those from out of town are lining up their seats through their out-of-town ticket brokers, or are putting the bee on contacts in the capital. And advance planning pays. The Center, a huge white marble palace at the Potomac waterline, is spacious, gracious, delightful. Its cream-white-red interiors glisten with imported mirrors and crystal, and provide acres of room to promenade, possibly to view the political luminaries who have been drawn like everybody else to the Center. Or (thanks to a special mandate from the Congress) a lover of the arts can sip cocktails between the acts, and—if installed in one of the poshier boxes—can chat with friends in a private anteroom almost as big as the den back home. Democratic is hardly the term. The Center's entrance is from a landscaped flagstone terrace off an extension of New Hampshire Ave., and it is no place to arrive at the last minute. Showgoers have a 300-ft. walk through flag-draped halls to the Center's river side where the foyers of the various auditoriums flow into the main thoroughfare. The 19-million cu. ft. structure has four theaters (including a film house to open next season), a cocktail lounge, and three quite pleasant restaurants. The acoustics?—There's not a "dead" seat in the house: this holds true for the largest auditorium, the concert hall, seating 2,760. A few of the big-name performers have been sold out for weeks, and scalpers in and about the town's big hotels are charging hold-up prices. But there remains an opportunity to obtain tickets conventionally for such items as: - The world premier this month of Wilderness Journal, a symphony by John Lemontaine, especially commissioned for the Kennedy Center. - A founding-artists series to raise money for show tickets for the poor; among the pop stars appearing are Diannan Carroll (Oct. 8), Pearl Bailey (Oct. 9), Victor Borge (Oct. 29), Tony Bennett (Nov. 13), Duke Ellington (Dec. 26). - Two top American ballet companies—the National, using the opera house the weeks of Oct. 10 and Dec. 5, and Ballet Theater, the Center’s own company, opening for two weeks on Dec. 19. - A refurbished version of Leonard Bernstein’s 1956 musical Candide, with a new book and changes in the score—a pre-Broadway run (Oct. 21-Nov. 19). - Concerts by many of the world’s great symphony orchestras, including not only such standards as the Philadelphia (Oct. 4), Boston (Oct. 25), Detroit (Nov. 5), and Chicago (Nov. 19). - An Afro-Asian festival built around dance troupes from Senegal, Iran, Cambodia, and Morocco, and featuring such theatricalism as the sword dance. Other star billings include Antal Dorati leading the National symphony and chorus, and various soloists, in a complete version of Haydn’s Creation (Nov. 2-4); England’s new Philharmonic, with Lorin Maazel conducting (Nov. 15); a new production of Verdi’s comic opera Falstaff (Dec. 2-6); Story Theater, Paul Sill’s unique combination of mime and fable-telling that clicked last season as a Broadway attraction for children (Dec. 6-Jan. 1); and the Moravian Folk Company of singers, dancers, and musicians from Czechoslovakia (Dec. 12). With as many as three major attractions in the Center the same night, the managers have tried to unmuddle the traffic crush by staggering curtain times. Still it can be a nerve-racking affair to have to inch a car into or out of the 1,600-space underground garage. Waiting for a cab for the after-show trip to your hotel can be a drag, too. So Washingtonians and repeat visitors have found that the best gambit is to arrive early and stay late, dining and drinking at La Grande Scene or at one of the two other restaurants in the Center. The Kennedy Center sits alongside Rock Creek parkway, with little else but trees and the river close by. One dining spot within wandering distance is the first-rate Watergate Terrace, in the apartment-hotel complex that has become so popular with the upper echelon of the Nixon administration. It is a well-lit two-block walk from the Terrace to the Center, and a number of showgoers have been dining there. There is also a Howard Johnson at hand. MOVING How to Talk to a Gorilla for Fun and Profit. A lot of people worry when they use a moving company. They think the moving man is a gorilla who drives a truck. So anything he has to say just naturally sounds like jungle gibberish. It’s full of strange, legal terms like “tare weight” and “bill of lading.” Which somehow always creep into the final bill. And make it add up to a lot more than they originally expected to pay. Well, now you can speak to the moving man in his own terms. There are new government regulations that carefully translate all of the gibberish into easy, everyday language. And help you to save time and money by explaining your rights when you use a moving company’s services. Lyon, one of the big, national moving companies, is offering to send you a free copy of these regulations under no obligation. By reading these new regulations, you’ll learn what to do when you think the bill is too high. What the mover’s liability is for lost or damaged household goods. How to avoid delays in delivery. Where to file a claim. And how generally to protect yourself and your possessions on moving day. Every state has its own regulations. And these are often difficult to learn. Because many states don’t print them for general distribution to the public. However, your local Lyon Moving and Storage agent will be happy to provide you with the rules that apply within your state. For a free copy of the regulations governing your move within or out of state, call (800) 243-6000 for your nearest Lyon agent. In Conn. call collect 1-800-942-0655. The calls are toll free, and there’s no obligation. You’re probably wondering why Lyon, a moving company, would want you to know about these new regulations. For a very simple reason. Lyon feels the more you know about the new rules that protect you, the more you’ll want to let Lyon guard your goods. LYON MOVING • STORAGE LYON VAN LINES, INC. The teen-ager, the parent—and alcohol "It isn't always pot, by any means," says Dr. Selden Bacon, director of the Alcohol Studies Center, Rutgers University. "Parents, some of them, anyway, ought to take a close look at the teen-ager versus alcohol." Why teen-agers drink, how much, when, where—and with what effects—are carefully explored in an excellent study, Teen-age Drinking, by Bacon and Jones (Crowell). It's a book that pulls few punches—and a lot of them are aimed at parents with dated misconceptions. The obvious point is made: By and large, a teen-ager's drinking habits are patterned after what he sees his parents do. Then the advice becomes less obvious. Some examples: Drugs: Don't relate the drinking experimentation of a teen-ager to the use of drugs. The two habits are quite different, and drinking rarely leads to drugs. Driving: Drinking is not a prime cause of teen-age auto accidents; teen-agers are generally well aware of the drinking-driving problem, often more so than their parents. Sex: Drop the notion, too, that teen-age drinking is a sure sign of an inner maladjustment; and as for sex, the evidence shows that most teen-agers are, again, well aware of the alcohol-sex relationship. Some evidence indicates that modest drinking actually may reduce experimentation in sex. Finally, the authors pose the question: Why are so many parents so ready to believe the worst about teen-agers? Tax scene: the smartest way to handle "T & E" Anybody who fails to keep clear, complete records of travel and entertainment ("T & E") expenses may face double jeopardy: They may miss out on the business expense reduction and pay a penalty besides. Lately the IRS has been undercutting deductions by 100% where records have been poor (and on appeal, the Tax Court has sided with IRS). Now, a hot stick has been poked. In a recent case, the Tax Court backed the IRS and disallowed an MD's professional entertainment costs due to faulty records—then slapped on a heavy penalty. Question of the deductions' legitimacy made no difference. Moral: Keep a detailed T & E tax diary, plus receipts for all outlays of $25 or over. . . . Brighter side: A new Tax Court case shows that travel expenses can be deducted where a taxpayer goes on a trip to inspect a piece of investment real estate—even in a resort-area. Estate tax trend: There is no question about taking the proceeds of your private life insurance out of your taxable estate—by giving all interest in the policy to your wife, with no strings attached. Now you can do the same thing with group life insurance (company, association, or the like)—if state law specifically allows it. States have been changing their laws on the point, and at latest count, Iowa and Missouri have brought the total to 39. . . . The hitch? You can't back out later on, and change beneficiaries. Jet-setters may not zoom down on Florida's brand new Disney World, but you'll find families with kids (and without) jamming the place. This $300-million, 27,000-acre spread of fun, fable, and Americana a la Disney, 15 mi. south of Orlando, inland—is fabulous. If you plan a visit, take this tip: Book a hotel weeks ahead, but don't count on those at the site or in Orlando. Drive to and from the show in a car, from Tampa (try the Causeway Inn, Causeway Inn South, or Manger Motor Inn), from Sarasota (Three Crowns, Frontenac), or St. Petersburg Beach (Treadway Inn). . . . Now is the time to book hotel space for the summer Olympics, Munich, West Germany (Aug. 26-Sept. 10, 1972). Recommended hotels are Bayerischer Hof (rooftop pool), Continental (excellent grill room at this writing), Regina Palast, Vier Jahres-zeiten, and the Germania. Send letter with check for two nights of a week's stay ($60); or, because Munich will be jammed, stay at a small town inn within commuting distance. Write 20 Olympiade, No. 8 Munchen, 13, Saarstrasse 7, Abteilung W, Munich. U.S. computer firms turn out for Soviet EDP show... Foreseeing a breakthrough in electronics trade with the Soviet bloc, more than 100 Western manufacturers of computers and related hardware have turned up at "Systemotechnika 71," a data processing and office equipment show now being held in Leningrad. Glahe International, a German trade-fair organizer instrumental in lining up the firms, regards the big turnout of U.S. companies—14 altogether—as the first major step toward an East-West rapprochement in the electronics sector. Among the 14 American firms are IBM, Western Electric, NCR, ITT, Bell & Howell, Rank-Xerox, and Memorex. Glahe International says that the lineup of Western hardware was selected to meet Soviet needs in computer and peripheral device applications, in air and road traffic control, in hospitals, schools, universities, statistical offices, and at various administrative centers. One requirement that the Soviets insisted upon was that Western peripherals be able to work in conjunction with Russian-made computers. ...as IBM moves closer to deal with U.S.S.R. The Leningrad show is another step in IBM's march on the Soviet Union's market for computers and related equipment. IBM is using an impressively large stand at the show and is displaying a 360/50—hitherto not sold in the U.S.S.R. Moreover, IBM recently disclosed that it had engaged the services of Satra Consulting Corp., a New York-based firm specializing in East-West trade, to handle some of its needs in developing eventual Soviet business. These two moves represent a further significant coming-together for IBM and Soviet computer planners. A year ago, a top-ranking delegation of IBM officials visited Moscow, but chairman Thomas Watson refused to comment on the prospects of dealing with the Soviet Union, pending the outcome of consultations with Administration officials. At that time, he only hinted at what he termed a "potential relationship." European firms seek to fill void left by RCA RCA Corp.'s abrupt scuttling of its computer manufacturing operations is intensifying negotiations toward cooperation among European electronics companies interested in building their own computers. Siemens AG, which was a partner with RCA in computer manufacturing [Electronics, International Newsletter, Sept. 27] is said to be moving toward alliance with the Netherlands' Philips to form a combine that could take second place in the European computer market behind IBM. The RCA move also is said to have given new impetus to talks among those two firms and France's CII and Britain's ICL, leading to a possible tie-in for computers. The companies are reported to be seeking some form of collaboration without government intervention to take up the slack left by RCA. ICL may be a big gainer in any such move; the company was said to have been working on a deal with Siemens about three months ago, but the German firm, at that time still linked with RCA, was reluctant to flash the green light. British firm builds 350-MHz counter A new high in frequency rating for a direct-gated frequency counter is claimed by Britain's Racial Instruments Ltd. The instrument gives 1-hertz resolution on a one-second sample to 300 megahertz under adverse conditions and 350 MHz in normal use. Available higher-frequency counters divide down the input frequency before gating a sample, so that last-figure resolution is lost, says the company. Racal uses MECL-3 ECL gates on multilayered thick film substrates to keep interconnections short and precise, plus a thick film preamplifier with 10-millivolt sensitivity and a programmable attenuator to adjust the input level. U.K. price will be $1,860. Higher-frequency counters are in development. Sony develops consumer Earom... A nonvolatile semiconductor memory that can be erased with impulses of reverse polarity from the writing pulses has been developed by engineers at the Sony Corp. The memory uses a single metal-alumina-oxide FET of the n-channel enhancement type to obtain read-access times in the tens of nanoseconds. However, write and erase times are too slow to allow use in read-write applications, so the new unit will be used as an electrically alterable read-only memory. Single-chip devices built thus far have had up to 256-bit capacities; larger configurations with decoding capabilities are planned. The Earom was developed in Sony's semiconductor production plant rather than in the laboratory, and so company engineers feel they can move it into products as early as next year. Applications may include programming radios or TVs to turn on a specified station at a predetermined time or programming jukeboxes. ...as GEC works on silicon gate associative memory In a joint effort, GEC Semiconductors Ltd. and GEC-Marconi Electronics Ltd. are developing a 128-bit silicon gate associative store for production next year. Experimental MOS content-addressable chips show that search time will not be longer than 10 nanoseconds and write time within 15 ns. The chips measure about 120 mil² in a dual in-line package. Normal arrangement will be 16 words of eight bits and the price objective for 1,000-and-up quantities is less than $30. GEC men envision a substantial market in parallel data processing and in high-speed correlation functions, particularly in radar where the existing trace can be held in the store and the incoming trace compared with it very quickly. Nixdorf takes award for Spanish terminal network Winning out over heavy competition, West Germany's Nixdorf Computer AG has landed a $1.5 million contract to set up a nationwide computer terminal network in Spain. The system will consist of 120 terminals and eventually tie together all revenue-collecting offices around the country. The terminals initially will be used in off-line operation but later will be put on-line to work in conjunction with a central computer—an IBM 370/135, located in Madrid. The terminals are Nixdorf 820/05 models. Addenda Interdata Inc. is buying an assembly plant in Britain for its new model 70 and 80 minicomputers, and plans to be fully operational by next summer. The New Jersey firm plans for eventual full manufacturing in Britain... The Swedish Ministry of Industry has named a commission to study the steps necessary to improve the native computer industry's competitive position, and to promote better use of EDP by industry... Australian sources report that nation's computer companies are being flooded with job applications from the U.S. following the demise of RCA's computer division. Read-only memory cut out for the job Diagonal slots in plane dividing word and sense lines couple lines when signal polarization is correct A read-only memory, claimed to be 10 times larger per module and four times faster than existing units, has been developed at Philips' Gloeilampenfabrieken's sprawling laboratories in Eindhoven, Holland. The unit uses a new matrix in which information is stored in linear inductive coupling elements between the word and the sense lines. The construction sandwiches a ground plane between a set of horizontal word lines and a set of vertical sense lines. To obtain the coupling, slits are etched in the ground plane at each crossing of word and sense lines. The slits are oriented in either of two 45° diagonal positions. The structure is completed with two shielding ground planes on either side of the sandwich. High-frequency components of the magnetic field of the word lines penetrate the ground plane through the slits. The direction of the field that penetrates the ground plane depends on the orientation of the slits. The result is not a yes/no system, as are most capacitance or inductive stores, but a coupling that is either positive or negative. In working out the concept over the past five years, Philips researchers were influenced by what they felt were two disadvantages of existing capacitance systems: "If you take a capacitive coupling element," says R.M.G. Wijnhoven, research scientist at the Philips lab, "you can have the situation where only one capacitance is loading the other word line." This loading variation results in variations in driving waveform, and, to give adequate tolerances on timing, elongates cycle time. The same situation occurs on the sense line where one or many, or even no, capacitances are coupled to it. This influences propagation characteristics, again resulting in variations in timing. Considerable research went into determining the best slit shape. To avoid parasitic capacitance couplings, the slit must be as narrow as possible near the crossing of the sense and word lines, yet there must be space for the magnetic field to penetrate the ground plane. Capacitance coupling is minimized in the Philips system by giving the word and sense lines a pitch of 1 millimeter. Also, inductive couplings are optimized by use of dumbbell shaped slits (see diagram). The biggest trouble, however, stemmed from noise due to the eddy current in the ground plane. Connections of the word lines to the ground plane at edges of the matrix were tricky because the ground plane current spread out to regions of neighboring slits. The solution was to allow for a 1.5 centimeter unused edge. The complete memory stack, consisting of 2,048 words, is made up of four submodules of 512 words each. These submodules are built up of two sense line sets, each 100 lines wide, on a flexible Mylar sheet, which is then glued to a 0.6-mm glass-epoxy board, with the ends of the sheet used to interconnect the sense lines via a pressure contact. Instead of an integral information-carrying ground plane, Philips opts for a plane made up of small pieces covering 64 words and 100 bits. These are mounted in a ground plane frame made of gold-plated copper on Mylar and containing windows in which fit the smaller information planes. The gold-plated edges of the small planes overlap the window edges and pressure contact makes an integral ground plane system. On top of the ground plane, perpendicular to the sense lines, are placed 512 word lines, etched on four 0.6-mm boards. The sense lines, information plane, and word line boards are positioned simply by using registration holes and pins. The pitch of 1 mm for word and sense lines is a practical minimum, fixed both by the limits of etching very long lines and the fact that the output signal becomes extremely small when the stack is reduced. "With this pitch," says Wijnhoven, "if you feed in 200 milliamperes with a rise time of 10 nanoseconds you get an output signal equal to, or larger than, 1.2 millivolts." Maximum propagation delay on the sense line is 16 nanoseconds and the attenuation is 4 decibels. Word line delay is 2 nanoseconds. In the electronics around the memory, Philips licked the small signal detection without major difficulties by using a preamplifier which boosts the signal up to about 20 millivolts. A post selection circuit using MOS/FET transistors condenses the internal 200-bit word to an external 100-bit word. Then comes a polarity detector with an input sensitivity of 10 millivolts. Detection chain delay is 11 ns. Siemens called the system a light-gate, field-selection matrix. The new technique, devised by Peter Salminger of the company's Munich-based data technology laboratories, offers several advantages to terminal designers. Since information selection can be done directly with the finger and without any space-consuming aids in front of the terminal, the equipment layout is considerably simplified. It's further simplified because the selection process is not incorporated into the display but is performed external to the screen. What's more, since the beams can be in the infrared range, the displayed information is not obscured, as it might be in selection schemes using an overlay of touch-sensitive wires. Matrix. In the prototype setup, the light beams that form the selection matrix are produced by a series of light transmitters arranged along two adjacent edges of a frame. On the opposite two edges are semiconductor light-detector elements, one for each transmitter. The light comes from a single gallium-arsenide source, the emission of which is distributed to individual transmitters through glass fiber light pipes. Lenses produce the required beam collimation. The beam pattern is made up of five vertical and six horizontal beams, giving a matrix with 30 possible selection fields. Of these, five are used as control fields and 25 as information fields. Thus, with a typical screen display capacity of 14 80-character rows, each information field can accommodate an information item, or descriptor, with up to 45 characters. The five control fields are used to initiate specific hardware functions such as transmission, reset, or display-erase, and thus correspond to special-function keys of keyboard-type control devices. VW harnesses a computer to diagnose bugs Despite the inroads electronic measuring instruments have made into automobile service shops, the mechanic's job of pin-pointing faults can still be an exasperating one. Now, Volkswagenwerk AG, Germany's biggest automobile maker, has introduced an electronic fault-location setup. Built around a small computer, the system links up with the test circuitry being installed on all new VW models and carries out up to 88 different checks—many of them fully automatically—in about half an hour. At the same time, the system prints out the test results to tell the repairman which parts are defective or are expected to go bad shortly. For the car owner that printout also serves as a record of his car's performance. To get maximum effectiveness from the system, VW is incorporating the test circuitry in all models in serial production. The internal circuitry, which comes at no extra costs to VW car buyers, will also be installed in the Audi models made at the company's Audi-Union division. On line. The first such computer-run diagnostic stands, will show up shortly at the company's service shops in Germany. By May all 2,500 VW shops there will be equipped. Later, the setup will be installed at VW shops elsewhere in Europe and overseas. The diagnostic stand costs around $3,000, complete with computer, the printer, a reader, and a small keyboard unit. To speed up hardware delivery for the systems, two firms have been chosen as the prime electronics suppliers—Siemens AG and Hartmann and Braun, a Frankfurt company. Hartmann and Braun is building the systems—about 1,500 of them—that will go abroad. The systems will be installed at VW shops on a lease basis; shop concessionaries must pay a rental fee of about $86 a month. The 88-point test will cost car owners between $3.60 and $4.20 depending on VW model. The first five diagnoses after a new car purchase will be free, however. To perform the tests the mechanic fits the plug at the end of the diagnosis cable into a multi-pole socket on the car. Next he replaces the oil dipstick with a temperature sensor, because oil temperature is one parameter to which some of the subsequent measurements are related. He then inserts into the system's reader a plastic punch card containing the test program. A number pasted next to the test socket tells the mechanic which card to use for the particular VW model being checked. The punch card contains all the data for the diagnosis program plus the nominal values of test parameters with which the actual ones obtained during the tests are compared. The test program begins with a few visual inspections, which include checking the play of the steering wheel, the clutch, the handbrake, and other mechanisms. During these checks the mechanic uses the portable keyboard unit pushing "plus" or "minus" buttons depending on whether results are good or bad. Next come the test items that the computer checks out automatically: brake lights, turn signals, rear window heaters, battery, and other electrical systems. In the battery check, for example, the state of charge is determined as well as the voltage drop it produces across various electrical devices. Compression is determined indirectly by measuring the amplitude of the starter current which is required to turn over the engine. In these and similar tests, the computer compares the nominal values with those fed in by the test circuitry. Some of the checks do demand a bit of work on the part of the mechanic, however. For instance, in toe-in, toe-out, and camber measurements of wheel suspension and alignment, the mechanic removes the front wheel hub caps and mounts a small mirror in their place. About two feet off to the side is a projector and an array of photocells. When the steering wheel is moved back and forth, a small cross of light formed by the projector is focused on the photocell array. The currents produced by the cells are added and the corresponding voltages are changed into degree and minute readings, which indicate whether or not limit values are exceeded. France Home-grown CAS system enters world competition French engineers have developed an aircraft collision avoidance system (CAS) that threatens to collide head-on in the international electronics market with American systems now being developed. The French have taken pains to make their version compatible with the systems in the United States so that they will be in position to capture a good share of the business—once airlines decide to equip all commercial craft with anticollision gear. Jean Besson, chief of the design team working on the system, says it can achieve distance accuracy of 3 meters and speed accuracy of 20 knots/second, a performance he claims surpasses that of U.S.-developed versions. The French have followed the standards set down by the U.S. Air Transportation Association, which has been encouraging development of CAS prototypes. McDonnell Douglas, Bendix and Sierra Research all have come up with proposed systems. The French prototype, now in its final development stages, is being handled by ONERA, the National Office of Aerospace Study and Research, located just south of Paris in the suburb of Chatillon. Final tests are planned for this autumn, Besson says, when the system will be tried out over the sea and mountainous terrain to check for echo problems and to increase the range of the system to 50 miles. Tests over flat terrain this spring were successful. "We have no reason to expect any obstacles," says Besson, who hopes to turn the test findings over to French industry for commercialization. ONERA believes that with the growing number of aircraft in the skies—and the imminent arrival of supersonic transport—CAS will soon be standard equipment for commercial and business aircraft. It already has assurances from the Concorde builders that the ONERA system will be a recommended Concorde accessory. Compatibility. ONERA's system would work only between planes equipped with the same or compatible gear. It would operate with a 50-watt transmitter on a frequency of 1.6 GHz. Besson says that when two aircraft enter the 50-mile range, they would pick up each other's signals. The distortion from the assigned frequency would permit equipment aboard to calculate the rate of approach by measuring the amount of Doppler effect, or change in frequency due to the movement of the transmitters toward the receivers. The distortion would be only 0.0001%, if the two craft were approaching at a rate of 300 meters per second, about 600 miles an hour. Besson adds: hence the need for precision equipment to analyze the signal. Due to the high frequency of the signal, however, Besson's team had to add a frequency divider to reduce it to 50 MHz, a measurable range. The distortion is then reamplified and measured to provide the exact tau factor—or rate of approach. Meanwhile, the precise distance between the crafts is being measured by timing the exact moment of reception. Transmissions from all equipped aircraft would be synchronized by precision oscillator clocks. ONERA tentatively plans to use a rubidium atomic clock developed by Thomson-CSF. The clock controls the emission of the radio signal every three seconds. Each three-second burst is divided into slots of 1,500 microseconds, allowing for each aircraft to read messages from many aircraft at once. Every emission contains the transmitting craft's altitude. These data are compared in a minicomputer with the altitude of the receiving craft. The computer sounds the alarm if it calculates a collision course and the pilot is advised by a flashing light to climb or dive immediately. Japan Coding opens way for semiconductor laser link Based on a small solid state laser, a communications system developed by Nippon Electric provides performance rivaling that of systems with large gas lasers—at a lower price. The system can transmit six telephone voice channels, or 300-kilobit-per-second data, up to 3 kilometers and in Tokyo's climate achieve error-free operation more than 99% of the time. If the distance is reduced to 1 kilometer, 99.9% operation is achieved. Price of the equipment, about $10,000 per unit, includes coder and decoder, laser transmitter and avalanche photodiode receiver, and receiving and transmitting antennas. The receiving antenna is an inexpensive-to-make fresnel lens, and the transmitting antenna is a camera telephoto lens specially coated for use at near infrared. The semiconductor laser can deliver reasonably large pulses, but its duty factor is low. Thus a highly efficient modulation scheme is needed. The laser is operated at a peak power output of 0.5 watts and a duty factor of 0.1% with a pulse length of 20 nanoseconds and repetition rate of 20 microseconds. With on-off modulation the data transmission rate is only 50 kilobits per second. This data rate can be increased by the use of pulse-position modulation. Each 20-microsecond interval is divided into a number of equal length time slots, and the slot in which the pulse appears can be made to vary with the input information. A 20-nanosecond pulse length gives up to 1,000 slots. Nippon Electric goes one better, though. It uses differential pulse-position modulation, which eliminates the need for synchronizing or other noninformation pulses. In this approach, each 20-microsecond interval is divided into, say, eight equal-length time slots repeating endlessly. As opposed to standard pulse-position modulation, where the signal for succeeding time frames may appear in, say, positions five, three, and four, in differential pulse position modulation the first pulse appears in slot five, but the following number, three, will appear three empty positions later, or slot one of the second frame. If no special precautions are taken, the following four would appear in time slot six of the same frame. This cannot be allowed because two pulses in the same frame exceeds the laser's allowable repetition rate. What's more, the average signal pulse rate would exceed the sampling rate, an impossible situation. Nippon Electric engineers solved that by shifting the pulse by the length of one frame, or eight time slots, an innovation that makes the approach practical. The actual system built by Nippon Electric can transmit six telephone channels or 300 kilobits per second and so requires 2" power or 64 time slots. It expects to use an even shorter pulse and build a system with 4,096 time slots. Great Britain Car's image displaces radar in speed checking gear Speed trap radars used by the police in Britain date back about a decade and need a heavy-duty battery to power a klystron. One candidate for the next equipment generation is the Gunn-diode powered radar, which is very much lighter. But the Home Office, which supplies speed trap equipment to the police, may abandon radar altogether and use the optical image of the vehicle. The principle is to look at the passing car through a fixed vertical grating and use a photodiode to register the frequency with which a bright spot in the image moves in and out from behind the grating bars. The fluctuations in diode output will be directly proportional to vehicle speed, and so provide an easy basis of measurement. Home Office men say the main advantage of the proposed equipment is that it can be very light: the solid state electronics are low voltage and use little power because the system uses no transmitted beam. It will also be used across the road at right angles, which will get rid of the doppler radar squint angle that sometimes produces ambiguous results and legal arguments. Peaks. Though an unambiguous waveform from the photodiode might seem to depend on non-uniform distribution of bright spots on the vehicle, engineers at Marconi Radar Systems Ltd., which is developing the system for the Home Office, say that in practice this is not a problem. If it did become a problem—that is if reflectivity is just too uniform—it could be dealt with by tracing the peaks of individual bright spots, they say. For night operation, it would probably be used along with a weak light pointed across the road. Marconi's prototype equipment obtains a grating effect by using 16 narrow, vertical mirrors side by side and slightly separated. The mirrors focus the vehicle's image on a single photodiode. IC logic is used for establishing the validity of the signal, frequency counting, and driving a three-figure readout, which may be solid state or liquid crystal. The aim is a system that will give an accurate reading from 20 to 150 miles per hour and fit in a box that can easily be carried around, reseted and put into service by one man. If we design it you can count on it being a winner. Take our new Ink Ribbon Printer. We'll customize it to your special requirements. What do we have to start with? A printer that's front loading, with two spools and 200 inches of true ink ribbon, and a reliable automatic feeding and reversing mechanism. It has a typewriter ribbon advancing action that assures longer ribbon life, doesn't require pressure sensitive tickets or tape, and gives you up to 30 active digits for use with ticket or tape. The rest is up to you to ask and us to design. Count on us. Write Product Manager, ITT General Controls, 801 Allen Avenue, Glendale, California 91201. GENERAL CONTROLS ITT Circle 147 on reader service card Celco Amplifiers 60 Volt 6 amps 9 amps 12 amps 35 Volt 4 amps 8 amps 12 amps 16 amps 20 Volt 4 amps 8 amps 12 amps for CRT DISPLAYS MAHWAH, N.J. 07430 UPLAND, CAL. 91786 CONSTANTINE ENGINEERING LABORATORIES COMPANY Circle 148 on reader service card The newest, fastest and easiest way to specify indicator lights, push button switches and readouts. Dialco's new 56-page product selector guide helps you select from over 1,500,000 visual indicators This book is the result of an all-out effort to provide you with fingertip data on all Dialight components and to make it very easy for you to locate the detailed specs and information you desire. Designers and engineers will find the "Product Selector Guide" invaluable in their work. Send for your copy today. Dialight Corp. 60 Stewart Ave., Brooklyn, N.Y. 11237. DIALIGHT A North American Philips Company Circle 135 on reader service card UNBELIEVABLE ADHESIVE It's a fact: EASTMAN 910® Adhesive can form remarkably strong, long-lasting bonds between just about any materials you can think of — and do it with surprising speed, without the need for any mixing, heat, solvents, catalysts or more than contact pressure. Hard to believe? Find out for yourself what this unique adhesive can do, and how it can help you cut assembly costs. Get your copy of the EASTMAN 910 Adhesive information kit by writing to Chemicals Division, EASTMAN CHEMICAL PRODUCTS, INC., Kingsport, Tennessee. SYNCHRON® When you need a lot of motor in a little package The Esterline Angus Minigraph recorder is so small (3⅝" x 5¾" x 4¾") that it can go practically anywhere—by itself or OEM. Some of its demanding applications include laboratories and the aerospace industry. So it's designed for precision and reliability as well as compactness. That includes specifying a Hansen Synchron motor to drive the chart and to activate the impact plate, which in turn causes a stylus to write a record consisting of dots. Precise. Reliable. Compact. That's Synchron. Maybe it's what you need. HANSEN MANUFACTURING COMPANY, INC. PRINCETON, INDIANA 47670 HANSEN REPRESENTATIVES: Carey & Associates, Houston and Dallas, Texas; R. S. Hopkins Co., Sherman Oaks, Calif.; Melchior Associates, Inc., San Carlos, Calif.; The Fromm Co., River Forest, Ill.; Orr Associates, Grand Rapids, Mich.; H. C. Johnson Agency, Inc., Rochester, N.Y.; Winslow Electric Co., Essex, Conn.; Kiley Electric Co., Villanova, Pa.; Herbert Mude Associates, Inc., Teaneck, N.J. Export Department: 2200 Shames Drive, Westbury, N.Y. 11590 Advertising Sales Staff Pierre J. Braudé [212] 971-3485 Advertising Sales Manager Atlanta, Ga. 30309: Charlton H. Calhoun, III 1075 Peachtree St. N.E. [404] 892-2888 Boston, Mass. 02116: James P. Pierce 607 Boylston St. [617] 262-1160 Chicago, Ill. 60611: Ralph Hannig Kenneth E. Nicklas 645 North Michigan Avenue [312] 785-4580 Cleveland, Ohio 44113: William J. Boyle [216] 586-1030 Dallas, Texas 75201: Richard P. Poole 1340 Republic National Bank Building [214] RI 7-9727 Denver, Colo. 80202: Harry B. Doyle, Jr. Tower 1000, 700 Broadway [303] 266-3863 Detroit, Michigan 48226: Ralph Hannig 2600 Penobscot Building [313] 962-1793 Houston, Texas 77002: Richard P. Poole 2277 Post Oak Blvd. [713] CA 4-8381 Los Angeles, Calif. 90017: Robert J. Reilly Bradley K. Jones 1125 W. 6th St. [213] HU 2-5450 Minneapolis, Minn. 55402: Kenneth E. Nicklas 1000 Nicollet Ave. Center [612] 332-7425 New York, N.Y. 10036 350 W. 42nd St. Warren H. Gardner [212] 971-3617 Michael J. Stipke [212] 971-3616 Philadelphia, Pa. 19103: Jeffrey M. Preston 6 Park Place, Suite 1000 [215] LO 8-6161 Pittsburgh, Pa. 15222: Jeffrey M. Preston 4 Gateway Center. [412] 391-1314 Rochester, N.Y. 14534: William J. Bryte 9 Genesee Street, Pittsford, N.Y. [716] 586-1040 St. Louis, Mo. 63105: Kenneth E. Nicklas The Clayton Tower, 7751 Carondelet Ave [341] PA 5-7265 San Francisco, Calif. 94111: Don Farris Richard A. Ferrara 425 Battery Street [415] 362-6004 Paris: Alan Offergeld 17 Rue-Georges Bizet, 75 Paris 16, France Tel: 720-73-01 Geneva: Alan Offergeld 1 rue de la Paix, Geneva, Switzerland Tel: 32-35-63 United Kingdom: Keith Martle Tel: Hyde Park 1451, 34 Dover Street London W1 Milan: Robert Sadel, Roberto Laureri 1 via Baracchini Phone 86-90-656 Brussels: Alan Offergeld 22 Chaussée de Wavre Bruxelles 1030 Belgium Tel: 13-16-30 Stockholm: Brian Bowers Office 11, Kontor-Center AB, Hagagatan 29 11 347 Stockholm Tel: 24 72 00 European Special Project Manager: Oliver Ball Tel: Hyde Park 1451, London 34 Dover Street Frankfurt/Main: Fritz Krusebecker Leeuustrasse 27c Phone 72 00 11 Tokyo: Masaru Wakeshima, McGraw-Hill Publications Overseas Corporation Kasumigaseki Building 2-5, 3-chome, Kasumigaseki, Chiyoda-Ku, Tokyo Japan [501] 188-1 Osaka: Akinori Kamemasa McGraw-Hill Publications Overseas Corporation, Kondo Bldg., 163, Umeda-cho Kitaku [662] 8771 Australia/Asia: Warren E. Ball, IPO Box 5106, Tokyo, Japan Business Department Stephen R. Weiss, Manager [212] 971-2044 Thomas M. Egan, Production Manager [212] 971-3140 Carol Gallagher Advertising Production Manager [212] 971-2045 Dorothy Carter, Contracts and Billings [212] 971-2908 Frances Vallone, Reader Service Manager [212] 971-6057 Electronics Buyers' Guide George F. Werner, Associate Publisher [212] 971-3139 Regina Hera, Directory Manager [212] 971-2544 ## Electronics advertisers \| Company Name \| Page \| \|--------------\|------\| \| Accelerators, Inc. \| 17 \| \| AEI Semiconductors Limited \| 35E \| \| MCR Advertising Limited \| \| \| Allen-Bradley Company \| 20-21 \| \| Hoffman-York Inc \| \| \| American Motors Corp. \| P B 8-9 \| \| Westinghouse Electric Inc \| \| \| Amphenol Controls Group. \| 53 \| \| Bunker Ramo Corporation \| \| \| Mansfield, Inc. \| \| \| AOIP \| 33E \| \| A. P. Circuit Corp. \| 126 \| \| Erickson American Advertising Agency \| \| \| Avionic & Digital Equipment Ltd. \| 8E \| \| Hughes Aircraft Co., Ltd \| \| \| Bausch & Lomb, Inc. \| 115 \| \| Wolf Assoe Inc \| \| \| Beckman Instruments, Inc., Hellpolt Div. \| 116 \| \| J. W. McDonald Inc \| \| \| Bell & Howell - Electronics & Instruments Group \| 35 \| \| Cogswell Advertising, Inc \| \| \| Birch Stolec Ltd. \| 25E \| \| Garfield & Co. Public Relations \| \| \| Boeing Electronic Products \| 48 \| \| N. W. Ayer, F E Baker Inc \| \| \| Bourne, Inc. \| 43 \| \| McLaughlin Assoc. Inc \| \| \| Business Week Guide \| P B 12 \| \| Bennett, Inc. \| \| \| By-Buk Company, Sub. of Webtek Corp. \| 126 \| \| J. R. Bloome Company \| \| \| California Computer Products \| 106 \| \| Quigley Advertising \| \| \| CELCO (Constantine Engineering Laboratory Co.) \| D-1 \| \| Stano Advertising \| \| \| Constantine Engineering Laboratory Co. (CELCO) \| D-1 \| \| Stano Advertising \| \| \| Cherry Electrical Products Corp. \| 11 \| \| G. L. H. Advertising Assoc., Inc \| \| \| Colorado Video, Inc. \| D-4 \| \| Culligan U S A \| 91 \| \| K. H. K. International \| \| \| Custom Electronics, Inc. \| 114 \| \| Atlas Advertising, Inc \| \| \| Data Control Systems, Inc. \| 6 \| \| Technical Industrial & Scientific Marketing, Inc \| \| \| Dialight Corporation \| D-1 \| \| McLaughlin Assoc. \| \| \| Eastman Chemical Products, Inc. \| \| \| Industrial Chemicals \| D-2 \| \| Wallace Thompson Company \| \| \| EDSYN, INC. \| 122 \| \| ADSYN Advertising Agency \| \| \| Elco Corporation \| 60 \| \| Mori Barish Associates, Inc \| \| \| Eldorado Electrodata Corp. \| 127 \| \| McLaughlin Assoc. \| \| \| Electro Motive Mig. Co., Inc. \| 96 \| \| Culver Adv., Inc \| \| \| Electronics \| 54-55 D-4 32E \| \| Electro Scientific Industries \| \| \| Hughes Advertising, Inc \| \| \| English Electric Valve Co. Ltd. \| 22E-23E \| \| C. R. Casson Ltd \| \| \| Erie Technological Products Company, Inc. \| 18-19 \| \| Almquist & Co. Inc Advertising \| \| \| Excellon Industries, Inc. \| 111 \| \| Elgin \| \| \| Fluke Manufacturing Co., John \| 15 \| \| Bonfield Associates \| \| \| GCA Corporation \| 24 \| \| Golin Advertising, Inc \| \| \| General Dynamics Corp. \| 22-23 \| \| Westinghouse Electric Corp \| \| \| General Electric Co., Miniature Lamp Div. \| 42 \| \| Batten, Barton, Durstine & Osborn Inc \| \| \| General Electric Co., Semiconductor Products Department \| 9 \| \| Advertising & Sales Promotion Syracuse Operation \| \| \| General Instrument Europe S. P. A. \| 17E \| \| Studer \| \| \| General Radio Co. \| 12-13 \| \| GPA Associates \| \| \| Georgia Department of Industry & Trade \| P B 12 \| \| Cargill, Wilson & Acree, Inc \| \| \| Georgia Department of Industry & Trade \| 10 \| \| Cargill, Wilson & Acree, Inc Advertising \| \| \| Gevaert \| 18E \| \| Gould, Inc./Brush Instruments \| 57 \| \| Carl Leggett Adv., Inc \| \| \| Grayhill, Incorporated \| 127 \| \| Carl Leggett Advertising, Inc \| \| \| GTE Sylvania, Parts Division \| 105 \| \| Sperry Rand Corp \| \| \| Gudebrod Bros., Silk Co., Electronics Division \| 120 \| \| Coastline Advertising \| \| \| Hansen Manufacturing Company \| 10 \| \| Kelly-Moore Paints \| \| \| Health Company, Sub. of Schlumberger Ltd. \| 120 \| \| Medical & Health Services \| \| \| Hermes Electronics Ltd. \| 8 \| \| Public & Industrial Relations Limited \| \| \| Hewlett Packard \| 2 \| \| James R. Hewlett, Inc \| \| \| Hewlett Packard, Colorado Springs Division \| 1 \| \| Tatjani-Yates Adv. \| \| \| Hewlett Packard, Cleveland \| 87 \| \| Tatjani-Yates Advertising \| \| \| Hewlett Packard \| 108 & 49 \| \| James R. Hewlett, Inc \| \| \| Hewlett Packard, San Diego Div. \| 119 \| \| Philips Ramsey Advertising \| \| \| Intersil \| 46-47 \| \| Jack Herrick Advertising, Inc \| \| \| Interstate Electronics Corporation \| 41 \| \| Leland Oliver Company, Inc \| \| \| ITT General Controls \| D-1 \| \| MacNamara, John & Adams, Inc. \| \| \| Kahle Engineering Co. \| 126 \| \| Douglas Turner, Inc. \| \| \| Keithley Instruments, Inc. \| 16E \| \| The Bayless-Kelley Co. \| \| \| Lee Electronics Corporation \| 56 \| \| Michael-Catther Inc \| \| \| Lemex S. A. \| 118 \| \| Jose Gonzalez K \| \| \| Litronix, Inc. \| 27 \| \| Litton Systems, Inc. \| \| \| Lyon Metal Products \| P B 7 \| \| Heincke Meyer & Finn, Inc \| \| \| Lyon Van Lines \| P B 15 \| \| W. A. Jorgensen / MacDonald, Inc \| \| \| McGraw-Hill Book Company \| 125 \| \| McGraw-Hill Professional Publishing, Inc \| \| \| Automotive Road Division \| 122 \| \| Magnetic Shield Division, Perfection Mica Co. \| 21 \| \| M. H. K. Engineering/Art \| \| \| MDS-Deutschland GmbH \| 29E \| \| Publicitas GmbH \| \| \| Microtrails \| 128 \| \| TAP (Printing-Advertising-Promotion) \| \| \| MOS Technology, Inc. \| 2nd cover \| \| National Technical Advertising, Inc \| \| \| Mostek Corporation \| 28 \| \| Continental Communications, Incorporated \| \| \| M. H. K. Communications and Electronics, Inc. \| 72 \| \| Brand Advertising Inc \| \| \| Motorola Semiconductor Products, Inc. \| 44-45 \| \| National Semiconductor Corp \| 7 \| \| Chai/Day Inc. Advertising \| \| \| Nikkei America \| 104 \| \| Non-Linear Systems, Inc. \| 125 \| \| Marketing Directions, Inc \| \| \| Northern Natural Gas Company \| 16 \| \| Rozen Advertising, Inc \| \| \| Nortronics Company, Inc. \| 34E \| \| National Semiconductor Corp \| \| \| Pa. Grade Crude Oil Assoc. \| P B 14 \| \| Meidrum & Fewsmit, Inc \| \| \| Philips ELA \| 14E-15E \| \| Philips Nederland International S. A \| 77 \| \| Philips Electrolotica Office Machine Division \| \| \| Philips Nederland delamar \| \| \| Philips GAD Elcoma \| 19E & 36E \| \| Philips GAD Elcoma \| 112 \| \| TAG - Intermarco delamar \| \| \| Philips N. V. Ph.T / & M. Division \| 2E \| \| Philips Nederland International S. A \| \| \| Pignone SUD \| 20E \| \| Lines SPN \| \| \| Plastics Engineering Company \| 59 \| \| Kuttner & Kuttner, Inc \| \| \| Polaroid Corp. \| 5 \| \| James Kane Bembach, Inc \| \| \| Portescap \| 27E \| \| Presse et Publicite \| \| \| Ralston \| 30E \| \| OSW - Werbeagentur \| \| \| Raytheon Semiconductor \| 50 \| \| Raytheon, Inc. \| \| \| RCA Electronic Components \| 4th cover \| \| Al Paul Letter Company \| \| \| RIEGELSTEIN (Investment Div.) \| 127 \| \| Josephson, Cuttler & Company \| \| \| Rohde & Schwarz \| 5E & 13E \| \| Sankyo Development Co., Ltd. \| 26E \| \| OCS Inc \| \| \| Sangamo Electric Company \| 58 \| \| Westinghouse Electric Company \| \| \| Scanbe Manufacturing Corp. \| D-3 \| \| W. A. Sheaffer Pen Co., A Textron Company \| \| \| Schaefer Manufacturing Corp. \| 100 \| \| Nolan, Keeler & Stiles \| \| \| W. A. Sheaffer Pen Co., A Textron Company \| P B 14 \| \| Schering Corporation \| \| \| Schlumberger \| 6E-7E \| \| B. Brown, Ltd \| \| \| Sierra International Corporation \| 14 \| \| B. P. Myers Associates, Inc \| \| \| Sodeco \| 31E \| \| Societe Industrie \| \| \| Sony Superscope \| P B 13 \| \| Sperry Advertising Assoc. \| \| \| Sorensen Corporation Raytheon Company \| 101 \| \| Provanpe, Eastwood & Lombardi, Inc \| \| \| Sperry Information Displays Division \| 3rd cover \| \| A. W. Williams Advertising Agency \| \| \| State Farm Insurance \| P B 2 \| \| National, Harper & Stires \| \| \| Sutton Donner Concord Instruments \| 95 \| \| Bonfield Associates \| \| \| Tektronix, Inc. \| 33 & 92 \| \| Tektronix, Inc. \| \| \| TRW Electronics, IRC Boone Division \| 39 \| \| Westinghouse Electric Company \| \| \| Uniltores Corporation \| 37 \| \| Impact Advertising, Inc \| \| \| Varadyne, Inc., Systems Division \| 102 \| \| W. A. DuPont Advertising \| \| \| V/O Techmashexport \| 28E \| \| V. M. Mongorakrema \| \| \| Wang Laboratories \| P B 11 \| \| W.L Associates, Inc \| \| \| Zeilec, Inc. \| 24E \| \| Advertising & Advertising \| \| ### Classified & Employment Advertising F. J. Beene, Manager of (213) 579-2303 EQUIPMENT (Used or Surplus New) For Sale \| Company Name \| Page \| \|--------------\|------\| \| E. L. K. International \| 121 \| \| Jensen Tools \| 121 \| \| Philip Fleiman \| 121 \| \| Port of New York Authority \| 121 \| \| Radiation Devices Co. \| 121 \| \| Radio Research Instruments Co. \| 121 \| For more information on complete product line see advertisements in the latest Electronics Buyer's Guide Advertisers in Electronics International How Can You Sort Useful Information from 100 Million Bits Per Second? Today the television camera is an extremely versatile data acquisition tool. It can give you around 100 million bits of information per second — spatial dimensions, brightness, color, time — with a wide range of sensitivities. Not all of what the camera sees is usable, however. How can you extract the exact, useful data you need from this source? Colorado Video has the Model 321 Video Analyser specifically designed to help you pinpoint the data you want and eliminate the redundant or useless bits. The 321 can be used on a "stand-alone" basis as a laboratory instrument for such applications as image analysis and camera-to-hard-copy interfacing. Or, the 321 can provide input to your computer for a variety of analysis functions. It can also be controlled by the computer. The 321 now is being used for such diverse operations as automatic instrument calibration via computer, cancer cell analysis and measuring changes in the diameter of human veins. Find out what the Colorado Video Model 321 Video Analyser can do for you. Write or call for information. Price: $2500. COLORADO VIDEO, INCORPORATED P.O. Box 928, Boulder, Colorado 80302. Phone (303) 444-3972 Video Data Acquisition • Processing • Display • Transmission How much free time do you have available for reading? Add up the number of magazines that come across your desk. Multiply them by an average reading time — minutes to scan, an hour or more to dig in. Most people can only spend six hours each week on all sources of occupational information. So how do you fill your professional needs and still keep your reading within reason? Leaf through this issue again. Notice how much of its editorial is designed to keep you aware of trends, show you how new technologies are being applied, and give you information on new products. If you want to keep yourself up to date and cut down on your reading time, do the same thing with every issue of Electronics. When you make Electronics your regular habit, you'll find you have only occasional need for other electronics magazines. More information, in less reading time, is one reason it pays you to have your own subscription to Electronics. Electronics. It's all you need to read. Use these handy postcards for more detailed information on products advertised, new products, new literature. Circle the number on the Reader Service postcard that corresponds to the number at the bottom of the advertisement, new product item, or new literature in which you are interested. Please print clearly. All written information must be legible to be correctly processed. If the cards have already been used, you may obtain the needed information by writing directly to the manufacturer, or by sending your name and address, plus the Reader Service number, to Electronics Reader Service Department, Box 444, Hightstown, N.J. 08520. All inquiries from outside the U.S. that cannot reach Electronics before the expiration dates noted on the Reader Service postcard must be mailed directly to the manufacturer. The manufacturer assumes all responsibilities for responding to inquiries. Electronics merely provides and clears requests for information from inquirer to manufacturer. Correct amount of post must be affixed for all mailings from outside the U.S. Multi-product advertisements For information on specific items in multi-product advertisements which do not have a specific Reader Service number indicated, write directly to manufacturer for information on the product in which you are interested. To subscribe to or to renew Electronics Fill in the subscription card adjoining this card. You need to send no money. Electronics will bill you at the address indicated on the card. Only Electronics editorial matter listed on this page available in reprint form: You may order any of the below listed reprints by key number. Discounts on quantities over 10. Bulk reprints of editorial matter can be ordered from current or past issues. The minimum quantity is 500 copies. Prices quoted on request: call 609-448-1700 Ext. 5494, or write to address below. Reshaping The EE For The Seventies 24 pages. $1.00 Key no. R-18 Special Report on New Look in LSI Packaging. Parts I & II. 16 pages. $2.00 Key no. R-15 Electronics Markets (1971). U.S.A. & European Combined Forecasts. 46 pages. 2 fold-out charts. Bound. $5.00. Key no. R-13 Computer-Aided Design. 148 pages. $4.00. Key no. R-11 Active Filter Chart & Infra-Red Detector Chart. 2 Electronics 1969 Special Guide Charts. $1.00 includes both 4-color charts. Key No. R-033 Active Filters. 96 pages. $4.00. Key no. R-032 Circuit Designers Casebook. 217 pages. $4.00. Key no. R-031 Consumer Hazards: Why they Happen, How they Can Be Fixed. 14 pages. $2.00. Key no. R-027 Special Report: Tomorrow's Communications. 32 pages. $3.50. Key no. R-023 Special Report on Ferrites. 16 pages. $2.00. Key no. R-017 Special Report on The Transistor: Two Decades of Progress. 48 pages. $3.00. Key no. R-016 Special Report on Gallium Arsenide. 7 parts. 32 pages. $4.00. Key no. R-012 Special Report on Large Scale Integration. 54 pages. $3.00. Key no. R-010 Field Effect Transistors. Parts I, II, and III. 64 pages. $4.00. Key no. R-64 1970 Electronics Index to Technical Articles and Authors. $1.00. Key no. R-32 To order reprints or for further information, please write to: Electronics Reprint Department P.O. Box 669 Hightstown, N.J. 08520 Only orders with cash, check, or money order will be accepted. No invoicing will be done. Sperry explodes the LED myth There has been a lot said in recent months about LED's representing the most significant advance in display technology and how they are destined to dominate the digital display market. We feel it's time to explode the myth and set the record straight. So, here's a direct, point-by-point, comparison of Sperry seven segment gas discharge planar displays vs LED displays. COST For the price of a single 1/4" LED digit you can buy three 1/2" or three 1/3" Sperry display digits. And, in the future, the Sperry displays should continue to be less expensive than LED displays. Gives you something to think about, doesn't it? SIZE Let the size speak for itself. READABILITY Have you tried to read a 1/6" or even a 1/4" LED display at 20'? On the other hand, the Sperry 1/3" display is easy to read at that distance and the 1/2" model can be read at up to 40'. See the difference! COLOR With LED's, you have the choice of red, red or red. Not so with Sperry. They come in an eye appealing orange — with amber and red available with filters. If you like red, why pay more for a LED? APPEARANCE Which do you prefer — looking at individual red dots on LED devices or at continuous unbroken Sperry figures? The choice is yours. BRIGHTNESS Sure you can read LED's indoors, but how about in bright light or direct sunlight? LED's fade fast while Sperry displays stay clearly legible with no appreciable loss in brightness. And, Sperry devices won't poop out when it gets hot! Sperry advantages don't stop here either. The small Sperry package is only a shade larger than a LED and nearly as thin. Sperry power dissipation is also significantly lower. And, Sperry reliability is so good that they have proven fail-safe in some of the world's toughest environments including the Boeing 747. There are no wire bonds to go bad, either. Don't just take our word for it. Arrange for a comparison demonstration and see for yourself what the difference will mean to your particular application. For complete technical information on Sperry displays, use this publication's reader service card or phone or write: Sperry Information Displays Division P.O. Box 3579, Scottsdale, Arizona 85257 Telephone (602) 947-8371 SPERRY INFORMATION DISPLAYS It's a whole new ball game in display devices! † Patents pending based on 1,000 digit quantity, and above. Sperry displays are available in 3 digit, 2 digit, and 1 1/2 (7 segment character and a with + and —) digit models in both 1/4" and 1/2" sizes. The easy way to communicate Just look at the charts. The basic facts they present dramatize RCA's single-source capability in total product for your communications needs. These devices are performance-proved. You get maximum dependability in communications applications. For more information, see your RCA Representative or call Microwave Marketing: (201) 485-3900; TWX 710 995-4474. International: RCA, 2-4 rue du Lièvre, 1227 Geneva, Switzerland, or Sunbury-on-Thames, U.K., or P.O. Box 112, Hong Kong. TWT's Choose RCA traveling-wave tubes. Select from one of the industry's widest lines, supported by a 20-year background in developing and field-testing state-of-the-art systems. \| RCA Type \| Power Output (W) \| Frequency (GHz) \| Application \| \|----------\|------------------\|-----------------\|-------------\| \| A1378 \| 20 \| 7.9 – 8.4 \| Government \| \| A1446 \| 20 \| 8.0 – 10.7 \| Government \| \| A1447 \| 5 \| 7.9 – 8.4 \| Government \| \| A1390 \| 20 \| 10.7 – 11.7 \| Common Carrier \| \| A1443 \| 15 \| 10.7 – 11.7 \| Common Carrier \| \| A1455 \| 20 \| 12.7 – 12.95 \| Community Antenna Relay System \| \| A1458 \| 20 \| 10.7 – 13.2 \| Common Carrier, Community Antenna Relay System \| \| A1460 \| 15 \| 13.0 – 14.0 \| Government \| TEA's Now you have the transferred-electron amplifier, introduced by RCA to the industry earlier this year. This is the first available octave bandwidth solid-state amplifier for C-band and above. \| RCA Type \| Power Output (mW) \| Frequency (GHz) \| Application \| \|----------\|-------------------\|-----------------\|-------------\| \| S384 \| 250 \| 6.0 – 8.0 \| Common Carrier \| \| S385 \| 250 \| 8.0 – 10.0 \| Government, Common Carrier \| \| S386 \| 120 \| 14.0 – 18.0 \| Community Antenna Relay System \| TEO's You have accepted the transferred-electron oscillator as another of RCA's major contributions to microwave design. Now you have a choice in standard products and subsystems for pulsed and CW applications from L-through Ku-band. \| RCA Type \| Power Output (mW) \| Frequency (GHz) \| Application \| \|----------\|-------------------\|-----------------\|-------------\| \| S409 \| 100 \| 6.0 – 8.0 \| Common Carrier \| \| S410 \| 30 \| 8.0 – 10.0 \| Government, Common Carrier \| \| S411 \| 30 \| 10.0 – 12.0 \| Common Carrier \| \| S412 \| 10 \| 12.0 – 14.0 \| Business Radio, Community Antenna Relay System \|
The production of pharmaceutical proteins from the milk of transgenic animals * LM Houdebine INRA, unité de différenciation cellulaire, 78352 Jouy-en-Josas cedex, France (Received 12 June 1995; accepted 7 September 1995) Summary — The preparation of recombinant proteins of pharmaceutical interest from the milk of transgenic animals is becoming a reality. No protein has reached the market yet but several have been prepared in large quantities not only from laboratory animals but also from ruminants (goat and sheep) and pigs. Rabbit appears more and more to be an intermediate animal well adapted for the preparation of limited amounts of proteins. Several problems remain to be solved to optimize the method. The expression level of genes of interest associated with milk protein gene control regions is usually unpredictable. The recombinant proteins secreted in milk are not always in a satisfactory biochemical form. Cleavage and glycosylation are not always carried out correctly. The problem of the possible presence of agents pathogenic for humans in proteins extracted from milk is not completely solved. Prions have not been found in mammary glands and other milk pathogens may be controlled using good practice in breeding. recombinant protein / transgenic animal / milk Résumé — Production de protéines d'intérêt pharmaceutique à partir du lait d'animaux transgéniques. La préparation des protéines recombinantes d'intérêt pharmaceutique à partir du lait d'animaux transgéniques est en train de devenir une réalité industrielle. Aucune protéine n'est encore sur le marché mais plusieurs sont produites en abondance non seulement par des animaux de laboratoire mais aussi par des ruminants (chèvre et mouton) ainsi que par des porcs. Le lapin apparaît de plus en plus comme un animal intermédiaire bien adapté pour la préparation de quantités limitées de protéines. Des problèmes restent encore à résoudre pour optimiser la méthode. Le taux d'expression des gènes d'intérêt associés aux éléments régulateurs des gènes des protéines du lait reste le plus souvent imprévisible. Les protéines produites dans le lait ne sont pas toujours sous une forme biochimique satisfaisante. Certains clivages ainsi que certaines glycosylations ne sont parfois réalisés que partiellement. Le problème de la présence éventuelle d'agents pathogènes pour l'homme dans les préparations de protéines recombinantes issues du lait n'est pas complètement résolu. Cet obstacle ne paraît pas infranchissable. Les prions ne semblent en effet pas présents dans la glande mammaire et le lait et la présence d'autres agents pathogènes peut probablement être contrôlée par de bonnes pratiques d'élevage. protéine recombinante / lait / animal transgénique * This review was written for the 6th European Congress of Biotechnology which took place in Nice, 20–24 February, 1995. The idea that the milk from transgenic animals could be the source of recombinant pharmaceutical proteins was suggested several years ago. The first experimental demonstration that this goal can be reached was given by Simons et al (1987) who succeeded in producing sheep β-lactoglobulin in the milk of transgenic mice. To obtain a protein not naturally secreted in milk, gene constructs containing the regulatory region of a milk protein gene fused to the coding region of the gene of interest must be prepared. The promoter regions from most of the milk protein genes from different species are used for this purpose. The promoter from αS₁-β-caseins, β-lactoglobulin, whey acidic protein (WAP), α-lactalbumin and even the long terminal repeat (LTR) from mouse mammary tumor virus (MMTV) have thus been fused to a variety of genes and used to generate transgenic mice, rats, rabbits, goats, sheep, pigs and cows. Probably more than 50 foreign proteins have been produced in this way in the milk of transgenic mammals. The published data was reported in a recent review (Houdebine, 1994). Since this review was published, additional proteins have been obtained in milk. Among these proteins are human lysozyme (Maga et al, 1994), human insulin-like growth factor-1 (IGF-1) (Maga et al, 1994), bovine chymosin (Brem et al, 1995), human plasminogen activator (Riego et al, 1993; Ebert et al, 1994), human growth hormone (Tojo et al, 1993; Ninomiya et al, 1994), human γ-interferon (Dobrovolsky et al, 1993), human lactoferrin (Platenburg et al, 1994), human erythropoietin (Hyttinen et al, 1994; Castro et al, 1995), human protein C (Drohan et al, 1994; Paleyanda et al, 1994; Wei et al, 1995), and human albumin (Barash et al, 1993; Hurwitz et al, 1994). Although little doubt remains on the use of transgenic animals as living fermentor, a certain number of problems are still to be solved before this method can become a common industrial process to produce recombinant proteins. THE GENERATION OF TRANSGENIC MAMMALS The first transgenic mouse was obtained 13 years ago. The method originally defined for mouse and relying on the direct microinjection of gene into the pronucleus of one-cell embryos has been extended to other mammals. Minor adaptations are sufficient to obtain many transgenic rats and rabbits. For farm animals, the yield of transgenics is dramatically low in most cases. Injecting a polycation-DNA complex instead of DNA into cytoplasm, although of moderate efficiency, leads to the generation of transgenic animals. This approach may improve the situation when pronuclei are not visible (Velander et al, unpublished data). The availability of embryos is a limiting factor for ruminants. It is now possible to generate a number of one-cell-stage embryos from cow ovary at a low cost by performing in vitro oocyte maturation and fertilization and to use them for transgenesis (Krimpenfort et al, 1991; Hyttinen et al, 1994). The in vitro generation of embryo from ovary has been extended successfully to sheep and goat (N Crozet et al, personal communication). The same is not true for porcine embryos at present. To tentatively reduce the number of recipient females, embryos can be cultured until the blastocyst stage. Only those that were not damaged by the micromanipulation remain at the end of the culture and are candidates to be transferred to females. The identification of transgenic embryos at the blastocyst stage is theoretically possible by using PCR, starting from a few blastomers. In practice, this method is poorly reliable, due to traces of remaining unintegrated DNA which gives numerous false positives (King and Wall, 1988; Ninomiya et Another method based on the discrimination of DNA methylation in the original insert and in the integrated gene may be used (Cousens et al, 1994). Fluorescence in situ hybridization, which proved to be a simple and reliable method to detect transgenes in newborn animals (Swiger et al, 1995), might be helpful in identifying transgenic embryos using isolated blastomers. The coinjection of reporter genes, such as that coding for vargula luciferase but not involving an invasive test on blastocysts, may be an attractive alternative (Thompson et al, 1995). This reporter gene has been associated with the heat stock protein 70 gene promoter and the matrix-attached regions (MAR) from human interferon-β gene. With such a gene construct, the vargula luciferase can be synthesized, secreted under heat stock and measured in the blastocyst culture medium. The presence of the MARs in the gene construct favours the expression of the integrated DNA and is expected to identify the blastocysts in which the microinjected DNA is integrated and to eliminate those in which unintegrated DNA is still present. A reliable and easy method to identify genuine transgenic blastocysts therefore remains unavailable for the time being. The transfer of foreign genes through embryonic stem (ES) cells or primordial germ cells (McLaren, 1992) remains impossible so long as these cells cannot be obtained in a reliable manner in species other than mouse. The recent observation that spermatogonia can colonize a foreign testis after a direct microinjection in this tissue and give birth to normal mice in good yield opens new avenues (Brinster and Avarbock, 1994; Brinster and Zimmermann, 1994). Gene transfer through spermatogonia awaits additional experiments showing that these cells can be cultured, transfected and retain their capacity to differentiate to functional spermatozoa after reimplantation into recipient testis. THE CONSTRUCTION OF EFFICIENT VECTORS Many experiments have led to the conclusion that native genes are often efficiently expressed as transgenes whereas their cDNA counterparts are not. The presence of intron, which is generally not required in cultured cells, is thus of paramount importance for transgenes (Brinster et al, 1988; Palmiter et al, 1991). Several studies did not reveal which combinations of promoters, introns and terminators are the best to obtain high and predictable expression of cDNA in transgenic animals (Palmiter et al, 1991). Selecting the best introns, 5'- and 3'-untranslated regions and terminators to be associated with the promoter of a given milk protein gene is a possible approach. In vitro and in vivo studies have thus shown that the intron from SV40 early genes is much less efficient than the intron from SV40 late genes and the SIS generic intron (Petitclerc et al, 1995). Some gene constructs containing a given cDNA associated with efficient introns and transcription terminators may however be much less efficient in vivo than the corresponding genomic fragment, although they were significantly more potent in cultured cells (Petitclerc et al, 1995). Introducing the foreign cDNA in the middle of milk protein genes, keeping some of their introns and exons in the cDNA situated before and after proved to be efficient in several cases (Brem et al, 1994; Drohan et al, 1994; Hyttinen et al, 1994; Maga et al, 1994; Brem et al, 1995). Using P1 phage or YAC vectors may be useful if long fragments of DNA surrounding milk protein genes or if long foreign genes are to be used. The coinjection of 2 overlapping complementary regions which recombine in vivo may also be performed. This simple procedure proved to be efficient in several species, including rabbit to regenerate functional genes from 2 pieces. At least 60% of the transgenic rabbits obtained after coinjections of overlapping DNA fragments containing the rabbit WAP promoters and human factor VIII cDNA harboured a reconstituted complete gene (Attal et al, unpublished results). Promoters from milk protein genes used so far show quite different potency. The most efficient promoters seem to be those from ruminant $\alpha$S$_1$-casein (Brem et al, 1994; Platenburg et al, 1994), $\beta$-casein (Ebert et al, 1994; Maga et al, 1994), sheep $\beta$-lactoglobulin (Barash et al, 1993; Brash et al, 1994; Hurwitz et al, 1994; Platenburg et al, 1994) and rabbit WAP (Bischoff et al, 1992; Devinoy et al, 1994). A distribution of the upstream regulatory regions specific of each species may explain some of these differences. The expression of transgenes is generally highly influenced by their chromatin environment. A few genes are expressed independently of the integrating site. This is the case for sheep $\beta$-lactoglobulin (Whitelaw et al, 1992) and rat WAP gene (Krnacik et al, 1995). This is not true when the $\beta$-lactoglobulin gene promoter is associated with a foreign gene (Barash et al, 1994). In most cases, the DNA sequences responsible for transgene insulation have not been defined. AT-rich MAR have been shown to amplify and insulate integrated genes. This was observed to some extent when chicken lysozyme MAR insulating region was associated with mouse WAP promoter (McKnight et al, 1992). No effect was seen when the MAR located in the 3'-OH of human apolipoprotein B 100 gene and in the SV 40 genome was added of both side of constructs containing the rabbit WAP gene promoter (Attal et al, unpublished data). Insulating transgenes therefore remains possible so far only in a limited number of cases. Introducing a foreign cDNA within a milk protein gene by using ES cells and homologous recombination should provide efficient promoter, introns, terminators and insulators. This technical approach was successfully used to replace mouse $\alpha$-lactalbumin and $\beta$-casein genes by homologous inactive mutated genes (Kumar et al, 1994; Stinnakre et al, 1994). These preliminary successes suggest that foreign proteins will be produced in milk of farm animals when functional ES cells are available. THE PREDICTION OF GENE CONSTRUCT EFFICIENCY Evaluating the potency of gene constructs in vitro before using them to generate transgenics would be very helpful. Indeed, a relatively long time elapses between the construction of the genes and the measurement of the foreign proteins in milk (especially of course when males are founders), even when transgenic mice are used. The mouse mammary cell line HC11 (Ball et al, 1988) can be utilized to evaluate the efficiency of gene constructs. All researchers admit however that the correlation between the potency of a construct in HC11 cells and transgenics is poor (Petitclerc et al, 1995). Transgenic mice are usually predictive of what will be observed in other transgenic mammals. In one case at least, however, a gene construct containing the mouse WAP promoter and the human protein C gene was poorly efficient in mice and of quite acceptable potency in the pig (Velander et al, 1992). Unexpectedly also, the same gene construct was ectopically expressed in sheep (Wall et al, 1995) but not in pig. Introducing a gene directly into the mammary gland is possible. This was achieved using cationic lipids (MA Sirard, personal communication), biolistics (Furth et al, 1992), and retroviral vectors (Archer et al, 1994). The expression of the gene was then low. It could be higher if gene transfer was carried out with the available potent adenoviral vectors. Whatever happens, this procedure is not expected to be able to replace transgenesis to express large amounts of foreign proteins in many animals and it has little chance of allowing a good prediction of the efficiency of gene constructs used as transgenes. Indeed, these methods are in vivo transfection and the foreign genes are not then submitted to the numerous modifications of chromatin which occur during embryo development and organogenesis. The direct gene transfer into functional mammary gland may however provide interesting information on the capacity of the mammary cell to correctly modify a given foreign protein post-translationally. THE CHOICE OF THE ANIMAL SPECIES For the production of very large amounts of foreign protein (more than 100 tons per year), transgenic cows seem the most appropriate. For quantities higher than 1 kg per year, sheep, goats or even pigs are valuably used. When no more than 1 kg of recombinant protein per year is needed, rabbits which can be milked with an adapted machine (Lebas, 1970; Duby et al, 1993), may be attractive, due to the relatively low cost of transgenic generation; rabbit also appears to be appropriate as a transition animal to obtain easily sufficient amounts of recombinant proteins for the study of their biochemical and biological properties before using larger animals to produce these proteins on an industrial scale. Mice can provide only very limited amounts of milk. Milking these animals is possible. It is also easy to collect the whole milk from mammary glands by keeping them on ice for a few hours (Stinnakre et al, 1992). THE BIOLOGICAL EFFECTS OF THE RECOMBINANT PROTEINS ON HEALTH OF TRANSGENIC ANIMALS Most of the recombinant proteins secreted in milk of transgenic animals are to be used potentially as pharmaceuticals for humans. Most of these proteins are biologically active in various mammals. The ectopic and non-temporally regulated expression on the transgenes proved to have detrimental effects on animals. Mouse WAP led to milchlos phenotype (Burdon et al, 1991). Human growth hormone (hGH) induced subfertility and gigantism in mice (Devinoy et al, 1994). Human erythropoietin cDNA fused to rabbit WAP promoter led to a marked accumulation of red blood cells, a delay in growth, a severe subfertility and a premature death of transgenic rabbits (Attal et al, unpublished data). Ectopic or non-temporally regulated expression of transgenes might be significantly attenuated by adding insulators to the constructs (see above) or by controlling the expression of the transgenes by an external inducer such as tetracyclin (Furth et al, 1994). An induced homologous recombination using the cre-lox system (Barinaga, 1994) might also contribute to reduced expression of the transgenes out of the lactation period. An improvement of the specificity of transgene expression might be obtained by replacing a milk protein gene by the gene of interest. This was achieved recently in mice expressing human $\alpha$-lactalbumin gene instead of their own (Stacey et al, 1994, 1995). It should be noted that this approach remains limited to mice so long as ES cells are not available in other species. Moreover, the in situ association of a foreign gene of interest with the regulatory sequence of a milk protein gene using homologous recombination may be not efficient in all cases. On the other hand, whey proteins such as $\alpha$-lactalbumin or WAP are spontaneously transferred from milk to blood during lactation (Grabowsky et al, 1991). This is also the case for recombinant proteins such as human growth hormone (GH) (Devinoy et In some cases, even when the expression of the transgene is perfectly controlled, the health of the animals may be severely altered by the recombinant proteins transferred from milk to blood. PURIFICATION OF THE RECOMBINANT PROTEINS FROM MILK Milk is not an exceedingly complex biological fluid. It contains a large quantity of a few proteins and only limited proteolytic activity. Conventional methods to purify proteins seem appropriate to isolate recombinant proteins from whey (Ebert et al., 1991; Wilkins and Kuys, 1992). In particular cases, when recombinant proteins are hydrophobic, they are secreted with lipid globules. This was the case for human cystic fibrosis transmembrane conductance regulator (CFTR) (Di Tullio et al., 1992). The major concern with recombinant proteins from animal origin is not the classical biochemical purity. The presence of pathogens in the purified proteins may be detrimental for humans as soon as the proteins are injected, and in some cases, such as albumin or haemoglobin, in very large amounts. The presence of prions is a potential problem. The capacity of these molecules to be pathogens for other species is not well documented. In case this problem cannot be solved the inactivation of the PrP gene by homologous recombination can be envisaged although presently only in mouse. In the mouse, the inactivation of this gene did not alter greatly the health of the animals. However, the problem may be not so crucial since no prion has been found so far in mammary gland extract and milk from cows suffering from bovine spongiform encephalopathy. Transgenic mice harbouring sheep PrP gene instead of their own might be used to detect the presence of prion in fractions isolated from sheep milk. THE BIOCHEMICAL STRUCTURE OF THE RECOMBINANT PROTEINS FROM MILK Animals cells rather than bacteria or yeast are used to produce a certain number of recombinant proteins to tentatively obtain polypeptides with all the post-translational modifications. It is by no means certain that the mammary gland can perform all these modifications in a convenient manner. This seems unfortunately to be the case. Human-α₁-antitrypsin from sheep milk (A Colman, unpublished data), human anti-thrombin III from goat milk (ES Cole, unpublished data) are not fully glycosylated. The in vitro enzymatic addition of the missing terminal sialic acid increases significantly the half-life of the protein in vivo without apparently altering its biological activity. Human protein C precursor from pig milk is not completely processed and not fully γ-carboxylated (Drohan et al., 1994). Most likely, the mammary cells of different species do not have exactly the same enzymatic equipment for protein maturation. It may be therefore hazardous in some cases to extrapolate from model animals such as the mouse and the rabbit to predict the structure of a given recombinant protein in pig or ruminants. Modifying the enzymatic equipment of the mammary cell through transgenesis to improve maturation of recombinant proteins is conceivable. It may however lead to disfunction of the mammary gland in some cases and a long investigation is needed before such an approach can be envisaged. The fact that recombinant proteins from milk do not always have the same structure as their native counterpart may be of limited importance. The possible side-effects of the recombinant proteins extracted from milk must thus be evaluated case by case. ALTERNATIVE PRODUCTION SYSTEMS Milk is presently considered as the best biological fluid to produce recombinant proteins from animals. Blood may be appropriate in some cases, when the recombinant proteins are not too unstable or too toxic for the animals. This was the case for human $\alpha_1$-antitrypsin which has been produced from liver at the concentration of 1 mg/ml with apparently a glycosylation similar to the native protein (Massoud et al, 1991). Human haemoglobin (Swanson et al, 1992) and peptides have also been produced in reticulocytes from transgenic animals. The purification of human proteins from the blood of transgenic animals may not be easy in some cases. Blood may thus be of limited interest in practice. Egg white might become, in the future, a good source of recombinant proteins by transgenic birds. Transgenic plants can also be the source of recombinant proteins. Hepatitis B surface antigen has been prepared in this manner (Mason et al, 1992). Plants may be appropriate when very large amounts of simple polypeptides must be prepared without any pathogens for humans. Human albumin and haemoglobin are good candidates to be extracted from transgenic plants. OTHER USES OF TRANSGENESIS TO EXPRESS FOREIGN GENES IN THE MAMMARY GLAND Gene transfer is used to study growth, differentiation (Jhappan et al, 1993; Pierce et al, 1993) and oncogenesis (Cardiff and Muller, 1993) of the mammary gland or other aspects of mammary gland function. Milk represents about 30% of the proteins of human food in rich countries. Its composition may be changed to improve yield or quality of the product for dairy industry. This may be achieved by modifying the concentration or amino acid composition of milk proteins. Milk is not only a food. It contains various growth factors, anti-pathogen proteins such as lysozyme, lactoferrin, antibodies. Milk may become a vehicle to provide humans with number of factors and it may thus become a novel food (Clark, 1992; Yom et al, 1993). Conventional gene transfer, targeted gene knock-out and mutation in farm animals may greatly contribute to reaching this goal. REFERENCES Archer JS, Kennan WS, Gould MN, Bremel RD (1994) Human growth hormone (hGH) secretion in milk of goats after direct transfer of the hGH gene into the mammary gland by using replication-defective retrovirus vectors. Proc Natl Acad Sci USA 91, 6840-6844 Ball RK, Friis RR, Schoenenberger CA, Doppler W, Groner B (1988) Prolactin regulation of beta-casein gene expression and of a cytosolic 120-kd protein in a cloned mouse mammary epithelial cell line. EMBO J 7, 2089-2095 Barash I, Baruch A, Nathan M, Shani M, Hurwitz DR (1993) HSA production by mammary explants of virgin transgenic mice: a reliable tool for predicting levels of secretion into milk. Anim Biotechnol 4, 203-215 Barash I, Faerman A, Ratovitsky T et al (1994) Ectopic expression of $\beta$-lactoglobulin/human serum albumin fusion genes in transgenic mice: hormonal regulation and in situ localization. Transgenic Res 3, 141-151 Bardinaga M (1994) Knockout mice: round 2. Science 265, 26-28 Bischoff R, Degryse E, Perraud F et al (1992) A 17.6 kb region located upstream of the rabbit WAP gene directs high level expression of a functional human protein variant in transgenic mouse milk. FEBS Lett 305, 265-268 Bowen RA, Reed ML, Schnieke A et al (1994) Transgenic cattle resulting from biopsied embryos: expression of c-ski in a transgenic calf. Biol Reprod 50, 664-668 Brem G, Hartl P, Besenfelder U, Wolf E, Zinovieva N, Pfaller R (1994) Expression of synthetic cDNA sequences encoding human insulin-like growth factor-1 (IGF-1) in the mammary gland of transgenic rabbits. Gene 149, 351-355 Brem G, Besenfelder U, Zinovieva N, Seregi J, Solti L, Hartl P (1995) Mammary gland specific expression of chymosin constructs in transgenic rabbits. Theriogenology 43, 175 Brinster RL, Avarbock MR (1994) Germline transmission of donor haplotype following spermatogonial transplantation. Proc Natl Acad Sci USA 91, 11303-11307 Brinster RL, Zimmermann JW (1994) Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci USA 91, 11298-11302 Brinster RL, Allen JM, Behringer RR, Gelinas RE, Palmiter RD (1988) Introns increase transcriptional efficiency in transgenic mice. Proc Natl Acad Sci USA 85, 836-840 Burdon T, Wall RJ, Shamay A, Smith GH, Henninghausen L (1991) Over-expression of an endogenous milk protein gene in transgenic mice is associated with impaired mammary alveolar development and a milchlos phenotype. Mech Dev 36, 67-74 Cardiff RD, Muller WJ (1993) Transgenic mouse models of mammary tumorigenesis. The Molecular Pathology of Cancer, Cancer Surveys 16 Castro FO, Aguirre A, Fuentes P, Ramos B, Rodriguez A, de la Fuente J (1995) Secretion of human erythropoietin by mammary gland explants from lactating transgenic rabbits. Theriogenology 43, 184 Clark AJ (1992) Prospects for the genetic engineering of milk. Mol Cell Biochem 49, 121-127 Cousens C, Carver AS, Wilmut I, Colman A, Garner I, O'Neill GT (1994) Use of PCR-based methods for selection of integrated transgenes in preimplantation embryos. Mol Reprod Develop 39, 384-391 Devinoy E, Thépot D, Stinnakre MG et al (1994) High levels of production of human growth hormone in the milk of transgenic mice: the upstream region of the rabbit whey acidic protein (WAP) gene targets transgene expression to the mammary gland. Transgenic Res 3, 79-89 Di Tullio P, Cheng SH, Marshall et al (1992) Production of cystic fibrosis transmembrane conductance regulator in the milk of transgenic mice. Bio/Technology 10, 74-77 Dobrovolsky VN, Lagutin OV, Vinogradova TV, Frolova LS, Kuznetsov VP, Larionov OA (1993) Human γ-interferon expression in the mammary gland of transgenic mice. FEBS 319, 181-184 Drohan WN, Zhang DW, Paileyanda RK et al (1994) Inefficient processing of human protein C in the mouse mammary gland. Transgenic Res 3, 355-364 Duby RT, Cunniff MB, Belak JM, Balise JJ, Robl JM (1993) Effect of milking on collection of milk from nursing New-Zealand white rabbits. Anim Biotechnol 4, 31-42 Ebert KM, Selgrath JP, DiTullio P et al (1991) Transgenic production of a variant of human tissue-type plasminogen activator in goat milk; generation of transgenic goats and analysis of expression. Bio/Technology 9, 835-838 Ebert KM, DiTullio P, Barry CA et al (1994) Induction of human tissue plasminogen activator in the mammary gland of transgenic goats. Bio/Technology 12, 699-702 Furth PA, Shamay A, Wall RJ, Hennighausen L (1992) Gene transfer into somatic tissues by jet injection. Anal Biochem 205, 365-368 Furth PA, St Onge L, Böger H et al (1994) Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter. Proc Natl Acad Sci USA 91, 9302-9306 Grabowsky H, Le Bars D, Chene N et al (1991) Rabbit whey acidic protein concentration in milk, serum, mammary gland extract, and culture medium. J Dairy Sci 74, 4143-4154 Houdebine LM (1994) Production of pharmaceutical proteins from transgenic animals. J Biotechnol 34, 269-287 Hurwitz DR, Nathan M, Barash I, Ilan N, Shani M (1994) Specific combinations of human serum albumin introns direct high level expression of albumin in transfected COS cells and in the milk of transgenic mice. Transgenic Res 3, 365-375 Hyttinen JM, Peura T, Tolvanen M et al (1994) Generation of transgenic dairy cattle from transgene-analyzed and sexed embryos produced in vitro. Bio/Technology 12, 606-608 Jhappan C, Geiser AG, Kordon EC et al (1993) Targeting expression of a transforming growth factor beta1 transgene to the pregnant mammary gland inhibits alveolar development and lactation. EMBO J 12, 1835-1845 King D, Wall RJ (1988) Identification of specific gene sequences in preimplantation embryos by genomic amplification: detection of a transgene. Mol Reprod Develop 1, 57-62 Krimpenfort P, Rademakers A, Eyestone W et al (1991) Generation of transgenic dairy cattle using in vitro embryo production. Bio/Technology 9, 844-847 Krnacik MJ, Li S, Liao J, Rosen JM (1995) Position-independent expression of whey acidic protein transgenes. J Biol Chem 270, 11119-11129 Kumar R, Clarke AR, Hooper ML et al (1994) Milk composition and lactation of β-casein-deficient mice. Proc Natl Acad Sci USA 91, 6138-6142 Lebas F (1970) Description d'une machine à traire les lapines. Ann Zootech 19, 223-228 Maga EA, Anderson GB, Huang MC, Murray JD (1994) Expression of human lysozyme mRNA in the mammary gland of transgenic mice. Transgenic Res 3, 36-42 Mason HS, Lam DMK, Arntzen CJ (1992) Expression of hepatitis B surface antigen in transgenic plants. Proc Natl Acad Sci USA 89, 11745-11749 Massoud M, Bischoff R, Dalemans W et al (1991) Expression of active recombinant human alpha-1-antitrypsin in transgenic rabbits. J Biotech 18, 193-204 McKnight RA, Shamay A, Sankaran L, Wall RJ, Henninghausen L (1992) Matrix-attachment regions can impart position-independent regulation of a tissue-specific gene in transgenic mice. Proc Natl Acad Sci USA 89, 6943-6947 McLaren A (1992) The quest for immortality. Nature (Lord) 359, 482-483 Ninomiya T, Hoshi M, Mizuno A, Nagao M, Yuki A (1989) Selection of mouse preimplantation embryos carrying exogenous DNA by polymerase chain reaction. Mol Reprod Dev 1, 242-248 Ninomiya T, Hirabayashi M, Sagara J, Yuki A (1994) Functions of milk protein gene 5'-flanking regions on human growth hormone gene. Mol Reprod Dev 37, 276-283 Paleyanda RK, Zhang DW, Henninghausen L, McKnight RA, Lubon H (1994) Regulation of human protein C gene expression by the mouse WAP promoter. Transgenic Res 3, 335-343 Palmiter RD, Sandgren EP, Avarbock MR, Allen DD, Brinster RL (1991) Heterologous introns can enhance expression of transgenes in mice. Proc Natl Acad Sci USA 88, 478-482 Petitclerc D, Attal J, Théron MC et al (1995) The effect of various introns and transcription terminators on the efficiency of expression vectors in various cultured cell lines and in the mammary gland of transgenic mice. J Biotechnol 40, 169-178 Pierce Jr DF, Johnson MD, Matsui Y et al (1993) Inhibition of mammary duct development but not alveolar outgrowth during pregnancy in transgenic mice expressing active TGF-beta1. Genes Dev 7, 2308-2317 Platenburg GJ, Kootwijk EPA, Kooiman PM et al (1994) Expression of human lactoferrin in milk of transgenic mice. Transgenic Res 3, 99-108 Riego E, Limonta J, Aguilar A et al (1993) Production of transgenic mice and rabbits that carry and express the human tissue plasminogen activator cDNA under the control of a bovine alpha S1 casein promoter. Theriogenology 39, 1173-1185 Simons JP, McClenaghan M, Clark AJ (1987) Alteration of the quality of milk by expression of sheep beta-lactoglobulin in transgenic mouse. Nature (Lord) 328, 530-532 Stacey A, Schnieke A, McWhir J, Cooper J, Colman A, Melton DW (1994) Use of double-replacement gene targeting to replace the murine $\alpha$-lactalbumin gene with its human counterpart in embryonic stem cells and mice. Mol Cell Biol 14, 1009-1016 Stacey A, Schnieke A, Kerr M et al (1995) Lactation is disrupted by $\alpha$-lactalbumin deficiency and can be restored by human $\alpha$-lactalbumin gene replacement in mice. Proc Natl Acad Sci USA 92, 2835-2839 Stinnakre MG, Devinoy E, Thépot D et al (1992) Quantitative collection of milk and active recombinant proteins from the mammary gland of transgenic mice. Anim Biotechnol 3, 245-255 Stinnakre MG, Vilotte JL, Soulier S, Mercier JC (1994) Creation and phenotype analysis of $\alpha$-lactalbumin-deficient mice. Proc Natl Acad Sci USA 91, 6544-6548 Swanson ME, Martin MJ, O'Donnell JK et al (1992) Production of functional human hemoglobin in transgenic swine. Bio/Technol 10, 557-559 Swiger RR, Tucker JD, Heddle JA (1995) Detection of transgenic animals without cell culture using fluorescence in situ hybridization. BioTechniques 18, 952-957 Thompson EM, Adenot P, Tsuji FI, Renard JP (1995) Real time imaging of transcriptional activity in live mouse preimplantation embryos using a secreted luciferase. Proc Natl Acad Sci USA 92 (in press) Tojo H, Tanaka S, Matsuzawa A, Takahashi M, Tachi C (1993) Production and characterization of transgenic mice expressing a hGH fusion gene driven by the promoter of mouse whey acidic protein (mWAP) putatively specific to mammary gland. J Reprod Dev 39, 145-155 Velander WH, Johnson JL, Page RL et al (1992) High-level expression of a heterologous protein in the milk of transgenic swine using the cDNA encoding human protein C. Proc Natl Acad Sci USA 89, 12003-12007 Wall RJ, Rexroad Jr CE, Powell A, Shamay A, McKnight R, Henninghausen L (1995) Mouse whey acidic protein gene is expressed ectopically in transgenic sheep. Theriogenology 43, 346 Wei Y, Yarus S, Greenberg NM, Whitsett J, Rosen JM (1995) Production of human surfactant protein C in milk of transgenic mice. Transgenic Res 4, 232-240 Whitelaw CB, Harris S, McClenaghan M, Simons JP (1992) Position-independent expression of the ovine beta-lactoglobulin gene in transgenic mice. Biochem J 286, 31-39 Wilkins RJ, Kuys YM (1992) Rapid beta-lactoglobulin genotyping of cattle using the polymerase chain reaction animal. Genetics 23, 175-178 Yom HC, Bremel RD (1993) Genetic engineering of milk composition: modification of milk components in lactating transgenic animals. Am J Clin Nutr 58, 299S-306S
PLAINTIFFS’ MOTION FOR LEAVE TO PRESENT FOURTH SUPPLEMENT TO MOTION FOR RECONSIDERATION REGARDING EMERGENCY TEMPORARY PRELIMINARY INJUNCTION AND SUPPLEMENT With reference to Federal Rules of Civil Procedure, Rule 15, the Plaintiffs Ethan Book and Ethan Book for U.S. Senate hereby present a Motion for Leave to Present Fourth Supplement to the pending Motion for Reconsideration Regarding Emergency Temporary Preliminary Injunction (#11), an unopposed pleading which was supplemented on October 6, 2010 (#23) and also on November 9, 2010 (#28) and November 12, 2010 (#30) (Of which the Plaintiffs to date have not received notice of Defendants’ opposition to any of these pending pleadings.). For simplicity and judicial economy, the same Fourth Supplement is incorporated together with this Motion for Leave. In this regard, pursuant to Rule 65(a) of the Federal Rules of Civil Procedure and in accordance with the claims for relief stated in the Amended Complaint (#27), the Plaintiffs hereby reaffirm their pending and unopposed pleadings to the effect to move that this Court issues a preliminary injunction to suppress the results of the portion of the August 10th Republican Primary Election pertaining to the position of U.S. Senator, to suppress the portion of the November 2nd General Election pertaining to the position of U.S. Senator and to direct Defendant Bysiewicz to schedule a special election four months from the date of such Court ruling to grant the Preliminary Injunction. As additional support for key areas of claims in support of the Plaintiffs’ position regarding a Preliminary Injunction, the Plaintiffs briefly describe here the various Exhibits which accompany this Motion and Supplement and then they discuss the relevant information as it supports the Plaintiffs’ claims: Exhibit 1: A formal Statement provided by the Plaintiffs on November 19, 2010 to the Honorable Nicholas Panuzio, Chairman of the Election Panel which was established by the City of Bridgeport to review issues of defects of the voting process experienced in the City during the general election of November, 2, 2010 (at A-1). Exhibit 2: A formal Supplement submitted by the Plaintiffs to Defendant Bysiewicz on November 22, 2010 (at A-16), a Supplement to the Petition presented by the Plaintiffs on November 3, 2010 (See Amended Complaint, #27, at its Exhibit 14, p. A-47.). Exhibit 3: A letter sent on November 23, 2010 to the Plaintiffs on by presumed Associate Attorney General Joseph Rubin (at A-26), a letter sent on behalf of Defendant Blumenthal as Attorney General, in response to matters including of Freedom of Information requests sent by the Plaintiffs to Defendant Blumenthal on November 21, 2010 (a letter which is an enclosure to Exhibit 2 at A-22). Exhibit 4: The full actual text of an e-mail message sent during the early afternoon of November 24, 2010 by Defendant Bysiewicz to the Plaintiffs (at A-27), that in response only to the Plaintiffs Supplemental Petition of November 22, 2010. Exhibit 5: The full actual text of an e-mail message sent by the Plaintiffs to Defendant Bysiewicz on November 28, 2010 (at A-28). The Plaintiffs hereby proceed to discuss the information of these various Exhibits as they pertain to Plaintiffs claims both of the pending pleadings for Preliminary Injunction as well as to related claims of the Amended Complaint. EXHIBIT 1: Exhibit 1 is a formal Statement made by the Plaintiffs to the Bridgeport Election Panel on November 19, 2010. It is a Supplement to an earlier Statement presented by the Plaintiffs on November 17, 2010. There are several relevant portions of credible and undisputed information provided and discussed in that Statement which are relevant to and which effectively merge with issues of the present. Prominent in the Statement is information regarding what the Plaintiffs describe as “weed-gate”, that is credible, corroborated and undisputed information regarding the participation of Christopher Dodd in the early 1980’s in the illegal importation of a plane-load of marijuana from Latin America to the United States, also a disturbing set of facts pertaining to “Lincoln-gate”, the unlawful action by a Bridgeport police officer on May 10, 2010 (curiously just prior to the Connecticut Republican Convention of May 22, 2010) to order unlocked, searched and to order seized a 2000 Lincoln Town Car which has been used in Plaintiff Book’s livelihood (Amended Complaint at Para. 13). The significance of “weed-gate” to the present is that, using the same legal references which have been used in the present for stating a claim (re: Pls. Opposition, #19, at pgs. 10 and 28 – 30), including that a claimant must present sufficient for a claim which is more than merely possible but rather is plausible, for Plaintiff Book’s role even before the curiously concurrent public announcements on January 7, 2010 by Senator Dodd that he would not seek re-election and of Defendant Blumenthal that he would seek the Democrat Party nomination for the position of U.S. Senate, there is more than a plausible showing that the emerging “weed-gate” was a factor in Senator Dodd’s decision. In addition, for what is discussed in Exhibit 1, it is at least plausible that the emerging potential political scandal of “weed-gate” was also a factor in the action of the Obama administration to include a $100 million political “favor” for Senator Dodd, that being a grant for the construction of a university hospital, just while the Obama administration in December of 2009 was frantically attempting to secure sufficient votes in the Senate for passage of the widely disputed and contested Obama health care legislation. This is a very important factor to understand as one of the starting points, a foundation made in the sand, of the political campaign of Defendant Blumenthal. It is in this setting that there also emerges a separate “Lincoln-gate”, that is the unlawful action on May 10, 2010 of Ofcr. John Gallant of the Bridgeport Police Department to order searched and seized a Lincoln Town Car owned by Plaintiff Book and used in his small transportation business, an alternative livelihood to Book’s disrupted banking career. The respective incident report indicates that the vehicle was suspected to have been stolen, however, the vehicle was not then stolen, Book had never reported it to be stolen, Book was not specifically contacted regarding such a matter, there was no credible information to satisfy any showing of probable cause that it was stolen, and further that at the time that Ofcr. Gallant was at the vehicle with a local private towing operator, a neighbor to Book called out, “Don’t take that car. It belongs to a man who is candidate for the U.S. Senate. He lives over there!” as she gestured to Book’s current temporary residence [a personal living situation which is a direct result of claims raised in Ethan Book Jr. v. Sgt. Kimberly Parks and the City of Bridgeport Police Department et al., Case No. 3:09-cv-0472(AVC), among others]. As the Plaintiffs asserted in their accompanying Statement (beginning at A-3), “it strikes me that my Lincoln may have been removed by Ofcr. Gallant, not simply despite the fact that I am a candidate for the U.S. Senate but rather because I am an active and threatening (for substantive issues which I raise and represent) Republican candidate for the U.S. Senate, and/or that there has been extended delay for the City officials resolving the matter because I am a Republican candidate for the U.S. Senate, among other issues”. Thus, it is at least plausible that some officials of the Bridgeport Police Department, that in a City where the municipal government has for a long-term been overwhelmingly Democrat, has for a number of years acted as an arm of the political campaign of Defendant Blumenthal, a campaign which very clearly began long before his public announcement of January 7, 2010 (additional supporting information available: Again, this is one area where claims of the present merge with proper claims presented in Book v. Parks et al.). Further, all the above in the full context of Exhibit 1, and of the credible claims of this lawsuit, this background also affords more than a plausible explanation of the high level of attention placed to the City of Bridgeport and the State of Connecticut election process by President Obama with his personal visit and public appearances on October 30, 2010 and by former President Bill Clinton with his personal visit on October 31, 2010, just prior to the recent general election (See discussion at A-4.). What the Plaintiffs have been doing is feared by those for whom the information and the issues is most suited for public exposure and review! EXHIBIT 2: Exhibit 2 is a Supplement to the formal Petition which the Plaintiff’s presented to Defendant Bysiewicz on November 3, 2010 (Amended Complaint at its Exhibit 14, p. A-47). In addition to the substantive legal issues of the lack of validity of the McMahon Certificate of Endorsement and the lack of validity Blumenthal Certificate of Endorsement (that for reasons including the failure of Defendant Bysiewicz to have obtained the statutorily-required approval of the legislative Regulation Review Committee for the form which was prepared in 2006 for Certificate of Endorsement), there are also additional issues of the failure of Defendant Bysiewicz to have obtained proper approval for the form which was prepared also in 2006 and distributed by municipal Registrars of Voters for Voter Registration, and further the failure of the municipal Registrars, that as per instructions of Defendant Bysiewicz, to reasonably verify citizenship and further the failure of local and state law enforcement to prosecute for false statements made on the Voter Registration forms, among other issues. For these and related reasons, there is both no legal validity to the election process for the position of U.S. Senator and also there can be NO PUBLIC CONFIDENCE in the election and voting processes statewide! In addition, the Plaintiffs give reference to yet another citizen of this State who has observedly been the object of other overreaching excess of statutory authority for Defendant Blumenthal, that where there was an apparent attempt at political showcasing (at Exhibit 2, its p. A-18, para. 2). This additional newly-obtained information for which the Plaintiffs are able to provide additional detail in a suitable forum, is corroborative to Plaintiffs’ claims against Defendant Blumenthal. As additional relevant support to such issues of statutory excesses and lack of minimal due process protections (all of which supplement and add to what the Plaintiffs presented to Defendant Bysiewicz on June 24, 2010; at Amended Complaint, its Exhibit 10 at p. A-36) and also of specific claims against Defendant Blumenthal (Amended Complaint at pgs. 12 – 19), more supporting information is observed of matters of the estimated $60 million paid by the State to three private law firms (including one firm which is a former employer of Defendant Blumenthal) for professional services in representing the State in the multi-state litigation against several tobacco companies (Amended Complaint at Para. 49). The Plaintiffs’ letter to Defendant Blumenthal of November 21, 2010 and which is an enclosure to Exhibit 2 (at A-22) gives good credible detail about Defendant Blumenthal’s chain of official errors, both with respect to that unauthorized contractual arrangement with the three private law firms as well as of his official conduct of evading his own public responsibility with respect to the Plaintiffs’ proper and timely requests for information. Additional discussion of these matters is found below regarding Exhibit 3. EXHIBIT 3: Exhibit 3 is a letter sent on November 23, 2010 by presumed Associate Attorney General Joseph Rubin on behalf of Defendant Blumenthal in response to the Plaintiffs’ letter of November 21, 2010 (beginning at A-22). Several points of detail deserve comments. The letter includes an important paragraph which explains that “[a]s you know, the Freedom of Information Act requires the production of existing non-exempt documents. It does not require state officials to answer your questions, nor to provide authorization or releases for others to provide you with documents”. However, the Plaintiff’s letter was not presented solely pursuant to the Freedom of Information Act. A quick review of the Plaintiffs’ letter shows that it contains material that, in addition to matters of the Freedom of Information Act, can be considered as a basic citizen request, expediting litigation and/or informal discovery in the present, and also considered as political campaign debate. The letter is in no way exclusive to the Freedom of Information Act! In addition, simply for the nature of earlier responses by Attorney Rubin and Defendant Blumenthal to prior related correspondence, they in fact previously did answer various questions. Therefore, what Exhibit 3 suggests is that the Plaintiffs have gotten Defendant Blumenthal into an uncomfortable corner in which he has no other proper responses but to admit serious error, that which he thereby elects to avoid and evade. Additional review is proper. Exhibit 3 is a partial response to prior requests whereby Attorney Rubin provides the Plaintiff with two contracts regarding representation on behalf of the State by the three private law forms in the multi-state tobacco litigation. One of those contracts is between the Attorney General and the law firm of Silver, Golub and Teitell. That law firm is known as a former employer to Defendant Blumenthal. Also, the contract reflects that the law firm was, through some unknown review process, designated to be the lead counsel for the three referenced law firms. The other contract is between the Attorney General and the three law firms as participants in that arrangement. The texts of the two contracts are essentially identical with the exception of one being for the lead law firm and one for all three participants. Several observations become notable. In the Plaintiffs’ Motion for Reconsideration of Court Ruling Regarding Articulation (#22), the Plaintiffs provided a copy of a formal complaint presented on September 26, 2010 against Defendant Blumenthal to the Freedom of Information Commission. At page 1, paragraphs 2 and 3 of that Complaint, the Plaintiffs point out that on August 19, 2010, they presented to Defendant Blumenthal a formal request for information in which they particularly sought “that documentation which would reflect the amounts paid [to the three contracted law firms] regarding their professional representation of state interests” in the multi-state tobacco litigation and that on August 22, 2010, Attorney Rubin sent a letter “in which he explained that the respective ‘tobacco counsel did not seek or receive any fees from the state’ as ‘[t]he state was not a party to that process and has no documents regarding the amount of payment of fees’”. However, Section 3 of the signed contract between Defendant Blumenthal and lead counsel Silver, Golub and Teitell deals with “Compensation and Reimbursement”. Sub-section 3.2 follows: Reimbursement of disbursements will be paid only after the submission of itemized documentation, in a form acceptable to the Attorney General, the Associate Attorney General or their respective designees. The Contractor agrees to maintain contemporaneous time and expense records. Upon the demand of the Attorney General or his designee, the Contractor shall submit statements to the Attorney General setting forth a detailed description of the services performed, the time spent in performance of each service and any disbursements to said date. This contractual material appears to be diametrically opposed to what was represented to the Plaintiffs by Defendant Blumenthal on August 22, 2010. Also, the Plaintiffs’ letter to Defendant Blumenthal of November 21, 2010 (enclosed with Exhibit 2) reflects that in response to a Plaintiffs’ request for documentation which would evidence “the identify of who or what entity has or had such documentation”, the official response given was that “[t]his office has no such documentation” (at A-22). However, the above quoted portion of the actual contract fully contradicts that statement! The Plaintiffs letter further deals with a proper issue regarding Freedom of Information law concerning an agency provision by law or by contract for access to public information held by others (at p. 2). Reference is made to a statement made by Defendant Blumenthal that “[t]his office is not aware of any law which entitles this Office to receive such documents”. Considering the above Section 3.2 of the actual contract, such a statement is, at the very least, materially incomplete and misleading! In addition, Sub-section 3.4 provides that “[t]he Contractor shall maintain accurate records and accounts of all expenditures under this Agreement as well as satisfactory evidence of payment to assure proper accounting. Such records and accounts shall be kept . . . and made available and furnished upon request of the Attorney General or his designee until six (6) years after completion of the Services under this Agreement or termination of this Agreement”. This provision is read in the context of Sub-section IX(c) of the Master Settlement Agreement (at p.56) which affirms the term of required payments under the Agreement “beginning April 15, 2000 and on April 15 of each year thereafter in perpetuity . . . .” Thus, the various participating law firms should, to this date, have the information which the Plaintiffs seek! It is also relevant that Sub-section 7.5 of the Contract between the Attorney General and the lead contracting law firm provides that “[t]he Contractor and all members of the Contractor Group have performed a detailed conflict of interest check prior to performing any Services . . . .” With other known information, one wonders how Defendant Blumenthal and the participating law firms define the terms that they use! Further, Sub-section 7.6 establishes that “all information or data, in any form, and all papers, recordings, documents and instruments generated or collected by the Contractor, any member of the Contractor Group . . . in the scope of his work under this Agreement shall be deemed to be the exclusive property of the State of Connecticut . . . .” This is quote potent summary reference! The significance of this newly-obtained information compared against prior representations made by or on behalf of Defendant Blumenthal and under his responsibility are major! Defendant Blumenthal early evaded public responsibilities regarding the disclosure of public information regarding the arrangements made between him and the three private law firms as he did not want public scrutiny of the full details as to what he had unlawfully and unethically entered into as he knew that such information, in the context of his other “misstatements” about his military service, could be very politically damaging to his campaign, and also detrimental to his ineffective defenses in this lawsuit. Finally, considering the Plaintiffs’ specific request for such documentation which would evidence the authorization for the agency decision to hire Attorney Robert Snook, the Defendants’ legal counsel in this matter, Attorney Rubin’s stated recent position of not answering questions effectively affirms what the Plaintiffs asserted in their letter of November 21, 2010 (enclosed to Exhibit 2 at p. 3, para. 2). For such a lack of proper statutorily-required authorization, that a matter which is repeated in state government, there is effectively no legal defense to the Defendants in this matter! EXHIBIT 4: This document is a full printed copy of an e-mail sent by Defendant Bysiewicz to the Plaintiffs at about 1:38 p.m. on Wednesday, November 24, 2010. That message was transmitted about one hour prior to Defendant Bysiewicz’ public announcement of having certified the November 2\textsuperscript{nd} election. In the e-mail message, she states simply that regarding merely the Plaintiffs’ correspondence of November 22, 2010 (Exhibit 2), she “will take it under advisement as we move forward and finalize the election results of 2010”. In other words, rather than address the substantive legal issues raised by the Plaintiffs’, Defendant Bysiewicz again chooses to ignore the law and simply to proceed with arbitrary and unrestrained political will! Considering the substance of what the Plaintiffs raised in their Petition of November 3, 2010 (at Amended Complaint, its Exhibit 14, p. A-47.), the related issues of the Plaintiffs’ Petition of June 24, 2010 (Amended Complaint, its Exhibit 10, p. A-36), and the multiple unaddressed issues of statutory authority and the statutorily required hearing provisions of the Uniform Administrative Procedure Act, there is observed a further continuation of extreme deliberate indifference by Defendant Bysiewicz, of violations of multiple constitutional provisions and of other legal errors! EXHIBIT 5: This is the Plaintiffs’ written e-mail response to Defendant Bysiewicz, a message transmitted on November 28, 2010. In this, the Plaintiffs correctly assert properly founded legal, jurisdictional and due process issues! As of the preparation of this pleading, the Plaintiffs have received no response. FURTHER DISCUSSION: Plaintiff Book has experienced in Connecticut a long-term pattern of political opportunism and of lack of restraint and lawlessness by those in positions of power. A proper analogy is the current recent ethics matter of New York Congressman Charles Rangle. For Plaintiff Book, this situation has been evidenced of officials state agencies and state courts not recognizing legitimate statutory and legal limitations such as is observed in his pursuit of public information regarding the bidding for the CRRA’s Bridgeport Resource Recovery Project (CRRA et al. v. Freedom of Information Commission et al., 19 Conn.App. 489), of limitations of governmental power regarding regulation of his small transportation business, of legitimate constitutional protections and also privilege for appeal in the misdemeanor matter administered in Stamford Superior Court in 2001, of jurisdictional issues of personal jurisdiction and corporate authority as pertains to a mortgage foreclosure lawsuit, that arising as a direct consequence of the mentioned petty misdemeanor matter, and other similar and related matters (See Amended Complaint.). Now, Defendant Bysiewicz again follows the pattern of acceding to political will to disregard the laws which have been established by and for the people! These Defendants including particularly Defendant Blumenthal and Defendant Bysiewicz have been riding waves of political advantage. For most public officials, there is a scope of legitimate public interest which is coupled with some element of personal, private interest. The general public hopes that the public portion of such interest far outweighs the personal side. What is apparent from this docket is that the personal side of the office function for these Defendants is much greater than would be desired or expected. In addition, as such deviations and dysfunction normally have a life cycle, such waves of advantage eventually play out and end. It is further apparent from this docket that these Defendants, similar to other public figures, errantly presumed that their waves would never end or that they were smart enough to ensure that the waves would continue. However, the patterns and realities of life and the fact of the Hand of God acting to ensure eventual justice have brought the waves to an end. Neither Defendant Blumenthal, Defendant Bysiewicz nor the general public, reasonably informed of these matters, would have any real confidence in Defendant Blumenthal now officially becoming U.S. Senator for this Constitutional State. Thus, the Plaintiffs have well established legal, equitable and ethical bases in support of their position with respect to the claims of this lawsuit as well as regarding their pending, unopposed pleadings regarding a Preliminary Injunction! For good and lawful causes, the Plaintiffs’ pleadings seeking a Preliminary Injunction can and should be promptly granted by this Court! Respectfully submitted for Ethan Book and Ethan Book for U.S. Senate Ethan Book P.O. Box 1385 Fairfield, CT 06825 Tel. (203) 367-8779 email@example.com November 23, 2010 Ethan Book P.O. Box 1385 Fairfield, CT 06825 Dear Mr. Book: I write in reply to your letter dated November 21, 2010 to Attorney General Richard Blumenthal. I acknowledge receipt of your check for $8.50 and accordingly enclose copies of the two contracts you requested. As you know, the Freedom of Information Act requires the production of existing non-exempt public documents. It does not require state officials to answer your questions, nor to provide authorizations or releases for others to provide you with documents. This office has no further additional public records to provide in response to your most recent letters. Very truly yours, Joseph Rubin Associate Attorney General JR:sd Enclosures Dear Mr. Book: I have reviewed your correspondence dated November 22, 2010 and will take it under advisement as we move forward and finalize the election results for 2010. Sincerely, Susan Bysiewicz Secretary of the State Dear Secretary of State Bysiewicz: Thank you for your note of early afternoon of November 24th. While I appreciate that you explain that you will take my correspondence of November 22nd under advisement while the State moves forward and finalizes the election results for 2010, given that my correspondence is a supplement to my formal Petition of November 3rd in which, for good lawful causes, I requested that you refrain from certifying the votes for the portion of the November 2nd election pertaining to the position of U.S. Senator and also given that issues raised in my November 3rd Petition raise legitimate jurisdictional issues, it is not proper for you to have attempted to certify the November 2nd general election without having prior fully answered the Petition. In addition, considering that the Petition and the supplement properly raise issues of voidness for multiple excesses of statutory authority and lack of minimal due process protections for key steps of the election process leading up to and including November 2nd, your attempt to certify the election on the afternoon of November 24th is equally void. For additional review of these matters, pursuant to the Freedom of Information Act, Conn. General Statutes, Section 1-200 et seq., please promptly provide me a copy of the document by which you attempted to certify on November 24th the election results. I will greatly appreciate your full, prompt cooperation. Sincerely, Ethan Book Tel. (203) 943-0045 www.ethanbookforussenate.org Constitutional integrity and individual freedom! CERTIFICATION: I hereby certify that a complete copy of the foregoing Plaintiffs’ Motion for Leave to Supplement and Supplement was mailed by 1st class regular mail in accordance with Rule 5(b) of the Federal Rules of Civil Procedure on November 30, 2010 to the following: Robert D. Snook Office of the Attorney General 55 Elm Street P.O. Box 120 Hartford, CT 06141-0120 firstname.lastname@example.org Also, notice of this pleading has been electronically presented to the legal representative of the Defendants as well also to other potentially interested persons such as Christopher Healy, Linda McMahon and Peter Schiff, among others. Ethan Book
Fate of Pyroxenite-derived Melts in the Peridotitic Mantle: Thermodynamic and Experimental Constraints S. LAMBERT\textsuperscript{1,2}, D. LAPORTE\textsuperscript{1,2}, A. PROVOST\textsuperscript{1,2} AND P. SCHIANO\textsuperscript{1,2} \textsuperscript{1}LABORATOIRE MAGMAS ET VOLCANS, CLERMONT UNIVERSITÉ, UNIVERSITÉ BLAISE PASCAL, BP 10448, 63000 CLERMONT-FERRAND, FRANCE \textsuperscript{2}CNRS, UMR 6524, IRD, R 163, 5 RUE KESSLER, F-63038 CLERMONT-FERRAND CEDEX, FRANCE RECEIVED FEBRUARY 7, 2011; ACCEPTED NOVEMBER 16, 2011 ADVANCE ACCESS PUBLICATION JANUARY 23, 2012 We performed a thermodynamic and experimental study to investigate the fate of pyroxenite-derived melts during their migration through the peridotitic mantle. We used a simplified model of interaction in which peridotite is impregnated by and then equilibrated with a finite amount of pyroxenite-derived liquid. We considered two pyroxenite compositions and three contexts of pyroxenite melt impregnation: (1) in a subsolidus lithospheric mantle; (2) beneath a mid-ocean ridge (MOR) in a subsolidus asthenospheric mantle at high pressure; (3) beneath a MOR in a partially molten asthenospheric mantle. Calculations were performed with pMELTS at constant pressure and temperature with a melt–rock ratio varying in the range 0–1. Concurrently, a series of impregnation experiments was performed at 1 and 1.5 GPa to reproduce the final stages of the calculations where the melt–rock ratio is 1. Incoming melt and host-rocks react differently according to the melt composition and the physical state of the surrounding mantle. Whereas clinopyroxene (Cpx) is systematically a reaction product, the role of olivine (Ol) and orthopyroxene (Opx) depends on the incoming melt silica activity $a_{\text{SiO}_2}$: if it is lower than the silica activity $a^{\text{eq}}_{\text{SiO}_2}$ of a melt saturated in Ol and Opx at the same pressure $P$ and temperature $T$, Opx is dissolved and Ol precipitates, and conversely if $a_{\text{SiO}_2} > a^{\text{eq}}_{\text{SiO}_2}$. Such contrasted reactions between pyroxenitic melts and peridotitic mantle may generate a large range of new lithological heterogeneities (wehrlite, websterite, clinopyroxenite) in the upper mantle. Also, our study shows that the ability of pyroxenite-derived melts to migrate through the mantle depends on the melting degree of the surrounding peridotite. The reaction of these melts with a subsolidus mantle results in strong melt consumption (40–100%) and substantial Cpx production (with some spinel or garnet, depending on $P$). This is expected to drastically decrease the system permeability and the capacity of pyroxenite-derived melts to infiltrate neighbouring rocks. In contrast, melt migration to the surface should be possible if the surrounding mantle is partially melted; although liquid reactivity varies with composition, melt consumption is restricted to less than 20%. Hence, magma–rock interactions can have a significant impact on the dynamics of melting and magma migration and should not be neglected when modelling the partial melting of heterogeneous mantle. KEY WORDS: experimental petrology; melt–peridotite interactions; MORB genesis; pMELTS; pyroxenite INTRODUCTION The upper mantle is likely to contain a significant fraction of pyroxenite: 2–5% according to Hirschmann & Stolper (1996), based on the proportion in orogenic peridotite massifs. Mantle pyroxenites show a large spectrum of bulk-rock compositions and modal proportions, presumably reflecting different origins [crystallization of asthenospheric melts percolating through the lithospheric mantle (Shervais \textit{et al.}, 1973); \textit{in situ} formation by metamorphic differentiation (Dick & Sinton, 1979); recycling of subducted oceanic crust into the asthenosphere (Polvé & Allègre, 1980); among others] and different degrees of interaction and re-equilibration with the surrounding peridotites. Whatever their origin, pyroxenites may play an important... role in controlling the chemical variability of mantle-derived melts (Sleep, 1984; Allègre & Turcotte, 1986; Sobolev et al., 2007). The role of pyroxenite lithologies in oceanic basalt generation has been studied in numerous geochemical (e.g. Le Roux et al., 2002; Sobolev et al., 2007) and experimental (e.g. Kogiso et al., 2004a, and references therein) studies. Even if the fraction of pyroxenite in the upper mantle is small in comparison with peridotite, pyroxenites may indeed contribute to a significant proportion of the total melt production owing to their lower solidus temperatures and higher melt productivities (e.g. Hirschmann & Stolper, 1996; Pertermann & Hirschmann, 2003; Lambart et al., 2009a). To fully describe partial melting processes in heterogeneous mantle, it is, however, necessary to take into account the interactions between the pyroxenites and/or pyroxenite-derived melts and the surrounding peridotite, which might have a significant role in melt extraction dynamics and, ultimately, in the preservation of a pyroxenite signature in aggregated melts erupted at the Earth’s surface. Pyroxenite–peridotite interaction processes depend on many chemical and physical parameters such as the size of the pyroxenite bodies (Kogiso et al., 2004b), the composition of the pyroxenites and pyroxenite-derived melts (e.g. Morgan & Liang, 2005), and the nature of the transport mechanism (pervasive porous flow, focused flow in dunite channels or magma transport in dikes). Accordingly, these processes are very difficult to model or simulate in the laboratory. A few experimental studies have been carried out to better understand the chemical interactions between silicate melts and host peridotite. Morgan & Liang (2003, 2005) performed kinetic experiments on interactions between basaltic melt and peridotite, and observed more or less significant modifications of the peridotite assemblage as a function of melt composition. In this type of disequilibrium experiment, however, it is difficult to quantify the reactions involved and their impact on melt and peridotite composition. Yaxley & Green (1998) and Pilet et al. (2008) performed ‘sandwich’ experiments, in which a layer of material with a low solidus temperature was packed between layers of peridotite. Yaxley & Green (1998) used a mid-ocean ridge basalt (MORB)-type eclogite (GAI) and performed their experiments at 3·5 GPa, whereas Pilet et al. (2008) used a hornblendeit (AG4) at 1·5 GPa. Unlike Morgan & Liang’s (2003, 2005) experiments, most of these experiments were run below the solidus temperature of peridotite, and so simulated the case of a partially molten pyroxenite body within a subsolidus peridotite host. These two studies led to opposite results: melts from GAI were consumed to produce orthopyroxene (Opx) whereas melts from AG4 dissolved Opx from the surrounding peridotite and precipitated olivine (Ol). Yaxley & Green’s (1998) and Pilet et al.’s (2008) studies deal chiefly with the effect of melt–rock reaction on melt composition, but a comparison of their results underlines the importance of the initial conditions (pressure and composition) for the reaction and shows that magma–rock interactions may have a significant impact on the lithological diversity of the mantle. A natural example is provided by pyroxenites from the Ronda Massif (SW Spain): according to Garrido & Bodinier (1999), some of these were produced by interaction between pyroxenite-derived alkali basalt melt and peridotite. A major aspect of the interaction of pyroxenite-derived melts with host peridotite is the coupled dissolution of Opx and precipitation of Ol (or vice versa). This can be described by a simple reaction such as \[ \text{Opx} + \text{Liq}_0 \leftrightarrow \text{Ol} + \text{Liq}_1. \] (1) In the case of a simple system at constant pressure and temperature, in which liquid, Ol and Opx are the only phases present, thermodynamic (Kelemen, 1990) and experimental (Daines & Kohlstedt, 1994) studies have shown that the mass ratio of precipitated Ol to dissolved Opx, $ M_{\text{Ol}} / M_{\text{Opx}} $, is $ \sim 0.7 $. Hence, Opx dissolution and Ol precipitation increase the proportion of melt in the system and facilitate its transport by increasing the porosity and permeability of mantle rocks. Inversely, Opx crystallization at the expense of Ol leads to a reduction of porosity and may slow down or even stop magma ascent by porous flow. This case highlights that basalt–peridotite interactions could significantly affect the rate of transport of magmas from their source region to the surface. In this study, we try to evaluate the fate of melts from pyroxenitic sources during their transport through the peridotite mantle as functions of their composition, $ P-T $ conditions and the physical state (subsolidus vs partially molten) of the surrounding mantle. Three cases are considered in order to discuss the impact of melt–rock interactions on the chemical composition of melts, the chemical and mineralogical compositions of the surrounding peridotite, and the ability of the melts to infiltrate the surrounding rock. (I) Partial melting of pyroxenites in a subsolidus lithospheric mantle. If heterogeneous lithospheric mantle is affected by a temperature increase (for example, in the context of asthenospheric upwelling and lithospheric extension), pyroxenites—with lower solidus temperatures than peridotite (e.g. Hirschmann & Stolper, 1996)—will melt first and pyroxenite-derived melts will tend to infiltrate and react with the surrounding peridotite. Here, we want to test whether peridotite impregnation by a pyroxenitic melt can produce a new generation of pyroxenites with specific features and so increase the ‘lithodiversity’ of the mantle, as suggested by Garrido & Bodinier (1999). Partial melting of pyroxenites in MORB-source mantle at high P (2–5 GPa). During mantle upwelling beneath mid-ocean ridges (MOR), pyroxenites start to melt at greater depths than peridotite owing to their lower solidus temperature. Here, our main objective is to determine whether the chemical exchange between the pyroxenites and subsolidus peridotite favour melt transport over relatively long distances or else lead to the isolation of partially molten pyroxenite bodies in unmolten peridotite. Partial melting of pyroxenites in the MORB-source mantle at low P (1 GPa). At low pressure beneath MORs, both pyroxenites and peridotite will be partially molten. In a previous experimental study (Lambart et al., 2009a), we showed that melts from pyroxenites and peridotites have similar major element compositions at 1–1.5 GPa, and so we predicted minimal interactions between pyroxenite-derived melts and the surrounding mantle and the preservation of a pyroxenite chemical signature during melt transport to the Earth’s surface. Our purpose here is to test this hypothesis. To model these three cases, we used a simplified model of interaction in which peridotite is impregnated by a finite amount of pyroxenite-derived melt. We used the thermodynamic model pMELTS (Ghiorso et al., 2002) and the Adiabat-lph front end (Smith & Asimow, 2005) to compute the proportions and compositions of solid and liquid phases in the impregnated peridotite after chemical re-equilibration, as a function of the composition and amount of pyroxenite-derived melt. Concurrently, we carried out impregnation experiments in a piston-cylinder apparatus to test the validity of the calculations. The thermodynamic and experimental approaches, the main results and their implications for the chemical and mineralogical evolution of pyroxenite-derived melts and peridotites within the upper mantle are presented below. THERMODYNAMIC AND EXPERIMENTAL STRATEGY* Peridotite–pyroxenite interaction model Impregnation model We consider a heterogeneous mantle composed of pyroxenite bodies dispersed in a peridotite matrix. In our calculations and experiments, peridotite–pyroxenite interactions are envisaged to occur in two stages: (1) impregnation of peridotite by a finite amount of pyroxenite-derived melt; (2) chemical re-equilibration of the impregnated peridotite at constant $P$ and $T$. Input parameters are the composition of the pyroxenite-derived melt and the melt–rock ratio (i.e. the mass ratio of pyroxenite-derived melt to peridotite), which varies in the range 0–1. Output parameters are the proportions and compositions of each phase in the impregnated peridotite after re-equilibration. Starting peridotite and pyroxenite compositions For the peridotite end-member, we chose a spinel lherzolite from Mont Briançon, French Massif Central (Bri2; Table 1). In terms of major-element composition this sample is intermediate between fertile mantle MM3 of Baker & Stolper (1994) and depleted mantle DMM1 of Wasyljenki et al. (2003). Two pyroxenites from the Beni Bousera ultramafic massif (northern Morocco) were used in both calculations and experiments: M5-40 and M7-16 (Table 1). The criteria used to select these two samples were discussed by Lambart et al. (2009a). M5-40 is a garnet (Gr) websterite, close to the mean composition of the natural pyroxenite population. M7-16 has a more extreme composition, with low SiO$_2$ and high FeO contents (43.6 and 14.5%, respectively; unless otherwise stated, all percentages in this paper—oxide concentrations, phase proportions, norms, etc.—are reported by weight). Both pyroxenites are olivine normative, but they show different degrees of silica undersaturation: M7-16 is nepheline (Ne) normative (2.3%) and M5-40 is hypersthene (Hy) normative (8.3%). P–T conditions We consider that the pyroxenite-derived melt reacts with the surrounding peridotite at the $P–T$ conditions of its formation and that impregnation takes place at constant $P–T$. To model the partial melting of pyroxenites in a subsolidus lithospheric mantle (case 1), we considered melts formed at 1250°C and 1.5 GPa. These conditions correspond to a subcontinental lithospheric mantle undergoing a thermal event owing to asthenospheric upwelling, such as the Ronda massif, SW Spain (Lenoir et al., 2001). To model peridotite–pyroxenite interactions in MORB mantle sources (cases 2 and 3), we assumed that the heterogeneous mantle follows an adiabatic decompression path with a potential temperature of 1350°C; that is, close to the mean potential temperature beneath MORs according to Herzberg et al. (2007). Using the parameterization of McKenzie & O’Nions (1991), we selected two $P–T$ couples lying on or close to this adiabatic decompression path: $P=2.5$ GPa and $T=1400^\circ$C to simulate the case where pyroxenites M5-40 and M7-16 are partially molten whereas peridotite Bri2 is subsolidus (case 2); $P=1$ GPa and $T=1290^\circ$C to simulate the case where both pyroxenites and peridotite are partially molten (case 3). The compositions of the partial melts of pyroxenites M7-16 and M5-40 at 1250°C and 1.5 GPa, and 1290°C and 1 GPa were determined by Lambart et al. (2009a) and are given in Table 2. The degree of melting is close to 10% in both pyroxenites at 1250°C and 1.5 GPa, and close or equal to 100% at 1290°C and 1 GPa. As discussed below, we performed two new experiments to determine the composition... Table 1: Peridotites Bri2, MM3 and DMM1 and pyroxenites M5-40 and M7-16, compared with the mean composition of group-A pyroxenites from Ronda Massif (SW Spain; Garrido & Bodinier, 1999) \| \| Bri2 \| M5-40 \| M7-16 \| Group A \| MM3 \| DMM1 \| \|----------\|--------\|--------\|--------\|---------\|--------\|--------\| \| SiO₂ \| 45·1 \| 48·53 \| 43·58 \| 47·1 \| 45·5 \| 44·9 \| \| TiO₂ \| 0·08 \| 0·52 \| 0·75 \| 0·51 \| 0·11 \| 0·04 \| \| Al₂O₃ \| 2·7 \| 12·37 \| 13·73 \| 13·79 \| 3·98 \| 2·38 \| \| Cr₂O₃ \| 0·27 \| 0·12 \| 0·07 \| — \| 0·68 \| 0·39 \| \| FeO \| 7·6 \| 9·02 \| 14·51 \| 9·76 \| 7·18 \| 8·34 \| \| MnO \| 0·12 \| 0·20 \| 0·30 \| 0·20 \| 0·13 \| 0·13 \| \| MgO \| 41·8 \| 16·64 \| 12·52 \| 14·05 \| 38·3 \| 41·59 \| \| CaO \| 2·2 \| 10·89 \| 13·77 \| 11·91 \| 3·57 \| 2·14 \| \| Na₂O \| 0·19 \| 1·65 \| 0·75 \| 1·32 \| 0·31 \| 0·055 \| \| K₂O \| 0·00 \| 0·06 \| 0·03 \| 0·02 \| — \| 0·006 \| \| Mg-no. \| 90·7 \| 76·7 \| 60·6 \| 72·0 \| 90·5 \| 89·9 \| \| Ne/Ol/Hy/\| 0/69·6/18·6/\| 0/27·7/8·3/\| 2·3/31·7/0/\| 0/28·4/8·6/\| 0/62·5/18·8/\| 0/69·2/20·5/\| \| Di/Fds* \| 3·5/8·1\| 22·4/40·5\| 28·2/36·3\| 22·5/43·5\| 6·6/12·1\| 3·5/6·8\| Bri2 is a spinel lherzolite from Mont Brianon, French Massif Central; M5-40 and M7-16 are pyroxenites from Beni Bousera ultramafic massif (Morocco). Bri2 was analysed by inductively coupled plasma atomic emission spectrometry (ICP-AES) at the Service d’Analyses des Roches et Minéraux (CRPG, Nancy, France); the K₂O content in Bri2 is <100 ppm. M5-40 and M7-16 were analysed by ICP-AES at Laboratoire Magmas et Volcans. MM3 is from Baker & Stolper (1994). DMM1 is from Wasyljenki et al. (2003). Mg-number is the molar ratio 100 Mg²⁺/(Mg²⁺ + Fe²⁺); we considered that all iron was as Fe²⁺. CIPW norms: nepheline (Ne)/olivine (Ol)/hypersthene (Hy)/diopside (Di)/anorthite + albite + orthoclase (Fds). of partial melts in pyroxenites M5-40 and M7-16 at 2·5 GPa and 1400 °C; under these conditions, the degree of melting is 5% in M5-40 and 22% in M7-16 (Table 2). Thermodynamic approach The impregnation process is simulated at constant $P$ and $T$ by adding up to 100 g of pyroxenite-derived melt, in increments of 5 g, to 100 g of peridotite. At each step, we computed the chemical composition of the impregnated peridotite and we used the thermodynamic model pMELTS (Ghiorso et al., 2002) to compute the equilibrium assemblage (phase compositions and proportions). From the evolution of phase proportions with increasing amount of added melt, we determined the modal reactions of impregnation using a method similar to that used by Baker & Stolper (1994) to compute melting reactions (see ‘Thermodynamic results’). The software package pMELTS was initially developed to model liquid–solid and solid–solid phase relations in peridotitic systems. The fundamental problem solved by pMELTS is to calculate an equilibrium assemblage of phases by minimizing an appropriate potential energy, subject to constraints on bulk composition, temperature, pressure or volume, enthalpy or entropy, and optionally on oxygen fugacity $f_{\text{O}_2}$. It is calibrated in the system $\text{SiO}_2-\text{TiO}_2-\text{Al}_2\text{O}_3-\text{Fe}_2\text{O}_3-\text{Cr}_2\text{O}_3-\text{FeO}-\text{MgO}-\text{CaO}-\text{Na}_2\text{O}-\text{K}_2\text{O}-\text{P}_2\text{O}_5-\text{H}_2\text{O}$ between 1 and 4 GPa. The interface Adiabat-lph (Smith & Asimow, 2005) allows an automation of pMELTS calculations. In particular, we used the ‘ADIABAT-ASSIMILATE’ variable of this interface to simulate the impregnation process, as it allows us to add a fixed mass of a second bulk composition after each calculation stage. Below, we list the choices and constraints that we used in our thermodynamic study so that our calculations can be reproduced exactly. (1) To simulate the impregnation of peridotite by 0–100 g of melt, we used 21 steps, starting from 100 g of pure peridotite and adding 5 g of melt at each step. (2) Calculations were made at $f_{\text{O}_2}=\text{FMQ}-1$ (i.e. the fayalite–magnetite–quartz buffer minus one log unit), close to the oxygen fugacity in our experiments (Laporte et al., 2004) and to the oxygen fugacity in the mantle source of MORB (Christie et al., 1986; Cottrell & Kelley, 2011). For comparison, we also performed some calculations at FMQ and FMQ−2 and we obtained very similar results. (3) Of the two garnet models (from Berman & Koziol, 1991) available in pMELTS, we used the older, uncorrected one, which is the default in Adiabat.lph. This Table 2: Compositions (wt %) of pyroxenite-derived and peridotite-derived melts used in the impregnation models \| $P$, T: \| Case 1 \| Case 2 \| Case 3 \| \|--------\|--------\|--------\|--------\| \| \| 1.5 GPa, 1250°C \| 2.5 GPa, 1400°C \| 1 GPa, 1290°C \| \| Source rock: \| M5-40 \| M7-16 \| M5-40 \| M7-16 \| M5-40 \| M7-16 \| Bri2§ \| \| Run no.: \| P40-15 \| P16-15 \| P40-25 \| P16-25 \| P40-10 \| P16-10 \| (83) \| \| Assemblage:† \| cpx, opx, cpx, ol, sp, liq \| cpx, gt, sp, liq \| cpx, gt, liq \| cpx, ol, liq \| cpx, opx, ol, liq \| \| $F$:† \| 13.5 (12) \| 9.4 (9) \| 4.5 (6) \| 21.8 (21) \| 84.6 (36) \| 100 \| 10.5 \| \| Element \| Case 1 \| Case 2 \| Case 3 \| \|---------\|--------\|--------\|--------\| \| SiO$_2$ \| 50.6 (5) \| 41.6 (5) \| 48.7 (6) \| 39.1 (5) \| 49.8 (4) \| 43.6 \| 49.2 \| \| TiO$_2$ \| 1.08 (4) \| 1.59 (16) \| 1.90 (17) \| 2.36 (18) \| 0.61 (12) \| 0.75 \| 0.50 \| \| Al$_2$O$_3$ \| 19.7 (4) \| 14.6 (3) \| 14.6 (3) \| 10.4 (3) \| 14.9 (3) \| 13.7 \| 16.83 \| \| Cr$_2$O$_3$ \| 0.00 (4) \| 0.01 (8) \| 0.01 (8) \| 0.01 (8) \| 0.11 (4) \| 0.07 \| — \| \| FeO \| 10.0 (4) \| 21.8 (5) \| 12.2 (4) \| 22.2 (5) \| 8.3 (3) \| 14.5 \| 6.5 \| \| MnO \| 0.15 (14) \| 0.36 (12) \| 0.19 (8) \| 0.37 (14) \| 0.19 (8) \| 0.30 \| 0.01 \| \| MgO \| 5.6 (2) \| 7.5 (2) \| 8.4 (2) \| 10.6 (4) \| 11.6 (3) \| 12.5 \| 13.1 \| \| CaO \| 6.7 (4) \| 9.8 (4) \| 8.4 (3) \| 12.8 (4) \| 12.5 (4) \| 13.8 \| 12.1 \| \| Na$_2$O \| 5.40 (2) \| 2.46 (19) \| 4.15 (25) \| 1.94 (16) \| 1.98 (14) \| 0.75 \| 1.71 \| \| K$_2$O \| 0.53 (10) \| 0.32 (10) \| 1.42 (16) \| 0.18 (8) \| 0.08 (4) \| 0.03 \| — \| \| Total \| 98.0 (8) \| 99.0 (13) \| 98.7 (6) \| 97.8 (8) \| 99.0 (5) \| — \| — \| \| Mg-no. \| 50.7 (11) \| 38.0 (6) \| 54.9 (10) \| 46.0 (14) \| 71.3 (12) \| 60.6 \| 78.2 \| \| Ne/Oi/Hy/ \| 5.7/21.4/0/ \| 8.3/35.8/0/ \| 8.3/22.3/0/ \| 8.9/39.9/0/ \| 0/16.2/9.2/ \| 2.3/31.7/0/ \| 0/15.2/13.1/ \| \| Di/Fds‡ \| 4.7/66.2 \| 17.6/35.4 \| 20.5/45.3 \| 20.7/19.1 \| 24.6/48.8 \| 28.2/36.3 \| 17.3/52.7 \| All compositions are normalized to a sum of 100% (the average total of electron probe analyses before normalization is reported in the column ‘Total’). For a given dataset, we calculated both the statistical dispersion (as measured by $2\sigma$, where $\sigma$ is the standard deviation) and the analytical error (following Ancey et al., 1978), and we selected the largest of these two values as an estimation of the error. The errors (in parentheses) are given in terms of the least unit cited; for example, 50.6 (5) and 1.90 (17) represent 50.6 ± 0.5 and 1.90 ± 0.17, respectively. The first term in the run number corresponds to the bulk composition (P40 for pyroxenite M5-40; P16 for pyroxenite M7-16), the second term to the pressure (in kbar); the run duration (in hours) is given in parentheses after the run number. Experiments P40-10, P40-15, P16-10 and P16-15 are from Lambart et al. (2009a) and correspond to their runs 40-A1, 40-E2, 16-A1, and 16-A2, respectively. †Equilibrium phase assemblage and degree of melting $F$ (wt %) in the run products. ‡Superliquidus experiment: the composition listed is the bulk-rock composition (from Table 1). §Theoretical equilibrium phase assemblage, melting degree, and melt composition of peridotite Bri2 at 1290°C, 1 GPa, and $f_{\text{O}_2}$ of FMQ – 1 as predicted by pMELTS (as pMELTS significantly underestimates the degree of melting of peridotites, the calculation was performed at 1350°C; see text for further explanation). ¶CIPW norms. model underestimates the grossular component in Gt (Smith & Asimow, 2005), but it predicts equilibrium of pyroxenite-derived melt with the same assemblage as in the experiments. Conversely, the new model yields a Gt-bearing paragenesis down to pressures as low as 1.5 GPa. (4) We considered that the pyroxenite-derived melt reacts with surrounding peridotite at the $P-T$ conditions of its formation. However, the degree of melting of the peridotite is underestimated by pMELTS calculations compared with experimental data (Ghiorsio et al., 2002). To circumvent this difficulty, our calculations were performed 60°C above the experimental temperature: for example, for melts experimentally produced at 1.5 GPa and 1250°C (case I), the impregnation reaction was calculated with pMELTS at 1.5 GPa and 1310°C. (5) We did not include Cr$_2$O$_3$ in the bulk compositions used for the calculations. Incorporation of Cr$_2$O$_3$ in solid phases is oversimplified by pMELTS (e.g. it is neglected in pyroxene and garnet; see Asimow et al., 1995); as a result, the stability range of spinel is strongly overestimated when Cr$_2$O$_3$ is included in the bulk composition. (6) We chose to consider MnO as a major element (rather than a trace) because this choice yields thermodynamic results more consistent with experimental ones. (7) Potassium was considered as a trace element because, as a major element, it is handled incorrectly in the subsolidus assemblage when feldspar is absent (Asimow & Ghiorso, 1998). Experimental approach* Experimental rationale Two series of experiments were performed in a piston-cylinder apparatus. The first investigated the partial melting of pyroxenites M5-40 and M7-16 to determine their melt compositions at 2·5 GPa and 1400°C (experiments P40-25 and P16-25, respectively, in Table 2). The second series was aimed at validating pMELTS calculations using an independent approach. This validation was required because pMELTS is calibrated only for peridotite compositions (Ghiorso et al., 2002). $P-T$ conditions and starting compositions were chosen to reproduce and complete our calculations of peridotite–pyroxenite interactions in a subsolidus lithospheric mantle (case I). Four experiments were performed: (1) Experiment Bri2-l5 was run to determine the initial mode (before impregnation) of peridotite Bri2 at 1·5 GPa and 1250°C. (2) Impregnation experiments I40-l5, I16-l5 and I16-l0 were carried out to characterize the equilibrium phase assemblage after impregnation of peridotite Bri2 with an equal mass of liquid from pyroxenite M5-40 (I40-l5) or M7-16 (I16-l5 and I16-l0). I40-l5 and I16-l5 were run at 1·5 GPa and 1250°C. I16-l0 was run at the same temperature but at a lower pressure (1 GPa) to cross the solidus of the peridotite mantle and to estimate the role of the physical state of the mantle (subsolidus vs partially melted) on reactions. Starting materials Fine powders of natural pyroxenites M5-40 and M7-16, and peridotite Bri2 (Table 1) were used as starting materials in experiments P40-25, P16-25, and Bri2-l5, respectively. The starting materials for the impregnation experiments were homogeneous mixtures in equal proportions of peridotite Bri2 and either gel 40 or gel 16 (Table 3). Synthetic gels 40 and 16 are close to the equilibrium partial melts of pyroxenites M5-40 and M7-16, respectively, at 1·5 GPa and 1250°C, as determined by Lambart et al. (2009a; experiments P40-l5 and P16-l5 in Table 2). As it is very difficult to prepare a gel composition exactly matching a given chemical composition, there are some differences between gels 40 and 16 and the glass analysed in experiments P40-l5 and P16-l5, respectively. Gels and experimental glasses share, however, the same principal features: a composition of alkali-rich and Ne-normative basalt for gel 40 and P40-l5 glass; a basanitic composition for gel 16 and P16-l5 glass. The two pyroxenites and the peridotite were crushed in an agate mortar, and then ground under ethanol in an agate micronizing mill to reduce the grain size to 2–4 μm. The two gels were prepared using the method of Luth & Ingamells (1965). Tetra-ethyl orthosilicate (TEOS) was used as a source for SiO$_2$ and titrated nitrate aqueous solutions for the other oxides. These solutions were obtained by dissolution in 10 M HNO$_3$ of either metals (Al, Fe, Mn, Mg) or carbonates (Ca, Na, K) or ammonium salts (Ti, Cr). Rock powders and synthetic gels were all fired for 6 h at 900°C in a CO$_2$–H$_2$ atmosphere with an oxygen fugacity between the magnetite–wüstite and iron–wüstite buffers ($f_{\text{O}_2} = 10^{-15.91}$ bar). The starting materials were stored under vacuum to reduce water adsorption. --- Table 3: Gels (40, 16) and peridotite–gel mixtures (I40, I16) used as starting materials in the impregnation experiments (all compositions are normalized to 100 wt %) \| \| Gel 40 \| I40 \| Gel 16 \| I16 \| \|----------------\|--------\|-------\|--------\|-------\| \| SiO$_2$ \| 52·17 \| 48·66 \| 39·82 \| 42·48 \| \| TiO$_2$ \| 1·12 \| 0·59 \| 1·70 \| 0·89 \| \| Al$_2$O$_3$ \| 18·41 \| 11·71 \| 11·8 \| 7·23 \| \| Cr$_2$O$_3$ \| 0·00 \| 0·14 \| 0·03 \| 0·15 \| \| FeO \| 11·07 \| 9·21 \| 24·64 \| 16·12 \| \| MnO \| 0·16 \| 0·14 \| 0·28 \| 0·201 \| \| MgO \| 5·32 \| 23·18 \| 8·49 \| 25·11 \| \| CaO \| 5·93 \| 3·51 \| 12·06 \| 7·14 \| \| Na$_2$O \| 5·24 \| 2·62 \| 1·01 \| 0·60 \| \| K$_2$O \| 0·57 \| 0·28 \| 0·18 \| 0·09 \| \| Mg-no. \| 46·1 \| \| 38·1 \| \| \| Ne/OI/Hy/ \| 1·1/22·4/0/ \| 4·6/39·0/0/ \| \| \| \| Di/Fds* \| 3·7/70·6 \| 23·7/27·1 \| \| \| Gel 40 reproduces the principal features (alkali-rich and Ne-normative basalt) of the glass composition analysed in the partial melting experiment of pyroxenite M5-40 at 1·5 GPa and 1250°C (P40-l5; Table 2); gel 16 reproduces the principal features (basanitic composition) of the glass composition analysed in the partial melting experiment of pyroxenite M7-16 at 1·5 GPa and 1250°C (P16-l5; Table 2). Gel compositions were analysed by ICP-AES at Laboratoire Magmas et Volcans. Starting mixtures I40 and I16 were prepared by mixing in equal proportions powdered peridotite Bri2 (Table 1) and either gel 40 or gel 16, respectively. CIPW norms. Experimental and analytical techniques Experiments were carried out either in a non-end-loaded, $ \frac{3}{4} $ inch piston-cylinder apparatus ($ P = 1-1.5 \text{ GPa} $), or in an end-loaded, $ \frac{5}{8} $ inch piston-cylinder apparatus ($ P = 2.5 \text{ GPa} $). From outside to inside, $ \frac{3}{4} $ inch assemblies consist of a NaCl cell wrapped in a lead foil, an outer Pyrex cylinder, a graphite furnace, and an inner Pyrex cylinder. To minimize the amount of adsorbed water, all the pieces of the piston-cylinder assemblies were fired at high temperature, and then stored at 150°C. We used double sample capsules made of graphite (inside) and platinum (outside). Capsules are placed inside the inner Pyrex cylinder, between a rod and a sheath of crushable MgO, and lie at the hot spot of the assembly. Temperature is controlled by a calibrated W$_{93}$Re$_{7}$–W$_{73}$Re$_{26}$ thermocouple. The thermocouple tip is stuck into the MgO sheath and is separated from the sample capsule by a 0.5 mm thick hard alumina disc [see Laporte et al., 2004] and Lambart et al. [2009b] for technical details]. The $ \frac{5}{8} $ inch assemblies are similar except that: (1) they do not include an inner Pyrex cylinder; (2) the thermocouple tip is separated from the platinum capsule by a 0.6 mm thick crushable MgO disc; (3) the capsule is placed into a thin-walled MgO sheath to avoid direct contact between the outer platinum container and the furnace. The ‘microdike’ technique (Laporte et al., 2004) was used to analyse the composition of liquids in equilibrium with mineral phases: this technique relies on the extraction of small volumes of liquid from the partially molten sample into fractures that formed inside the inner graphite container at the beginning of the experiments. A few microdikes were observed in all experiments (Fig. 1), both at the top and at the bottom of the sample chamber. As the thermal gradient in our assemblies is small ($ \sim 5^\circ \text{C across the sample chamber} $), liquid compositions are homogeneous throughout the sample chamber and there are no significant differences between glass analysed in upper and lower microdikes (Lambart et al., 2009b, and unpublished data). Microdike sizes vary from tens to hundreds of micrometres in length and from a few micrometres to hundreds of micrometres in width. These lengths are smaller than the thickness of the graphite walls (0.8–0.9 mm in $ \frac{3}{4} $ inch assemblies and 1.4–1.2 mm in $ \frac{5}{8} $ inch assemblies) making the contact between melt and platinum container very unlikely. A major advantage of the microdike technique is that it works down to very low degrees of melting ($ \leq 1\% $; Laporte et al., 2004) as the volume of melt extracted into the microdikes is small compared with total sample volume. At the end of an experiment, the capsule was enclosed in epoxy, sectioned lengthwise, polished and carbon-coated. Textures and phase assemblages were characterized using a JEOL JSM-5910 LV scanning electron microscope. Phase compositions were analysed with a Cameca SX100 --- Fig. 1.* (a) Backscattered electron image showing microdikes in a high-pressure experiment carried out in a $ \frac{5}{8} $ inch piston-cylinder apparatus (run P40-25; 2.5 GPa and 1400°C); the graphite container is black and the partially molten pyroxenite grey; the basaltic microdikes contain large pools of glass and large crystals. (b) Close-up view of white square no. 1 in (a): the phase assemblage of the partially molten pyroxenite is Cpx + Gt + Liq; the interstitial melt is not perceptible owing to the relatively low melt fraction (4-5%) and to the growth of quench Cpx at the end of the experiment. (c) Close-up view of the white square no. 2 in (a) showing a large pool of glass not affected by the growth of quench crystals. Scale bars: 200 µm in (a); 10 µm in (b) and (c). electron microprobe. A 15 kV accelerating voltage, a 15 nA beam current, counting times of 20 s for Ni, Cr and Ti, and of 10 s for other elements, and a focused beam were used for crystalline phases. For glass analyses, the beam current was lowered to 8 nA and a beam size of 5 μm was used to minimize sodium loss. In most cases, the analytical totals were good (98–100 wt %), typical of basaltic glasses in nominally anhydrous piston-cylinder experiments. The relatively low analytical total of glass in the II6-15 run (97.3 wt %) could be explained by a more significant amount of dissolved volatiles owing to (1) the low amount of melt (6.6%), and (2) the use of gel in the starting material, a highly hygroscopic material that, despite the great care taken, is almost impossible to maintain in a fully dry state. The proportions of liquid and solid phases best fitting the bulk composition of the starting material were calculated using a mass-balance program modified from Albarede & Provost (1977). Table 4 summarizes the experimental conditions and the phase proportions and compositions in the run products. THERMODYNAMIC RESULTS Computing the impregnation reactions Impregnation calculations consist in adding from 5 to 100 g of pyroxenite-derived liquid to 100 g of peridotite, in steps of 5 g. After each impregnation step, we compared the equilibrium phase assemblage predicted by pMELTS with the initial phase assemblage of peridotite Bri2. Equilibrium phase assemblages at the end of the impregnation process (100 g of liquid added to 100 g of peridotite, i.e. a melt–rock ratio of unity) are given in Table 5 for the two pyroxenite-derived liquids and the three $P-T$ sets considered in this study; the initial phase assemblages are shown for comparison. To illustrate the evolution of the phase assemblage during impregnation, we plotted the mass variation, $\Delta m$, of each phase as a function of the mass $m_{\text{Liq}_0}$ of added liquid (Fig. 2); $\Delta m$ is the difference between the mass of a phase in the impregnated peridotite and its initial mass in peridotite Bri2. The impregnation reactions were retrieved from the plots in Fig. 2 using a method similar to that used by Baker & Stolper (1994) to compute melting reactions in a fertile peridotite with increasing temperature. Retrieval is easy in the case of impregnation by the liquid from pyroxenite M5-40 (Fig. 2a, c, e) because the phase assemblage remains the same from 0 to 100 g of added liquid and each $\Delta m$ evolves almost linearly with $m_{\text{Liq}_0}$. In this case, the impregnation process may be described by a single reaction whose coefficients are the slopes of the best-fit lines in Fig. 2: a negative slope indicates that the phase is consumed and a positive slope that it is produced. Let us consider, for instance, the impregnation of peridotite Bri2 by the melt from M5-40 at 2.5 GPa and 1400°C, in which case there is massive production of Cpx, to a lesser extent of Gt and Opx, and Ol dissolution (Fig. 2c). Writing the reactions in grams for 1 g of added liquid, the impregnation process may quantitatively be described by the following reaction: $$1 \text{ g Liq}_0 + 0.30 \text{ g Ol} \rightarrow 0.26 \text{ g Liq}_1 + 0.17 \text{ g Opx} + 0.68 \text{ g Cpx} + 0.19 \text{ g Gt}$$ (2) where Liq$_0$ is the initial melt (from pyroxenite) and Liq$_1$ is the residual melt after equilibration with peridotite; reaction (2) is valid from 0 to 100 g of added liquid. The case of the addition of the liquid from pyroxenite M7-16 is not so simple because the plots of $\Delta m$ vs $m_{\text{Liq}_0}$ show slope breaks and steps (Fig. 2b, d and f) that indicate a change of the impregnation reaction in relation to a change of the phase assemblage (exhaustion of a solid phase and/or precipitation of a new phase). At 2.5 GPa and 1400°C (Fig. 2d), for instance, a first reaction operates from 0 to 25 g of added melt and is characterized by the dissolution of a large quantity of Opx and the precipitation of Cpx, Ol, and Gt: $$1 \text{ g Liq}_0 + 0.70 \text{ g Opx} \rightarrow 0.00 \text{ g Liq}_1 + 0.41 \text{ g Ol} + 1.15 \text{ g Cpx} + 0.14 \text{ g Gt}.$$ (3a) Reaction (3a) results in the exhaustion of Opx at a value of $m_{\text{Liq}_0}$ between 25 and 30 g. Then the added liquid interacts with the Opx-free residue according to the reaction $$1 \text{ g Liq}_0 \rightarrow 0.25 \text{ g Liq}_1 + 0.19 \text{ g Ol} + 0.46 \text{ g Cpx} + 0.09 \text{ g Gt}$$ (3b) which is valid from 30 to 100 g of added liquid. It should be noted that the production of a phase during the impregnation process is not systematically linked with an increase of its mass fraction in the final assemblage. Indeed, the mass fraction of a given phase (A) increases only if the mass of phase A produced during impregnation divided by the mass of added liquid exceeds the initial mass fraction of A in the peridotite, $a_0$. If $c_A$ (in gram per gram of added liquid) is the coefficient of phase A in the impregnation reaction, then the mass fraction of A increases only if $c_A > a_0$. For instance, let us consider the case of reaction (3a), which produces 0.41 g of Ol, 1.15 g of Cpx, and 0.14 g of Gt per gram of added liquid. The initial mass fractions of Ol, Gt, Opx, and Cpx in Bri2 at 2.5 GPa and 1400°C are equal to 0.652, 0.034, 0.206, and 0.108, respectively (Table 5), so $c_A$ is larger than $a_0$ for Cpx and Gt, but smaller than $a_0$ for Ol. Using reaction (3a), we can compute the masses after impregnation of 100 g of peridotite by 25 g of liquid: 75.45 g of Ol, 6.9 g of Gt, 3.1 g of Opx, and 39.55 g of Cpx; that is, in terms of mass fractions, 0.604, 0.035, 0.025, and 0.316 for Ol, Gt, Opx, and Cpx, respectively. As expected, the mass fractions of Cpx and Gt increased during impregnation, but that of Ol decreased slightly. This is the explanation for the slight Table 4: Summary of run information: pressure, temperature, duration, phase proportions and compositions (wt %) \| n \| Mode \| SiO₂ \| TiO₂ \| Al₂O₃ \| Cr₂O₃ \| FeO \| MnO \| CaO \| MgO \| Na₂O \| K₂O \| Total \| Mg-no. \| (Fe–Mg) \| K_D* \| \|---\|----------\|--------\|-------\|-------\|-------\|--------\|-------\|-------\|-------\|-------\|-------\|-------\|--------\|---------\|-------\| \| Run P40/25 (2.5 GPa, 1400 °C, 15 h) \| \| liq (10) \| 45 (6) \| 48.69 (56) \| 1.90 (17) \| 14.61 (23) \| 0.01 (18) \| 12.22 (33) \| 0.19 (8) \| 8.36 (24) \| 4.15 (25) \| 1.42 (16) \| 98.7 (6) \| 54.9 (10) \| 0.33 \| \| cpx (8) \| 71.8 (23) \| 51.44 (42) \| 0.45 (6) \| 8.83 (70) \| 0.14 (7) \| 7.74 (38) \| 0.19 (6) \| 13.38 (58) \| 1.79 (18) \| 0.00 (0) \| 99.3 (6) \| 78.7 (10) \| 0.46 \| \| gt (6) \| 23.7 (22) \| 41.24 (95) \| 0.38 (13) \| 23.1 (11) \| 0.17 (18) \| 11.58 (30) \| 0.32 (5) \| 17.22 (74) \| 5.90 (76) \| 0.09 (21) \| 100.5 (5) \| 72.6 (13) \| 0.46 \| \| Run P45/25 (2.5 GPa, 1400 °C, 30 h) \| \| liq (12) \| 21.8 (21) \| 39.09 (48) \| 2.36 (18) \| 10.39 (25) \| 0.01 (18) \| 22.23 (48) \| 0.37 (14) \| 10.63 (43) \| 12.78 (39) \| 0.18 (18) \| 98.1 (8) \| 46.0 (14) \| 0.31 \| \| cpx (8) \| 46.2 (20) \| 48.66 (60) \| 0.54 (6) \| 9.80 (73) \| 0.09 (3) \| 8.31 (34) \| 0.16 (6) \| 17.24 (43) \| 18.80 (38) \| 0.90 (9) \| 100.2 (8) \| 73.2 (11) \| 0.54 \| \| gt (5) \| 32.0 (17) \| 40.26 (39) \| 0.38 (31) \| 22.80 (25) \| 0.07 (3) \| 14.8 (17) \| 0.45 (5) \| 13.10 (94) \| 8.11 (17) \| 0.01 (2) \| 100.9 (7) \| 61.3 (18) \| 0.54 \| \| Run B12/15 (1.5 GPa, 1250 °C, 137 h) \| \| ol (6) \| 65.0 (35) \| 41.08 (40) \| 0.01 (4) \| 0.10 (6) \| 0.08 (4) \| 9.30 (23) \| 0.13 (6) \| 49.41 (44) \| – \| 0.21 (5) \| – \| 90.4 (3) \| 0.27 \| \| apx (7) \| 21.7 (48) \| 54.23 (66) \| 0.12 (6) \| 5.86 (90) \| 0.78 (11) \| 5.70 (20) \| 0.12 (6) \| 31.25 (60) \| 1.85 (41) \| – \| 100.0 (5) \| 90.7 (2) \| 0.33 \| \| cpx (8) \| 13.1 (25) \| 51.91 (63) \| 0.33 (8) \| 6.78 (51) \| 1.12 (20) \| 3.55 (50) \| 0.11 (10) \| 18.11 (33) \| 17.0 (20) \| 1.09 (12) \| 0.01 (1) \| 90.1 (9) \| 0.37 \| \| sp (3) \| 0.2 (6) \| 0.32 (7) \| 0.18 (9) \| 55.1 (28) \| 13.8 (32) \| 8.72 (38) \| 0.11 (7) \| 21.70 (79) \| 0.03 (5) \| 0.01 (3) \| – \| 89.8 (8) \| 0.54 \| \| Run A16/15 (1.5 GPa, 1250 °C, 176 h) \| \| liq (9) \| 31.8 (19) \| 51.41 (58) \| 1.13 (41) \| 19.43 (41) \| 0.05 (19) \| 6.96 (31) \| 0.09 (8) \| 5.84 (40) \| 6.74 (40) \| 7.44 (43) \| 0.91 (13) \| 98.1 (18) \| 63.3 (17) \| 0.27 \| \| ol (5) \| 16.1 (20) \| 40.31 (38) \| 0.05 (6) \| 0.10 (4) \| 0.11 (9) \| 12.86 (31) \| 0.14 (6) \| 46.10 (24) \| 0.20 (7) \| – \| 100.5 (5) \| 86.3 (3) \| 0.27 \| \| apx (6) \| 42.7 (33) \| 52.70 (78) \| 0.35 (12) \| 7.4 (12) \| 0.19 (18) \| 9.38 (25) \| 0.16 (7) \| 27.80 (72) \| 1.99 (34) \| – \| 100.1 (7) \| 84.1 (6) \| 0.33 \| \| cpx (6) \| 67.1 (15) \| 51.45 (42) \| 0.68 (18) \| 9.28 (22) \| 0.28 (7) \| 6.86 (32) \| 0.16 (7) \| 18.19 (25) \| 11.55 (40) \| 1.54 (11) \| 0.01 (2) \| 100.3 (7) \| 82.5 (3) \| 0.37 \| \| sp (4) \| 2.8 (7) \| 0.22 (5) \| 0.35 (12) \| 59.4 (37) \| 8.5 (41) \| 11.19 (85) \| 0.14 (7) \| 20.11 (61) \| 0.05 (5) \| 0.00 (0) \| – \| 98.7 (9) \| 0.54 \| \| Run B16/15 (1.5 GPa, 1250 °C, 97 h) \| \| liq (10) \| 66.6 (6) \| 44.42 (55) \| 3.99 (20) \| 15.29 (30) \| 0.02 (8) \| 15.23 (42) \| 0.16 (8) \| 7.82 (28) \| 8.52 (32) \| 3.33 (20) \| 1.22 (16) \| 97.3 (11) \| 47.8 (12) \| 0.27 \| \| ol (7) \| 43.2 (14) \| 38.47 (39) \| 0.07 (6) \| 0.09 (4) \| 0.03 (9) \| 23.04 (30) \| 0.25 (6) \| 37.72 (24) \| 0.35 (7) \| – \| 100.0 (9) \| 99.8 (13) \| 0.31 \| \| apx (6) \| 1.0 (30) \| 51.7 (15) \| 0.51 (20) \| 6.6 (15) \| 0.21 (18) \| 13.27 (88) \| 0.20 (7) \| 2.02 (40) \| 2.55 (13) \| – \| 100.7 (9) \| 77.4 (29) \| 0.27 \| \| cpx (10) \| 46.6 (24) \| 49.29 (68) \| 1.21 (17) \| 8.77 (81) \| 0.22 (5) \| 9.80 (40) \| 0.20 (6) \| 16.19 (86) \| 15.3 (13) \| 0.86 (8) \| 0.01 (2) \| 100.1 (1) \| 74.6 (7) \| 0.31 \| \| sp (4) \| 3.6 (4) \| 0.31 (12) \| 0.36 (17) \| 58.0 (43) \| 4.7 (42) \| 19.35 (83) \| 0.12 (6) \| 16.46 (72) \| 0.11 (7) \| 0.01 (3) \| – \| 96.5 (9) \| 60.3 (20) \| 0.60 \| \| Run B16/10 (1 GPa, 1250 °C, 84 h) \| \| liq (10) \| 41.4 (7) \| 46.26 (71) \| 1.99 (17) \| 15.05 (77) \| 0.10 (9) \| 14.47 (57) \| 0.19 (9) \| 8.78 (11) \| 11.33 (37) \| 0.25 (8) \| 1.56 (14) \| 99.9 (18) \| 55.5 (40) \| 0.31 \| \| ol (14) \| 49 (8) \| 38.95 (38) \| 0.03 (4) \| 0.08 (4) \| 0.03 (9) \| 20.27 (32) \| 0.22 (8) \| 3.94 (31) \| 0.36 (6) \| – \| 100.3 (6) \| 77.9 (4) \| 0.31 \| \| apx (14) \| 95 (13) \| 49.84 (68) \| 0.65 (15) \| 7.35 (19) \| 0.73 (15) \| 8.47 (55) \| 0.17 (8) \| 17.14 (65) \| 15.33 (35) \| 0.32 (7) \| 0.00 (2) \| 100.3 (9) \| 73.3 (10) \| 0.30 \| \| cpx (14) \| 41.4 (7) \| 46.26 (71) \| 1.99 (17) \| 15.05 (77) \| 0.10 (9) \| 14.47 (57) \| 0.19 (9) \| 8.78 (11) \| 11.33 (37) \| 0.25 (8) \| 1.56 (14) \| 99.9 (18) \| 55.5 (40) \| 0.31 \| \| sp (6) \| 0.1 (2) \| 0.19 (9) \| 0.44 (6) \| 56.8 (18) \| 9.2 (18) \| 16.36 (25) \| 0.13 (10) \| 16.78 (27) \| 0.13 (9) \| 0.01 (3) \| 98.9 (4) \| 64.7 (3) \| 0.59 \| For all experiments, the run number is built as follows: the first term is the bulk composition and the second term is the pressure (in kbar). In this way, runs P40-25 and P45-25 are partial melting experiments on pyroxenites M5-40 and M7-16, respectively. Run B12-15 is a subsolidus experiment to determine the mode of peridotite B12 at 1.5 GPa and 1250 °C. Re-penetration experiments A4-15, 116-15, and 116-10 were performed to characterize the equilibrium phase assemblage after re-equilibration of peridotite B12 by an actual mass of liquid in a given sample. We calculated the statistical dispersion of the dataset (as measured by 2σ, where σ is the standard deviation and the analytical error following Ancey et al., 1978), and we selected the largest of these two values as an estimation of the error. The errors (in parentheses) are given in terms of the least unit cited. Modes (in wt %) are calculated using a mass-balance program modified from Albarede & Provost (1977). The number in parentheses is 2σ, given in terms of the least unit cited. Fe–Mg exchange coefficients between mineral and melt: $ K_D = \frac{(\text{Fe}/\text{Mg})_{\text{mineral}}}{(\text{Fe}/\text{Mg})_{\text{melt}}} $. Table 5: pMELTS results: masses of solid phases and liquid in the peridotite before impregnation (columns ‘Bri2’) and after impregnation by an equivalent mass of melt from pyroxenite M5-40 (columns ‘Bri2 + M5-40’) or M7-16 (columns ‘Bri2 + M7-16’) \| $P$, $T$ ($T_{pMELTS}$):* \| Case 1 \| Case 2 \| Case 3 \| \|----------------------------------\|--------\|--------\|--------\| \| \| Bri2 \| Bri2 + M5-40 \| Bri2 + M7-16 \| Bri2 \| Bri2 + M5-40 \| Bri2 + M7-16 \| Bri2 \| Bri2 + M5-40 \| Bri2 + M7-16 \| \| OI \| 63·0 \| 25·6 \| 65·6 \| 65·2 \| 33·7 \| 89·2 \| 66·4 \| 64·3 \| 80·7 \| \| Gt \| 0 \| 0 \| 0 \| 3·4 \| 21·6 \| 10·6 \| 0 \| 0 \| 0 \| \| Opx \| 25·0 \| 71·6 \| 0 \| 20·6 \| 35·8 \| 0 \| 21·8 \| 18·8 \| 0 \| \| Cpx \| 10·3 \| 35·4 \| 100·2 \| 10·8 \| 80·5 \| 80·4 \| 1·3 \| 12·8 \| 18·8 \| \| Sp \| 1·7 \| 8·0 \| 6·6 \| 0 \| 0 \| 0 \| 0 \| 0 \| 0 \| \| Liq \| 0 \| 55·4 \| 26·0 \| 0 \| 28·4 \| 18·4 \| 10·5 \| 104·1 \| 100·5 \| \| Reaction† \| 4 \| 5a and 5b \| \| 2 \| 3a and 3b \| \| 6 \| 7a and 7b \| \| For Bri2, masses are given for 100 g of rock; for mixtures Bri2 + M5-40 and Bri2 + M7-16, masses are given for 200 g of rock. $P$, $T$ conditions of the three cases considered. Temperature in parentheses is the temperature at which the calculations were performed: as pMELTS significantly underestimates the degree of melting of peridotites in comparison with experimental data (Ghiorso et al., 2002), the calculations were made 60 °C above the nominal temperature. †This line gives the number of the impregnation reaction in the text. decrease of the OI percentage between 0 and 25 g of added liquid in the OI–Opx–Cpx triangle (Fig. 3b). Finally, we note that in some cases, the choice of a linear approximation may be questioned. In Fig. 2b, Cpx and Opx show a variation of their reaction coefficients (i.e. slope variation) not correlated to a change of assemblage. This highlights a limitation for the method of calculation. This particular case will be discussed subsequently. Case 1 (1·5 GPa and 1250 °C)* Case 1 corresponds to a situation where bodies of partially molten pyroxenite are surrounded by unmolten peridotite that prevents extraction of the pyroxenite-derived melts at low degrees of melting. At 1·5 GPa and 1250 °C, both M5-40 and M7-16 have degrees of melting close to 10% (Table 2). We stress, however, that the main conclusions below do not depend on the exact degree of melting of the pyroxenite (and so on temperature) as long as it lies between 0% and 15–20%; within this range, the partial melts from pyroxenites M5-40 and M7-16 are Ne-normative and keep the same compositional characteristics, namely silica- and alkali-rich for M5-40 and silica-poor for M7-16 (see table I in Lambart et al., 2009a). In terms of temperature, $T$ must be lower than $\sim$1270 °C (at 1·5 GPa) to satisfy the constraint that the surrounding peridotite is subsolidus. Impregnation of peridotite Bri2 by pyroxenite-derived melts at 1·5 GPa and 1250 °C leads to major modifications of the modal composition of the solid assemblage. In the case of M5-40 melt (Fig. 2a), the impregnation leads to OI dissolution and production of Opx, Cpx, and a little spinel (Sp). The reaction remains unchanged from 0 to 100 g of added liquid: \[ 1 \text{ g Liq}_0 + 0·44 \text{ g OI} \rightarrow 0·53 \text{ g Liq}_1 \\ + 0·57 \text{ g Opx} + 0·28 \text{ g Cpx} + 0·06 \text{ g Sp}. \] It should be noted that almost half the added melt is consumed in this reaction. The modal compositions of the impregnated rocks are projected onto the OI–Opx–Cpx triangle in Fig. 3a. The evolution is dominated by the drop of the OI component, which leads to a transition of the solid assemblage from the lherzolite field to the OI websterite field; this transition occurs at a melt–rock ratio of 0·40 (i.e. $m_{\text{Liq}_0} = 40 \text{ g}$). In the case of the impregnation by the melt from M7-16, the evolution is markedly different (Fig. 2b). From 0 to 45 g of added melt, Opx is dissolved, a large quantity of Cpx and small quantities of OI and Sp are produced, and most of the added liquid is consumed. After the disappearance of Opx (between 45 and 50 g of added melt), OI switches from the product side to the reactant side of the reaction. However, before the Opx exhaustion, the evolution lines of Opx and Cpx mass variations are curved (Fig. 2b). To explain this nonlinear evolution, Fig. 4a illustrates the mass variation of bulk pyroxene (Cpx + Opx) as a function of the mass of added liquid between 0 and... Fig. 2. Thermodynamic results: impregnation of 100 g of peridotite Bri2 by 0–100 g of liquid (Liq$_0$) from pyroxenite M5-40 (upper row) or pyroxenite M7-16 (lower row). (a, b) Case 1: 1.5 GPa and 1310°C; (c, d) case 2: 2.5 GPa and 1460°C; (e, f) case 3: 1 GPa and 1350°C. The calculations were made 60°C above nominal temperatures (i.e. 1250, 1400, and 1290°C for cases 1, 2, and 3, respectively) because pMELTS underestimates the melting degree of peridotite at a given $P$ and $T$ (Ghiorso et al., 2002). The abscissa is the mass $m_{\text{Liq}_0}$ of liquid Liq$_0$ added to the system. The ordinate is the mass variation, $\Delta m$, of the different phases; for a given mass of added liquid, $\Delta m$ is the difference between the mass of a phase in the impregnated peridotite and its initial mass in peridotite Bri2; line $\Delta m = 0$ is shown by dashes. Symbols are as follows: crosses, residual melt after equilibration (Liq$_0$); squares, Cpx; triangles, Opx; circles, Ol; diamonds, Sp; stars, Gt. Straight lines are best-fit lines used to compute the impregnation reactions; the slope is positive when the phase is produced, negative when it is consumed; a slope break marks a reaction change in relation to the exhaustion of a solid phase and/or the precipitation of a new phase; for M7-16 three sets of best-fit lines are plotted, corresponding to reactions (5a'), (5a'') and (5b) (see text). 45 g. The almost perfect linear correlation ($R^2 > 0.99$) shows that the mass changes in Cpx and Opx are strongly interdependent. We also plotted Ca/(Ca + Mg + Fe) in Cpx as a function of $\Delta m$ for Cpx (Fig. 4b): the correlation between the mass variation of Cpx and its composition is linear. Hence, Fig. 4 emphasizes a limitation of the method: when a solid phase undergoes important compositional changes, its mass variation is no longer proportional to the amount of added melt. It should be noted, however, that, except in the case discussed above, all regression coefficients of Cpx calculated for a given phase assemblage are higher than 0.9. Accordingly, the linear approximation is appropriate for the calculation of impregnation reactions in most cases. In the case of Fig. 2b, we subdivided the curve into parts with $R^2 > 0.9$ and we estimated that the reaction evolves from $$1 \text{ g Liq}_0 + 0.03 \text{ g Opx} \rightarrow 0.16 \text{ g Liq}_1 + 0.06 \text{ g Ol} + 0.74 \text{ g Cpx} + 0.08 \text{ g Sp} \quad (5a')$$ between 0 and 30 g of added melt, to $$1 \text{ g Liq}_0 + 1.31 \text{ g Opx} \rightarrow 0.21 \text{ g Liq}_1 + 0.02 \text{ g Ol} + 2.00 \text{ g Cpx} + 0.08 \text{ g Sp} \quad (5a'')$$ from 35 g of added melt up to Opx exhaustion. After the disappearance of Opx (between 45 and 50 g of added melt), mass variations evolve almost linearly according to the reaction $$1 \text{ g Liq}_0 + 0.06 \text{ g Ol} \rightarrow 0.37 \text{ g Liq}_1 + 0.62 \text{ g Cpx} + 0.06 \text{ g Sp}. \quad (5b)$$ In the Ol–Opx–Cpx triangle (Fig. 3a), reaction between melt from M7-16 and peridotite Bri2 results in a strong increase of the Cpx/Opx ratio, and the solid assemblage moves from the lherzolite field into the wehrlite field (at a melt–rock ratio of $\sim 0.45–0.50$). With reaction (5b), the solid assemblage drifts along the Ol–Cpx side of the triangle toward the Cpx corner, reaching the field of Ol-bearing clinopyroxenites at a melt–rock ratio of $\sim 0.90$. Case 2 (2.5 GPa and 1400°C) According to pMELTS, the phase assemblage of peridotite Bri2 at 2.5 GPa and 1400°C is a subsolidus Gt lherzolite (mass of peridotite Bri2 is 100 g). Magma–rock ratios are indicated in italics. In (a), the black circles correspond to group C pyroxenites from Ronda (Remaidi, 1993; Garrido & Bodinier, 1999; Bodinier et al., 2008; the grey field corresponds to peridotites from Ronda (Remaidi, 1993; Van der Wal & Bodinier, 1996) and the grey square and triangle correspond to the equilibrium phase assemblages computed for bulk compositions I16 and I40, respectively, at 1.5 GPa and 1310°C (see text for further explanation); D, dunite; H, harzburgite; L, lherzolite; Wl, wehrlite; OPX, orthopyroxenite; CPX, clinopyroxenite; Wb, websterite. In these $P-T$ conditions, melts from pyroxenites M5-40 and M7-16 are both in equilibrium with Cpx and Gt (runs P40-25 and P16-25, Table 4). As in the previous case, impregnation of peridotite by pyroxenite-derived melts at 2.5 GPa and 1400°C results in a strong modification of the solid phase proportions. In the case of the partial melt from pyroxenite M5-40, impregnation leads to a massive production of Cpx and to a lesser extent of Gt and Opx, and to Ol dissolution (Fig. 2c): for 100 g of added liquid, the impregnation reaction produces 68 g of Cpx, 19 g of Gt, and 17 g of Opx, it consumes 30 g of Ol, and the mass of residual liquid is 26 g [see reaction (2) above]. The evolution of impregnated rocks is marked by a major drop of the Ol component and a significant increase of the Cpx/Opx ratio; the solid assemblage changes from a lherzolite to an Ol websterite at a melt–rock ratio of $\sim 0.50 - 0.55$ (Fig. 3b). The mass fraction of Gt increases from 3.4% in peridotite Bri2 to 10.8% at a melt–rock ratio of unity (Table 5). Impregnation of peridotite Bri2 by the melt from pyroxenite M7-16 is also characterized by the crystallization of a large amount of Cpx and a smaller amount of Gt [1.15 and 0.14 g, respectively, per gram of added liquid; reaction (3a)]. As in case 1, however, the behaviours of Ol and Opx are the opposite of those observed with the melt from M5-40: the addition of 1 g of liquid from M7-16 consumes 0.70 g of Opx and precipitates 0.41 g of Ol. Reaction (3a) rapidly leads to the exhaustion of Opx (at a $m_{\text{Liq}}$ value between 25 and 30 g; Fig. 2d); it is then replaced by reaction (3b), in which the products are Cpx, Ol, and Gt (0.46, 0.19, and 0.09 g, respectively, per gram of added liquid). Residual liquid is absent in reaction (3a) and its mass is small in reaction (3b) (0.25 g). The evolution of impregnated rocks (Fig. 3b) is similar to that described in case 1, with first an increase of the Cpx/Opx ratio at a nearly constant Ol mode [reaction (3a)], and then a decrease of the Ol mode along the Ol–Cpx side of the Ol–Opx–Cpx triangle [reaction (3b)]; the transition from the lherzolite field to the wehrlite field occurs at a melt–rock ratio slightly larger than 0.2. Case 3 (1 GPa and 1290°C) A distinctive feature of case 3 is that peridotite Bri2 is then partially molten: according to pMELTS, at 1 GPa and 1290°C, Bri2 is made of 10.5% liquid in equilibrium with a harzburgitic residue (66.4% Ol, 21.8% Opx, and 13% Cpx; Table 5). Another distinctive feature is that the pyroxenites are totally or almost totally molten at these $P-T$ conditions (Table 2). We ran two sets of calculations for case 3. In the first set, we directly added the pyroxenite-derived liquid to the bulk peridotite, as in cases 1 and 2. It may be unrealistic, however, to assume that a large body of partially molten peridotite with a degree of melting of $\sim 10\%$ would not undergo any solid–liquid separation. Indeed, melt migration through peridotites proceeds at degrees of melting from 0.1–1 vol. % (Kohlstedt, 1991; Wark et al., 2003) to 2–3 vol. % (Faul, 1997), depending on the permeability–porosity relationship. To test the effect of melt extraction from the peridotite, we made a second set of calculations, in which most of the melt is removed from the peridotite at each step, just 2% melt being left in the system before the next impregnation step by 5 g of pyroxenite-derived melt. In this way, we simulate the situation where the partial melt from the peridotite is mostly extracted and replaced by the pyroxenite-derived melt (as set 2 yields results very close to those of set 1, only the results of set 1 are shown in Fig. 2e and f). Set 1 Contrary to all previous calculations, the impregnation of peridotite Bri2 by the melt from M5-40 at 1 GPa and 1290°C results in only minor modifications of phase masses. Small quantities of Opx and Ol are dissolved, a small quantity of Cpx is precipitated, and the mass of liquid remains almost unchanged (Fig. 2c): \[ 1 \text{ g Liq}_0 + 0.03 \text{ g Opx} + 0.02 \text{ g Ol} \rightarrow 0.93 \text{ g Liq}_1 \\ + 0.11 \text{ g Cpx}. \] (6) As the coefficients of Ol and Opx in reaction (6) are small, their proportions in the solid assemblage change little with impregnation: they decrease from 75.1% and 24.5% in Bri2 to 70.1% and 18.2% respectively at a melt–rock ratio of unity. Concurrently, the Cpx proportion increases from 0.3 to 11.7%, and the solid assemblage shifts from the harzburgite to the lherzolite field (Fig. 3c). The minor effect of impregnation predicted by pMELTS in the present case is consistent with the weak compositional contrast between melts from M5-40 and Bri2 at 1 GPa and 1290°C (Table 2): the melt from pyroxenite M5-40 has a composition close to the peridotite-derived melt except for slight differences in CaO and Al$_2$O$_3$ contents owing to a higher melting degree of the pyroxenite, and a lower Mg-number. The melt from M7-l6 has a much lower SiO$_2$ content and a much higher FeO content than the peridotite melt (Table 2), so we anticipate a more significant effect of impregnation than for M5-40. Indeed, the addition of liquid from M7-l6 consumes Opx and precipitates Cpx and Ol (Fig. 2f): \[ 1 \text{ g Liq}_0 + 0.32 \text{ g Opx} \rightarrow 0.86 \text{ Liq}_1 \\ + 0.18 \text{ g Ol} + 0.28 \text{ g Cpx}. \] (7a) After the disappearance of Opx (between 55 and 60 g of added melt), the effect of impregnation drops dramatically, and the reaction becomes \[ 1 \text{ g Liq}_0 + 0.03 \text{ g Cpx} \rightarrow 0.94 \text{ g Liq}_1 + 0.09 \text{ g Ol}. \] (7b) In the Ol–Opx–Cpx triangle (Fig. 3c), the main effect of reaction (7a) is to increase the Ol and Cpx modes at the expense of Opx so that the solid assemblage in the impregnated peridotite goes from the harzburgite field, through the lherzolite field, to the wehrlite field. After the disappearance of Opx, the solid assemblage undergoes only minor changes. Set 2 In this set of calculations, we extracted partial melt from the bulk composition leaving 2 wt % of melt before launching the next impregnation step. For the melt from M5-40, the reaction with the residual peridotite is exactly the same as that obtained in the first set [reaction (6)]. For M7-l6, the impregnation reactions are \[ 1 \text{ g Liq}_0 + 0.37 \text{ g Opx} \rightarrow 0.80 \text{ g Liq}_1 \\ + 0.20 \text{ g Ol} + 0.36 \text{ g Cpx} \] (8a) until the disappearance of Opx (between 50 and 55 g of added melt), and then \[ 1 \text{ g Liq}_0 + 0.12 \text{ g Cpx} \rightarrow 0.93 \text{ g Liq}_1 + 0.19 \text{ g Ol} \] (8b) Despite a slightly higher melt consumption and Ol production in the second set, these two reactions are similar to reactions (7a) and (7b). Hence, the great similarity between the results of sets 1 and 2 suggests that calculations and experiments performed in a closed system where melt stays in equilibrium with the solid assemblage can be applied to the more realistic case where melt is continuously extracted from its source, as expected beneath mid-ocean ridges (e.g. Langmuir et al., 1992). EXPERIMENTAL RESULTS Attainment of equilibrium Phase assemblages, modes and run durations are given in Table 4. The durations of our experiments (76–137 h) are significantly longer than those in most high-pressure and high-temperature studies (Hirschmann et al., 2008). Although the present set of experiments was not reversed, the following observations indicate that chemical equilibrium was closely approached: (1) despite the presence of unreacted cores ($\leq 5 \mu$m) in some garnets, most solid phases show no detectable compositional zoning and their compositions are homogeneous throughout the sample chamber; (2) a high degree of microtextural equilibration is observed even at low degrees of melting; (3) Fe–Mg exchange coefficients (Table 4) between solid phases and liquid are consistent with those from the literature; (4) temperatures calculated on the basis of Cpx–liquid equilibrium (Putirka et al., 1996, 2003) are in good agreement with the experimental temperatures (see below). The close approach of chemical equilibrium in our experiments is due to the long run durations, the fine grain size of the starting materials and the highly reactive nature of gel when it is present. Partial melting experiments of pyroxenites M5-40 and M7-l6 The phase assemblage in experiments P40-25 and P16-25 consists of Liq + Cpx + Gt (Fig. 1; Table 4). In run P40-25, the melt fraction is lower (4.5%) than in run P16-25 (21.8%), and the Cpx/Gt ratio is significantly higher (3.0 and 1.4, respectively). Cpx crystals (20–30 $\mu$m in size) show no detectable compositional zoning and the compositional variability in a single sample is small. In contrast, Gt sometimes presents unreacted cores and its composition... in a single sample is more variable, probably as a result of incomplete equilibration. Fe–Mg exchange coefficients between Cpx and melt and Gt and melt are $0.32 \pm 0.01$ and $0.50 \pm 0.04$, respectively (Table 4); these coefficients are broadly consistent with those measured in previous pyroxenite partial melting experiments at $P > 2$ GPa (e.g. Kogiso et al., 2003; Kogiso & Hirschmann, 2006). Because Fe and Mg diffusion rates are generally higher than the diffusion rate of other elements (Na, Al, Ca), equilibrium tests based solely on Fe and Mg are not sufficient (e.g. Rhodes et al., 1979; Putirka, 1999). Accordingly, we used Putirka et al.’s (1996, 2003) geothermometer, which takes into account most major elements, to test equilibrium between Cpx and melt in our experiments (Fig. 5). There is a good agreement between calculated ($T_{\text{calc}}$) and experimental ($T_{\text{exp}}$) temperatures, as $(T_{\text{calc}} - T_{\text{exp}})$ differences are well within the error of the thermometer ($\pm 45^\circ$C; Putirka, 2008). Impregnation experiments: modes, textures, and tests for equilibrium Composition Bri2 equilibrated at 1.5 GPa and 1250°C is a subsolidus lherzolite with 65% Ol, 22% Opx, 13% Cpx and traces of Sp (run Bri2-l5 in Table 4). The phase assemblages in impregnation experiments I40-l5 and II6-l5 are qualitatively the same: Ol + Cpx + Sp + Opx + Liq. with 31.8% melt in I40-l5 and 6.6% in II6-l5. The proportions of the solid phases are very different in the two samples, however: the main solid phase in I40-l5 is Opx (42.7%), followed by Ol (16.1%), Cpx (6.7%), and Sp (2.8%); the main solid phases in II6-l5 are Cpx and Ol (46.6% and 43.2%, respectively), followed by Sp (3.6%) and Opx (trace). The phase assemblage in impregnation experiment II6-l0 is Ol + Cpx + Sp + Liq. Compared with II6-l5 (same composition and temperature, but a higher pressure of equilibration), sample II6-l0 has a larger melt fraction (41.3% vs 6.6%), a slightly larger Ol fraction, and much lower Cpx and Sp fractions (Table 4). In runs Bri2-l5 and II6-l5 (either subsolidus or with a low melt fraction), grain growth did not play a prominent role, and grains grew only from a few microns in the starting mixture to 2–15 μm in the final product. However, a high degree of textural equilibration was attained owing to the very small grain size of our starting material: microtextures are characterized by polygonal grain shapes with plane or slightly curved grain boundaries, and grain-boundary angles close to 120° as expected in a well-annealed polycrystal (Smith, 1964). In runs I40-l5 and II6-l0 in which the melt fraction is significantly higher, Ol crystals exceed 20 μm and pyroxene grains can reach a diameter of 10–15 μm. Fe–Mg exchange coefficients between Ol and melt are equal to $0.30 \pm 0.02$, perfectly consistent with values predicted by Töpflis’ (2005) model ($0.31 \pm 0.01$). Fe–Mg exchange coefficients between Opx and melt and between Cpx and melt are equal to $0.30 \pm 0.04$ and $0.33 \pm 0.04$, respectively. These values are in agreement with those reported in the peridotite system at the same pressure (Kushiro, 1996; Robinson et al., 1998; Falloon & Danyushevsky, 2000). Finally, the differences between temperatures calculated using Putirka et al.’s (1996, 2003) geothermometer and the experimental temperature are within the error bar of the thermometer (Fig. 5), indicating that Cpx–liquid equilibrium is approached in our experiments. Effect of impregnation on melt composition: experimental insight The experimental approach allowed us to determine the composition of the residual melt after the impregnation of a subsolidus peridotite by a pyroxenite-derived melt. Melt compositions in runs I40-l5, II6-l5, and II6-l0 are given in Table 4. The main differences between the two initial liquids (high concentrations of SiO$_2$, Al$_2$O$_3$ and alkalis, and low concentrations of FeO, CaO and MgO in gel 40 compared with gel 16; Table 3) are still visible, though reduced, after equilibration with the peridotite. Thus, the partial melt in run I40-l5 is a basaltic trachyandesite according to the total alkali–silica classification (Le Bas et al., 1986) whereas the partial melt in run II6-l5 is a basanite and the partial melt in run II6-l0 is a low-SiO$_2$ basalt close to a picro-basalt. In comparison with the initial melt compositions, the liquid phase in all experiments is strongly depleted in FeO and, as MgO shows only little... variation, the impregnation process results in a strong increase of melt-Mg-number: from 46·1 in gel 40 to 63·3 in run I40-15, from 38·1 in gel 16 to 47·8 in run II6-15, and to 55·5 in run II6-10. Other major differences between the initial liquid compositions and the liquid compositions at the end of the impregnation experiments include: (1) an increase of the Na$_2$O and K$_2$O concentrations, especially in run II6-15; (2) a strong enrichment in SiO$_2$ and Al$_2$O$_3$, and a depletion in CaO in runs II6-15 and II6-10; (3) a strong increase of TiO$_2$ in the melt from sample II6-15 (4% vs 17% in gel 16). The behaviour of elements and element ratios in melts during impregnation may be explained as follows. (1) Alkalis, Al$_2$O$_3$, and TiO$_2$ behave as incompatible elements and are concentrated in the melt as crystallization proceeds: their larger concentrations in the experimental melts compared with the initial gel compositions simply reflect the high melt consumption during the impregnation reaction. This effect is specially marked in run II6-15 owing to the small fraction of residual melt after equilibration (6·6%). (2) The behaviour of CaO in the melt is mainly controlled by the amount of Cpx precipitated in the impregnation reaction. Thus the constant CaO content in run I40-15 (compared with gel 40) is explained by the near-null participation of Cpx in the reaction [reaction (9a) below]. Conversely, the strong precipitation of Cpx in run II6-15 [reaction (10a) below] resulted in a major decrease of CaO in the residual melt (from 12·1 to 8·5%). (3) The strong silica enrichment in melt from runs II6-15 and II6-10 is due to Opx dissolution, Ol precipitation, and melt consumption during impregnation [reactions (10a) and (11) below]. In spite of this SiO$_2$ increase, the degree of silica undersaturation is larger in run II6-15 (8% normative Ne) than in gel 16 (5% normative Ne) owing to the significant increase of the alkali content in the melt (from 1·2 to 4·6%). In contrast, the melt in run II6-10 is Hy-normative (∼10%), a feature that we relate to the smaller consumption of melt during the reaction and the consequent smaller alkali content in the melt (1·8%). In run I40-15, the silica content in the melt remains nearly constant, but the alkali content increases significantly; hence, the degree of silica undersaturation strongly increases (from 1% normative Ne in gel 40 to 17% in melt I40-15). (4) Mg-number increase is related to equilibration with peridotite (Mg-number = 90·7). Partial melts in equilibrium with a solid matrix rich in high-Mg-number Ol, such as mantle peridotites, have FeO contents that evolve within a narrow band at a given $P$ owing to the olivine buffering effect (e.g. Kushiro, 1996; Médard et al., 2006): from ∼4 to 10·5% FeO at 1·5 GPa (Kushiro, 1996; Robinson et al., 1998; Falloon & Danyushevsky, 2000). In comparison, the initial melts in our experiments are significantly enriched in FeO: 11·1% in gel 40 and 24·6% in gel 16. Hence, they underwent a major drop of FeO content during their equilibration with peridotite Bri2: from 11·1% in gel 40 to 7·0% in I40-15, and from 24·6% in gel 16 to 15·2% in II6-15. In contrast, their MgO content varied little during interaction with peridotite. Indeed, at a given temperature, melts from pyroxenites and from peridotites have similar MgO contents (Lambart et al., 2009a). Impregnation experiments: comparison with pMELTS calculations The experimental phase proportions are compared with the proportions calculated with pMELTS in Table 6. The calculations were performed with the same starting compositions (Bri2, I40, and II6; Tables 1 and 3) and at the same pressure as in the experiments, but at a temperature 60°C higher than the experimental temperature (i.e. 1310°C vs 1250°C). This is to take into account the temperature shift between pMELTS predictions and the experimental data, as discussed by Ghiorso et al. (2002). The theoretical modes in Table 6 are slightly different from those for case 1 in Table 5 because the synthetic gels 40 and 16 do not exactly match the compositions of the pyroxenite-derived melts used in the thermodynamic approach (Tables 2 and 3). The differences are relatively small, however, compared with the modal changes owing to the impregnation reaction: in the Ol–Opx–Cpx triangle (Fig. 3a), the modes computed for bulk compositions I40 and II6 plot close to the modes obtained after impregnation of 100 g of peridotite Bri2 by 100 g of melt from pyroxenites M5-40 or M7-16. As in the thermodynamic section above, phase abundances are given as masses, not as percentages: for run Bri2-15 (equilibration experiment on peridotite Bri2 at 1·5 GPa and 1250°C), we give the mass of each phase for 100 g of peridotite; for the three impregnation experiments, which simulate the equilibration of 100 g of peridotite with 100 g of melt, we give the mass of each phase for a bulk mass of 200 g. In this way, the behaviour of a phase is obvious: its mass increases if it is on the product side of the impregnation reaction, it decreases if it is on the reactant side, and the difference between its final mass in a 200 g bulk mass and its initial mass in 100 g of peridotite gives its coefficient in the impregnation reaction (written in grams for 100 g of added liquid). It should be noted that we can only write a global impregnation reaction corresponding to the addition of 100 g of melt: unlike with the thermodynamic approach, we cannot detect a switch in the impregnation reaction in relation to the exhaustion of Table 6: Comparison between the modes measured in the impregnation experiments and the theoretical modes calculated with pMELTS \| \| Br2-15 \| \| I40-15 \| \| Gel 16 \| I16-15 \| \| Br2-10* \| \| I16-10 \| \| \|----------------\|--------\|--------\|--------\|--------\|--------\|--------\|--------\|---------\|--------\|--------\|--------\| \| \| exp \| calc \| exp \| calc \| exp \| calc \| exp \| calc \| exp \| calc \| \| \| ol \| 65·0 \| 64·3 \| 32·2 \| 7·7 \| 86·3 \| 80·3 \| 65·0 \| 98·0 \| 83·4 \| \| \| opx \| 21·7 \| 24·4 \| 85·4 \| 104·1 \| 0·3 \| — \| 23·7 \| 0 \| 0 \| \| \| cpx \| 13·1 \| 10·5 \| 13·3 \| 25·5 \| 93·1 \| 106·5 \| 7·1 \| 19·0 \| 32·2 \| \| \| sp \| 0·2 \| 0·7 \| 5·6 \| 7·8 \| 7·1 \| 8·0 \| 0·3 \| 0·2 \| 0 \| \| \| liq \| — \| — \| 63·5 \| 54·9 \| 13·3 \| 5·1 \| 3·8 \| 82·7 \| 84·3 \| \| \| $a_{\text{SiO}_2}$ \| 0·44 \| 0·35 \| 0·31 \| 0·35 \| \| \| \| \| \| \| Modes are given as masses, not percentages, and are expressed in two ways: for the initial states Br2-15 (peridotite Br2 at 1·5 GPa and 1250°C) and Br2-10 (peridotite Br2 at 1 GPa and 1250°C), we give the mass of each phase for 100 g of peridotite; for the final states I40-15, I16-15, and I16-10, which correspond to the equilibration of 100 g of peridotite with 100 g of melt, we give the mass of each phase for a bulk mass of 200 g. The calculations were performed at the same pressure as the experiments ($P = 1·5$ GPa for Br2-15, I40-15, and I16-15; $P = 1$ GPa for Br2-10 and I16-10), but at a temperature 60°C higher than the experimental temperature (1310°C vs 1250°C) to take into account the temperature shift between pMELTS predictions and experimental data (Ghiorso et al., 2002). Oxygen fugacity was set to FMQ – 1 in all calculations. As we did not run an experiment with peridotite Br2 at 1 GPa and 1250°C, we give only the theoretical mode for the initial state Br2-10. †The impregnation reaction and thus the evolution of phase proportions are controlled by the silica activity, $a_{\text{SiO}_2}$, in the initial liquids (gel 40, gel 16). Silica activities were calculated using the MELTS supplemental calculator (Ghiorso & Sack, 1995; Asimow & Ghiorso, 1998) at 1·5 GPa and 1310°C. a solid phase and/or the precipitation of a new phase. As the solid phase assemblage in runs I40-15 and I16-15 is the same as in peridotite Br2, this difficulty concerns only experiment I16-10 in which we do not know the mass of melt required to exhaust Opx. The theoretical mode computed before impregnation is in good agreement with the mode measured in experiment Br2-15: the masses calculated for Opx and Sp are overestimated and the mass of Cpx is underestimated, but they remain within the uncertainty of the mass-balance calculations in Table 4. The theoretical modes after impregnation are also consistent with those observed in experiments I40-15, I16-15, and I16-10, in spite of some systematic deviations (underestimation of Ol fraction and overestimation of Cpx and Sp fractions in the calculations). The comparison of the global impregnation reactions predicted by pMELTS with the reactions inferred from experimental data supports this agreement: I40-15 (exp): $$100 \cdot 0 \text{ g Liq}_0 + 32 \cdot 8 \text{ g Ol} \rightarrow 63 \cdot 5 \text{ g Liq}_1 + 63 \cdot 7 \text{ g Opx} + 0 \cdot 2 \text{ g Cpx} + 5 \cdot 4 \text{ g Sp}$$ (9a) I40-15 (calc): $$100 \cdot 0 \text{ g Liq}_0 + 56 \cdot 6 \text{ g Ol} \rightarrow 54 \cdot 9 \text{ g Lid}_1 + 79 \cdot 7 \text{ g Opx} + 15 \cdot 0 \text{ g Cpx} + 7 \cdot 1 \text{ g Sp}$$ (9b) I16-15 (exp): $$100 \cdot 0 \text{ g Liq}_0 + 21 \cdot 4 \text{ g Opx} \rightarrow 13 \cdot 3 \text{ g Liq}_1 + 21 \cdot 3 \text{ g Ol} + 80 \cdot 0 \text{ g Cpx} + 6 \cdot 9 \text{ g Sp}$$ (10a) I16-15 (calc): $$100 \cdot 0 \text{ g Liq}_0 + 24 \cdot 4 \text{ g Opx} \rightarrow 5 \cdot 1 \text{ g Liq}_1 + 16 \cdot 0 \text{ g Ol} + 96 \cdot 0 \text{ g Cpx} + 7 \cdot 3 \text{ g Sp}$$ (10b) I16-10 (calc): $$100 \cdot 0 \text{ g Liq}_0 + 23 \cdot 7 \text{ g Opx} + 0 \cdot 3 \text{ g Sp} \rightarrow 80 \cdot 5 \text{ g Liq}_1 + 18 \cdot 4 \text{ g Ol} + 25 \cdot 0 \text{ g Cpx}.$$ (11) In the case of I16-10, we give only the thermodynamic impregnation reaction because we have no experimental constraints on the mode of peridotite Br2 at 1 GPa and 1250°C. In Fig. 6, we present the compositional difference between melts calculated with pMELTS and the experimental melts. This figure highlights that pMELTS significantly underestimates the CaO contents and overestimates the MgO and FeO contents of melts in comparison with the experimental melts. It should be noted that Na$_2$O is also significantly overestimated in calculations on I40-15 and I16-15. In the case of I16-10, on the contrary, Na$_2$O is slightly underestimated by pMELTS (133%) compared with the experiment (157%). These discrepancies are due to the limitations of pMELTS as discussed by Ghiorso et al. (2002), and are consistent with the overestimation of Cpx production in the calculations compared with the experiments (see above), the much lower coefficients for the residual liquid in reactions (9b) and (10b) than in reactions (9a) and (10a) respectively, and the much larger Ol coefficient in reaction (9b) than in reaction (9a). Despite these differences, there is a perfect qualitative agreement between the thermodynamic and experimental approaches as the reactions predicted by pMELTS have the same form as those deduced from the experiments: the same phases are on the reactant side of the reactions (Ol for I40-15; Opx for I16-15) and on the product side (Opx, Cpx, and Sp for I40-15; Ol, Cpx, and Sp for I16-15). The agreement is also good quantitatively as in most cases the coefficients in the thermodynamic reactions are close to those in the experimental ones. In addition, it is possible that part of the difference between the melt fractions predicted by pMELTS and those measured in the experiments is linked to a small amount of dissolved volatiles in the experimental glasses. This hypothesis is supported by the case of I16-10: the experimental glass in I16-10 is presumably volatile-free as its analytical total is 99.9%, and this is the only case in which the quantity of melt measured in the experiment is equal (within error) to the quantity predicted by pMELTS. We conclude that pMELTS is a reliable tool to estimate qualitatively and quantitatively the impregnation reactions between pyroxenitic melts and surrounding peridotite. In particular, the experimental study supports one of the main conclusions of the thermodynamic study (for cases 1 and 2): melts from pyroxenites are strongly consumed by interaction with subsolidus peridotite (Fig. 2), a factor that should tend to limit further melt infiltration and to isolate the pyroxenite bodies from the surrounding peridotites. DISCUSSION AND PETROLOGICAL IMPLICATIONS* Synthesis of thermodynamic and experimental results Four major results emerge from our calculations and experiments on the impregnation of subsolidus peridotites by pyroxenite-derived melts (cases 1 and 2): the massive production of clinopyroxene, the contrasted behaviour of Opx and Ol depending on the composition of the added melt, the transformation of lherzolites into Ol websterites or wehrlites, and the important consumption of the added melt. A fifth result comes from our calculations on case 3; namely, the minor effect of pyroxenitic melts on partially molten peridotites at low pressure. These results are summarized below. (1) Massive production of clinopyroxene. As a rule, large amounts of Cpx (and small quantities of Sp at 1.5 GPa or Gt at 2.5 GPa) are produced during the impregnation of subsolidus peridotites by pyroxenite-derived melts at 1.5–2.5 GPa. In the most extreme cases, the mass of pyroxene produced is larger than the mass of added liquid: for instance, 1.15 g of Cpx per gram of added liquid in reaction (3a). (2) Contrasteds behaviours of Ol and Opx. Both calculations and experiments show that the behaviours of Ol and Opx depend on the composition of the added liquid. In the case of melt from pyroxenite M5-40, Opx is on the product side of the impregnation reaction and Ol on the reactant side, at both 1.5 and 2.5 GPa [reactions (2), (4), (9a) and (9b)]. Conversely, in the case of melt from M7-16, Ol is on the product side and Opx on the reactant side [reactions (3a), (5a), (10a) and (10b)]. In the next section, we show that the main factor controlling the behaviour of Ol and Opx during peridotite–pyroxenite interactions is the silica activity of the infiltrating melt, not its degree of silica undersaturation as the liquids from M7-16 and M5-40 have similar percentages of normative nepheline (Table 2). (3) Fate of pyroxenite-derived melts in a subsolidus environment. All impregnation reactions at 1.5 and 2.5 GPa are accompanied by an important consumption of pyroxenitic melt: from ~35–45% [reactions (4), (9a) and (9b)] to ~85–100% [reactions (3a), (10a) and (10b)] of the added melt is lost during the equilibration with peridotite. In particular, we observe that reactions (3a), (5a), (10a) and (10b), which have Ol on the product side, yield only very little residual liquid owing to the massive precipitation of Cpx. Hence, in our experiments and calculations for cases 1 and 2, peridotite–pyroxenite interactions tend to consume a significant fraction of the liquids produced by the partial melting of pyroxenites and to precipitate large quantities of pyroxenes. These effects are expected to decrease drastically the permeability of the system and the capacity of pyroxenite-derived melts to infiltrate neighbouring rocks. (4) From lherzolites to Ol websterites or wehrlites. The impregnation process results in major changes of the modal composition of the solid assemblage (Fig. 3a and b). In the case of melts from M5-40, the solid assemblage goes from the lherzolite field to the Ol websterite field owing to the precipitation of Opx and Cpx, and the dissolution of OL. In the case of the melts from M7-16, the evolution is from the lherzolite field toward the wehrlite field owing to the precipitation of Cpx and the dissolution of Opx. (5) Minor effect of peridotite–pyroxenite interactions at low pressure. Lambart et al. (2009a) showed that many pyroxenites produce liquids that are similar to peridotite-derived melts for most major elements (SiO$_2$, Al$_2$O$_3$, CaO, MgO, and FeO) at 1 GPa and 1250–1300 °C. Accordingly, a restricted effect of the impregnation by pyroxenite-derived melts may be the rule for most pyroxenites at the $P-T$ conditions corresponding to our case 3, as we observed for the melt derived from M5-40. In particular, we want to emphasize that for both pyroxenites M5-40 and M7-16, the mass of residual liquid is close to the mass of added liquid [reactions (6), (7a) and (7b)]; that is, only a small fraction of the added liquid is consumed during the impregnation process. This behaviour is the opposite of that observed in cases 1 and 2 above. Hence, in the conditions of case 3, the impregnation process will not ‘clog’ the melt pathways in the peridotite. This result suggests that when the overlying peridotites are partially molten, the pyroxenitic melt can be extracted and percolate from its source through the mantle. Factors controlling the dissolution or precipitation of Ol and Opx In many theoretical (e.g. Lundstrom et al., 2000; Kogiso et al., 2004b), experimental (e.g. Keshav et al., 2004; Kogiso & Hirschmann, 2006) and petrological (e.g. Griffin et al., 1999; Aulbach et al., 2007; Piccardo & Vissers, 2007) studies, it is assumed that a silica-undersaturated melt interacting with peridotite consistently dissolves pyroxenes and precipitates Ol. The liquid from pyroxenite M7-16, which is Ne-normative (Table 2), does indeed dissolve the Opx from peridotite Bri2 and precipitate Ol. However, the liquid from pyroxenite M5-40, which is also silica-undersaturated and Ne-normative (Table 2), dissolves Ol and precipitates Opx. In fact, the melts from pyroxenites M5-40 and M7-16 have similar percentages of normative Ne (from 6 to 9%; Table 2), but contrasted silica contents: the melt from M5-40 contains 50-6% SiO$_2$ at 1.5 GPa and 48-7% at 2.5 GPa whereas the melt from M7-16 contains 41-6% SiO$_2$ at 1.5 GPa and 39-1% at 2.5 GPa. Accordingly, the main factor controlling the behaviour of Ol and Opx during peridotite–pyroxenite interactions is the silica content of the infiltrating melt, or more strictly the silica activity, not the degree of silica undersaturation. In most impregnation calculations, the liquid is in equilibrium with a lherzolitic assemblage: Ol+Opx+Cpx+Sp/Gt (except after Opx exhaustion; Fig. 2). The silica activity of Ol+Opx-saturated melts, $a^0_{\text{SiO}_2}$, is buffered by the reaction $\text{Ol} + (\text{SiO}_2)_{\text{liquid}} \leftrightarrow 2 \text{ Opx}$. Therefore, at fixed temperature and pressure, $a^0_{\text{SiO}_2}$ will be nearly constant, even though liquid compositions may vary considerably (Carmichael et al., 1970). Hence, in an interaction process at constant $P$ and $T$ as described in this paper, the silica activity of the added melt will condition the behaviour of Opx and Ol during the reaction: if the added liquid has a silica activity lower than $a^0_{\text{SiO}_2}$, which is the case of the melts from M7-16 (see below), it will dissolve Opx and precipitate Ol until its silica activity equals $a^0_{\text{SiO}_2}$ (or until Opx exhaustion). If the added melt has a silica activity higher than $a^0_{\text{SiO}_2}$, which is the case of the melts from M5-40, then it will dissolve Ol and precipitate Opx until its silica activity equals $a^0_{\text{SiO}_2}$ (or Ol is exhausted). To illustrate this point, we computed the silica activities in melts from experiments I40-15 and II6-15, and in initial melts gel 40 and gel 16 (silica activities were calculated using the MELTS supplemental calculator; see Table 6). As expected, liquids from runs I40-15 and II6-15, which are both in equilibrium with Ol+Opx (+Cpx+Sp), have similar silica activities: $\sim$0.35. In contrast, silica activities in the initial melts are different: 0.44 in gel 40 vs 0.31 in gel 16. These calculations confirm the major control exerted by silica activity on the impregnation reaction: impregnation of peridotite by a melt with a low silica activity, such as gel 16, leads to Opx dissolution and Ol precipitation, while impregnation by a liquid with a high silica activity, such as gel 40, results in Ol dissolution and Opx precipitation. To highlight the relationship between silica activity, the evolution of melt compositions and phase relations during melt–rock interactions, melt compositions at the beginning (gels 16 and 40) and at the end (experimental glasses in runs II6-15 and I40-15) of the impregnation process at 1.5 GPa are projected from diopside (Di) onto the Figure 7 allows us to display the evolution of the liquid phase and of its solid matrix (‘the residue’) during the impregnation experiments. Runs I40-I5 and I16-I5 show the same behaviour for CaTs and Fo (with an enrichment in CaTs and a depletion in Fo in both the liquid and the solid residue) but a contrasted behaviour for Qz. To attain the Ol–Opx cotectic, the liquid in run I40-I5 must become poorer in the Qz component through Ol dissolution and Opx precipitation; in parallel, the solid residue becomes richer. As a result, the tie line joining the solid and the liquid and passing through the bulk composition rotates anticlockwise during the equilibration process (arrows in Fig. 7). The relationships are exactly the reverse in the case of experiment I16-I5: the liquid must become richer in Qz to reach the Ol–Opx cotectic, the solid residue becomes poorer and this is achieved by Opx dissolution and Ol precipitation. Implications for the lithological diversity of the upper mantle: the case of Ronda Our calculations and experiments show that the interaction between melts derived from pyroxenites and surrounding rocks in a heterogeneous mantle may generate new rock-types, including cpx-rich peridotites, wehrlites and ol-websterites (Fig. 3). In this section, we show that the ultramafic massif of Ronda, SW Spain, provides a natural example where a new generation of pyroxenites was produced by interactions between peridotites and percolating melts derived from older pyroxenites. Ronda pyroxenites are divided into four groups (A–D; Garrido & Bodinier, 1999), of which groups A and C are of special interest for our study. Pyroxenites from group A are mainly Gt pyroxenites with a strong radiogenic Os isotopic signature whereas group C pyroxenites are Sp-bearing websterites and Ol websterites with $^{187}$Os/$^{186}$Os ratios intermediate between group A pyroxenites and the surrounding peridotites (Reisberg et al., 1991). On the basis of petrological and geochemical observations, Garrido & Bodinier (1999) proposed a model in which group C pyroxenites were produced by interactions between the surrounding peridotites and a reacting melt with a component derived from the partial melting of group A pyroxenites; the latter component has high $^{187}$Os/$^{186}$Os ratios. We compared the modal proportions of Ol, Opx, and Cpx in group C pyroxenites with those predicted by pMELTS for our case 1 (Fig. 3a); it should be noted that the $P$–$T$ conditions of case 1 ($1250^\circ$C and 1.5 GPa) are close to those estimated for the thermal event in Ronda: $T \geq 1200^\circ$C, $P \approx 1.5$ GPa (Lenoir et al., 2001). Group C pyroxenites have modes consistent with the impregnation of peridotite by melts with compositions close to the melt from M5-40 or intermediate between melts from M5-40 and M7-16 at 1.5 GPa and 1250°C: melts with a high silica activity, such as the melt from M5-40, are good candidates to generate the most Opx-rich websterites (with Opx/Cpx ratios approaching... between M5-40 and M7-16 (though closer to M5-40; Table 1, Fig. 8). Accordingly, the conclusion that partial melts derived from group A pyroxenites are a major component of the mixture of melts that reacted with peridotite to yield group C pyroxenites (Garrido & Bodinier, 1999) is supported by our study. Finally, based on trace element distribution in Cpx, Bodinier et al. (2008) estimated that the magma/rock ratio required to produce group C pyroxenites is between 0.5 and 1.7. These estimations agree with those inferred from our calculations: $\geq 0.4$ for Ol websterites and significantly greater than 1 for true websterites (Figs 3a and 8). As a whole, there is a good qualitative and quantitative agreement between the petrological and geochemical studies of Bodinier and collaborators and the present sets of calculations and experiments. We conclude that interactions between percolating melts, especially those derived from pyroxenites, and host peridotites can increase the lithological and compositional diversity of the mantle, and produce new types and generations of peridotites and pyroxenites (for instance, peridotites with less than 50% Ol and/or low Mg-number, or websterites with a high Opx/Cpx ratio). It should be noted that this type of process is also likely to occur in the source regions of basalts. Indeed, the formation of pyroxenite via liquid–rock interactions has recently been proposed to explain the compositions of oceanic basalts (e.g. Sobolev et al., 2007), and the origin of iron-rich peridotites (e.g. Tommasi et al., 2004). Implications for the extraction and transport of pyroxenite partial melts at high pressure The solidus temperatures of pyroxenites are lower than those of peridotites: accordingly, pyroxenites begin to melt deeper than peridotites in a heterogeneous mantle ascending adiabatically. In addition, Gt remains stable at lower pressures in pyroxenites than in peridotites. These properties were used by Hirschmann & Stolper (1996) to model magma genesis at mid-ocean ridges and to reinterpret the garnet signature in MORB: if bodies of pyroxenite are present in the ascending mantle, then melts with a garnet signature (i.e. with compositions indicating the presence of residual Gt in the source region, such as a high LREE/HREE or a decoupling of Hf–Nd isotope systematics) can be produced at lower pressures than in the case of a homogeneous peridotitic mantle. If we admit that the trace-element garnet signature in MORB is supplied by pyroxenite-derived melts, then the reaction between melts once in equilibrium with Cpx + Gt and the surrounding peridotite must be limited, so that they can preserve their geochemical characteristics en route to the surface. We discuss below the case of isolated pyroxenite bodies in a peridotite matrix and the mechanisms that would allow the extraction and transport of pyroxenite-derived melt with minimal interactions with the peridotite. Pyroxenite bodies in a subsolidus mantle At high pressure, between the pyroxenite solidus and the peridotite solidus, the melting degree of pyroxenites is low to moderate while the surrounding peridotites are unmolten (case 2 in Table 2; the considerations below also apply if the peridotite crosses its solidus as long as its degree of melting remains very low). Interactions at the peridotite–pyroxenite interface result in the consumption of a large part of the pyroxenite-derived melt and the crystallization of Cpx + Gt + Opx or Cpx + Gt + Ol (depending on pyroxenite composition; Fig. 2c and d). Hence, a low-permeability layer develops around the pyroxenite body, favouring a (near) closed-system evolution. In this context, gravity-driven solid–liquid separation could lead to the formation of melt-rich lenses at the top of the pyroxenite bodies. As such accumulations are mechanically unstable, the next step in the evolution of the system could be the periodic expulsion of melt into veins or channels (Kogiso et al., 2004b), allowing the rapid transport of pyroxenite-derived melts to the surface and the preservation of their garnet signature. Pyroxenite bodies surrounded by partially molten peridotite Now we assume that the peridotite is partially molten and that its permeability is sufficiently large to allow the percolation of pyroxenite-derived melts. This situation is similar to our case 3 in Table 2, and corresponds to pressures close to the mean pressures of melting beneath mid-ocean ridges (1–1.5 GPa; e.g. Langmuir et al., 1992). At these conditions, pyroxenites have high to very high degrees of melting and many of them (such as M5-40) yield partial melts that are similar to peridotite-derived melts for most major elements (see Fig. 6 in Lambart et al., 2009a). This small compositional contrast results in a very limited reactivity between pyroxenite-derived melt and peridotite [reaction (6); Figs 2e and 3c] that is favourable to the preservation of their trace element signature. Special pyroxenite compositions such as M7-l6 Lambart et al. (2009a) showed that, at a given pressure, pyroxenite partial melts may have strongly contrasted silica contents depending on whether they are saturated in Opx or not. In particular, low-SiO$_2$ clinopyroxenites such as M7-l6 produce liquids that are strongly depleted in SiO$_2$ in comparison with peridotite-derived melts. Our calculations and experiments indicate that interaction between such low-SiO$_2$ melts and the surrounding peridotite leads to Opx dissolution [reactions (3), (5), (7), (10a) and (10b)] and formation of wehrlite or Opx-free clinopyroxenite as solid residues (Fig. 3). At low pressure (1–1.5 GPa), when peridotites are partially molten (our case 3), melt consumption is limited [$\sim$10% in reaction (7); Fig. 2f], and thus low-SiO$_2$ pyroxenite-derived melts can be extracted from their source. However, as they are very different from typical peridotite melts, they will interact with the surrounding peridotite and undergo significant chemical changes, as exemplified by gel l6 in experiment II6-10 (Tables 3 and 4). In particular, a strong enrichment in silica of the percolating melt, associated with Opx dissolution in the solid matrix, is expected, to increase its silica activity and to approach the Ol–Opx cotectic, as discussed above (Fig. 7). Moreover, if pyroxenite-derived melts are enriched in iron compared with peridotite partial melts, they will also become depleted in iron. The importance of these chemical changes will decrease with additional melt infiltration because Opx and then Cpx will disappear from the solid matrix, leaving a pure Ol residue. Opx is expected to be rapidly exhausted (at a melt–rock ratio between 0.5 and 0.65 in our calculations). Extending our calculations in Fig. 2f to high melt/rock ratios, we computed that a ratio of $\sim$8 is needed to exhaust Cpx. As pMELTS overestimates pyroxene production at the expense of Ol, the true melt–rock ratio required for Cpx exhaustion is probably much lower. Hence, pyroxene-free dunite regions could be formed by circulation and focusing of SiO$_2$-poor pyroxenitic melts within peridotite. Such melts could therefore contribute to the initiation and formation of a large-scale network of dunite channels, although their role is probably limited by the small proportion of low-SiO$_2$ pyroxenites in the upper mantle. Once a dunite network is established, additional melt could circulate without significant compositional change (Kelemen et al., 1995). A sketch of MORB petrogenesis in a heterogeneous mantle The main conclusions of our study are summarized in Fig. 9. The conceptual model presented here does not take many dynamic parameters into account (such as the solid flow rate or the rate of melt extraction relative to the rate of matrix melting). However, it illustrates how our results can be applied to the natural case, summarizing (1) the effect of melt–peridotite interactions on melt transport and mantle lithological diversity, and (2) the role of pyroxenites in magma genesis at mid-ocean ridges (MOR). In Fig. 9, we consider a heterogeneous mantle made of veins of pyroxenite within a peridotitic matrix. As mantle ascends adiabatically beneath a MOR, it first crosses the pyroxenite solidus at high pressure: at this point, the pyroxenites give rise to Ne-normative liquids, which react with the surrounding peridotite (still subsolidus) to produce new rock-types (Gt-bearing hybrids intermediate between peridotites and pyroxenites). As peridotite–pyroxenite interactions consume a large fraction of the pyroxenite-derived liquids, they tend to stop further melt infiltration and to isolate the pyroxenite bodies from their host. At this stage, solid–liquid separation can lead to the accumulation of melt-rich lenses at the top of the pyroxenite bodies, a situation favourable to the expulsion and rapid transport to the surface of melts carrying the geochemical signature of garnet. At some lower pressure (which depends on the local potential temperature of the mantle), the ascending mantle crosses the peridotite solidus, and an interconnected network of melt develops at the grain scale in the peridotite matrix. Rapidly, peridotite- and pyroxenite-derived melts begin to migrate upwards by diffuse porous flow. Most pyroxenite-derived melts should be able to percolate through the peridotite matrix with minimum interactions, thus preserving their isotopic and trace element signature. Because of their low silica activity, some pyroxenitic melts will, however, react with the peridotite and dissolve away Opx and possibly Cpx, a process that may initiate the formation of dunites and lead to a switch from pervasive porous flow to focused transport into a network of high-permeability dunite channels (Kelemen et al., 1995). Such dunite channels provide the magma with a rapid pathway to the surface and a chemical isolation from the surrounding mantle, thus promoting the preservation of information about the melting process at depth (e.g. melting conditions, primary melt compositions, etc.). Finally, we come back to processes of magma transport beneath MORs. The available geochemical data (e.g. O’Hara, 1968; Stolper, 1980; Johnson et al., 1990) are explained by the fact that the migration of mantle melts beneath a MOR occurs through focused flow along chemically isolated dunite channels (Kelemen et al., 1997). However, the transition from diffuse porous flow to focused flow in channels remains unclear. The following models have been proposed to explain melt focusing: (1) formation of self-organized dissolution channels by reactive infiltration instability (e.g. Daines & Kohlstedt, 1994; Spiegelman et al., 2001); (2) dike formation by hydraulic fracturing (e.g. Sleep, 1988; Takahashi, 1992); (3) mechanical stress-driven localization of melt (Holtzman et al., 2003); (4) presence of a barrier to upward melt migration at the base of the oceanic lithosphere (e.g. Sparks & Parmentier, 1991; Ghods & Arkani-Hamed, 2000; Katz et al., 2006). As a variant of (4), we propose that some mantle heterogeneities may evolve into permeability barriers able to trigger melt focusing. A body of partially molten pyroxenite isolated in the subsolidus peridotitic mantle could differentiate into a melt lens at the top and a layer of strongly compacted cumulates at the bottom. If the body is wide enough (>100 m; e.g. Spiegelman et al., 2001), it could form a barrier to melt ascension, forcing the melt to focus along its sides. This assumption needs to be further studied and quantified, but could provide a viable mechanism for the onset of focusing. This is consistent with some previous studies: permeability barriers (Herbert & Montési, 2010) and compaction of the solid matrix (e.g. Spiegelman et al., 2001) are two factors that can promote melt focusing. Fig. 9. Sketch of MORB petrogenesis in the case of a heterogeneous mantle composed of pyroxenite veins (dotted, folded layers at the bottom) in a peridotite matrix (see text). Processes acting at a given depth are listed in the boxes on the right: melting processes in light grey boxes, and peridotite–pyroxenite interactions in dark grey boxes. The outcome of each process for the chemical and mineralogical evolution of mantle is summarized in the left column. The grey triangle is the melting zone: in its lower part, only pyroxenites are partially molten; in its upper part (cross-hatched area), both pyroxenites and peridotites are partially molten. Black arrows above the peridotite solidus mark the onset of penetrative porous flow; white arrows represent melt focusing into high-permeability channels. CONCLUSIONS To predict the fate of pyroxenite-derived melts during their migration through the peridotitic mantle, we performed a thermodynamic (with pMELTS) and experimental study of the impregnation of a peridotite by pyroxenite-derived melts. Three contexts were considered: (1) in a subsolidus lithospheric mantle; (2) at high pressure beneath a mid-ocean ridge, in a subsolidus asthenospheric mantle; (3) at lower pressure beneath a mid-ocean ridge, in a partially molten asthenospheric mantle. Despite some mismatches observed between model outputs and observations (Cpx overestimation, temperature offset) showing that pMELTS still needs to be improved, our work shows that the software is a suitable tool to estimate qualitatively the impregnation reactions between pyroxenitic melts and surrounding peridotite. More importantly, this study highlights the importance of magma–rock interactions for MORB petrogenesis and for the development of mantle lithological diversity, and, in particular, their significant impact on magma extraction processes. Reactions are varied—they depend on many parameters such as melt composition, pressure and the melting degree of surrounding rock—and experimental data are still too sparse. The scientific challenge is now to successfully incorporate these processes into a dynamic melting model of a heterogeneous mantle. ACKNOWLEDGEMENTS The authors are grateful to Jacques Kornprobst, Emeritus Professor at Université Blaise Pascal, Clermont-Ferrand, for providing the two pyroxenites from Beni Bousera used in this study. This study has benefited from discussions with Michel Pichavant, Othmar Müntener, Etienne Médard and Muriel Laubier. Special thanks are due to the following persons: Jean-Louis Fruquière and Frank Pointud for manufacturing the piston-cylinder assemblages and for maintenance operations; Jean-Luc Devidal for technical assistance with the electron microprobe and for the preparation of synthetic gels; Jean-Marc Hénot for technical assistance with the scanning electron microscope; Mhammed Benbakkar for the ICP-AES analyses of pyroxenites and synthetic gels. We also thank Tetsu Kogiso, Bruno Scaillet and an anonymous reviewer for their constructive reviews. FUNDING This work was supported by the program DyETI of the Institut National des Sciences de l’Univers (INSU-CNRS) and by the Agence Nationale de la Recherche ‘MIME’ grant (ANR-07-BLAN-0130–01 to A.P). REFERENCES Albarède, F. & Provost, A. (1977). Petrological and geochemical mass-balance equations: an algorithm for least-square fitting and general error analysis. Computers and Geosciences 3(2), 309–326. Allègre, C. J. & Turcotte, D. L. (1986). Implications of a two-component marble-cake mantle. Nature 323(6084), 123–127. Ancey, M., Bastenaire, F. & Tixier, R. (1978). Application des méthodes statistiques en microanalyse. In: Maurice, F., Meny, L. & Tixier, R. (eds) Microanalyse, microscopie à balayage. Orsay: Les éditions du physicien, pp. 325–347. Asimow, P. D. & Ghiorso, M. S. (1998). Algorithmic modifications extending MELTS to calculate subsolidus phase relations. American Mineralogist 83, 1127–1132. Asimow, P. D., Hirschmann, M. M., Ghiorso, M. S., O’Hara, M. J. & Stolper, E. M. (1995). The effect of pressure-induced solid–solid phase transitions on decompression melting of the mantle. Geochimica et Cosmochimica Acta 59(21), 4489–4506. Aulbach, S., Griffin, W. L., Pearson, N. J., O’Reilly, S. Y. & Doyle, B. J. (2007). Lithosphere formation in the central Slave Craton (Canada): plume subcretion or lithosphere accretion? Contributions to Mineralogy and Petrology 154, 409–427. Baker, M. B. & Stolper, E. M. (1994). Determining the composition of high-pressure mantle melts using diamond aggregates. Geochimica et Cosmochimica Acta 58(15), 2811–2827. Berman, R. G. & Koziol, A. M. (1991). Ternary excess properties of grossular–pyrope–almandine garnet and their influence in geothermobarometry. American Mineralogist 76, 1223–1231. Bodinier, J.-L., Garrido, C. J., Chaneo, I., Brugruier, O. & Gervilla, F. (2008). Origin of pyroxenite–peridotite veined mantle by refertilization reactions: evidence from the Ronda peridotite (Southern Spain). Journal of Petrology 49(5), 999–1025. Carmichael, I. S. E., Nicholls, J. & Smith, A. L. (1970). Silica activity in igneous rocks. American Mineralogist 55, 246–263. Christie, D. M., Carmichael, I. S. E. & Langmuir, C. H. (1986). Oxidation states of mid-ocean ridge basalt glasses. Earth and Planetary Science Letters 79, 397–311. Cottrell, E. & Kelley, K. A. (2011). The oxidation state of Fe in MORB glasses and the oxygen fugacity of the upper mantle. Earth and Planetary Science Letters 305, 270–282. Daines, M. J. & Kohlstedt, D. L. (1994). The transition from porous to channeled flow due to melt-rock reaction during melt migration. Geochemistry, Geophysics, Geosystems 2(2), 145–148. Dick, H. J. B. & Sinton, J. M. (1979). Compositional layering in Alpine peridotites: evidence for pressure solution creep in the mantle. Journal of Geology 87, 403–416. Fallon, T. J. & Danyshevsky, L. V. (2000). Melting of refractory mantle at 1.5, 2 and 2.5 GPa under anhydrous and H$_2$O-undersaturated conditions: implications for the petrogenesis of high-Ca boninites and the influence of subduction components on mantle melting. Journal of Petrology 41(2), 257–283. Faul, U. H. (1997). Permeability of partially molten upper mantle rocks from experiments and percolation theory. Journal of Geophysical Research 102(B5), 10299–10311. Garrido, C. J. & Bodinier, J.-L. (1999). Diversity of mafic rocks in the Ronda peridotite: Evidence for persuasive melt–rock reaction during heating of subcontinental lithosphere by upwelling asthenosphere. Journal of Petrology 40, 729–754. Ghiorso, M. S. & Sack, R. O. (1995). Chemical mass transfer in magmatic processes IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology 119(2), 197–212. Ghiorso, M. S., Hirschmann, M. M., Reiners, P. W. & Kress, V. C., III (2002). The pMELTS: A revision of MELTS for improved calculation of phase relations and major element partitioning related to partial melting of the mantle to 3 GPa. Geochemistry, Geophysics, Geosystems 3(5), doi:10.1029/2001GC000217. Ghods, A. & Arkan-i-Hamed, J. (2000). Melt migration beneath MORs. Geophysical Journal International 140, 687–697. Griffin, W. L., O’Reilly, S. Y. & Ryan, C. G. (1999). The composition and origin of subcontinental lithospheric mantle. In: Fei, Y., Bertka, C. M. & Mysen, B. O. (eds) Mantle Petrology: Field Observations and High-pressure Experimentation. Geochemical Society Special Publications 6, 13–45. Herbert, L. B. & Montési, L. G. J. (2010). Generation of permeability barriers during melt extraction at mid-ocean ridges. Geochemistry, Geophysics, Geosystems 11(12), doi:10.1029/2010GC003270. Herzberg, C., Asimow, P. D., Arndt, N., Niu, Y., Lesher, C. M., Fitton, J. G., Cheadle, M. J. & Saunders, A. D. (2007). Temperature in ambient mantle and plumes: Constraints from basalts, picrites and komatiites. Geochemistry, Geophysics, Geosystems 8(2), doi:10.1029/2006GC001390. Hirschmann, M. M. & Stolper, E. M. (1996). A possible role for garnet pyroxenite in the origin of the ‘garnet signature’ in MORB. Contributions to Mineralogy and Petrology 124(2), 185–208. Hirschmann, M. M., Ghiorso, M. S., Davis, F. A., Gordon, S. M., Mukherjee, S., Grove, T. L., Krawczynski, M., Medard, E. & Till, C. B. (2008). Library of experimental phase relations (LEPR): A database and Web portal for experimental magmatic phase equilibria data. Geochemistry, Geophysics, Geosystems 9(3), doi:10.1029/2007GC001894. Holtzman, B. K., Groebner, N. J., Zimmerman, M. E. & Ginsburg, S. B. (2003). Stress-driven melt segregation in partially molten rocks. Geochemistry, Geophysics, Geosystems 4(5), doi:10.1029/2001GC000253. Johnson, K. T. M., Dick, H. J. B. & Shimizu, N. (1990). Melting in the oceanic upper mantle: an ion microprobe study of diopside in abyssal peridotites. Journal of Geophysical Research 95(B3), 2661–2678. Katz, R. F., Spiegelman, M. & Holtzman, B. (2006). The dynamics of melt and shear localization in partially molten aggregates. Nature 442, 676–679. Kelemen, P. B. (1990). Reaction between ultramafic rock and fractionating basaltic magma I. Phase relations, the origin of calc-alkaline magma series, and the formation of discordant dunite. Journal of Petrology 31(1), 51–98. Kelemen, P. B., Shimizu, N. & Salters, V. J. M. (1995). Extraction of mid-ocean ridge basalt from the upwelling mantle by focused flow of melt in dunite channels. Nature 375, 747–753. Kelemen, P. B., Hirth, G., Shimizu, N., Spiegelman, M. & Dick, H. J. (1997). A review of melt migration processes in the adiabatically upwelling mantle beneath oceanic spreading ridges. Philosophical Transactions of the Royal Society of London 355(1723), 283–318. Keshav, S., Gudfinnsson, G. H., Sen, G. & Fei, Y. (2004). High-pressure melting experiments on garnet clinopyroxenite and the alkalic to tholeiitic transition in ocean-island basalts. Earth and Planetary Science Letters 223(3–4), 365–379. Kogiso, T. & Hirschmann, M. M. (2006). Partial melting experiments of bimineralic eclogite and the role of recycled mafic oceanic crust in the genesis of ocean island basalts. Earth and Planetary Science Letters 249(3–4), 188–199. Kogiso, T., Hirschmann, M. M. & Frost, D. J. (2003). High-pressure partial melting of garnet pyroxenite: possible mafic lithologies in the source of ocean island basalts. Earth and Planetary Science Letters 216(4), 603–617. Kogiso, T., Hirschmann, M. M. & Pertermann, M. (2004a). High-pressure partial melting of mafic lithologies in the mantle. Journal of Petrology 45(12), 2407–2422. Kogiso, T., Hirschmann, M. M. & Reiners, P. W. (2004b). Length scales of mantle heterogeneities and their relationship to ocean island basalt geochemistry. Geochimica et Cosmochimica Acta 68(2), 345–360. Kohlstedt, D. L. (1991). Structure, rheology and permeability of partially molten rocks at low melt fractions. In: Phipps Morgan, J., Blackman, D. K. & Sinton, J. M. (eds) Mantle Flow and Melt Generation at Mid-ocean Ridges. American Geophysical Union, Geophysical Monographs 71, 103–121. Kornprobst, J. (1970). Les péridotites et les pyroxénolites du massif ultrabasique des Beni Bouchera: une étude expérimentale entre 1100 et 1550 °C sous 15 à 30 kilobars de pression sèche. Contributions to Mineralogy and Petrology 29, 290–309. Kumar, N., Reischig, L. & Zindler, A. (1996). A major and trace element and strontium, neodymium, and osmium isotopic study of a thick pyroxenite layer from the Beni Bousera Ultramafic Complex of northern Morocco. Geochimica et Cosmochimica Acta 60(8), 1429–1444. Kushiro, I. (1996). Partial melting of a fertile mantle peridotite at high pressures: an experimental study using aggregates of diamond. In: Basu, A. & Hart, S. (eds) Earth Processes: Reading the Isotopic Code. American Geophysical Union, Geophysical Monographs 95, 109–122. Lambart, S., Laporte, D. & Schiano, P. (2009a). An experimental study of pyroxenite partial melts at 1 and 1.5 GPa: implications for the major-element composition of mid-ocean ridge basalts. Earth and Planetary Science Letters 288, 335–347. Lambart, S., Laporte, D. & Schiano, P. (2009b). An experimental study of focused magma transport and basalt–peridotite interactions beneath mid-ocean ridges: implications for the generation of primitive MORB composition. Contributions to Mineralogy and Petrology 157, 429–451. Langmuir, C. H., Klein, E. M. & Plank, T. (1992). Petrological systematics of mid-ocean ridge basalts: constraints on melt generation beneath oceanic ridges. In: Phipps Morgan, J., Blackman, D. K. & Sinton, J. M. (eds) Mantle Flow and Melt Generation at Mid-ocean Ridges. American Geophysical Union, Geophysical Monographs 71, 183–280. Laporte, D., Toplis, M., Seyler, M. & Devidal, J.-L. (2004). A new experimental technique for extracting liquids from peridotite at very low degrees of melting: application to partial melting of depleted peridotite. Contributions to Mineralogy and Petrology 146(4), 463–484. Le Bas, M. J., Le Maitre, R. W., Streckeisen, A. & Zanettin, B. (1986). A chemical classification of volcanic rocks based on the total alkali–silica diagram. Journal of Petrology 27, 745–750. Lenoir, X., Garrido, C. J., Bodinier, J.-L., Dautria, J.-M. & Gervilla, F. (2001). The recrystallization front of the Ronda peridotite: evidence for melting and thermal erosion of subcontinental lithospheric mantle beneath the Alboran Basin. Journal of Petrology 42(1), 141–158. Le Roux, P. J., Le Roex, A. P. & Schilling, J.-G. (2002). MORB melting processes beneath the southern Mid-Atlantic Ridge (40°–55°S): a role for mantle plume-derived pyroxenite. Contributions to Mineralogy and Petrology 144, 206–229. Lundstrom, C. C., Gill, J. & Williams, Q. (2000). A geochemically consistent hypothesis for MORB generation. Chemical Geology 162(2), 105–126. Luth, W. C. & Ingemarss, C. O. (1965). Gel preparation of starting materials for hydrothermal experimentation. American Mineralogist 50, 255–258. McKenzie, D. & O’Nions, R. K. (1991). Partial melt distributions from inversion of rare earth element concentrations. Journal of Petrology 32(5), 1021–1091. Médard, E., Schmidt, M. W., Schiano, P. & Ottolini, L. (2006). Melting of amphibole-bearing wehrites: an experimental study on the origin of ultra-calcic nepheline-normative melts. Journal of Petrology 47(3), 481–504. Morgan, Z. & Liang, Y. (2003). An experimental and numerical study of the kinetics of harzburgite reactive dissolution with applications to dunite dike formation. Earth and Planetary Science Letters 214(1–2), 59–74. Morgan, Z. & Liang, Y. (2005). An experimental study of the kinetics of lherzolite reactive dissolution with applications to melt channel formation. Contributions to Mineralogy and Petrology 150(4), 369–385. O’Hara, M. J. (1968). Are ocean floor basalts primary magma? Nature 220(5168), 683–686. Pearson, D. G., Davies, G. R. & Nixon, P. H. (1993). Geochemical constraints on the petrogenesis of diamond facies pyroxenites from the Beni Bousera peridotite massif, North Morocco. Journal of Petrology 34(1), 125–172. Pertermann, M. & Hirschmann, M. M. (2003). Partial melting experiments on a MORB-like pyroxenite between 2 and 3 GPa: Constraints on the presence of pyroxenite in basalt source regions from solidus location and melting rate. Journal of Geophysical Research 108(B2), 2125, doi:10.1029/2000JB000143. Piccardo, G. B. & Vissers, R. L. M. (2007). The pre-oceanic evolution of the Erro-Tobbio peridotite (Voltri Massif, Ligurian Alps, Italy). Journal of Geodynamics 43(4–5), 417–449. Pilet, S., Baker, M. B. & Stolper, E. M. (2008). Metasomatized lithosphere and the origin of alkaline lavas. Science 320(5878), 916–919. Polvé, M. & Allegre, C. J. (1980). Orogenic lherzolite complexes studied by $^{87}$Rb/$^{87}$Sr: a clue to understand the mantle convection processes? Earth and Planetary Science Letters 51, 71–93. Putirka, K. D. (1999). Clinopyroxene + liquid equilibrium to 100 kbar and 2450 K. Contributions to Mineralogy and Petrology 135, 151–163. Putirka, K. D. (2003). Thermometers and barometers for volcanic systems. In: Putirka, K. D. & Tepley, F. J., III (eds) Minerals, Inclusions and Volcanic Processes, Mineralogical Society of America and Geochemical Society, Reviews in Mineralogy and Geochemistry 69, 61–120. Putirka, K. D., Johnson, M., Kinzler, R. J. & Walker, D. (1996). Thermobarometry of mafic igneous rocks based on clinopyroxene–liquid equilibria, 0–30 kbar. Contributions to Mineralogy and Petrology 123, 92–108. Putirka, K. D., Ryerson, F. J. & Mikaelian, H. (2003). New igneous thermobarometers for mafic and evolved lava compositions, based on clinopyroxene + liquid equilibria. American Mineralogist 88, 1542–1554. Reisberg, L. C., Allègre, C. J. & Luck, J. M. (1991). The Re–Os systematic of Ronda ultramafic complex of southern Spain. Earth and Planetary Science Letters 105, 196–213. Remaidi, M. (1993). Étude pétrologique et géochimique d’une association de péridotites réfractaires-pyroxénites dans le Massif de Ronda (Espagne). Ph.D Thesis, Université de Montpellier II, 437 p. Rhodes, J. M., Dungan, M. A., Blanchard, D. P. & Long, P. E. (1979). Magma mixing at mid-ocean ridges: evidence from basalts drilled near 22°N on the Mid-Atlantic Ridge. Tectonophysics 55, 35–61. Robinson, J. A. C., Wood, B. J. & Blundy, J. D. (1998). The beginning of melting of fertile and depleted peridotite at 1.5 GPa. Earth and Planetary Science Letters 155(1–2), 97–111. Shervais, J. W., Wilshire, H. G. & Schwarzman, E. C. (1973). Garnet clinopyroxenite xenolith from Dish Hill, California. Earth and Planetary Science Letters 19, 120–130. Sleep, N. H. (1984). Tapping of magmas from ubiquitous mantle heterogeneities: An alternative to mantle plumes. Journal of Geophysical Research 89, 10029–10041. Sleep, N. H. (1983). Tapping of melts by veins and dykes. Journal of Geophysical Research 93, 10255–10272. Smith, C. S. (1964). Some elementary principles of polycrystalline microstructure. Metallurgical Reviews 9, 1–48. Smith, P. M. & Asimow, P. D. (2005). Adiabat.lph: A new public front-end to the MELTS, pMELTS, and pHMELTS models. Geochemistry, Geophysics, Geosystems 6(1), doi:10.1029/2004GC000316. Sobolev, A. V., Hofmann, A. W., Kuzmin, D. V., Yaxley, G. M., Arndt, N. T., Chung, S.-L., Danyushevsky, L. V., Elliott, T., Frey, F. A., Garcia, M. O., Gureenko, A. A., Kamenetsky, V. S., Kerr, A. C., Krivolutskaya, N. A., Matviienko, V. V., Nikogosian, I. K., Rocholl, A., Sigurdsson, I. A., Sushchekavskaya, N. M. & Teklay, M. (2007). The amount of recycled crust in sources of mantle-derived melts. Science 316(5823), 412–417. Sparks, D. W. & Parmentier, E. M. (1991). Melt extraction from the mantle beneath spreading centers. Earth and Planetary Science Letters 105(4), 368–377. Spiegelman, M., Kelemen, P. B. & Aharonov, E. (2001). Causes and consequences of flow organization during melt transport: The reaction infiltration instability in incompatible media. Journal of Geophysical Research 106(B2), 2061–2077. Stolper, E. M. (1980). A phase diagram for mid-ocean ridge basalts: Preliminary results and implications for petrogenesis. Contributions to Mineralogy and Petrology 74(1), 13–27. Takahashi, N. (1992). Evidence for melt segregation towards fractures in the Horoman mantle peridotite complex. Nature 359(6390), 52–55. Tommasi, A., Godard, M., Coromina, G., Dautria, J.-M. & Barsczus, H. (2004). Seismic anisotropy and compositionally induced velocity anomalies in the lithosphere above mantle plumes: a petrological and microstructural study of mantle xenoliths from French Polynesia. Earth and Planetary Science Letters 227(3–4), 539–556. Toplis, M. J. (2005). The thermodynamics of iron and magnesium partitioning between olivine and liquid: criteria for assessing and predicting equilibrium in natural and experimental systems. Contributions to Mineralogy and Petrology 149(1), 22–39. Van der Wal, D. & Bodinier, J. L. (1996). Origin of the recrystallisation front in the Ronda peridotite by km-scale pervasive porous melt flow. Contributions to Mineralogy and Petrology 122, 387–405. Wark, D. A., Williams, C. A., Watson, E. B. & Price, J. D. (2003). Reassessment of pore shapes in microstructurally equilibrated rocks, with implications for permeability of the upper mantle. Journal of Geophysical Research 108(B1), 2050, doi:10.1029/2001JB001575. Wasylekien, L. E., Baker, M. B., Kent, A. J. R. & Stolper, E. M. (2003). Near-solidus melting of the shallow upper mantle: partial melting experiments on depleted peridotite. Journal of Petrology 44(7), 1163–1191. Yaxley, G. & Green, D. H. (1998). Reactions between eclogite and peridotite: mantle refertilisation by subduction of oceanic crust. Schweizerische Mineralogische und Petrographische Mitteilungen 78(2), 243–255.
THEORETICAL SOLUTIONS FOR CALCULATING LEAKAGE THROUGH COMPOSITE LINER SYSTEMS R.K. Rowe Department of Civil and Environmental Engineering The University of Western Ontario London, Ontario, Canada N6A 5B9 Tel: 519-661-2126; Fax: 519-661-3942 Email: firstname.lastname@example.org and John R. Booker* Department of Civil Engineering University of Sydney Sydney 2006, N.S.W., Australia Deceased Geotechnical Research Centre Report GEOT-18-98 November 1998 ISSN 0847-0626 ABSTRACT A new semi-analytic solution for the leakage of fluid through a circular hole in an otherwise essentially impermeable geomembrane underlain by a clay liner is presented. This solution covers the full range of layer thickness between very thin (one-dimensional flow) and infinitely thick (Forchheimer's solution). It demonstrates that, in general, the flow is greater than that predicted by the limiting cases. The solution can be used for a wide range of practical problems where the radius of the hole may range from a pinhole to a large quasi-circular wrinkle in a perforated geomembrane (i.e. where the "hole" is considered to be the area where there is no contact between the geomembrane and clay). The solution assumes perfect contact between the geomembrane away from the hole but does allow consideration of hydraulic anisotropy of the clay layer. Using the proposed theory, a very simple, analytic, approximate expression is obtained. This solution can be used in hand calculations to establish the leakage rate in many practical design situations. KEYWORDS Semi-analytic solution, leakage, geomembrane, circular hole, composite liner, design INTRODUCTION Composite liner systems involving a geomembrane (e.g. high density polyethylene) over a clay liner (e.g. compacted clay, CCL, or a geosynthetic clay liner, GCL) play an important role in the design of modern waste and other fluid impoundment facilities. The intact geomembrane provides an essentially impermeable barrier to water. Thus, leakage of water through the composite liner will be controlled by the number and size of holes in the geomembrane. These holes can often be approximated as circular and a number of limiting solutions have been obtained for a circular hole overlying a clay layer. For example, if one assumes that there is perfect contact between the geomembrane outside the hole then the flow $Q$ through a hole on a semi-infinite isotropic soil deposit is given by Forchheimer's (1930) Equation $$Q = 4r_o h_d k \quad (1)$$ where $r_o$ is the radius of the hole, $k$ is the hydraulic conductivity of the clay liner, and $h_d$ is the head loss and is generally taken to be the height of fluid above the hole. An alternative approach, adopted by Jayawickrama et al. (1988), was to consider a transmissive zone between the geomembrane and the clay. They developed a solution that considered axi-symmetric horizontal radial flow at the interface and 1-D vertical flow through the clay between the transmissive layer below the geomembrane and a transmissive layer below the clay layer. They assumed that the head in the underlying transmissive layer was at the level of the geomembrane (i.e. $h_a = D$ in Figure 1). This solution has been widely used (e.g. Giroud & Bonaparte 1989; Giroud et al. 1994, 1998) in practice. However, as useful as they are, both solutions have limitations. Equation 1 is limited to very deep layers; a situation generally not encountered. The solution developed by Jayawickrama et al. (1988) considers the case of a layer of finite depth but breaks down where the transmissivity of the interface between the geomembrane and liner is very low (e.g. when there is good contact as considered numerically by Walton and Sager (1990) and Walton et al. (1997)). The objective of this paper is to provide a relatively simple solution methodology for a number of additional cases that may be of practical significance. In particular, attention is focussed on the case of a geomembrane in good contact with the soil for $r > r_o$ (i.e. no transmissive layer between the geomembrane and the soil) and a layer of finite depth. No restriction is placed on the size of the hole that may be either small or large relative to the thickness of the clay layer. This has important potential applications for holes in a geomembrane over a GCL liner (that may be only 5-10 mm thick) as shown schematically in Figure 1a or when simulating an approximately circular perforated wrinkle above a compacted clay liner as shown schematically in Figure 1b. THEORY - GENERAL CASE* Considering the general case shown in Figure 1, we seek a solution to the mixed boundary value problem associated with the specified head at the hole \[ h = h_d = (D - h_a) + h_w \quad 0 \leq r \leq r_o \quad \text{at the surface } (z=D-h_a) \] (2) and specified zero vertical flux where there is intact geomembrane \[ v_z = k_v \frac{dh}{dz} = 0 \quad r > r_o \quad \text{at the surface } (z=D-h_a) \] (3) and $ k_v $ is the vertical hydraulic conductivity. Suppose that the velocity is not known but that the total head, $ h_o(x,y) $ is known in some region R, and assume that the velocity $ v_o(x,y) $ (at $ z=D-h_a $) on R can be approximated by an expression of the form \[ v_o(x,y) = c_o \phi_o(x,y) + ...c_n \phi_n(x,y) \] (4) where we will seek basis functions $ \phi_j $ and coefficients $ c_j $ that will provide the best approximation to the actual velocity distribution. A solution can be achieved in transform space and so introducing the Fourier transforms gives \[ h(x,y,z) = \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} e^{i(\alpha x + \beta y)} H(\alpha, \beta, z) d\alpha d\beta \] (5a) \[ H(\alpha, \beta, z) = \frac{1}{4\pi^2} \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} e^{-i(\alpha x + \beta y)} h(x, y, z) \, dx \, dy \] \[ v(x, y, z) = \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} e^{i(\alpha x + \beta y)} V(\alpha, \beta, z) \, d\alpha \, d\beta \] \[ V(\alpha, \beta, z) = \frac{1}{4\pi^2} \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} e^{-i(\alpha x + \beta y)} v(x, y, z) \, dx \, dy \] where \[ V(\alpha, \beta, z) = -k_v \frac{\partial H(\alpha, \beta, z)}{\partial z} \] For a horizontally layered deposit it is easy to establish (see Appendix A) that \[ H(\alpha, \beta, z) = L(\alpha, \beta, z)V_o(\alpha, \beta) \] where \[ V_o = \frac{1}{4\pi^2} \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} e^{-i(\alpha x + \beta y)} v_o(x, y) \, dx \, dy \] \[ = \frac{1}{4\pi^2} \int_{R} \int_{R} e^{-i(\alpha x + \beta y)} v_o(x, y) \, dx \, dy \] (where $ v_o = 0 $ outside $ R $). For steady state conditions and a layer of thickness $ D $ underlain by a permeable base it can be shown (Appendix A) that: \[ L(\alpha, \beta, z) = \frac{\sinh\left[\rho(h_a + z)\sqrt{\frac{k_h}{k_v}}\right]}{\rho\sqrt{k_h k_v} \cosh\left[\rho D \sqrt{\frac{k_h}{k_v}}\right]} \] and for $ z = D - h_a $ (i.e. at the top of the layer) \[ L_o(\alpha, \beta) = \frac{\tanh\left[\rho D \sqrt{\frac{k_h}{k_v}}\right]}{\rho\sqrt{k_h k_v}} \] and for $D \to \infty$, $$L_o(\alpha, \beta) = \frac{1}{\rho \sqrt{k_h k_v}}$$ \hspace{1cm} (10b) where $\rho = \sqrt{\alpha^2 + \beta^2}$. It follows from Eqs. 4 and 8 that $$V_o(\alpha, \beta) = \sum_{j=0}^{n} c_j \Phi_j(\alpha, \beta)$$ \hspace{1cm} (11) where $$\Phi_j = \frac{1}{4\pi^2} \int_R \int e^{-i(\alpha x + \beta y)} \phi_j dxdy$$ \hspace{1cm} (12) Now from Eqs. 5a, 7 and 11, $$h(x, y, z) = \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} L(\alpha, \beta, z)V_o(\alpha, \beta)e^{i(\alpha x + \beta y)} d\alpha d\beta$$ \hspace{1cm} (13) $$= \sum_{j=0}^{n} c_j \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} L(\alpha, \beta, z)\Phi_j(\alpha, \beta)e^{i(\alpha x + \beta y)} d\alpha d\beta$$ and at $z=D-h_a$ (top of the layer) $$h_o(x, y) = \sum_{j=0}^{n} c_j \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} L_o(\alpha, \beta)\Phi_j(\alpha, \beta)e^{i(\alpha x + \beta y)} d\alpha d\beta$$ \hspace{1cm} (14) and hence $$\int_R \int h_o \phi_\ell(x, y)dxdy = \sum_{j=0}^{n} c_j \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} [L_o(\alpha, \beta)\Phi_j(\alpha, \beta) \int_R \int e^{i(\alpha x + \beta y)} \phi_\ell(x, y)dxdy]d\alpha d\beta$$ \hspace{1cm} (15) $$= 4\pi^2 \sum_{j=0}^{n} c_j \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} \Phi_\ell^* L_o(\alpha, \beta)\Phi_j d\alpha d\beta$$ where $\Phi_\ell^$ is the complex conjugate of $\Phi_\ell$. Hence, \[ \sum A_{\ell j} c_j = b_\ell \] (16) for a given set of basis functions $ \phi_j $; this can be solved for the coefficients $ c_j $ that provide the best approximation to the velocity distribution where \[ A_{\ell j} = 4\pi^2 \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} \Phi_\ell^ L_0(\alpha, \beta) \Phi_j d\alpha d\beta \] (17) \[ b_\ell = \int_R \int h_o(x, y) \phi_\ell(x, y) dx dy \] (18) SOLUTION FOR A CIRCULAR HOLE ON A HALF-SPACE To illustrate the approach for a problem with a known solution, consider the special case of a circular hole on a half-space and suppose \[ \phi_o = \left(1 - \frac{r^2}{r_o^2}\right)^{-0.5} \] (19) and hence from Eq. 4 \[ v_o = c_o \left[1 - \frac{r^2}{r_o^2}\right]^{-1/2} \quad 0 < r < r_o \] (20a) \[ = 0 \quad r \geq r_o \] (20b) then (after integration) it follows from Eq. 12 \[ \Phi_o = \frac{1}{2\pi} \frac{r_o}{\rho} \sin \rho r_o \] (21) and hence from Eqs. 10b and 14, \[ h_o(x, y) = \int_{-\infty}^{\infty} \int_{-\infty}^{\infty} c_o \frac{e^{+i(\alpha x + \beta y)}}{\rho \sqrt{k_h k_v}} \frac{r_o \sin \rho r_o}{2\pi \rho} d\alpha d\beta \] \[ = \frac{c_o r_o}{\sqrt{k_h k_v}} \int_0^\infty J_o(\rho r) \frac{\sin \rho r_o}{\rho} d\rho \] \[ = \frac{c_o r_o}{\sqrt{k_h k_v}} \frac{\pi}{2} \quad 0 < r < r_o \tag{22a} \] \[ = \frac{c_o r_o}{\sqrt{k_h k_v}} \sin^{-1}\left(\frac{r_o}{r}\right) \quad r_o < r < \infty \tag{22b} \] Hence, since $ h_o(x, y) = h_d $ is a constant for $ 0 < r < r_o $, from Eq. 22a \[ c_o = \frac{2h_d \sqrt{k_h k_v}}{\pi r_o} \] and the flow $ Q $ is given by \[ Q = 2\pi c_o \int_0^{r_o} r (1 - \frac{r^2}{r_o^2})^{-0.5} dr = 2\pi r_o^2 c_o \] \[ = 4r_o h_d \sqrt{k_h k_v} \tag{23} \] that reduces to Forchheimer's (1930) solution for an isotropic soil with $ k = k_h = k_v $ (see Eq. 1). Thus in this case only one term is required in Eq. 4 to obtain the exact solution. SOLUTION FOR A CIRCULAR HOLE ON A LAYER OF FINITE DEPTH The functions $ \phi_0, \phi_1, \phi_j $ etc. in Eq. 4 provide the basis for mixed boundary value problems through the functions $ \Phi_0, \Phi_1, \ldots, \Phi_j $ given by Eq. 12. Consider basis functions of the form \[ \phi_o = (1 - \frac{r^2}{r_o^2})^{-1/2} \tag{24a} \] \[ \phi_{\mu} = (r_o^2 - r^2)^{\mu-1} \quad \mu = 1, 2, \ldots \] (24b) In general, one could use as many basis functions as one wishes, and the coefficients $ c_j $ will eliminate those not needed. However, it was found that for all cases examined, three basis functions $ \phi_0, \phi_1 $ and $ \phi_2 $ were sufficient. Thus, it follows from Eq. 12 that for these axi-symmetric conditions, \[ \Phi_0 = \frac{r_o}{2\pi} \frac{\sin(\rho r_o)}{\rho} \] (25a) \[ \Phi_{\mu} = \frac{2^{\mu-1}}{2\pi} \frac{r_o^{\mu} \Gamma(\mu)}{\rho^{\mu}} J_{\mu}(\rho r_o) \quad \mu = 1, 2, \ldots \] (25b) where $ \Gamma(\mu) = (\mu-1)! $ and $ \Gamma(1) = 1 $ and hence the coefficients $ A_{\ell,j} $ ($ \ell = 0, 1, 2; j = 0, 1, 2 $) can be calculated numerically from Eq. 17. The values of $ b_{\ell} $ are given by Eq. 18. \[ b_0 = 2\pi r_o^2 h_d \] (26a) \[ b_{\mu} = \frac{\pi r_o^{2\mu} h_d}{\mu} \quad \mu = 1, 2, \ldots \] (26b) for a constant head $ h_0(x,y) = h_d = D-h_a+h_w $. Once the coefficients $ A_{\ell,j} $ and $ b_{\ell} $ are evaluated, Eq. 16 can be solved for the constant $ c_j $ that match the head distribution at the hole and hence the velocity distribution can be deduced from Eq. 11. Finally, the flow can be evaluated. \[ Q = \int_R \int v_o(x, y) \, dx \, dy \] (27) and from Eqs. 4 \[ Q = \int_R \int \sum_{j=0}^{n} c_j \phi_j \, dx \, dy = \sum_{j=0}^{n} c_j \int_R \int \phi_j \, dx \, dy \] (28) Hence, from Eqs. 18 and 28 \[ Q = \sum_{j=0}^{n} \frac{c_j b_j}{h_d} \] (29) for constant head $ h_d = h_w + D - h_a $ over $ R $, where $ b_j $ is known from Eq. 26 and $ c_j $ are known by solving Eq. 16. NUMERICAL CONSIDERATIONS The solution presented above is essentially analytic, however, the number of basis functions required to converge to the exact solution is not known a priori and so the number of terms, $ n $, may be varied. However, as noted earlier, it was found that in all situations examined, only three terms were required. For deep deposits ($ r_o/D \to 0 $), one term was sufficient. The evaluation of the coefficients $ A_{ij} $ in Eq. 17 involves numerical integration. In general, this is a two-dimensional integration, however, for the case of circular symmetry considered in this paper, Eq. 17 reduces to \[ A_{ij} = 8\pi^3 \int_0^\infty \Phi_i(\rho)L_o(\rho)\Phi_j(\rho)\rho d\rho \] (30a) where $ \Phi_i, \Phi_j $ are given by Eq. 25 and $ L_o(\rho) $ is given by Eq. 10. Note also that the infinitely deep case ($ r_o/D \to 0 $) arises naturally from Eq. 10a that reduces to Eq. 10b in a numerically stable manner. The integration required in Eq. 30 can be readily performed by Gauss Quadrature and is approximated by \[ A_{ij} = 8\pi^3 \int_0^W \Phi_i(\rho)L_o(\rho)\Phi_j(\rho)\rho d\rho \] (30b) where the width $ W $ can be increased until $ A_{ij} $ converges to a unique value. Due to the fact that there is no discretization error (i.e. unlike finite element or finite difference solutions), this approach readily allows considering a wide range of combinations of layer thickness $D$ and hole size $r_o$. LIMITING CASES As previously noted, for the limiting case of a deep layer ($r_o/D \to 0$) is given by Eq. 23 $$Q = 4 r_o h_d \sqrt{k_h k_v}$$ \hspace{1cm} (23) and for a very shallow layer ($r_o/D \to \infty$) the solution becomes one-dimensional and $$Q = \pi k_v r_o^2 h_d / D = \pi \frac{r_o}{D} r_o h_d k_v$$ \hspace{1cm} (31) Between these two limits the flow increases and it may be hypothesized that the solution for intermediate values of $r_o/D$ may be given by $$Q = \left(4 + \frac{Fr_o}{D}\right) r_o h_d k_v$$ \hspace{1cm} (32) where one seeks a function $F$ that will provide a reasonable approximation for all values of $0 \leq r_o/D \leq \infty$. Examination of limiting cases and Eq. 32 would suggest that flows through a "hole" could be presented in terms of a dimensionless flow $M$: $$\frac{Q}{r_o h_d k_v} = M = \left(4 + \frac{Fr_o}{D}\right)$$ \hspace{1cm} (33) and it will be shown in the following section that to sufficient accuracy (better than 3%), $F$ can be given by $$F = 2.455 + 0.685 \tanh[0.6 \ln(r_o / D)]$$ \hspace{1cm} (34) and hence $ F \rightarrow 1.77 $ as $ r_o/D \rightarrow 0 $ and $ F \rightarrow \pi $ as $ r_o/D \rightarrow \infty $. APPLICATION TO COMPOSITE LINER INVOLVING A GCL Geosynthetic clay liners (GCLs) with powdered bentonite in the upper geotextile have been developed. Thus, provided that the geomembrane liner is in intimate contact with the GCL, there will be negligible horizontal flow at the interface between the geomembrane and GCL. Under these circumstances, the transmissivity, $ \theta $, of the interface between the geomembrane and GCL away from the hole is very low ($ \theta = 0 $) and the solution presented in the previous section may be used for two important practical cases. The first involves a "hole" in the geomembrane that is a physical perforation (e.g. a puncture) and the geomembrane is in intimate contact with the GCL at all points near the puncture (hole). The second involves a wrinkle with a perforation where the size of the "wrinkle" (i.e. zone of geomembrane that is not in intimate contact with the GCL) defines the size of the "hole". Holes and wrinkles that are approximately two-dimensional (e.g. waves) may be modelled as described by Rowe (1998). This present paper focuses on holes and wrinkles that are either circular or have dimensions such that they can be reasonably approximately as circular. In the case where there is a perforation in a wrinkle, the flow will be controlled by the minimum of either (a) the capacity for flow through the perforation (which is controlled by Bernoulli's equation; see Giroud & Bonaparte 1989a); or (b) the flow through the GCL in a zone where there is no intimate contact between geomembrane and GCL. Condition (a) can be readily checked on a hand calculator as described by Giroud and Bonaparte (1989a) or Rowe (1998) and is not considered further here. Condition (b) may be checked using the theory presented in this paper. Analyses were initially performed for a GCL with a hydrated thickness of $D=0.01$ m, $k \approx 2 \times 10^{-10}$ m/s, $h_w = 0.3$ m and $h_d = D + h_w$. These represent typical conditions of a composite liner consisting of a geomembrane over a GCL over a secondary leachate collection (leak detection) system at relatively low applied vertical stress ($\approx 30$ kPa) and allowing for the effect of clay-leachate interaction in assessing the $k$ value (see Rowe 1998). Results were obtained for a range of hole sizes, $r_o$, and the results were plotted in terms of the dimensionless flow, $M$, per hole versus $r_o/D$ and the results are plotted in Figure 2. It can be seen that the deep solution (Forchheimer 1930) is only really valid (to better than 1%) for very small holes, $r_o/D < 0.02$ or $r_o < 0.0002$ m in this specific case, and that to better than about 5% accuracy is only valid for $r_o/D \leq 0.1$ ($r_o \leq 1$ mm). This corresponds to a "small hole" (i.e. perforation in the geomembrane). A large hole (as defined by Giroud and Bonaparte 1989a,b) corresponds to $r_o/D \approx 0.56$ ($r_o = 5.64$ mm) and in this case the flow exceeds that predicted by Forchheimer's Equation (Eq. 23) by 30%. Thus, for most practical cases involving perforations and wrinkles, Forchheimer's (1930) equation is not adequate. It is also evident that the solution only approaches the one-dimensional case (Eq. 31) when $r_o/D > 20$ ($r_o > 0.2$ m in this case). This would be relevant to a large wrinkle or puncture. Thus there is a need for an intermediate solution for a wide range of practical cases. From the results presented in Figure 2 and Eq. 33, the factor $F$ can also be deduced and is plotted against $\ell n r_o/D$ as shown in Figure 3. It can be seen that there is a transition from $F \approx 1.77$ for small values of $r_o/D$ to $F = \pi$ for large values of $r_o/D$. In order to find a simple approximate equation that could be used with a hand calculator, we seek a simple analytical approximation to the variation in $F$ with $r_o/D$. The nature of the variation suggests a simple function of the form $$F = 2.455 + 0.685 \tanh[w \ell n(r_o/D)]$$ (35) where $w$ is a fitting parameter selected to minimize the root mean squared error of the fit to the results (shown as data points in Figure 3) obtained using the theory presented in the previous section. This was achieved for $w=0.6$ giving the functional form shown by the solid line in Fig. 3 and given by Eq. 34. As can be seen, this provides an excellent fit with a maximum deviation of less than 3% and typically less than 0.5%. Equation 34 has considerable potential for use in practical engineering calculations, however, it is necessary to establish how well it works for situations other than the specific case for which it was established. Firstly, consider a similar liner system but at a higher applied stress ($\approx 100 \text{ kPa}$) that results in a decreased thickness, $D = 0.007 \text{ m}$, but also a lower hydraulic conductivity, $k = 6 \times 10^{-11} \text{ m/s}$ based on Rowe (1998). All other parameters are the same as previously assumed. The results, presented in terms of the factor $F$, are also shown in Fig. 3 and it can be seen that they are the same as for the previous case. APPLICATION TO A COMPOSITE LINER INVOLVING A CCL The theory presented in this paper is applicable to a full range of layer thicknesses $D$ and head losses $h_d$. To illustrate its application to a composite liner system involving a geomembrane over compacted clay, consider a geomembrane over a $D=0.75 \text{ m}$ thick compacted clay liner (CCL) with $k=10^{-9} \text{ m/s}$ as specified for both the primary liner in single liner systems and for both the primary and secondary liner in double liner systems by MoE (1998). It is again assumed that $h_w=0.3 \text{ m}$ (a typical design value) and $h_d=D+h_w$. This latter condition may be met where there is an active secondary leachate collection system (leak detection system) below a primary liner. It may also be met for either a single composite liner system or the secondary liner in a double composite liner system when the potentiometric surface is at the bottom of the compacted clay. liner. It was found that for a given value of $r_o/D$, the dimensionless flow $M$ was identical to that obtained for the GCL systems described in the previous section (see Fig. 2). Likewise, the fit to the approximate curve for $F$ (Eq. 34) was the same (Figure 3), illustrating the applicability of this simple solution (to within 3%, which is adequate for most engineering applications). Of course, the theory and method of analysis presented in this paper can be used when more accurate results are required. EFFECT OF ANISOTROPY Based on Eqs. 23 and 31, it can be inferred that for $r_o/D \rightarrow 0$ any anisotropy in the clay liner will increase flow in proportion to $\sqrt{\frac{k_h}{k_v}}$ but that for $r_o/D \rightarrow \infty$ the effect would be negligible ($k_v$ controls). The proposed theory allows one to analyze the effect of any potential anisotropy. To illustrate the interaction between $r_o/D$ and the effect of $\sqrt{\frac{k_h}{k_v}}$, a series of analyses was initially performed for $k_h/k_v = 10$. Figure 4 shows a plot of the ratio of the dimensionless flow calculated for the anisotropic case ($M_a$) divided by that for the isotropic case ($M$). As expected, the ratio approaches $\sqrt{10} = 3.16$ for small $r_o/D$ and unity for large $r_o/D$. Based on these results, and results for other values of $\sqrt{\frac{k_h}{k_v}}$, it was found that the ratio $M_a/M$ could be approximated by the function $$\frac{M_a}{M} = 1 + f \sqrt{\frac{k_h}{k_v}}$$ (36a) where \[ f = 0.5 \left(1 - \sqrt{\frac{k_v}{k_h}}\right) \left[1 - \tanh <\gamma + \delta \ln(r_o / D)> \right] \tag{36b} \] and the parameters $ \gamma $ and $ \delta $ are fitting parameters obtained to minimize the root mean square error to the approximation to the actual calculated value of $ M_a/M $ obtained using the theory presented in this paper. It was found that over the practical range $ 1 \leq k_h/k_v \leq 100 $, an excellent fit was obtained for \[ \delta = 0.6 \tag{36c} \] \[ \gamma = -0.167 - 0.0073 \sqrt{\frac{k_h}{k_v}} \tag{36d} \] To illustrate the fit, Figure 4 shows the calculated values of $ M_a/M $ using Eq. 36 as a solid line and the actual values calculated using the rigorous theory as data points. It can be seen that Eq. 36 provides an excellent approximation and would be sufficient for all practical purposes. Equation 36 was only checked over the range $ 1 \leq \sqrt{\frac{k_h}{k_v}} \leq 100 $ but may be valid beyond the range. Checking would be required to confirm this. CONCLUSION A new semi-analytic solution to the problem of leakage through a hole in a geomembrane and the underlying clay has been presented. This solution covers the full range of layer thickness between a very thin and very thick layer (relative to hole size). The solution also allows consideration of anisotropy in the clay. By using the proposed theory, it has been shown that the leakage, $ Q $, through a circular hole of diameter $ r_o $ and an underlying isotropic clay layer with thickness $ D $ and hydraulic conductivity $ k_v $ can be approximated by a simple equation. \[ \frac{Q}{r_o h_d k_v} = M = (4 + F \frac{r_o}{D}) \] (33) where \[F = 2.455 + 0.685 \tanh <0.6(\ell n r_o / D)>\] (35) and $h_d$ is the head loss across the system. For anisotropic soils with horizontal hydraulic conductivity $k_h$, it was found that the dimensionless flow, $M_a$, through the hole and soil could be related to the dimensionless flow, $M$, through an isotropic soil with $k = k_v$ by the simple expression: \[ \frac{M_a}{M} = 1 + f \sqrt{\frac{k_h}{k_v}} \] (36a) where \[f = 0.5 \left(1 - \sqrt{\frac{k_v}{k_h}}\right) [1 - \tanh <\gamma + 0.6 \ell n(r_o / D)>] \] (36b) and \[\gamma = -0.167 - 0.0073 \sqrt{\frac{k_h}{k_v}} \] (36d) While Eqs. 33, 35 and 36 were obtained by curve fitting to the results of the more rigorous analysis, they have the advantage that they can be readily implemented in hand calculations as part of the design processes of composite landfill liners. ACKNOWLEDGEMENTS The work reported in this paper was conducted while R.K. Rowe was on sabbatical leave at the University of Sydney in November 1997. The work was substantially completed prior to the untimely death of Dr. John Booker, however, he did not have the opportunity to review the final manuscript. The work was funded by the Natural Sciences and Engineering Research Council of Canada. REFERENCES Forchheimer, P. (1930). "Hydraulik," 3rd Edition, B.G. Teubner Ed., Leipzig und Berlin, Germany, 596p. Giroud, J.P. and Bonaparte, R. (1989a). "Leakage through liners constructed with geomembranes - Part I. Geomembrane liners," Geotextiles and Geomembranes, Vol. 8, No. 1, pp. 27-67. Giroud, J.P. and Bonaparte, R. (1989b). "Leakage through liners constructed with geomembranes - Part II. Composite liners," Geotextiles and Geomembranes, Vol. 8, No. 2, pp. 71-111. Giroud, J.P., Soderman, K.L., Khire, M.V. and Badu-Tweneboah, K. (1998). "New developments in landfill liner leakage evaluation," Proceedings of 6th International Conference on Geosynthetics, Atlanta, Industrial Fabrics Association International. MoE (1998). Ontario Ministry of Environment Landfill Standards Guidelines on the Regulatory and Approval Requirements for New Landfilling Sites (see also Ontario Regulation 282/98). Rowe, R.K. (1998b). "Geosynthetics and the minimization of contaminant migration through barrier systems beneath solid waste," Keynote Paper, Proceedings of the 6th International Conference on Geosynthetics, Atlanta, March, Vol. 1, pp. 27-103. Walton, J.C. and Sagar, B. (1990). "Aspects of fluid flow through small flaws in membrane liners," Environmental Science and Technology, 24, pp. 920-924. Walton, J.C., Rahman, M., Casey, D., Picornell, M. and Johnson, F. (1997). "Leakage through flows in geomembrane liners," Journal of Geotechnical and Geo-environmental Engineering, ASCE, 123(6), pp. 534-539. APPENDIX The general solution for transient conditions must satisfy $$k_h \left( \frac{\partial^2 h}{\partial x^2} + \frac{\partial^2 h}{\partial y^2} \right) + k_v \frac{\partial^2 h}{\partial z^2} = m_v \frac{\partial h}{\partial t} \quad (A1)$$ and taking the Laplace and Fourier Transforms, this reduces to $$\frac{\partial^2 \bar{H}}{\partial z^2} = \mu^2 \bar{H} \quad (A2)$$ where $\mu^2 = \frac{sm_v + \rho^2 k_h}{k_v}$ and $\rho^2 = \alpha^2 + \beta^2$ Equation A1 has a solution $$\bar{H} = A \cosh \mu(h_a + z) + B \sinh \mu(h_a + z) \quad (A3)$$ and considering a layer of depth $D$ with $\bar{H} = 0 @ z = -h_a$ gives $$\bar{H} = \frac{\bar{V}_o \sinh \mu(h_a + z)}{k_v \mu \cosh \mu D} = \bar{V}_o \bar{L} \quad (A4)$$ and at the top of the layer, $z = D-h_a$, $$\bar{H}_o = \bar{V}_o \frac{\sinh \mu D}{k_v \mu \cosh \mu D} = \bar{V}_o \frac{\tanh \mu D}{k_v \mu} = \bar{L}_o \bar{V}_o \quad (A5)$$ that, for steady-state conditions, reduces to \[ L_o = \frac{\tanh\left[\rho D \sqrt{\frac{k_h}{k_v}}\right]}{\rho \sqrt{k_h k_v}} \] and for a very deep layer this reduces to \[ L_o = \frac{1}{\rho \sqrt{k_h k_v}} \] Figure 1 Problem definition showing (a) hole of radius $r_0$ in geomembrane in intimate contact with clay liner and (b) wrinkle with perforation - effective "hole" radius $r_0$ is for zone where geomembrane is not in intimate contact with the clay liner (GCL or CCL). Note: $h_w =$ fluid (e.g. leachate) head on geomembrane $D =$ thickness of clay liner $k_v =$ vertical hydraulic conductivity of clay liner $k_h =$ horizontal hydraulic conductivity of clay liner $h_a =$ potentiometric level in permeable layer relative to bottom of clay liner $h_d = D + h_a + h_w =$ head drop across composite liner Figure 2 Variation in dimensionless flow $M = Q/(r_0 h k_v)$ with ratio of hole size, $r_0$, to clay liner thickness, $D$. Figure 3 Variation in thickness factor $F$ with $\ln$ of ratio of hole radius $r_0$ and layer thickness $D$. Data points were calculated from rigorous analysis. Solid line is a plot of Eq. 34. Figure 4 Ratio of dimensionless flow for an anisotropic soil ($M_a$) to that for an isotropic soil (M) against $\ln$ of ratio of hole radius $r_0$ to liner thickness D for $k_H/k_V=10$. Data points are results from rigorous analysis. Solid line is a plot of Eq. 36.
REGULAR URBAN RENEWAL AGENCY MEETING Monday, March 12, 2012 at 12:00 p.m. Ketchum City Hall, Ketchum, Idaho Present: URA Chairman Mark Eshman URA Vice Chair Trish Wilson Commissioner Charles Friedman Commissioner Randy Hall Commissioner Nina Jonas Commissioner Curtis Kemp Commissioner Trina Peters Also Present: Gary Marks, Executive Director Lisa Horowitz, Community and Economic Development Director Attorney Stephanie Bonney Lisa Enourato, Associate Planner/Assistant to City Administrator 1. CALL TO ORDER The meeting was called to order by Chairman Eshman at 12:01 p.m. 2. COMMUNICATIONS FROM THE BOARD OF COMMISSIONERS None 3. COMMUNICATIONS FROM THE PUBLIC Phyllis Shafran noted a correction on the minutes of December 5, 2011. She also discussed the costs involved for the URA using city staff and sharing the city attorney and the possibility of the URA having their own staff and attorney. She also asked for more detailed and honest reports from the URA. Chairman Eshman responded that the URA budget would have to be increased with a change in the administration of the agency; however, he suggested that there be a separate agreement with current counsel. Gary Marks described the Redevelopment Association of Idaho report as being one that was not a report about “Ketchum’s URA” but was developed to provide RAI information to outside groups. 4. PUBLIC HEARINGS a) 2011 Ketchum Urban Renewal Annual Report Gary Marks stated that the agency is required to file an annual report before March 31. This has been prepared and has identified the major undertakings of the agency for FY2011; the subject of the annual report. The statute requires a public hearing to take comment on the report. Gary Marks agreed to update the tenant improvement amounts in the report as suggested by Commissioner Jonas. She also stated that in the City of Ketchum, what the URA is doing within the district doesn’t benefit the population we serve. The Board disagreed and pointed out that the benefit for any URA is to stimulate economic activity and bring in more jobs, which benefits the greater population. Commissioner Jonas had the following questions on the budget: - Remodel improvements of $34,000 as fixed asset, costs was $56,000 This was determined by Dennis Brown, CPA to conform in how to report capital improvements. - Provide a subtotal line as it’s difficult to calculate the human resources line from salaries through state unemployment insurance The Board agreed - What is administrative expense as compared to salaries Gary Marks stated that the administrative expense is the agency’s share of the light bill and other operational costs of maintaining a building as a percentage of the city budget, which is an accepted practice throughout the nation in dividing expenses within a municipal environment. Commissioner Jonas urged the Board to review the costs and try to trim them by having their own staff or facilities. Gary Marks explained that professional services include legal fees, consultant fees, auditor, etc. Sandy Cady, URA Secretary/Treasurer, will provide a detailed breakdown of the charges included in professional services. Commissioner Hall pointed out that the reason why staff and attorney are shared with the city is for efficiencies and maximizing the revenues generated from the district. He suggested that the Board put out an RFP for the services to compare with the efficiencies that are gained and the cost savings as a result of partnering with the City of Ketchum. Commissioner Hall believes the numbers and the relationship are fair. The Board requested the following changes to the annual report: - Add a subtotal line - Change narrative for “tenant improvements that were three times” - Change the word “represent” with “benefits” The Board also requested the following, which is not part of the report: - Clarification for the fixed asset on 491 Sun Valley Road - Provide breakdown of professional services Commissioner Peters approved the annual report as amended, seconded by Commissioner Friedman, unanimously approved. 5. COMMUNICATIONS FROM COMMUNITY & ECONOMIC DEVELOPMENT DIRECTOR a) Recommendation to negotiate an agreement with Allstar Property Management for property management services at 491 Sun Valley Road Lisa Horowitz spelled out that the tenants are responsible for the janitorial work and the landlord/owner is responsible for the maintenance of the building. The joint recommendation of CED and the Parks Department was that the Board consider having a professional property manager to handle the maintenance. Allstar Property Management was chosen from three bids. Lisa stated that calls have been coming in weekly and there are certain levels of service that the URA is not doing because they are not familiar with those types of services. Jeff Engelhardt described the janitorial and maintenance services of the Allstar Properties contract. The billing fee is $581 per month. He noted that there are some areas in the lease that should be ironed out so that the landlord and the tenant know who is responsible for what so that Allstar is doing what they are supposed to be doing. Commissioner Hall reviewed the proposal to determine what types of expenses there would be over and above the monthly fee and confirmed that the URA will be receiving a payment in lieu of taxes (PILT) from the tenant. He would like to resolve the confusion in the lease. Commissioner Jonas asked for clarification as to who was taking care of the common area. Lisa Horowitz stated that if something breaks, the URA pays for it, but the tenant pays for product such as paper towels. The common area and services that are provided to that area were discussed. Other property management companies that did not respond were Matt Engel, High Country Management and Sun Country Management. Allstar Properties will provide references as requested. Stephanie Bonney stated that the tenant is responsible for janitorial and there is no confusion on that but that when something crosses from janitorial to repair, a lease can’t cover every situation. There will be some give and take on discussions like these. The expertise of a property manager can determine what is typically janitorial and what is repair. Parking and dumpster issues in the alley were discussed along with the tables and chairs that have been stored by Starbucks at Town Square. Commissioner Hall noted that the storage issue is part of the property management. He asked for a map of the inside of the property showing the lease space and the common area noting what the URA and the lessee responsibilities are. Commissioner Jonas requested removing “decorating” from section 10. She also discussed her concern over the monies set aside for property improvements that will now become maintenance and how that is reflected in the budget. There were also questions as to how bills will be received from the property manager. Commissioner Hall described the relationship that the URA has with City employees and that the URA is getting a deal by having staff to keep it going. He suggested that in terms of saving money, to have Gary Marks provide a report so that it is known what areas can be cut. Commissioner Hall motioned to make an agreement with Allstar Property Management for property management services at 491, seconded by Commissioner Kemp, unanimously approved. b) Recommendation to prepare quarterly Payment in Lieu of Taxes (PILT) billing statements for Cairde Group related to the property located at 491 Sun Valley Road Lisa Horowitz described the report as showing the procedure to collect the payment in lieu of taxes. Allstar Properties previously pointed out that the figure in the lease was not applied to the proper percentage of the building leased. It was also suggested by Allstar that the PILT be billed monthly, not quarterly. Ms. Horowitz suggested waiting until the third broker presented their estimate of the building value. Commissioner Kemp motioned to direct staff to prepare monthly PILT billing statements for Cairde Group pursuant to the lease agreement, seconded by Commissioner Friedman, unanimously approved. 6. COMMUNICATIONS FROM EXECUTIVE DIRECTOR a) Update on urban renewal legislation Gary Marks reviewed the items pending legislation that address urban renewal, which was produced by the Redevelopment Association of Idaho. Legislation pending before the Senate Local Government and Taxation Committee: - House Bill 506 – to abolish the authority of eminent domain for urban renewal agencies; a power that is limited already in statute. A result of this bill would be to lose the ability to issue tax exempt bonds. - House Bill 507aa – this would limit urban renewal plans to specific projects. This doesn’t describe what is meant by “specific project”. - House Bill 519 – provides an exemption for site improvements where there would not be any real property tax levied on those improvements until other improvements, such as the construction of a facility or the developer conveys the property to another property. - House Bill 562aa – removes community college property tax levies on the increment from flowing to an agency. Legislation pending before the House Local Government Committee: - House Bill 580 – requires 2/3 voter approval at a countywide election for any agency indebtedness. - House Bill 597 – establishes provisions for an elected urban renewal agency board. Gary Marks noted that the school district levy is not active now. Commissioner Hall asked what the RAI is doing to continue to educate the legislators on the value of URAs. Mr. Marks described the report that RAI created of a composite of major projects of agencies throughout the state and the Code of Ethics; a best practice code for all agencies. This was meant to address events that occurred in the past with urban renewal agencies around the state. An objective is to pull the agencies into best practice through the Code of Ethics. The Board requested the Code of Ethics and the report be placed on the KURA website. b) Update on FY2011-12 budget. KURA is trying to find a $50,000 match to a grant secured by Mountain Rides for the Transit Center. Is there a way that the City and KURA can come up with the $50,000. Mr. Marks presented and outlined 3 analyses: 1. Resources Analysis 2. Historical Cash Flow Chart 3. Budget vs Year End Projective Mr. Marks then moved on to a requirements analysis: 1. Operations and Capital Expenses 2. Bond Reserve Transfer a. Debt payment on 2010 Bond b. Required reserve 3. Program Ending Fund Balance 4. Budget vs Year End Projection Mr. Marks brought these together to determine whether or not there was a surplus. The actuals to date are 42%. Mr. Marks provided an explanation that the increase in the tax rolls came from properties that were in existence, but were not previously occupied and therefore, were not on the tax rolls until this was brought to the attention of the Blaine County Assessor. Mr. Marks noted that at this point in the budget the capital project is done, the transfer on the debt is done and they have set aside the ending fund balance. He also stated that the budget will be spent. His suggestion on the need for the $50,000 was for an agreement to be reached between the URA and the City (to be reviewed in August) and that between the two, the $50,000 will be made available to the Transit Center. Jon Duval stated that Walkability will be making a presentation for those that haven’t heard it yet and there might be an estimated ask, but not one set in stone. They would like to have some of the improvements in place by July 4; however the scope has not yet been determined. Mr. Marks advised not to put money into that project until after July. Commissioner Hall suggested splitting the contribution between two fiscal years. Commissioner Wilson noted that she would not like to give specific commitments until it is in the reserve. Also, the transit center, while it is in the City of Ketchum, it is a broad benefit to the entire county as it serves the whole community. Another concern is that as an organization, items have not been prioritized that the URA has set forward to do as a project over the ten to twenty years. She believes the revenue stream is not sound. Commissioner Peters agreed as to project priorities and recommended a process by which they are evaluated. Chairman Eshman recommended the URA host a visioning session with the community in order to get a better idea as to which projects money should be spent on in the eyes of the taxpayers. The smaller projects that come up in the short term can be responded to accordingly. 7. CONSENT CALENDAR a) Approval of minutes from the February 21, 2012 URA Board Meeting Stephanie Bonney responded to Chairman Eshman’s comment relating to his signature on the minutes. She stated that he is signing the minutes with the current title, not what his title was during that meeting. b) Recommendation to approve current bills Commissioners Jonas and Hall suggested taking the administrative expenses out and breaking them down and itemizing them with regard to reimbursement to the City fund. The Board also noted other miscellaneous expenses they were unaware of. Commissioner Kemp moved to approve the consent agenda minutes and the bills, which in the future will provide a detailed description of reimbursement to the City General Fund, seconded by Commissioner Jonas, unanimously approved. Commissioner Friedman commented on a letter from Bob Crosby to the Ketchum URA and the Mayor and City Council and asked that it receive a written response. It suggests we use the Washington and First property for a parking lot. Commissioner Hall noted that after his conversation with Bob, if there were private citizens or anyone interested in renting the property, they should put together a proposal and present it to the URA. The URA originally purchased the property for specific purposes and suggested that the Sawtooth Board of Realtors make their request to the Board. In addition, the KCDC has a lease option until September 2012 and will be reapplying to IHFA. The Board would like to address a formal response to the letter. 8. EXECUTIVE SESSION to discuss personnel, litigation and land acquisition pursuant to Idaho Code 67-2345 1(a), (b), (c) and (f) No Executive Session 9. Adjourn Commissioner Wilson motioned to adjourn the meeting, seconded by Commissioner Hall. The motion was unanimously approved. The meeting was adjourned at 2:50 pm. Mark Eshman Chairman ATTEST: Sandra E. Cady, CMC Secretary/Treasurer
The usage of the terms "diagnosis" and "prescription" in the fields of medicine and electronic troubleshooting is reviewed, and a common structure for diagnosis-prescription is proposed. The diagnosis-prescription decision sequence is outlined. Prescription-without-diagnosis and diagnosis-without-prescription in education is discussed. The implications of the diagnosis-prescription structure for instructional management and instructional program development are explored. A bibliography is appended. (SK) TITLE: DIAGNOSIS AND PRESCRIPTION IN THE CONTEXT OF INSTRUCTIONAL MANAGEMENT AUTHOR: Ronald Rea ABSTRACT The usage of the terms "diagnosis" and "prescription" in the fields of medicine, education, and troubleshooting is reviewed and a common structure for the "diagnosis-prescription" problem is suggested. Implications for the fields of instructional management and instructional program development are discussed. U.S. DEPARTMENT OF HEALTH EDUCATION AND WELFARE NATIONAL INSTITUTE OF EDUCATION The terms diagnosis and prescription appear frequently in descriptions of instructional strategies and in instructional management models. Diagnosis has been defined as "a description of student performance characteristics relevant to the instruction at hand" (Glaser and Nitko, 1971). Diagnosis is described as a process "...to gather information that may be helpful in making treatment decisions" (Della-Fama, 1968). Silver (1970) uses the term diagnosis in the context of interpreting individual performance on achievement tests and prescribing remedial assignments. Glaser and Nitko (1971) offer a second definition of diagnosis as a decision concerning teaching technique appropriate for an individual. Kriewall (1969) includes a broader class of decisions under the rubric "diagnosis and prescription," that being "to categorize learners into groups on the basis of their common requirement for instructional treatment." Bloom (Bloom, Hastings, and Madaus, 1971) accounts for some of these variations by distinguishing between two purposes of diagnosis, "...to place the student properly at the outset of instruction" and, "...to discover the underlying causes of deficiencies in student learning as instruction unfolds." These references are but typical of the usage of the terms "diagnosis and prescription" in educational literature and, of necessity, have been quoted out of context to clarify the usage of these terms, several questions can be posed. Are the concepts diagnosis and prescription inherently interdependent? Is diagnosis best considered to be a process, a decision, or both? How is diagnosis related to placement? Background pertinent to these questions was sought in the usage of these terms in the fields of medicine and electronic troubleshooting. Educational usage typically has been influenced by medical rhetoric since the teacher-student relationship is often viewed as akin to the doctor-patient relationship. Engineering referents also appear worth consideration, particularly in an educational R&D context. Diagnostic decision models have been developed and tested in both medicine (Linsberg and Greenland, 1968; Betaque and Gorry, 1971; and Croft, 1971) and electronic troubleshooting (Fremer and Teplitzky, 1969). Comparison of the meanings of the concepts in these fields to educational usage should facilitate the development of diagnostic procedures and algorithms in education. Problems found in implementing diagnostic decision models in medicine (Croft, 1971) may be particularly relevant to current attempts to assist teachers with computer-generated instructional prescriptions. The Medical Diagnosis-Treatment Problem Assume that a patient comes to a physician exhibiting or able to describe some set of symptoms. The set of feasible actions that might be taken immediately consists of diagnostic tests, treatments, or simply doing nothing. The purpose of a diagnostic test is to verify and quantify the described symptoms and to obtain further information concerning possible determinants of the symptoms in order to reduce uncertainty in the predicted effectiveness of each potential treatment. Assume that a diagnostic test cannot effect a cure, but may involve risk. There diagnostic tests of my kind. This is done under several circumstances: (1) The observed symptoms or syndrome are an immediate threat to the patient's eventual recovery, (2) a particular low risk treatment usually alleviates the symptoms and whatever underlying causes exist and little risk is incurred in delaying diagnosis if, in fact, the patient's condition is not typical, and (3) the risk associated with potential diagnostic tests exceeds the risk of an incorrect treatment selection. Prescription-without-diagnosis certainly occurs in medicine—and may be quite appropriate. The opposite case, diagnosis-without-prescription, is also common. The autopsy is a clear instance of diagnosis-without-prescription or treatment. Let us pursue this no further at this point, later the question will be raised whether there is an educational analogy to the autopsy. The Electronic Troubleshooting Problem The electronic troubleshooting problem is representative of diagnostic-decision problems dealing with electro-mechanical systems (televisions, automobiles, airplanes, etc.) that begin with some knowledge that the performance is malfunctioning. The diagnostician has initial evidence that a non-normal state exists but lacks information about specific determinants. Cremer and Replitzky (1964) describe diagnosis in electronics as a multi-stage process: Stage I: Test-selection strategy. Stage II: Identification of symptom or noise from a received piece of datum. Since the recognition of malfunction from symptom or noise is not always independent, the decision process may cycle through the three stages several times if the diagnostician should regard additional diagnostics as necessary to reduce uncertainty in predicting malfunctions. As with medical decisions, the course of action taken may be prescription and treatment without diagnosis. It may be more economical to replace components that have relatively short lifetimes (vacuum tubes, spark plugs, etc.) than to change in a diagnostic process. If such treatment-without-diagnosis fails to eliminate the symptoms, diagnosis may then be required. Diagnosis-without-prescription is a less common but important occurrence. The detailed investigation into the cause of an airplane crash is a familiar example. No repair of the airplane is feasible or intended, but a diagnosis is conducted to learn why the crash occurred in the chance that a future accident may be prevented. The area of process control (Friedell, 1973) is a less publicized case of diagnosis-without-treatment. Investigating the determinants of failure in an irreparable system may result in improving the manufacturing process. A Common Structure for Diagnosis-Prescription The following structure of the diagnosis-prescription problem, which emphasizes the similarities of decision processes in the three fields, is suggested. The problem comes to the attention of a decision maker by the evolution of a syndrome—a readily apparent set of symptoms. A symptom is defined as a measurable deviation of a system's behavior from what is considered to be normal (Ackoff, 1970). A treatment is any process to which the system may be assigned. A prescription is a decision to assign a system to a treatment. It is assumed that each of the possible causes for the syndrome has identifying characteristics--a set of symptoms. Each possible cause (disease, malfunction, or disability) is treated as a state; the system is assumed to be in one of the states. The syndrome is an indication that the actual state is an undesirable one; syndrome evidence may delimit the set of possible states, but typically several states remain as probable causes. Diagnosis is a cyclic, three-stage process of (1) test selection, (2) symptom detection, and (3) probable state evaluation. Tests are selected that are valid measures of symptoms which discriminate among the states currently considered as possible causes. Symptom detection is a decision that a symptom is or is not present; it may in some instances be expressed as a probability or a confidence measure. Probable state evaluation is a listing of states currently considered probable and an estimation of the relative likelihood of these states. A prescriptive procedure or algorithm uses as input the current diagnostic information and considers the cost, risk (if any), and predicted effectiveness of each potential treatment. In addition to the set of potential treatments, one alternative is to engage in another cycle of the diagnostic process and a second is to do nothing. The entire process may be terminated if the selected treatments do not cause the system to assume a desirable state. Diagnosis-Prescription Decision Sequence Diagnostic Cycle Evidence of a Syndrome Performance Evaluation Relative likelihood of probable system states (a) Predicted effectiveness of each treatment, conditional on system state. (b) Cost and risk of treatments. (c) Resource constraints. Assignments to Treatments No Treatment Treatments Prescriptive Algorithms or Procedures Re-enter diagnostic cycle Symptom Detection Test Reliability Probable State Evaluation Test Selection (a) Test Validity (b) Test Cost (c) Test Risk Involvement in diagnosis Yes No The usage of "the diagnostic" as a verb is a special case of this general, more descriptive doctor problem. It is permissible to speak of diagnosis as a decision model: the diagnostic cycle terminates with the identification of a single probable cause, or (2) if the prescriptive procedure considers only the most likely system state. Examples can readily be envisioned in which a prescription which considers only the most likely cause produces disastrous results. A doctor will frequently prescribe a "safe" treatment to avoid the risk of a relatively improbable disease. Educational Prescription Without Diagnosis What would be the consequences of adopting the preceding definitions of diagnosis and prescription for educational usage? It is apparent that one of the current educational applications of the diagnostic metaphor—placing the student at the outset of instruction—does not fit the definitions. There is no parallel in electronic troubleshooting to this placement problem. For the doctor-patient relationship, the area of preventive medicine seems most comparable. There seems to be a fundamental difference between educational decisions and prescriptions in medicine or troubleshooting. Educational decisions are made in an arrangement context, i.e., in the context of making effective use of school resources to achieve instructional goals. Prescriptions made by a doctor or a technician are usually made in an interventionist context. Decisions based on diagnosis tend to be selection decisions—the assignment of an individual to one of a set of feasible treatment...placement, persons are divided among two or more groups which are assigned to different treatments (Cronbach and Gleser, 1965). The problem of placing the student at the outset of instruction is a general assignment problem, only in very special cases can it be treated as either pure selection or placement. It may eventually be feasible to make many different types of assignment decisions simultaneously (Besel, 1971), but at present, placement and selection decisions must be made sequentially. The process of gathering information to be used in making placement decisions can be given the name, student characterization. Student characterization may involve the measurement of a wide range of psychological variables: aptitudes, learning style, traits such as intelligence, achievement indicies. These form a description of individual student strengths and individual differences rather than a summary of deficiencies and disabilities. Diagnosis is concerned with the pathological--with an answer to the question, "Why has an instructional treatment failed?" The characterization process seems sufficiently different from the diagnostic process to merit separate conceptualization and terminology. A second case of prescription-without-diagnosis is the assignment of students to remedial instruction based solely on lower-than-desired scores on a performance measure. This is a result of treating the performance measure as a measure of a syndrome rather than as part of the diagnostic process. The performance measure thus may result in a proficiency (desired state) or non-proficiency (undesirable state) decision. At this point, either prescription of remedial instruction or engaging in a diagnostic process to investigate underlying causes for the syndrome are feasible alternatives. The concept of a "presymptom" (Ackoff, 1970) leads to a third instance of prescription-without-diagnosis. A presymptom is a predictor of a future symptom. If a student, during an instructional activity, gives some verbal or non-verbal indication that he has failed to grasp a concept, it may be a valid predictor of non-mastery performance on a future performance measure. At that instant, further investigation of symptoms may be neither possible nor needed. The prescription of remedial activities, review, or repetition of instruction may be feasible without intervening diagnosis. Systematic identification of valid presymptoms is a potentially fruitful area for educational research. Educational Diagnosis-Without-Prescription Kubie (1967) has asserted that "...any forward movement in education waits for education to develop its own corps of pathologists with courage to study its failures." He credits accountability at the autopsy table as a motivating force—resulting in rapid progress in identifying diseases and discovering cure and preventatives. It seems simplistic to assume that schools can institutionalize an educational analog of the autopsy. A school rarely knows the extent of its failures; to systematically assess the successes and failures of a school would impose a monumental task. Even if a criterion for "success" could be developed, identifying causes for failure within the complex and uncontrollable environment of a school would require an unreasonable expenditure of effort. A more likely event is the realization that diagnosis-without-prescription is present in instructional development (Schutz, 1970). When an instructional program is in an early developmental stage, it is unlikely that the required key skills are accurately known and the measures needed to indicate students for placement have not been validated or even developed. It should be expected that some students using the program will fail to achieve the desired performance standards, i.e., exhibit syndrome. But it is most unlikely that diagnostic tests and procedures which will identify the cause of learning disability will already exist or be developed in time to help the students first exhibiting the syndrome. It would, furthermore, be entirely impractical to develop remedial instructional treatments in anticipation of learning problems which rarely occur. Diagnosis must, of necessity, be done piecemeal--with whatever data are available--with no expectation of making prescriptions for the particular students involved. Within the relatively narrow context of an instructional program, detailed investigation of individual student failures is practical and profitable. The development of decision procedures for an instructional program may be viewed as an evolutionary process. Early in development, there is likely to be predominantly diagnosis-without-prescription and prescription-without-diagnosis (prescribing remediation directly from evidence of non-mastery). As diagnostic procedures are developed and alternative instructional treatments are validated, the occurrence of diagnosis-without-prescription will diminish accompanied with an increase in diagnosis-prescription. A placement procedure--based on knowledge of required entry skill and recovery of differential treatment/learner-characteristics relation improve—are developed, the occurrence of prescriptions based on diagnosis may be reduced in frequency. However, the fact that the required entry skills are known and can be measured does not imply that they must always be assessed prior to instruction. If a deficiency in a particular entry skill is infrequent for the student population, it may prove to be cost-effective to begin instruction, and with frequent performance evaluation, diagnose those students exhibiting the syndrome characteristic of deficiency in the entry skill. It is either naive or egotistical to assume that developmental stages can be bypassed permitting initial implementation of valid placement procedures. Reliance on placement in instructional management assumes that instruction can function effectively as an "open-loop" system. Such an approach requires the capability of assigning pupils to instructional activities with guaranteed success. Valid placement procedures require predictive capability; instability or change in student population, instructional materials or procedures tend to reduce this capability. Diagnosis-prescription procedures are characteristic of "closed-loop" instructional systems. Proper implementation of diagnosis-prescription requires valid corrective or remedial instructional activities. Effective placement procedures can reduce the amount of student time spent in corrective activities; retaining diagnosis-prescription procedures can soften the consequences of placement errors and permit variations or revisions in instructional materials and methods. It is probably possible to develop an effective instructional system which has strong placement and weak diagnosis-prescription capabilities or vice-versa, but if efficient use of student time is considered, a combination of placement and diagnosis-prescription promises to be cost-effective. Diagnosis and Prescription for Instructional Groups Up to this point it has been implied that diagnosis-prescription refers to decisions made about individuals. The defined structure for a diagnostic problem would seem to apply to an instructional group, i.e., the group is the system for which prescriptions are made rather than the individual. While there is no apparent medical analogy, there are instructional goals for which assessment at the individual student level is either impractical or not meaningful. Group process goals, attitudes, and interests are cited as examples. It may also be impractical or counterproductive in terms of socialization goals to continually regroup students or rely on individualized modes of instruction. Group prescription of review, drill and practice with specified content, increased redundancy or altered pace of instruction could all be based on group diagnostic information. In fact, this group diagnostic information may be more reliable than comparable information concerning individual learning disabilities. Educational Diagnosis-Prescription in Perspective While it could well be that conceptualizations of instructional systems can avoid the diagnosis-prescription paradigm, this approach to Instructional management does appear to have cost-effective promise. Emphasis on diagnosis, with or without prescription, during program development should accelerate progress as compared to test and discard procedures. Diagnostic tests in education are relatively cheap both to construct and to administer and rarely raise the side-effect hazards of medical diagnosis. The management context of instruction which complicates the assignment-to-treatment problem has a bright side: measurement already needed for performance evaluation (quality control) and program evaluation (process control) may, if designed properly, yield diagnostic information--if so analyzed. Treatments in the troubleshooting context can generally be classified as either replacement or repair of a malfunctioning component. Preventive maintenance is the special case of replacement in the absence of syndrome evidence. In medicine, most surgical treatments are comparable to component repair; usages of artificial limbs and organ transplants are akin to replacement of components. Medical treatments which have no troubleshooting analog include the injection of antibiotics, special diets and group psychiatric therapy. There would seem to be educational analogs, within the scope of instructional activities, to most of these troubleshooting and medical treatments. The unrealized potential exists that instructional treatments could be generated--perhaps with computer assistance--rather than selected. An instructional program with a validated diagnosis-prescriptive capability would appear most able to capitalize on this potential. Students could be assigned to initial instruction geared to a rapid pace and low redundancy. For those students requiring additional instruction or practice, diagnostic information could form the basis for computer generated "second instruction." Various combinations of diagnosis-prescription, prescription-without-diagnosis and diagnosis-without-prescription provide viable alternatives to the much publicized: pretest $\rightarrow$ placement $\rightarrow$ individualized instruction $\rightarrow$ posttest paradigm. The choice is not diagnosis-prescription or pretest-placement. The challenge is to design instructional systems which exploit both paradigms. Bibliography Ackoff, R. L. "Steering and Branching Through Corporate Planning," in Proceedings of the Fifth International Conference on Operational Research, Ed. by J. Lawrence, Tavistock Pub., 1970. Besel, R. R. "A Linear Model for the Allocation of Instructional Resources," paper presented at the 12th American Meeting of the Institute of Management Sciences, Detroit, October, 1971. Betaque, N. E. and Gorry, G. A. "Automating Judgmental Decision Making for a Serious Medical Problem," Management Science, 17 (April, 1971), pp. 421-434. Bloom, B. S., Hastings, J. T., and Madaus, G. F. Handbook on Formative and Summative Evaluation of Student Learning, McGraw Hill, 1971. Cremer, R. H. and Teplitzky, F. "An Investigation of a Multi-Stage Characterization of Electronics Troubleshooting," Report 70-3, Department of Industrial and Systems Engineering, University of Southern California, 1969. Croft, D. J. "Mathematical Medical Diagnosis: A Status Report." paper presented at the 39th National Meeting of the Operations Research Society of America, May, 1971. Cronbach, L. J. and Gleser, G. C. Psychological Tests and Personnel Decisions, 2nd Ed., University of Chicago Press, 1965. Della-Piana, G. M. Reading Diagnosis and Prescription, Holt, Rinehart, and Winston, 1968. Ginsberg, A. S. and Offensend, F. L. "An Application of Decision Theory to a Medical Diagnosis-Treatment Problem," RAND Corporation, Santa Monica, California, 1968. Glaser, R. and Nitko, A. J. "Measurement in Learning and Instruction," Educational Measurement, 2nd Edition, ed. by R. L. Thorndike, American Council on Education, 1971. Kriewall, T. E. "Applications of Information Theory and Acceptance Sampling Principles to the Management of Mathematics Instruction," Technical Report No. 103, Wisconsin Research and Development Center for Cognitive Learning, 1969. Kubie, L. S. "The Utilization of Preconscious Functions in Education," Behavioral Science Frontiers in Education, ed. by E. M. Bower and W. G. Hollister, Wiley, 1967. Schutz, R. E. "Programmatic Instructional Development," Professional Paper 11, Southwest Regional Laboratory, 1970. Silver, G. A. "The Computer as a Diagnostic Aid in a City College," Journal of Educational Data Processing, 8 (1971) pp. 1-7.
Linkage analysis with chromosome 15q11–13 markers shows genomic imprinting in familial Angelman syndrome E J Meijers-Heijboer, L A Sandkuijl, H G Brunner, H J M Smeets, A J M Hoogeboom, W H Deelen, J O van Hemel, M R Nelen, D F C M Smeets, M F Niermeijer, D J J Halley Abstract Angelman syndrome (AS) and Prader-Willi syndrome (PWS) have become the classical examples of genomic imprinting in man, as completely different phenotypes are generated by the absence of maternal (AS) or paternal (PWS) contributions to the q11–13 region of chromosome 15 as a result of deletion or uniparental disomy. Apparently, most patients are sporadic cases. The genetic mechanism underlying familial AS has remained enigmatic for a long time. Recently, evidence has been emerging suggesting autosomal dominant inheritance of a detectable or undetectable defect in a gene or genes at 15q11–13, subject to genomic imprinting. The present report describes an unusually large pedigree with segregation of AS through maternal inheritance and apparent asymptomatic transmission through several male ancestors. Deletion and paternal disomy at 15q11–13 were excluded. However, the genetic defect is still located in this region, as we obtained a maximum lod score of 5-04 for linkage to the GABA receptor locus GABRB3 and the microsatellite DNA marker D15S10, which have been mapped within or adjacent to the AS critical region at 15q11–13. The size of the pedigree allowed calculation of an odds ratio in favour of genomic imprinting of 9.25 x 10^6. This family illustrates the necessity of extensive pedigree analysis when considering recurrence risks for relatives of AS patients, those without detectable deletion or disomy in particular. (J Med Genet 1993;30:853–7) The main features of Angelman syndrome (AS) are severe mental retardation, absent speech, paroxysms of laughter, abnormal gait, seizures or EEG abnormalities, microcephaly, brachycephaly, macrostomia, and prognathism. Diagnosis during the first year of life may be difficult because facial dysmorphism and seizures are not typical. The incidence of AS is estimated to be around 1 in 20 000. More than 50% of the patients have a cytogenetically visible deletion or rearrangement of chromosome 15q11–13 and 75 to 80% of the patients have molecular deletions. The deletion always involves the maternally inherited chromosome 15. A small percentage of non-deletion AS patients lack the maternal 15q11–13 region as a result of paternal disomy. The same cytogenetically defined region of chromosome 15 is involved in Prader-Willi syndrome (PWS). PWS is phenotypically very different from AS as PWS patients display infantile hypotonia, childhood hyperphagia and obesity, mental retardation, and hypogonitalism. In contrast to AS, in PWS the deletion of chromosome 15q11–13 is of paternal origin whereas uniparental disomes are of maternal origin. This difference parental origin of 15q deletions and disomes resulting in different phenotypes indicate that genes in this region show differential expression on maternal versus paternal chromosomes (genomic imprinting). The report of a family with a chromosomal translocation involving chromosome 15 leading to unbalanced translocations with deletions in 15q in the offspring supports the role of genomic imprinting in both syndromes, as identical unbalanced karyotypes in the offspring either produced AS or PWS, depending on the sex of the transmitting parent. Recently, molecular analysis showed that the smallest regions of overlap (SRO) of the deletions in both syndromes are distinct, the SRO of AS residing between D15S11 and D15S10 and the SRO of PWS between D15S9 and D15S11. This suggests that different genes are involved in the aetiology of both syndromes. It is conceivable that a defect in inhibitory neurotransmission mediated by the GABA\textsubscript{A} receptor plays a role in the pathogenesis of AS. Therefore, the GABA\textsubscript{A} ($\gamma$-aminobutyric acid) $\beta$ subunit 3 receptor (GABRB3) gene, which has recently been localised to the SRO of AS, has been proposed as a candidate gene for AS. The vast majority of patients with AS are apparently sporadic cases. Several familial AS patients have been reported and, unlike sporadic cases, detectable abnormalities of chromosome 15q are rare. Since most of the familial AS patients are sibs, it has been hypothesised that AS may be inherited as an autosomal dominant trait. However, Niermeijer et al reported a family in which three AS sibs did show a deletion of 15q, which they shared with their healthy mother and grandfather. Recently, Wagstaff et al reported three unaffected brothers producing offspring without... detectable abnormalities involving 15q, most likely representing autosomal dominant inheritance and genomic imprinting at 15q. In this family a maximum lod score of 2.91 at $\theta = 0.00$ with $GABRAS(CA)-1^{19}$ was obtained. Marker $GABRAS(CA)-1$ was reported to be localised distal to the SRO of AS.$^{19}$ We present a family of exceptional size as it spans five generations with AS occurring in several sibships. Maternal inheritance was apparent in each case. Asymptomatic transmission through several male ancestors could be inferred from the pedigree. Cytogenetic analysis was performed to detect chromosomal rearrangements in the patients and their mothers. Molecular and statistical analyses were applied to investigate subtle deletions or uniparental disomy involving 15q, linkage between $AS$ and markers on 15q, and to determine the mode of inheritance. Materials and methods PATIENTS The pedigree of a five generation Caucasian family with eight AS patients is shown in fig 1. Seven of the patients showed typical features of AS, including severe mental retardation, absent speech, and frequent seizures (table 1). Before this family presented to us, the only patient in generation III (fig 1) had died at the age of 53 years. Her medical records and photographs confirmed that she was also affected with AS. Patient V.3 was seen at the age of 10 months and all other patients and family members investigated were older than 10 years. Seven patients had EEG abnormalities, while two patients, IV.10 and III.x, had seizures. Strabismus was seen in patients V.1, V.3, IV.6, and III.x. Fig 2A and B show the development of the facial dysmorphism in patient IV.10. Fig 2C shows patient V.1 at the age of 8 years. There was no other family history of congenital malformations, developmental delay, or epilepsy. Consanguineous marriages were not recorded. CYTOGENETIC ANALYSIS Chromosome analysis on lymphocytes of patients V.1, V.3, IV.6, IV.7, IV.15, and IV.16 was performed using standard techniques. High resolution GTG banding was applied to study one of the patients (IV.10) in more detail. In order to enable detection of a familial translocation of chromosome 15 in its unbalanced form, the patients’ mothers, IV.2, III.8, III.10, and III.13, were also karyotyped. DNA ANALYSIS Blood samples for DNA analysis were collected from 38 subjects, as indicated in fig 1, representing three generations and including all available nuclear families with affected offspring. DNA from leucocytes was isolated according to Miller et al.$^{20}$ Dinucleotide (CA) repeat markers at $GABRB3$ and $D15S10$ were analysed using radiolabelled PCR conditions essentially as previously described.$^{21,22}$ RFLP --- Figure 1 Family pedigree. Filled symbols = affected. All numbered subjects were typed for $GABRB3-CA$ (alleles I–8) and $D15S10-CA$ (alleles a–d), as well as for RFLPs at $D15S18$, $D15S13$, $D15S9$, $D15S11$, $D9S10$, and $D8S12$ (not shown). III.x is the dead patient. Table 1 Clinical symptoms in the eight AS patients. \| \| V.1 \| V.3 \| IV.6 \| III.x \| IV.7 \| IV.10 \| IV.15 \| IV.16 \| \|---------------------\|-----\|-----\|------\|-------\|------\|-------\|-------\|-------\| \| Mental retardation \| + \| + \| + \| + \| + \| + \| + \| + \| \| Absent speech \| + \| + \| + \| + \| + \| + \| + \| + \| \| Paroxysms of laughter\| + \| + \| + \| + \| + \| + \| + \| + \| \| Abnormal gait \| + \| + \| + \| + \| + \| + \| + \| + \| \| Seizure-like abnormalities \| + \| ND \| + \| + \| + \| + \| + \| + \| \| Microcephaly \| + \| + \| + \| + \| + \| + \| + \| + \| \| Brachycephaly \| + \| + \| + \| ND \| + \| + \| + \| + \| \| Macrostomia \| + \| + \| + \| + \| + \| + \| + \| + \| \| Prognathism \| + \| – \| + \| + \| + \| + \| + \| + \| + = present, – = absent, ND = no data available. analysis by Southern blotting and hybridisation using standard methods involved the following probes from 15q11–13: pIR39 (locus D15S18), pTD189.1 (D15S13), pML34 (D15S9), pIR4-3R (D15S11), pTD3-21 (D15S10), and pIR10-1 (D15S12).11 LINKAGE ANALYSIS All linkage calculations were carried out with the MLINK option of the LINKAGE package of computer programs (version 5.03).23 Equal allele frequencies were assumed for the marker loci. The disease allele was given a frequency of 0.001 in the calculations, but it was assumed that only a single AS mutation segregated in this family. Calculations were carried out under two models: one with 50% penetrance for all subjects heterozygous for the autosomal dominant disease allele (absence of the disease phenotype in gene carriers is explained by random effects leading to 50% penetrance), and another model with complete penetrance for subjects who had inherited the disease allele from their mother, while paternal inheritance was assumed never to lead to expression of the mutation (absence of the disease phenotype in gene carriers is solely the effect of genomic imprinting). Odds for imprinting were calculated by comparing the maximum likelihood obtained under the model of imprinting and linkage with the maximum likelihood allowing for linkage but not for imprinting. Subjects V.1, V.3, IV.6, III.x, IV.7, IV.10, IV.15, and IV.16 were considered to be affected, while all other family members were considered to be unaffected. In the multipoint analysis the distance between GABRB3 and D15S10 was fixed at 1 cM, based on an observation of a single recombination event between these loci in this pedigree. Recombination frequencies in males and females were assumed to be equal. Results CYTOGENETIC AND DNA ANALYSIS Chromosome analysis showed normal karyotypes in all seven investigated patients and their mothers, with no cytogenetically visible deletions or translocations involving chromosome 15q. DNA marker analysis showed heterozygosity at GABRB3 in all patients and at D15S10 (CA repeat) in two patients (fig 1). Heterozygosity was also observed at D15S11 in the DNA of four patients when analysing the NsiI and RsaI polymorphisms detected by IR4-3R (not shown). The hybridisation patterns obtained by Southern blotting using other RFLP markers from chromosome 15q11–13 were all in agreement with the presence of two alleles in the patients’ DNA. Hence, we were unable to detect a deletion in the chromosomal region we investigated at the molecular level. Moreover, the heterozygous patterns at GABRB3 were of unequivocal biparental origin in five of the patients (fig 1). The paternally derived allele could be distinguished with certainty in the four patients who were heterozygous at D15S11 (not shown). Assuming that one genetic defect segregates in this family this rules out paternal disomy at these loci as the cause of AS in this family. LINKAGE ANALYSIS We noticed that the same allele, numbered 3 in fig 1, at GABRB3 was found in all seven patients investigated, their mothers, and the only living grandparent (III.1), whereas this allele was not observed in unaffected sibs of patients (fig 1). Next, we analysed whether the genetic defect in this family showed linkage to GABRB3 and D15S10. Table 2 summarises the lod scores obtained at different 0, loci, and models. No recombinations between AS and either D15S10 or the GABRB3 locus were observed. A maximum Table 2 Lod scores between AS and CA repeats at the GABRB3 locus and at D15S10 at various recombination fractions. Lod scores were calculated assuming two different models: one with 50% penetrance of all heterozygotes for the disease allele (no imprinting), and one with complete penetrance for subjects who had inherited the disease allele from their mother, while paternal inheritance was assumed never to lead to expression of the mutation. \| \| 0 \| 0-01 \| 0-05 \| 0-1 \| 0-2 \| 0-3 \| \|------------------\|-----\|------\|------\|------\|------\|------\| \| GABRB3-CA \| \| \| \| \| \| \| \| D15S10-CA \| 4-28\| 4-18 \| 3-81 \| 3-36 \| 2-53 \| 1-71 \| \| GABRB3-CA + D15S10-CA with imprinting \| 5-01\| 2-95 \| 2-70 \| 2-37 \| 1-67 \| 0-95 \| \| GABRB3-CA + D15S10-CA without imprinting \| 5-40\| 3-30 \| 3-81 \| 3-04 \| 2-36 \| 2-26 \| A multipoint lod score of 5-40 was obtained at θ = 0-00 between AS and GABRB3 under the assumption of genomic imprinting. The alternative model assumed incomplete (50%) penetrance instead of genomic imprinting and yielded a substantially lower lod score. The ratio between the maximum likelihoods of the two models determines the odds for genomic imprinting. Assuming complete linkage of the AS gene(s) with D15S10 and GABRB3, an odds ratio in favour of imprinting of 9 × 25 × 10^3 was calculated. Discussion The AS patients in the family reported here showed no evidence for a deletion or other chromosomal abnormality involving 15q11–13. Paternal disomy of chromosome 15 was excluded as the cause of AS in this family. The maximum lod score of 5-40 in favour of linkage with GABRB3 and D15S10 provides strong evidence for a dominant mutation in 15q11–13. For the first time the role of genomic imprinting in familial AS could be established by statistical analysis, as an odds ratio of 9-25 × 10^3 in favour of genomic imprinting versus reduced penetrance was achieved. The pattern of imprinting in this family was such that no father passed on the disease, but only the females II.2, III.4, 8, 10, 13, and IV.2 (fig 1). The data indicate that the mutation causing AS in the eight family members originates from the male founder in generation I.1 (fig 1). Apparently, the mutation was transmitted through as many as three generations by male carriers without any phenotypic expression. This fact is of great importance in view of the risk assessments for even distant female relatives, who may have a 50% risk of affected offspring depending on their position in the pedigree (fig 1). The AS families reported by Hamabe et al. and Wagstaff et al. are in agreement with the concept of a dominant mutation in one or more genes in 15q11–13 subject to imprinting as the cause of familial AS. Two other AS families have been reported that were investigated with DNA markers. In one case the affected sibs inherited the same maternal 15q11–13 region. The affected sibs in the other family inherited at least partly different maternal regions of 15q11–13, but a possible recombination event between the SRO of AS and the tested markers made the latter result inconclusive. We conclude that the present molecular data from familial AS patients are either in favour of a dominant genetic defect on 15q11–13 with phenotypic expression depending on imprinting or do not contradict this hypothesis. Further molecular genetic analysis of non-deletion AS patients will be essential to determine the gene(s) involved in the pathogenesis of AS. Our data do not contradict a possible role of the GABRB3 gene in the pathogenesis of AS as no recombinations were detected between AS and this locus in the entire family. The consistent lack of PWS in the offspring of AS patients carrying an AS mutation adds to the evidence that distinct defects are involved in the aetiologies of AS and PWS. In our family non-penetrance as an explanation for the lack of PWS offspring was highly unlikely as multiple transmissions of the mutation occurred through male meioses without phenotypic expression of PWS. The influence of genomic imprinting on the expression of mutation has been described in a growing number of human genetic diseases. Hereditary paraganglioma is another example where a role for genomic imprinting in the phenotypic expression in an extended family has been shown to be absolute. The recent implication of genomic imprinting in familial AS may complicate genetic counselling. In familial AS cases, an X linked mode of inheritance may falsely be deduced when the syndrome diagnosis is overlooked. On the other hand, imprinting may obscure the hereditary nature of the defect owing to the lack of phenotypic expression in offspring of male carriers. If a genetic defect at 15q11–13 causes autosomal dominant AS, as seems to be a universal phenomenon in familial AS, unaffected sibs of AS patients would have no increased risk for AS in their offspring. Risk assessments for more distant female relatives of AS patients, particularly those without detectable genetic defects, would necessarily involve an extensive search of the pedigree for AS. We thank Dr B A van Oost for critical reading of the manuscript. 1 Angelman H. 'Puppet' children. Dev Med Child Neurol 1965;7:681–6. 2 Clayton-Smith J, Pembrey ME. Angelman syndrome. J Med Genet 1992;29:412–3. 3 Fryns JP, Berg WR, Lirnzenen V. Diagnosis of Angelman syndrome. Am J Med Genet 1991;38:58–64. 4 Kaplan LC, Wharton R, Elias E, et al. Clinical outcome of deletions in the long arm of chromosome 15: report of 3 new cases and their possible significance. Am J Med Genet 1988;28:101–10. 5 Deogratias TA. Small molecular deletions on chromosome 15q11.2 are encountered in four Prader-Willi and Angelman syndrome patients. Hum Genet 1989;82:101–4. 6 Knoll JHM, Nichols RD, Magenis RE, et al. Angelman and Prader-Willi syndromes caused by different chromosome 15 deletion but differ in parental origin of the deletion. Am J Med Genet 1989;33:281–9. 7 Maitland NJ, Clayton-Smith J, Nicholls M, et al. Uniparental paternally in Angelman’s syndrome. Lancet 1989;i:1009–11. 8 Smeets DFCM, Hamel BCJ, Nelen MR, et al. Prader-Willi syndrome and Angelman syndrome in cousins from a family with a translocation between chromosome 6 and 15. N Engl J Med 1992;326:807–11. 9 Ledbetter DH, Riccardi VM, Aihart SD, et al. Deletions of chromosome 15 in patients of the Prader-Willi syndrome. N Engl J Med 1981;304:527–31. 10 Prader A, Labhart A, Willi H. Ein Syndrome von Adipositas, Kleinwuchs, Kryptorchismus und Oligophrenie nach myoneurologischen und endokrinologischen Gesichtspunkten. Med Wochenschr 1956;86:1266–1. 11 Nicholls RD, Knoll JHM, Glatt KA, et al. Restriction fragment length polymorphisms within proximal 15q and their use in molecular cytogenetics and the Prader-Willi syndrome. Am J Hum Genet 1989;33:104–13. 12 Nicholls RD, Knoll JHM, Butler MG, et al. Genetic imprinting suggested by maternal heterozygosity in non-deleted Prader-Willi syndrome. Lancet 1989;ii:1267–8. 13 Hultin M, Armstrong S, Challine P, et al. Genomic imprinting in the Prader-Willi/Angelman translocation family. Lancet 1991;338:638–9. 14 Hamabe J, Kurada Y, Imaiizumi K, et al. DNA deletion and its association with the Prader-Willi syndrome. Am J Med Genet 1991;41:64–8. 15 Hamabe J, Tsuchiya K, Harada N, et al. Molecular study of the Prader-Willi syndrome: deletion, RFLP, and phenotype analyses of 50 patients. Am J Hum Genet 1991;49:1000–10. 16 Wagstaff J, Knoll JHM, Fleming J, et al. Localization of the gene for the GABA(A) receptor β3 subunit to the Angelman/Prader-Willi region of human chromosome 15. Am J Hum Genet 1991;49:330–7. 17 Santoro M, Superti-Furga A, Dallapiccola B, et al. Familial Angelman syndrome caused by imprinted submicroscopic deletion encompassing GABA(A) receptor β3-subunit gene. Lancet 1992;339:360–2. 18 Barraister M, Patton M, Lam ST, et al. The Angelman (happy puppet) syndrome: is it autosomal recessive? Clin Genet 1989;35:19–24. 19 Wagstaff J, Knoll JHM, Glatt KA, et al. Maternal but not paternal transmission of 15q11–13-linked nondeletion Angelman syndrome leads to phenotypic expression. Am J Hum Genet 1992;51:126–32. 20 Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 21 Mutirangura A, Ledbetter DA, Kozak CA, et al. Dinucleotide repeat polymorphism at the GABA(A) receptor β3 (GABRB3) locus in the Angelman/Prader-Willi region (AS/PWS) on chromosome 15. Hum Mol Genet 1992;1:101–6. 22 Litt M, Lai R, Kimmel S, Wong C. A dinucleotide repeat polymorphism of D15S10 in the Prader-Willi chromosome 15 region (PWC1R). Nucleic Acids Res 1991;19:5449–50. 23 Lathrop GM, Lalouel JM. Easy calculations of lod scores and recombination risks on small computers. Am J Hum Genet 1984;36:495–502. 24 Knoll JHM, Glatt KA, Nicholls RD, et al. Chromosome 15 uniparental disomy is not frequent in Angelman syndrome. Am J Hum Genet 1991;48:16–21. 25 Hall JG. How imprinting is relevant to human disease. Development (suppl) 1991;141:8–18. 26 Horsthemke B, van den Vekel AGC, van Bokkum JJ, et al. A gene subject to genomic imprinting and responsible for hereditary paraplegia maps to chromosome 11q23-qter. Hum Genet 1992;89:17–21.
The first step in the process is to identify the specific type of cancer and its stage. This information helps determine the most appropriate treatment options. Treatment for pancreatic cancer can include surgery, chemotherapy, radiation therapy, or a combination of these approaches. The goal is to remove the tumor completely or shrink it to make it easier to manage with other treatments. Surgery is often used when the cancer is still in an early stage. However, it may not be possible in more advanced cases due to the location and size of the tumor. Chemotherapy uses drugs to kill cancer cells throughout the body. Radiation therapy involves using high-energy beams to destroy cancer cells. In some cases, doctors may recommend a combination of these treatments to achieve the best results. It's important to note that each person's experience with pancreatic cancer will be unique, and the treatment plan will depend on individual factors such as age, overall health, and the specific type and stage of the cancer. If you or someone you know has been diagnosed with pancreatic cancer, it's essential to work closely with a healthcare team to develop a personalized treatment plan. They can provide guidance on the best course of action based on your unique situation. Library of Congress. Chap. BR125 Shelf N75 Copyright No. UNITED STATES OF AMERICA. THE FINE ARTS A PRACTICAL COURSE OF STUDY FOR THE USE OF ART STUDENTS AND ART LOVERS BY WILLIAM H. WOOD PUBLISHED BY THE AUTHOR 1902 THE TELESCOPE: OR, SACRED VIEWS OF THINGS PAST, PRESENT, AND TO COME. BY SAMUEL NOTT, JUN. It is good that we transplant the instruments of fancy into religion.—Jeremy Taylor. Mine eyes he closed, but open left the cell Of fancy, my internal sight.—Milton. BOSTON: PUBLISHED BY PERKINS & MARVIN, 114, Washington Street. 1832. Entered according to act of Congress, in the year 1832, By Perkins & Marvin, in the Clerk's Office of the District Court of Massachusetts. 1364 CONTENTS. The Voice of the Grave: or, Youthful Forecast, 5 Boston, on her Two Hundredth Anniversary, 19 A Vision of the Last Night in the Year, 37 The Heir of the World, 46 Philadelphia, and the Sunday School Union, 63 The Field of Death, 83 New York, and the American Bible Society, 90 The Box Opened, 105 The Heir of Heaven, 112 London; a Retrospect of the British and Foreign Bible Society, 136 The Ganges, and the Mississippi, 163 THE TELESCOPE. THE VOICE OF THE GRAVE: OR YOUTHFUL FORECAST. The grave at which we are going to listen, was closed in the year 1758. The history of the man, who was then buried out of sight of the living, gives to its silence, and darkness, and corruption, a sacred eloquence. He who has been hidden so long from the eyes of men, came into being about the beginning of the last century, in one of the ancient villages on the banks of the Connecticut. Almost one hundred and thirty years have passed, since his parents rejoiced over a new born son; how helpless in that infant frame! how ignorant in that infant mind! Fond parents sheltered and cherished and guided him in infancy and childhood, and blessed his youth with the means and opportunities of knowledge and religion. From that helpless infancy he arose into life, endowed with powers of thought, which made him, for thirty years, the ornament of his country, and of his kind. After fifty-five years spent on this earth, he passed suddenly from the sight of men; leaving a name which has not yet lost its lustre, nor passed from the mouths of men. It is yet an early stage in the progress of his immortal spirit. Not yet has he doubled his earthly career; but even now we may pause upon his grave, and hear from its silent chambers, his monitory, his encouraging voice: if, like the long-dead Abel, he yet speaketh. It was from the height of honor that he went down to the grave. Having stood among the great men of his day, he had reached one of the most responsible and honored stations in the land, when in a few months he fell a victim to a loathsome and contagious disease, and went down to the grave suddenly, as helpless as the dying infant of a day; as separate from the congregation of the living; as confined to his own narrow house, his solitary cell; yet in honor still, amidst the clustered worthies of former and later times. Turn traveller aside, wait a stage or two, that thou mayest walk and meditate in that retired and lonely cemetery. Sons of science and the prophets, spend your vespers there, or listen to the matins of the songsters, who at early dawn, sing to their Maker, as if about their graves they caught the songs of just men made perfect. In that hour of stillness and solitude, when the evening shades are closing thee in from the near-by village; or when the dawn reveals slowly the record on those chiseled tablets, ere yet the observant world has waked from its slumbers, choose this grave-yard walk: ambitious youth, place thy knee upon the turf which parts the monuments of the learned dead, and recline thy head upon that which bears the date, 1758....and thou wilt hear a mysterious voice like that from the first of human dead. Listen; thou wilt learn thy frailty: thou wilt hear report of the undying spirit, by what choice it rose to a house not made with hands, eternal in the heavens. One voice thou wilt hear in harmony with the clustered just. Here for nearly fourscore years the grass has grown and withered. Near thrice three thousand times the sun has arisen and poured its light over this dark grave; that eye has not been pleased to behold the sun; nor that body sprung forth to enjoy "the cool, the fragrant, and the silent air." The eye, the limbs, the whole frame have become dissolved. The corpse has no coffin, no death-dress; the bones are crumbled or crumbling into dust. If the spade should violate that grave, nought would be found but dust and fragments of the great and good, whose name was upon the tongue of our fathers four-score years ago. Now, he speaketh, LISTENER, in thine ear, at least this word, "Corruption, thou art my father; worm, thou art my sister and my mother. The grave is my house. I have made my bed in darkness. Whatsoever thy hand findeth to do, do it with thy might, for thou art hastening to the grave, where there is neither wisdom, knowledge, nor device." But hark again! Within that silent chamber lie the remains of a frame whose plan was written in the book of the Almighty; made in secret, curiously wrought, fearfully and wonderfully made: fitted for the abode and the growth of an immortal spirit. Sitting within its secret chamber, that spirit maintained, by mysterious connection with the corporeal senses, an intercourse with surrounding scenes; with the very thoughts of men; with the character, and deeds, and relations of the Eternal God. In its very boyhood, it kenned the secrets of philosophy, which sages since have discovered and divulged. During the short space of fifty-five years, the man arose, from the ignorance of new-born infancy, to an extensive knowledge of things human and divine: to a range and power of thought, which gave him, both living and dead, a commanding influence over men: which promises to reach to all nations, and to all times. Of the stroke which laid in the dust this curious frame, we have the biographical record, the chiseled memorial; and of that spirit to which it ministered, which grew beyond its growth, his works will repeat, "Secundus nemini mortalium,"* when this marble record has wasted into dust. Where is that mighty spirit? Is it lost amidst the dust of its former dwelling? Listener: dost thou not hear its voice from amidst the spheres? The dust has returned to the earth as it was, but the spirit unto God who gave it. Listener! hast thou read the earthly history of the spirit which, almost fourscore years ago, returned to God who gave it?—the testimonies of a moral and spiritual life; of a fountain which * Second to no mortal. sprung up, strong and clear, of everlasting life; of a heart humble and contrite, already visited and revived by the High and Lofty One! Canst thou doubt that when that spirit parted from its clayey tabernacle, it spread its wings and soared away into holier regions, and dwells fast by the throne of God, with angels and the spirits of just men made perfect? Amidst this morning silence, sure there is a voice, in sweet harmony with these carols all around: Though the earthly house be dissolved, I have a building of God, a house not made with hands, eternal in the heavens. Yet hearken again, beloved youth: the sun is not yet risen, though the dawn brightens in the eastern sky. The villagers have not yet come forth from their chambers. Listen yet again to the voice of the grave; and hear of that faith by which the dead obtained testimony that he pleased God; by which he offered himself up to God, a living sacrifice. Hark! soft as the singing of the birds; fresh as the dew of the morning; cheerful as the light of the sun after the darkness of the night; the voice is mellow as the tones of youth;—it is the voice of youthful piety, choosing, in the dawn and morning of existence, a day without a night, an everlasting day. Is it memory? or is the voice repeated in thine ear amidst the beauties of the morning? "Resolved, That I will do whatsoever I think to be most for the glory of God, and my own good, profit and pleasure, in the whole of my duration, without any consideration of the time, whether it be now, or never so many myriads of ages hence." Sublime resolve! What forecast for futurity, for being endless and progressive! for a soul to grow forever as it grew from infancy, to this power of thought, and forecast, and decision! Be it thine: that when thy body shall lie moulder ing and mouldered, thy spirit may dwell on high, rich in the fruits of forethought so sublime! Ah listener: thy heart revolts. Thy fancy has already soared away to some mountain height, whence thou canst see all the glories of the world, and call them thine. Pause, then, and hold controversy with thy deceived heart, until thy voice can sound in harmony with the sainted dead. He held that controversy, and gave example of victory. Think not, he found decision easy. Had he no sinfulness? or lived he when there was no tempter to encourage and beguile? Had he no aided wing of fancy, no bewildered eye? Did not the tempter try to deceive the youth with blessings which must perish with his day? No doubt he looked, and wished for a moment, and again and again chose the fleeting present, and forgot the ever-during future, and preferred a momentary glare to the whole weight of future glory. No doubt some plain demand of conscience and of faith seemed so grievous, as to incline his heart to give up the kingdom of heaven, as that young man did who went away sorrowful, from the Saviour, because his great possessions were demanded. How fearful of a frown! How fond of fame! How ready to give himself the glory of his growing powers! How cheered with the thought of higher and higher honors! How ready was the young immortal, for a morsel, to sell his everlasting birthright! and yet how often was he disappointed; mocked by lying vanities—sick with desires unsatisfied—tossed with the waves of sinful passion; of pride, or envy, or self-love, or anger, or sensual desires! Thus Satan foiled himself, and opened the door of blessings to the soul. The inexperienced youth saw the deception, and triumphed over the deceiver. As he looked around from his fancied height, and saw the devil's picture of the world, the divine reply sprung to his lips, "Get thee hence, Satan; for it is written, Thou shalt worship the Lord thy God, and him only shalt thou serve." Blessed season of discovery—of triumph over the usurped government of the prince of the power of the air! when the kind angels, always hovering over him, and helping the fearful struggles of the young immortal, found opportunity to aid the wings of fancy, as she flew up to the "Delectable Mountains," and to hold steadily the prospect-glass as the eye looked away into the distant future! when the spirit, long striving to enter, was welcomed to a fixed abode in the troubled heart. What aids were there, who can tell? What hosts of witnesses to the Spirit's entry to the heart, melted to humbleness and contrition by his rising beams. Would that the grave might whisper the hour when the resolve kindled into flame to shine through the ages of eternity! Perhaps it was when weariness and night clouded his soul with gloom; or when lightning and tempest filled him with awe and terror; or when, in that pleurisy, he felt justly seized by the hand of the Almighty, and shaken over the devouring pit, and was forced to seek, "as he never sought before;" or rather, when listening to the still small voice of rebuke and kindness, amidst the silent grove, to which afterwards he listened so often, so attentive. There, alone with God, seeing Him among the trees, in the cool of the day, in every moving branch, in every rustling leaf, in every blade of grass, and in the flood of light—hearing his reproving, winning voice, in the soft breeze, as it passed through the trembling forest—when air and sky, and growing nature, and the vocal birds, gave token of Him who filleth immensity with his presence:— In such a calm and quiet solitude, when the tempests of the outer world were lulled to sleep, then, perhaps a calm came in upon the tempest of the soul, a sweet and awful stillness in which God was heard speaking with a father's tenderness, "My son give me thine heart:" and eternity was seen proceeding in all its endlessness from the passing moment, and receiving its everlasting character from the passing thoughts. Then perhaps the conflict ended in that sublime decision, I will do, at every fleeting moment, that which shall be most for God's glory, and my own good, whether now, or never so MANY MYRIADS OF AGES HENCE. How lovely, then, how heavenly, must that youthful face have shone, which bore in its maturity, a youthful loveliness, an aspect fit for heaven, which even human art has been able to preserve, in its ten thousand copies! Methinks the angels must have paused a moment in joy and admiration, ere they flew from the field of conflict where that victory was won: ere they bore away the golden copy of that Resolve, and made heaven's arches ring with their triumphant songs. Sure, as he returned, and went forward in the pursuits of mortal life, filled with the high ambition, of seeking God's glory and his own best good, FOR MYRIADS AND MYRIADS OF AGES, ministering spirits kept about his path; a Father's kindness chastened and cheered him that he might be a partaker of his holiness; the HIGH AND LOFTY ONE, who inhabiteth eternity, made that humbled spirit his dwelling place. No doubt the daily prayer was lifted up, "Lead me not into temptation, but deliver me from evil:" until, having been guided by the counsel of a Father, he was received to glory. While he lived, how steady, how bright, how increasing, were his piety towards God and his good will to men! What comforts cheered him in perplexity, and care, and sorrow! And when amidst the brightest promise of his honored life, his earthly prospects were suddenly darkened in death, how pleasantly he died with these last words upon his lips, Trust in God and ye need not fear! Listener! does thy heart revolt? Let thy fancy soar: Angels will bear thee up. Set thy feet once on the delectable mountains. Look beyond thy funeral day, when thy limbs will be stiff, thy eyes closed, thy senses vanished:—beyond thy mouldered body. Take forethought for the coming century, and that which shall come afterwards and yet again. Think how thou wilt cheer thyself in the dark valley and shadow of death; how thou wilt be joyful when thou hast just escaped from the body of corruption; how thou wilt employ thyself, when earth’s service shall be paid for a thousand years, and earth’s hymns mingle in the sweet harmony with the hymns above; how thou wilt rejoice amidst the wonders of the last day, and along the endless, endless path, on which thy immortal spirit has begun to travel! Ah! how thy mind changes. Methinks I hear thy voice, soft as the singing of the birds, fresh as the dew of the morning, cheerful as the light of the sun, in the mellow tones of youthful piety—I will live to day and hereafter, as shall be most for God's glory, and my own best good, whether now, or never so many myriads of ages hence. Look! the sun is bursting from the east. Thine angel is speeding his way on the wings of the morning, and thy resolve, written in the golden beams, will be registered in a moment in the archives of heaven. Go back to thy chamber, and pray, and study, and live for eternity! Sons of Nassau! As ye float by an unbroken stream of youth, on your way to water and refresh the land, listen to the voice of the grave. Or if ye fear at early dawn and under the day star to listen to that mysterious voice, at midnight hour look upon the wall! Perhaps amidst the gloom, thou wilt see the hand of Jonathan Edwards, writing the sublime resolution of his youth; and amidst that stillness thy heart will rise to a resolve for myriads of ages! Sons of the Prophets, listen to that voice. Read that writing on the wall: and as ye go forth, oh men of God, follow after righteousness, godliness, faith, love, patience, meekness. Fight the good fight of faith, lay hold on eternal life. TRUST IN GOD AND YE NEED NOT FEAR. Youthful reader, wherever thou art, at the desk, in the field, on the exchange, at the toilette, at the social party—listen to THE VOICE OF THE GRAVE: or if amidst temptation thou miss that voice, in the calm hour of night read thy duty and thy safety in living characters around thy bed,'and resolve for ETERNITY. Traveller! if thou turn aside to visit the grave of Edwards, amidst the illustrious Presidents of the College of New Jersey; obey the counsel inscribed on the marble almost fourscore years ago. ABI VIATOR, ET PIA SEQUERE VESTIGIA. Go, traveller, and follow his footsteps; seek God's glory and thy best good, now, and for myriads and myriads of ages hence. BOSTON, ON HER TWO HUNDREDTH ANNIVERSARY, SEPTEMBER 17TH, 1830. As the sun arose, the years that were passed began to be numbered by the artillery. The repeated thunder seemed the knell of the departed century. The silence that followed, was yet more awful; leaving without memorial the preceding century mingled in the common mass of passed time; the preceding generations lost in the common multitude of the dead. At the hour appointed for the rendezvous of the living population, I joined the multitude who kept this holiday;—the very thought of which raised in my imagination visions of decay and change, of past and future, strange as if they were dreams, and yet as true and real as the ever changing crowd before me. I had scarce taken my stand, midway in the crowd, in front of that noble pile, which shows its elevated dome far above the common height of the city habitations, towering above the ancient summit of Beacon hill, when the clock struck nine: announcing to the gathered and gathering throng, that another century was already speeding its flight! that ere the pageantry began, unwearied time had been measuring off the hours which were hastening the century to its close! The Mall, made for spectacle; an amphitheatre formed by nature for the display and sympathies of multitudes; raised yet higher with its artificial galleries, its windows, piazzas, porticos, reflecting the splendors of that brilliant morn, and receiving the reflection of the surrounding slopes, covered with gardens and villas and villages and farms; the mall, from its highest pinnaclces of art down its gentle slope, and in scattered, changing groups over its noble lawn, presented a living miniature of the living world. There, were the servants of the church and state, with church and state themselves; the sciences and the arts with their professors and their practitioners; commerce and husbandry; the aged, as if to bid their last adieu to the living world; the young, in all their gaiety and beauty, with the four thousand blossoms* of the early morning, to meditate on life's fleeting scene, and to say, * The public Schools. "We too must die." As I looked over this field, scattered with decaying plants, covered with ripened fruit, and spring and summer blossoms in such bloom and richness, as betokened immortality, I knew these bodies could not be incorruptible until first they had wasted into dust; and I said, surely in sympathy with living thousands, "We must die, as those who lived in centuries before us." Then, fevers, and fluxes, and consumptions, and apoplexies, and accidents, and old age, and lightning, and tempest, seemed flitting before my imagination; and all the living laid unknowing and unknown in the dust of death. Yes, long, long before the morning of another century shall arise upon the living world, we shall be dead, and our spirits which cannot die, will have been borne by the angels to the bosoms of our ancestors in glory, or left in their chosen misery and sin. At that moment of sympathy, I seemed but to see what was revealed to every eye; and to feel but the sentiment of every soul. There we stood, numbering our days even to four score years, and calling them a dream! What wisdom sprung up from that one discovery! The moralist looked upon his robe, and it was filthy rags: The miser saw the canker corroding his silver and gold: The ambitious, aiming higher in paths as various as the multitude before me, humbled their high looks and bowed down their haughtiness: The sensualist turned his thoughts inward in search of an appetite which could feed forever on imperishable feasts: The caviller laid aside his questions, and with the eagerness of a dying man grasped after eternal life; and even the infidel stood aghast, looking earnestly after Him who brought life and immortality to light. The whole mass of men seemed waking to the simple prayer, "So teach us to number our days that we may apply our hearts to wisdom. Oh satisfy us early with thy mercy, that we may rejoice and be glad all our days. Let thy work appear unto thy servants, and thy glory unto their children: And let the beauty of the Lord our God be upon us!" As I recovered from the reverie into which these reflections cast me, I looked around for some relics of the hand of time; some abiding memorial of the century remembered; some means of sympathy with the departed age. Not one, who lived a hundred years ago, not one, was there; not one, on whom I could fix my eyes and say, You saw the last centennial morning as I see this. All who arose that morning from their beds, active and vigorous, now lie deep in darkness in the city cemeteries, or in surrounding hills and valleys, or scattered through the regions of the West, or in the ocean’s depths, or on foreign shores. In the soft breathings of that gentle morn, I seemed to hear the whisper from the four winds of heaven, “We were, but are not!” Yet I saw abiding memorials of the departed age. These majestic elms, methought, stand in the unchanging glory of their kind, unmodernized amidst the changeful displays of human art; grown from beauty to sublimity, amidst the decay and renewal of human life. Our fathers planted them, admired their young beauty, and regaled themselves under their tiny shades, talking of the distant times, when a crowded population would be refreshed under their towering tops, and hold sweet converse along these embowered walks. These rolling lands, these hills and valleys, also, in beauty as lasting as the earth, bear over their swelling, sinking bosoms, my sympathy with the generations who are gone. And this glowing sun, shining from its ancient firmament, melts all my passions into fellowship with our earliest fathers; who chose this spot by its light; and were cheered amidst their gloom in establishing for the benefit of future times and of other lands, the Metropolis of New England. Oh that I could indeed sympathise with the Christian dead; that I could glow with their pious ardor, with their love of God and man! These hillocks, these shady walks were not wont, in the ancient days, to be the way of the ungodly and the seats of the scornful: but here, alone or in company, sainted spirits walked, meditating, delighting in the law of the Lord. Here many an hour of Christian fellowship was enjoyed, by those who now have fellowship with angels and the spirits of the just made perfect. Behind these gentle swells, in the secresy of that vale where those willows grow, by the side of this still water, canopied by the starry host, was the closet of joyful contemplation, the chamber of angels' visits to the soul, the temple of secret prayer, and fellowship with God. Here the Mathers, and Sewals, and Princes, and here, too, names unknown to fame, impressed the paths with sanctity, and tinctured the air with devotion; and in this open oratory, the great master of Christian eloquence, offered up the supplications of listening thousands, and proclaimed the offers of the gospel. But these majestic trees, these everlasting hills and vales, this bright and glowing sun, and this blue firmament, are not the only representatives of the centuries remembered. The fathers are living now, in the persons of their children; the good seed which they sowed has not all perished in the soil; neither have the tares come up so numerous and thick, as to prevent the growth and fullness of the wheat. Even now, Boston is blessed with the spirit and the power of her early fathers. I could almost see again Mather's aged saint, and hear renewed, after more than a hundred years,* the exultation, "I am now going to heaven, and I will there tell the faithful which are gone long since from New England, that though they who gathered our churches are all dead and gone, yet the churches are still alive, with as numerous flocks of Christians as ever were among them." Moreton seemed again to say, "Out of these small beginnings, even greater things have been produced by his hand who made all things out of nothing: and as one small candle may light a thousand, so the light here kindled * Magnalia, vol. I. p. 83. hath shone unto many, yea in some sort unto this whole nation; and even unto the whole world."* As I stood lost in the crowd, I saw many gifted persons who were serving God and their generation after the example of the fathers of New England; striving earnestly and prayerfully, that at home, the grace of God may not be received in vain, and that its saving power may be carried forth to all heathen nations. Yes, the fathers are living again in the persons of their children: they are here, who came into the wilderness to build the church of the living God; who labored to gather into her hallowed courts, the surrounding Indian tribes. Yes, the Lord hath arisen in this latter day, yet more gloriously, upon this favored city, and his glory may now be seen upon her, now that He has made her the almoner of the whole land to the Pagan world. "The Gentiles are coming to her light, and kings to the brightness of her rising." Methought I could hear the angels whispering, or softly singing in the air, "Lift up your eyes round about and see! all they gather themselves together and come to thee—thy sons are coming from afar, and thy daughters are nursed at thy side." * See New England's Memorial. As I lifted up my eye, the crowd around seemed lost before me, and far distant scenes opened on my view. "Thy towers, Bombay," gleamed bright upon my eye. Thy sea-girt shores, thy crowded city, thy palm-sheltered suburbs, thy ever-busy crowds; the faces, and the costumes, and the languages of all nations, were before me. Thy noble plain, open from sea to sea, was thronged with tens and tens of thousands, paying their pagan gratitude to Parasu Rama, in endless offerings to the calmed ocean. Then again thy multitudes amidst unceasing din, were forcing their way through all thy streets to the sacred water, mystically drowning their perishable gods. Then again thy thousand fires, glaring in the faces of tens and tens of thousands, unblushing amidst their indecent songs: fit preludes to the universal pantomime of lust. But as I looked, my ear caught the dirge of Christian compassion sounding plaintive amidst that long-continued desolation: Who hath believed our report, and to whom hath the arm of the Lord been revealed? Yet amidst those plaintive tones I heard the voice of prayer, in that lowly strain, in which with earnest longings it was sent up eighteen years ago. The grass has withered. The flower has faded. The suppliants are scattered. Yet the ever-living word renews and continues the spirit of prayer and praise; even now procuring blessings for Bombay. Amidst those vain idolatries, those deluded crowds, thy messengers, Boston! lay the foundations of the Temple of the Great Supreme. The murmur of thy Christian schools is heard, and instruction dropping as the rain, distilling as the dew, amidst the softer tones of Christian music, in that oratory which America has built on the coast of Malabar. Ah! the vision changes. They are hastening now to the God who heareth prayer. The harsh discords, fit for the praise of gods who cannot hear, are changing into the soft tones of Christian harmony, coming up acceptably before the Lord God of Hosts. The very streets are cleared of the noisy, unfeeling gabble, and common speech sounds in the melody of faith and hope and love; and faces, meaningless before, glow with the conceptions, and designs, and hopes, of heaven-born minds! Suddenly, I glanced the spicy groves of Ceylon, where seed buried long in dust, has sprung up already, making glad thy reapers, Boston: with Fabricius and Swartz, from whose vigorous hand seed sown in Tanjore fell even on the shores of Jaffna. Then I glanced the Pacific isles—belting their mountain fires, with fertile lands. How strangely did ye make yourselves ready to greet the messengers from Boston! How strangely did ye bring your sons and your daughters to be nursed at our side! Thou, too, Palestine! And ye countries and islands of the inland sea: Boston shines on you. The sun which went down in your western sky, arose on us, and now it dawns on you with the promise of the morning. As fancy in that rapid flight turned homeward, it rested on the fearful, trembling ones, seen through the thickets of the forest; not with the tomahawk and scalping-knife, but at the plough, at the distaff, and the loom; and by night and by day, meditating in the law of the Lord. Ye, too, have been fostered by the alms which Boston has gathered from the north and from the south. Ye, too, have been sheltered by her love. Christian Rulers! if they be the least of those whom the Saviour will call his brethren, and place at his right hand in the last great assembly, be it your honor and your joy to hear Him say, I roamed the forest, and ye gave me a home—I was a sufferer, and ye cheered me with your love—I was lost, and ye helped those who came to seek and save me. Inasmuch as ye did it unto one of the least of the red men, ye did it unto me. As I recalled my wandering fancy, and again stood fixed in attention upon the scenes before me, the chiseled tablets, which used to meet the spectator, as he surveyed Boston and its environs from the top of Beacon Hill, seemed to enlarge their claim upon the sons of the Pilgrims. "Americans! while from this eminence scenes of luxuriant fertility, of flourishing commerce, and the abodes of social happiness meet your view, forget not those who have secured to you these blessings." — As you cast your eyes to-day over all nations, revive in your bosoms the Christian kindness of your Fathers: and let them live again in their faithful children. Ah! methinks, within the bosoms of the crowd around, the future scene is hidden. We shall beautify or mar the prospect which future and again future ages will call up. The seeds which we plant will yield a harvest on the next, and on the next centurial day, in this and other lands. Oh help us, Father of mercies, by thine angel ministries, by thine indwelling Spirit, to know for ourselves, and to display to the world the grace of God which bringeth salvation. Make us the instruments of turning the wilderness into a fruitful field, until glories which exceed our thought shall adorn the heritage of our Fathers, and from every quarter of the world, multitudes shall be joyful in that light which, two hundred years ago, our Fathers kindled on the hills of Boston! Here, patriotism caught me suddenly on its wings, or raised me, soaring on its bubble of pride, where I saw the glory of my country,—transforming, by its magic freedom, all ignorance to knowledge, all vice to virtue; even party strife, the monster of its own production, into love; calumny to kindness; and ruling all the elements of discord into harmony and peace; carrying me upward until I could descry a multitude, saying one to another, let us, unasking, and unaided by Him who rules the nations, build for ourselves, and for the admiration of all people, a self-sustained fabric which shall last forever, and gather around its honored base, the copied governments of mankind! Suddenly, I saw tongues confounded, plans defeated, hope blasted, and the boasters scattered, who thought that nothing was restrained from them which they had imagined to do. Oh for such diligence, and earnestness, and self-distrust, and reliance upon the King of nations, as can alone make our nation happy in herself, or a light to the world. Oh for the spirit of our Fathers, purified, revived, bringing the nation, whose foundations they laid in piety and prayer, to God as the author of all blessings! Then, prophecy caught me on its wings, and raised me whence I could see all the coming glory. The whole earth seemed spread out before me in one brilliant panorama. Nothing marred its beauty. No storm or tempest disturbed the three oceans and numerous seas, all melted and flowing into one; no thunders of war roared amidst the stillness, and troubled the sails and steam of commerce. There were no morasses, or wilderness, or desert. Sahara was a garden. Empires lay spread before me, with signs of peace in all their borders; distinct in government, but bound in universal love. The swords were ploughing in the field, the spears were gathering clusters from the vine, and the powers of nature were employed in providing comforts for man, and in binding all nations into neighborhood. The whole earth was a paradise. For all flesh had come to Him who heareth prayer. The vision was distinct; the beauty was surpassing; the accomplishment sure. Yet while I would have known the times and seasons, I perceived they were wisely hidden beneath the obscure symbols of prophecy. But as I looked, the assembled crowd before me seemed covered with the reflected glory of the scene; some, radiant with faith and hope; some, hiding their faces at so near a view of Him, whose countenance is like the sun shining in his strength; who, coming in utmost mercy, wears on his vesture and on his thigh the Name by which he will assure to himself the conquest of the world, and to the world its rest of a thousand years. Me-thought there were angelic whispers in the breeze, caught and repeated by the tongue of man. "Kiss the Son, lest he be angry, and ye perish from the way." "Blessed are all they who put their trust in Him." It was not in my power to join the interesting convocation in the temple of the Lord: there, to unite in the public covenant to serve the Lord God of our Fathers, and to hand down to posterity our inherited blessings. As I was conveyed around the mall, the centurial procession seemed like a funeral march over the graves of six generations: while the thought sunk to the heart, how low we must lie ere this pageantry is renewed again; ere another centurial dirge strikes up; another centurial pomp covers these fields from whence we now retire! The hills will remain, and the valleys, which, as I pass, now hide, and now display this floating crowd. That central tree, whose bending branches salute a fifth or six generation as they pass, will greet that other day; and piazzas, windows, porticoes will glow with beauty, and gaze with ten thousand eyes, when we are faded away! With these reflections I left the scene, and, hastening through the uncrowded streets, I found myself in a few moments surrounded by the beauties of Dorchester; and from its gentle heights, turning me round, I saw Boston shining in the splendors of her two hundredth birth-day: rising on her everlasting hills, shorn of their summits, with her lofty spires, her towering dome; and joyful in her healthful and prosperous population. I had passed away from the hum of tongues, and the noise of her busy streets no longer reached my ears. As I fixed my eyes upon the metropolis of New England, the fountain-head of blessings descended from our Fathers, and of blessings flowing over the world, how could I help breaking the solemn silence and exclaiming: Boston! thy foundations were laid in piety and prayer, in the name of the Father, and the Son, and the Holy Ghost! Thou hast suffered the common lot of temptation, but thou hast not yet utterly fallen; thy prosperity has not yet ripened thy self-confidence, and wrought thy folly and thy downfall; thy candlestick is not yet removed out of its place, but shineth still on thee and on the remotest tribes of men. Thy Saviour is the Saviour of the world. He fulfils his promise and is with thee, and shows Himself the God of the whole earth. Be steadfast to the covenant of thy fathers. Cleave to Him, by whose power thou hast conquered Satan and turned darkness to light in the pagan world. Answer the wisdom of men by thy continued and growing victories in the name of the Father, the Son, and the Holy Ghost. Thy prosperity has been cherished by thy kindness to others; thy piety has been renewed and increased amidst thy care of the pagan world. Thy Saviour victorious abroad, has been received as their God, by increasing crowds at home! Pursue thy work. Guide the nation in publishing the gospel to the world, and thy Redeemer's growing victories will turn the hearts of the children to the fathers, and the hearts of the fathers to the children; and thou shalt be called a holy people, the redeemed of Jehovah, a city not forsaken! A VISION OF THE LAST NIGHT IN THE YEAR. Amidst the musings natural to the last night of the year, I retired to my bed; meditating on the mortality of man, and often repeating to myself the words of the prophet: "All flesh is grass, and all the goodliness of man as the flower of the field:" and longing earnestly to receive within me "the word of the Lord, which endureth forever." I have no recollection of intervening thought until I found myself far distant from my place of rest, standing at the great gateway of the Park of our commercial metropolis. A peculiar and death-like silence reigned through the extensive avenue of which I commanded the entire view—a portentous sign of silence in all its tributary streets. There was no rattling of carriages; no hurrying and jostling of the crowd; no intermingling of voices; no hum of distant business, though the sun had but just passed the meridian, and was pouring its full and unclouded light upon all the haunts and ways of men. I saw only the public and private buildings which human art had reared, as if to mock the frailty of the hands which built them; the trees stripped of their summer foliage, and the withered grass, nature's yearly lesson of mortality to living men; and a mysterious preparation of hearses and mourning carriages, as far as the eye could reach, as though the city were sitting in silent waiting for a universal funeral! I did not muse long upon the scene before me when a general knell struck upon my ear from every dome in the city; speaking in deep and varied tones the general calamity, and leaving minutes of silence more death-like than before, the mournful, meditative silence of 200,000 souls. What, thought I, can be the meaning of this awful silence, this pause of motion and business, this mysterious preparation, this universal knell? Has some fearful pestilence made havoc of the people, some angel of destruction smitten the first-born and changed the joyous city into a scene of mourning and wo? While I was musing, fixed in astonishment, the whole city, as by one consent, seemed to be put in motion. The narrow houses of the dead, apparently innumerable, were brought out from the abodes of the living; I could hear the sounds of universal weeping and lamentation, and felt unutterable sympathy in the public agony. Immediately the death-march commenced, to the different cemeteries, of various processions passing in different directions, without disorder or confusion, moving slowly to the general chime of tolling bells. I attempted to hasten away from a scene which filled me with horror; but I could not escape. Wherever I went the funeral was there; in every avenue, in every street, the same death-like order and stillness, and weeds of mourning, and tolling bells, the same flow of a smitten people to their graves;—to which abodes of silence, the living were everywhere consigning their dead, as it seemed to me past numbering. With the rapidity of thought I found myself transported from one part of the city to another, but everywhere amidst mourning and wo. At one time, I was crushed amidst the crowds of Trinity, until my imagination was bewildered, and I seemed to see strange sights of the long slumbering dead, rising, pallid and half-skeletons, to bid a welcome to this new year's levee. Then I was in the Bowery, arrested by the confluences from the east and west, and south and north, slow moving, whither the dead might find their last, long home. Then I was at the Potter's field. The crowds were immense. The whole city seemed to be flowing to that great field of death. The earth seemed fresh-dug and unfrosted, and covered with the living crowds. The living were as silent as the dead; no sound broke upon the ear but the rattle, "earth to earth, ashes to ashes, dust to dust!" I would have asked the meaning of a scene of wo so mysterious, but I saw it written in lines of anguish, and remorse, and repentance, and resignation, and resolution, and faith, in the varied faces of the living crowds. I seemed to have a sympathy with the common heart; to know the sentiment which the looks expressed. I saw not indeed such marks, as I expected, of overcoming horror for a public and universal calamity, but rather the universal expression which we are wont to see in each particular instance of bereavement; of sentiments which few are able to banish for the season, how few are willing to retain! What convictions of the frailty of man! what feelings of self reproach! what promises and vows and prayers did I see in the souls which passed before me, transparent as the pellucid lake or river! Here and there I saw faith, meekly seeking for the guidance and blessing of an un-upbraiding Father:—looking through its tears as if it wept not, and fixing a cheerful hope far beyond the fleeting fashion of this world, amidst the things unseen. Overcome at length, I hastened away that I might find a place of quiet thought in the winter loneliness of that beautiful promenade, skirted with water on the west and south; wont of a summer’s evening to be thronged by cheerful groups of young and gay in innocent recreation. But the funeral was there! The clear, transparent waters gilded with the sun now hastening to set, showed not their ordinary display of craft of all sorts sporting by wind and steam as if to decorate a holiday. The shipping, moored at the wharves or anchored in the stream, showed no other signs of living beings but colors at half-mast. Here and there, scattered sails and steam-boats, covered with coffins and decorated dismally with palls and filled with mourners, were apparently carrying their bewailed friends to be buried among grandsires and parents and kindred who were gone before: here and there also a few vessels of larger size, from distant voyages, with dead on board, now disembarking, how differently from their hope, when they put merrily to sea! I felt that there was no escape from the horrors by which I was surrounded; no avoiding this awful funeral, this universal knell, still sounding in softened and distant tones, upon my ears; and I sat myself down to give vent to my sorrows in a flood of tears. As I was weeping, I felt a gentle touch upon my shoulder, such as a kind friend might have given who had become an accidental spectator of my grief. I turned and saw a face so lovely, so benignant, as seemed to me more than human: a countenance which could never have been ruffled with anger nor radiant with pride: surely I thought, a ministering spirit; some holy angel, come to unfold the mystery before me, to sooth the anguish of my heart and to aid me in learning some lesson of salvation; as I trust, unseen, he has often whispered instruction and consolation to my afflicted spirit. "What you have seen to-day," said he, "you may be surprised to know is nothing new. All that is uncommon in the scene before you, is, that by my aid the funerals of 365 days have been clustered before your imagination into one. The knell you have heard was the knell of five thousand, the victims of death's daily and common work. No other evil has befallen the city of New York but its usual mortality of one hundred a week. No fearful pestilence, no overwhelming calamity has filled the city with mourning or caused this universal knell. Health and prosperity have cheered the past year. The thousands whose obsequies have passed in vision before you, met their death by the common varieties of human calamity and disease. When the sun cast the shadows last as you now see them, the greater part were in health, and had no reason to expect themselves to be the victims of death. Rapid fevers, and fluxes and lingering consumptions have wasted and destroyed multitudes of the strong active and blooming, who have gone to their graves instead of the infirm and aged whom they were expecting to follow. Some fell down dead suddenly amidst their walks, or conversation, or daily toil; or were blasted by lightning or steam. Some alone and without forewarning, breathed away their lives amidst the quiet slumbers of the night; and heard not the lingering morning call, as it fell again and again upon their dead ear; nor the cry of astonishment and wo, which burst from their friends at the sight of their lifeless corpse. "Some as they died, no matter where or how, were met by the angels. No shock came so suddenly, no blast so terribly, as to elude the care of those ministering spirits who have daily, nightly charge of covenanted souls. Even in the storm and tempest, in darkness and alone, the charged angels covered them with their shields, until they were fitted for their upward flight, then speeded and aided them to the regions of purity and love." I was waiting in anxiety approaching to agony to hear my heavenly guide, speak of those who died unused to prayer; who had never accepted the offered covenant of their Maker, nor welcomed the Spirit sent down by their exalted Saviour: but the foreboding awoke me. As I awoke I found myself still longing to receive within me the ever-living, life-giving word: and saying again, "The living, the living he shall praise thee." "Whatsoever my hand findeth to do I will do it with my might, for I am hastening to the grave." "I will be steadfast, immovable, always abounding in the work of the Lord, for as much as I know that my labor shall not be in vain in the Lord." THE HEIR OF THE WORLD: Matt. xvi. 25. What an hour of disappointment was, the morning of the resurrection; the morning of an undone eternity! I had forgotten that time was to end. I had outlived a multitude of generations, the only survivor of those which had passed away. I had forgotten the word of Him with whom one day is as a thousand years, and a thousand years as one day. I thought within myself as I had done for centuries; "To-morrow shall be as this day, and yet more abundant;" until I became assured that amidst the wreck of human generations there would be no limit to my life; to the enjoyment of the whole world which had fallen to my single lot. I dreamed that night, the last night of my earthly existence, of rest for ages to come. I saw centuries rolling unto me filled with earthly delights; myself admired and served by succeeding crowds of dying men. I dreamed of earthly immortality. As I awoke, I saw the sun arise as I had seen him for thousands of years; and as he ascended the eastern sky, I hailed his presence and traced his path as the witness and the harbinger of my boundless prosperity. My servants shed, as before, the choicest perfumes through all my halls. I was clothed again in robes ornamented with gold and diamonds, and I sat down to a table covered with the choicest viands of the most fruitful climates of the globe; with an appetite as fresh, as in the earliest days of my unchanging and glowing youth. I could command, that morning, the services of the men of every nation. Even the wisdom of the dead ministered unto me; for I was encircled with all the conveniences that art ever devised. The inventions and discoveries which had mocked their authors in all ages, were perfected and left as my inheritance; and adorned and blessed the palace and the domain, where dwelt in unceasing prosperity the heir of the world. I had passed the ordinary limits of human wisdom; and as far excelled the race which died around me in the scope of my mind as in the circle of meaner enjoyments. I had mastered the wisdom of the sages of all times, and had triumphed over their temporary follies. I knew the theories of matter and mind—the philosophy of mineral, vegetable, animal and intellectual nature; and in the practice of centuries had trained my mind to such habits of easy thought and recollection, that I could call up my boundless stores of knowledge or dismiss them at my pleasure. I imagined that I could unite the highest happiness of the angelic, with all the delights of the human nature; and vainly considered myself the favorite of both worlds. At the moment when I was ready to perish, I imagined myself the heir of boundless and endless enjoyments, which should make me happier than either man or angel, than either earth or heaven! I was a stranger to disappointment, sorrow, or pain: I was a stranger to anxiety and fear. I had nothing to awake me from my delusive dream. As the stream of generations floated by me, I remained firm as a rock; fruitful and flourishing forever, like a tree standing by the waters. I saw before me boundless, endless enjoyments, and assured myself of ETERNITY as THE HEIR OF THE WORLD. Yet amidst the profusion of my blessings, and in full expectation of their continuance, I was not absolutely and perfectly happy. I was heir of the world; I had obtained the utmost limits of desire; I had even gained as a perpetual gift what to mortals was only an occasional blessing. I had an appetite that was never palled; a self glorying that was never put to shame; a hope of good to come that had never been disappointed, and which had no misgivings for the future. What was wanting to fill up the measure of my bliss? I wondered what: why I felt inwardly less happy than the heirs of poverty and pain, whose faces shone often with a glory, which showed them while on earth the heirs of heaven. I had all possible prosperity; I even believed my prosperity unchanging; but neither enjoyment nor expectation made my bliss complete. I felt a craving which was unsatisfied, and which sometimes made my expectations of future and boundless prosperity a burden. Yet I had all conceivable means of happiness, increasing around me, every century I lived: and all the happiness which was possible to a mind not filled with the love of God and the love of man: all that was possible to a mind, which had chosen the creature and forsaken the all-sufficient Creator. No gifts of nature or of art could have added to my bliss: nothing could have made me happy without deficiency, but to have chosen God as my Portion and Lord, and to have been a co-worker with Him, in kindness to his creatures. Oh that I had looked upward in season to the Giver of my blessings; and that I had learned to be a ministering spirit, happiest in serving others! . . . . Yet, all that the world could give, it gave: all that the Creator could bestow, upon the soul which departed from Him, He gave to me: and for thousands of years, more in each than in a single one was ever given to any other member of the family of man. Guarded from evils, loaded with blessings, new every morning and repeated every moment through successive ages, I had all possible good, and all possible duration; for my prosperity never ceased until the last sun arose; until the last trumpet sounded, and summoned the HEIR OF THE WORLD, in an instant, poor and miserable, and blind and naked and destitute of all things! It was my guilty and fatal choice, to have my portion in the world. As I rose into life I saw the world inviting a possessor, and all craving to possess it. Some were ambitious of a limited dominion, and bounded domains, under their regal control, for a score or two of years! I made as I thought the wiser choice, of blessings in abundance and without other limit than the duration of time;—in the fullest sense to have my portion in the world. My wish was granted: I sought and I received the world, as my inheritance: without the perplexities and fatigues of power. I perceived at first that my happiness would at some distant period come to an end, and the prospect marred my enjoyment; but prosperity and ease soon blinded my eyes and satisfied my heart: and wherever I was, and in whatever era of the history of men ever crowding to the tribunal of their Judge, I was still unmindful that my prosperity would be closed at last by a day of account, which would make all my past enjoyments a vanity and a curse. I became myself, my God. I thought I was Almighty to secure my existence and to provide the conveniences and luxuries which adorned and blessed my abode. Whatever I desired for gratification or use came at my bidding, until I forgot that the Most High ruled in the affairs of men, and could bring desolation upon the heart of the heir of the world. The casual exclamation of the king of Babylon, became the sentiment of my existence: and for ages I said daily within myself, Have I not gained the whole world, by the might of my power, and for my increasing and endless enjoyment? Nor was I undeceived until ages rolled me to that morning, when I learned, in the twinkling of an eye, the weakness and sinfulness of man and the dominion of God! I awoke as from a dream. From the building of the tower of Babel, to the morning of the resurrection, seemed when it was passed like a watch in the night, or even like the passage of a weaver's shuttle. I could scarcely believe that I had lived longer than the multitudes, who came and passed so rapidly in the short lived generations of men. I can never forget the sudden horror of the coming of the Son of man. I heard the trumpet, and was startled from a dream of thousands of years; I had not time to collect my thoughts before myriads were gathering around me, from every wind of heaven, to meet the Lord in the air. My first thought was that the dead only were summoned; for while the living stood in admiration and awe, countless multitudes arose from their graves. The sea and the land gave up their dead. But another blast shook the heavens; and in an instant I had travelled all the distance between the earth and the seat of judgment. I looked back upon my inheritance with anguish and dismay. Fires were bursting from its mountains and wrapping all in one mass of flame. I was petrified with horror. I cast my eyes around, wishful for a refuge. But all was gone. I was surrounded by multitudes innumerable. All that had ever lived from the morning to the night of time, were now gathered around the proud heir of the world; how humbled now! Not one could befriend, or serve, or even soothe me now. My memory traced the whole track of time, but I could not bring back a moment of lost and departed ages. I could not recover a morsel of my boundless stores, which but a few moments past I held as my everlasting possession. I was as desolate as the heirs of poverty; as helpless as an infant. I looked around me. The throng was endless: downwards, upwards, around, the whole arch of heaven was crowded with the multitudes of the living and the dead! But my thoughts centered on myself. Amidst multitudes unnumbered, I felt alone. My soul was so fixed upon itself, so overwhelmed with the wants of the hour of doom, that I forgot the presence of the whole family of man, and even their existence. All possible anxiety was centered in the soul of the disappointed heir of the world! In that sudden horror, I had no eye to see but one of the human family; that mysterious One whom I had neglected in the season of my boundless prosperity; that mysterious One who was the temple of God's merciful presence with mankind; by whom, I and all had been summoned suddenly to judgment. In my hour of agony, I turned my eye to this Brother of the human family; to the Almighty friend of man; and I expected that He who sought so intently the most unworthy of men, and even turned earth into heaven for a thousand years, would smile kindly upon me. But as I looked, his face was turned away from me. I saw his right hand stretched forth to a multitude with shining faces, with palms in their hands and crowns of victory on their heads. I saw that I had no part nor lot in the friend of man. For the first time I wished that affliction had been my portion, that I had been tried in the furnace and brought forth as gold. I stood aghast, expecting the Judge to turn to those on the left hand. Alas that day! the day of judgment! It is called a day. I had no measure of its duration; for I had left on flame the ball which had measured time in its daily and annual revolutions: but that day seemed longer than thousands of years preceding. It was a life of horror: one moment of which was enough to overbalance the profit of him, who, having gained the whole world, was come to the day of the loss of the soul! I was filled with sudden, bitter, incurable remorse. For thousands of years I had felt no compunction. I had not imagined myself guilty of a crime. I had lost the apprehension of the law and the Lawgiver. I had become supreme to myself, and my desires were the rule by which I measured my heart and my life. I thought for myself, I lived for myself, I could conceive of no sin except against myself. By the peculiarity of my condition, by the curse which I had chosen as my blessing, I could not even sin against myself; for as I gathered around me the choicest blessings of the world, I was taught instinctively to use them only to the point of their highest and most enduring gratification; and to pause within the boundary of their abuse. From the hour of my fatal choice, abused nature never reminded me of the folly and the guilt of departing from the source of all good. I was left to the final breaking up of my inheritance, to learn that I had sinned against God, and thus against myself. As I saw the throne of judgment, and the Friend of Man seated on it, as the Judge, my moral sense awoke. All the claims of my Creator, Preserver and Redeemer rushed upon my mind. Alas! I had passed my opportunity. I had refused to take shelter under the wings which were spread out to protect me. I had doomed myself to sin and anguish and despair. I was self-condemned. I was guilty before God. Yet when I looked, I saw in the hands of the Judge the prints of the nails which bore witness of his love. When I saw those hands pointing to the countless myriads of the redeemed, the celestial road to the kingdom prepared from the foundation of the world, I wished for a moment to mingle in that tide of happy beings, which was beginning to flow towards the ocean of eternal love. But in an instant I was shocked at the thought, and I felt that if the Redeemer should turn towards me with the words of invitation which I heard echoing from the tongues of angels along the ranks of the redeemed, I should be unwilling to accept the blessing; that the heir of the world could not become the heir of heaven. I felt that I was a moral ruin. I could call up no feelings which betokened restoration; which seemed the germ of a recovery. I had been satisfied with myself because nature and art and earth and heaven conspired to render me happy. I sought to quell the tempest of my soul; to be again proud, and selfish, and happy, as I was when all nature and all time ministered unto me. But the ruling principle was gone. None ministered unto me. I was left to myself; with habits and desires which craved to have the Almighty and all his creatures the ministers of my happiness; and vainly to call it all my own. I had destroyed myself. I had chosen the whole world and lost the vigor and the virtue of my soul. I suddenly discovered that I was a victim of my folly,—I was vile and unhappy; a troubled sea, casting up mire and dirt! But I was roused from my misery by the voice of my judge. I saw that his power and justice were to give form to the curse, which I had wrought for myself, and sunk into a deeper despair, because I was to be doomed an outcast from heaven. I heard the charge and joined in the general plea, "when saw we thee a hungered, or athirst, or a stranger, or naked, or in prison, and did not minister unto thee." Alas! I had been too blind to see the Redeemer in his suffering disciples. I had been too busy in seeking the provisions and embellishments of my palace, to perceive his secret presence in the hearts of men. Even when he reigned over the earth for a thousand years, and united all nations in a harmony never known before, I saw that glory rise and prevail without ministering to the Prince of Peace, for that glory added nothing to the splendors which had encompassed before the Heir of the World. I remembered my neglect and unkindness. I cast my eyes along the ranks of the redeemed, and knew again those whom I had known before; but I could not see one whom I had urged or encouraged onward in the path to glory; one to whom I could appeal for testimony against the dreadful charge. Alas! it was my miserable choice to be ministered unto: I had no time—no means—no inclination to minister to others. No looks of gratitude for kindness from me, shone forth along those ranks of the redeemed; yet there were looks of pity, which showed me that if it were possible they would even then have ministered unto me. I rose almost to hope at that look of deep compassion, until as I turned I saw the gathering tempest of reproach and hatred, which covered the ranks of the condemned. It might not have been so; but it seemed to me that every eye was fixed on me with bitter reproach and hatred. I know not who spoke—or if all spoke—but it seemed to me as if the whole arch of heaven rung with their curses—and with threats of bitter and eternal vengeance upon me, as foremost in commending the fatal choice which millions made: who, ever exposed to the common lot, lived and acted as if their houses and their lands and their prosperity would never have an end. It seemed as if hell was beginning, in those looks and cries and threats, which began to fill the miserable throng; that midheaven would be its place, if there we were permitted to remain. But the Almighty Judge did not suffer us to disturb the harmony of his kingdom: he had descended to the world to save us; he had died to prove his love and to open the way of salvation and to set our feet in the way of peace; he had risen from the dead to commend his power; he had ascended on high and poured out his spirit; his spirit had strived with us that we might be saved; but alas! we had destroyed ourselves. No prison was prepared for us; but we were driven away by the gentle Shepherd into everlasting fire, prepared for the Devil and his angels. I was carried onward in the throng. It was a tumultuous sea; billows rolled on billows: surges of a countless, restless multitude, driven as by a tempest towards the place prepared for the Devil and his angels. I felt eternal misery settling on my soul. I became frantic with the sympathies of countless wretches. Within I was a troubled sea; without, the wicked were a troubled sea on which no calm could come, no quiet sun arise. Oh to have avoided the horrors of that hour—if hour it was: as it passed it seemed eternity. My inheritance of the whole world had vanished like a dream. I was awake and I was wretched. I was absorbed in the misery of the present. Eternity seemed wrapped up in the very morning of its being. The certain anticipation brought the horrors of boundless futurity to my bosom. All that I could conceive of duration of sin and suffering was present to my soul. In that first agony I thought I felt eternal wretchedness. The prison doors were closed! The sentence, the prison, the company, the prospect, renewed and increased the agony; especially the distant view of the glory of the blessed. I extended my eye across the impassable gulf. I heard, or thought I heard, the heavenly harmony. The pang was unutterable when I turned my thought within, and saw the desolation of my soul. Wretched in my prison, whither could I flee? I saw that if heaven were brought nearer, and I were its inmate, that harmony could never soothe my soul. Which way I looked, was hell—myself was hell. I felt within a worm that cannot die; a fire that cannot be quenched. I fed the worm; I kindled and fanned the fire. I saw that I had bred in my soul the passions which now raged beyond control, beyond recovery. Yet the thought of relief did for a moment come in amidst the first agony; and I looked again to find the Great Shepherd—the loving friend of mankind. I could see far distant the seat where he used to be exalted as a Prince and Saviour; but he had resigned it. There remained no signs of intercession—no descending Spirit—no gift of repentance. I know not where I am in the progress of eternity. I know not how long I have been wretched. I feel as if eternities were passed since I closed the thousands of years which are vanished like a dream. Alas! ETERNITIES, ETERNITIES are still before me. What is my profit, to have gained the WHOLE WORLD and lost my own soul? PHILADELPHIA AND THE SUNDAY SCHOOL UNION, MAY 24, 1831. At the distance of more than three hundred miles, let me in my chamber catch the hallowed sympathies, and form the holy resolutions, and pour the earnest prayers which adorn the annual festival of the American Sunday School Union. Let me hear and obey the soft whisper, borne by the gentle breezes over all the land, Suffer the little children to come unto me and forbid them not, for of such is the kingdom of heaven. It is not without visible design, that divine Providence has allotted to each of the three principal cities of the United States, one of the three leading religious enterprises of the day. These enterprises are managed probably with more simplicity and energy than could be expected, if they were all seated at our principal metropolis. The most gifted and devoted citizens of each, can now exert the full vigor of purpose and enterprise, unweakened by excessive labors and undistracted by multifarious cares: devoting to one high design their best affections and thoughts. Each of these fountains of moral influence, also, is thus furnished with the best means of inward purity and fullness. The main organs of the public life are kept in health by the purposes and the doings of an active benevolence, and may thus transmit vigor and virtue to the extremities of the republic. There is a beautiful fitness also in the actual allotment, worthy of the wisdom of its Divine Author. Boston, with New England, of which she is the metropolis, inherited from her ancestors, the principles and habits which qualified her to lead in the work of Foreign Missions. Her means of education, and her ancient habit of applying those means to a preparation for the ministry, have furnished her with a more abundant supply of the ministers and agents of religion. Her examples of early regard to the heathen tribes, and of systematic and noble charity, have raised up among the rich many prepared to give birth and growth to enlarged plans of education at home and missions abroad; and her general population to unite their smaller alms into broad and deep streams of Christian charity. At the same time, Boston needed to see the Redeemer go forth and prove Himself the God of the whole earth. Let her light shine forth more and more upon the Pagan world, and, as hath been for the last twenty years, will it be reflected more and more with healing on herself! New York, is the proper centre of the Bible Society; sitting at the confluence of many waters from every extremity of the country, she is the proper organ of all classes who bear the Christian name, in giving and receiving the word of life. Sitting by the side of the sea, the metropolis of a great Christian Republic, she offers herself as an Almoner to all nations of the only charter of our liberty, The Bible! As to Philadelphia, her very name befits her chosen office, of training successive generations of the young, to live in love on earth, to dwell in the regions of love forever. Her founder, the benevolent Penn, prepared in the western world a city for the device of the benevolent Raikes; styling her, as if with prophetic foresight of her future glory, Philadelphia, the city of brotherly love. As that name occurs, one seems to see the gathering crowds of savages filling the forest of Coaquannock, coming to meet the man of peace and love, under the wide spreading branches of the elm; and William Penn, unarmed and fearless amidst a warlike host. One seems to hear his unsworn promise of love and kindness, and their assurance of love returned to Penn and to his children, as long as the sun and moon endure! As that crowd of savages retired, loving and beloved, Philadelphia arose, how worthy of her name; how worthy of that office to be conferred upon her future sons, loving and beloved, in deeds of Christian kindness to the young, until all shall know the Lord, from the least to the greatest! Is it a fancy of the writer, or is there not something in the very approach to Philadelphia, which prepares a stranger to sympathize with her in her maternal care of the rising race? I have never approached that city from the south and west; but I need not tell those who are familiar with the journey from New York to Philadelphia, how agreeable the contrast after the jolting and dust of a hurried ride from the Raritan, when one finds himself passing rapidly down the Delaware in a swift and mighty chariot on the waters; now gliding in the channel, now shooting to the western, now to the eastern shore; discerning villages, and villas, and farms, and gardens, in all the beauty of the most verdant season of the year. The entranced traveller forgets the horrors of the road, as he rides triumphantly on the waters amidst the ever changing beauties of nature and art, growing on his eye, until he enters calmed and quieted, the city of BROTHERLY LOVE. Is it fancy again, in sympathy with the calm and quiet spirit of her founder and her first settlers, or is there not a fitness for her lovely office, in the regularity of her arrangement, in the evenness of her surface, in the uniformity of her aspect, in the gentleness and beauty of her whole appearance, sitting on her quiet waters; and in that fine arrangement by which she pours from her side, through all her habitations, streams of cleanliness and health and security? Wears she not thus, an air of dignity and grace, suited to her noblest character as the quiet, loving, faithful mother of the rising race of our great Republic? Lovely city, PHILADELPHIA! Lament not thy loss of the Capitol; nor long to be the seat of political power. Divine Providence has given thee a nobler office, a more commanding influence, a higher glory. Thy worthiest work of power was the promulgation of the Bible to our needy land:* thy lasting glory shall be to prolong and perfect that work in thy present exaltation. Let thy doctrine drop as the rain, and distil as the dew, as the small rain upon the tender herb. Thus shalt thou overspread the land with piety and peace, and be more honored and beloved, than to have dwelt for ages the civil metropolis of our growing Republic. Thy seat is Mahanaim, whence go forth over all the land, seen and unseen ministries from the Lord's Host. Angels hover o'er thee, and amidst thee, and with thee, guiding and aiding the labors of thy love. Thy founder, Penn, thine exemplar, Raikes, it may be, visit thee, as their chosen spot of heavenly ministry! Yet they see thee not, thine angelic guides and guardians see thee not, altogether such as * The first English Bible ever printed in the United States, was published by Atkins, at Philadelphia, 1781, a time when the supply from England was interrupted by the war. It was recommended to the people by an Act of the American Congress, signed by their Secretary, Charles Thompson. A copy of this Bible, was used by the gentleman who stated the fact to me, and was worn out in the service of his family; he cherishes the most lively regard for this Bible of his childhood, and has purchased and preserves a copy. I trust the present good work of Philadelphia will appear in like ripe fruits fifty years hence, and in far greater abundance. thy founder wished, as angels can rejoice over. But amidst the scenes which now delight my fancy, I will not know thy spots; save that I cannot hide that pile, which opposite seems to claim pre-eminence over the temple of the Sabbath schools;* herself a temple for the young, patroned by adult and hoary devotees at her honored shrine. Tell me then, ye citizens of Philadelphia, ye ministers of mercy and of justice, is she also a tender, nursing mother to the young, blessing them for time and for eternity? Or like her sister Theatres, does she also hide eternity behind the curtains of time; and even involve the interests of time in a dream of pleasure; and in that dream of pleasure is she the pander to drunkenness and debauchery; the tempter to all crimes, the mother of all ruin, ever covering her base imposture under the pretence of morals and refinement? Is she like her sister at the Park of our great commercial metropolis, who faces the hall of justice, with bold front, while she fills her purlieus with temptation and sin, and nightly trains the successive victims of her evenhanded neighbor? or like Tremont, look- * Chesnut Street Theatre, opposite the Sabbath School Union House. ing askance on the green fields of Boston, as half ashamed, to have broken her promise to the metropolis of the Pilgrims? Methinks, as I ask these questions, I see the lovely youth, fresh from his mother's bosom, from his father's counsels, listening to the call of this rival temple. In loud and stubborn tones, and with an impudent face, (how unsuited Philadelphia to thy name of love, to thy mild and gentle manners,) she proclaims, "I have peace offerings with me, this day have I paid my vows, therefore came I forth to meet thee, diligently to seek thy face and I have found thee. Come with me, and I will show thee good from evening to the midnight hour—I have sights for the eye, music for the ear, a feast for the soul, and with me all that appetite can crave or lust desire." Alas! he stops and listens; her words are already music to his ear. He enters, and again he enters, and again, the hall of temptation; at first weakened in his mind, and perverted in his moral sentiment, by late hours, excessive amusement, and the unchristianised, paganised, profane drama; above all, by the voluntary loss of all opportunity for sacred reading, meditation and prayer; at length, the spectator and the partner of intemperate revelry, and the victim of the harlot’s wily invitations. Alas! temptation produces desire, desire ripens into crime; and health, and morals, and youthful hope, are sold for the pleasures of a moment; are bartered for misery and ruin! It may not be so with every youthful visitant. The grosser temptations may assail, and not overcome. The inmate of the Theatre may remain safe from intemperance and debauchery; may prosper in business, acquire wealth, and live and die on the heights of prosperity:—still an inmate of the Theatre! But will he be a Christian? Dear youth, what will be thy calamity, if thy evenings should be too busy and too happy for prayer, and thy mornings too feeble and worn;—thy Bible a sealed book, thyself without religion and without God! without faith and without hope! What though prosperity should shine upon thy path, and pleasure gladden it, and pride should tell thee that the glory is all thine own? What will soothe thee when sickness or calamity cometh suddenly upon thee? Where wilt thou make thy refuge until the storm be overpast? Or if sickness and calamity should spare thee, whither wilt thou flee when the almond tree hath blossomed, and the grasshopper has become a burden, and desire has failed, and thy imagination sees before thee, thy long home? Where then will be thy refuge? Who then will be thy friend? Or, when the last trial comes, in youth, maturity or old age, what shall comfort thee in that dark hour, if the Lord is not thy shepherd, if thou hast not a Saviour? Or at the last great day, what will cheer thee, when the Saviour says—"Thou didst not minister unto me." Or in eternity, whence will flow thy bliss, if God is not thy God, and sin remains a never dying worm, a never quenched fire! But list! another voice: "My son, keep my words, and lay up my commandments with thee. Keep my commandments and live, and my law as the apple of thine eye. Bind them upon thy fingers, write them upon the table of thine heart. Say unto wisdom thou art my sister, and call understanding thy kinswoman." Oh, there is a wisdom in the providence of God, which places the antidote adjacent to the evil. Rather what wisdom in that arrangement, which has extended the call of wisdom from this blessed centre, over the whole nation, and invites and allures the young before they meet the temptations to folly, intemperance and debauchery; before they have learned in the toil of pleasure, to pervert nature's seasons of meditation and prayer; the morning and evening, before they have lost nature's opportunity for religion; the bloom of youth! Lovely temple! Best ornament of Philadelphia, the city of brotherly love! Thy ministries are ministries of kindness; pure, peaceable, gentle, easy to be entreated, full of mercy and good fruits, without partiality and without hypocrisy! Thou causest short, hasty life to rejoice in the blessings of industry, prudence, and piety, and its greatest sufferings to prove short means of everlasting holiness. Child of the Sabbath school; youth, taught in the word of God, in the law of the Lord! meditate therein day and night; then thou shalt be like a tree planted by the rivers of water, which bringeth forth its fruit in its season; his leaf also shall not wither, and whatsoever he doeth shall prosper. The storms of life may thicken over thee, and the tempests threaten thy destruction; but thy roots shall be strengthened, and thy branches refreshed, and thy fruits ripen for immortality. Who shall harm thee in the hour of peril and calamity? What shall affright thee in the hour of death? And when the last breath of life departs, who shall hinder thy angel friends from bearing thee to the bosom of Abraham and the saints made perfect? And when thy spirit shall enter its glorified body, who shall forbid thy standing at the right hand of the Redeemer, or shut against thee the gates of the kingdom of heaven? Yes, Raikes, that angelic whisper, Try, heard by thee in the streets of Gloucester,* has been renewed from the city of Penn to tens of thousands. Holy resolution has been roused; faith working by love is carrying Christian invitation and instruction to every city and village and hamlet, turning the hearts of the children to the parents, and the hearts of the parents to the children, and the disobedient to the wisdom of the just. At that word, from the haunts of poverty and vice, children are called forth into the school of Christ; and returned to win their parents by their gentleness and obedience. From the abodes of a mere earthly economy, and neatness and thrift, where parents and children together, are neglectful of the Bible, indifferent to * "I can never (he remarked to Mr. Lancaster) pass by the spot where the word 'Try' came so powerfully to my mind, without lifting up my hands and heart to heaven in gratitude to God for having put such a thought into my heart." eternity, the children are called forth to consider their latter end, and returned unconscious prophets of another world. Even from the temples of domestic piety, the children are invited to receive the testimony of beloved friends, to those precious truths which they have been taught from their infancy, and to be aided by the sympathy of their companions. Happy the families thus preserved from examples and sympathies abroad, which formerly were wont to hinder or to blast the utmost care at home. Blessed schools, which restore the family to its proper work, or with skilful hand, feed and guide the well tended lambs of the parental fold. Try! Let that word which gave existence to the Sabbath schools, perfect their power! Experiment, patient and untiring! What discoveries in science has it revealed! What power has it prepared in the works of art! From the simplest elements of thought, it has furnished a measure for the spheres. From the first rude inventions of untutored man, it has contrived the finished machines which have multiplied ten thousand fold the conveniences of life! Try! Teachers! Let that word speak to your very soul, as it spoke to the soul of Raikes. What wonders will it work in you; what wonders spread around you! Make experiment on yourselves, and see what stores of knowledge you can amass; what aptness you can acquire to win and teach the tender mind. Examine the works of God; above, below, around. Range the walks of history. Study the Bible, and with it nature and man, by which it displays the character of God. School memory, to ready and skilful recollection; imagination to the conception of all that God reveals, and let faith guide it to the substance of things hoped for, to the reality of things not seen. Let reflection prepare and lay up wisdom in its deep treasury. Invention, let it never sleep, until all is contrived for the benefit of man, nor industry slacken until all is done. Piety, humble, fearful, trembling in its weakness, let it try the strength of the Almighty! Shall a successor of Raikes be dull, and dole away the hour which he and his pupils alike wish to be over! Shall the volunteer instructor of an immortal mind, be satisfied with the doing of a task? and not try to train himself to his high calling? Who is already wise? Whose sanctified fancy holds clearly enough the things unseen; or reason judges, or invention reaches to their utmost limit? Where is the Sabbath school which does not fail to interest and bless the young immortals, for want of intelligent, interested, conscientious, believing teachers:—because Try has not fallen upon their souls as it fell upon the very soul of Raikes? Parents, try! That angel in the streets of Gloucester was no demon of temptation; no patron of parental ignorance and sloth; no wolf to tear away the lambs from the parental fold. The shepherds were not caring for their lambs; did not gather them with their arms, nor carry them in their bosoms, but left them to stray through brake and brier and slough, far away from the green pastures and the still waters of salvation. Raikes saw them, and opened to receive them a little pasture beside a living spring; that they might not perish, and that nature's shepherds might be warned and won to the care of their tender lambs. Parents! be aroused and encouraged by this tenderness and care. Hasten to restore the fold, and open the pastures and fountains of domestic piety; and all the Sabbath and all the week carry the lambs in your arms and gather them in your bosom, and feed and refresh them around your door. Let home, sweet home, be the school of knowledge and religion, furnished with their richest treasures, with their aptest teachers. Let the sitting in the house, the walking by the way, the evening's lying down, and the morning's rising up, find the law of God dwelling in your heart, and flowing forth upon your children. Bow the knee in prayer; and draw down upon the shepherds and their lambs a blessing, even life forever more. But imagination bears me to-day to the most interesting single scene which Philadelphia ever displays—the annual convocation of the American Sunday School Union. The sun, great instrument of light, renovator of the seasons, shines into my chamber, while at this instant it cheers the path of the multitude, as they are thronging the streets of PHILADELPHIA: while a choice number from all parts of our land are hastening to the house of God. Could there be a fitter place for this annual convocation, than this sacred solitude; amidst, yet far from the haunts of men; amidst, yet secluded from the din and bustle of the crowded city? As we go up the ascent to the house of God,* we turn and survey one of the finest fields, where nature has shed her choicest beauties over the handy work of art; covered with verdure unexcelled waving in the soft breeze; and studded with trees and shrubs in every variety of beauty. From every stem and leaf and flower the unclouded sun shines forth with gentle radiance upon the cheered eye of the gathering crowds. Oh who could look for a moment upon his dazzling brightness, as he now shines in glory from the heavens. Yet with what a cheered eye and gladdened heart we receive his rays, reflected from this beautiful and brilliant parterre—from this blue canopy above. Even so within these courts of the Lord, shall we receive the healing light of the Sun of Righteousness, reflected on our sight from the nurseries and gardens of youthful piety. Even so will we, as their Patrons and Cultivators, shed forth the healing lustre of Christian example and influence. In our place and station we will receive the glory of the risen sun, and shed its softened radiance on the world. Here we enter thy courts, King of kings! We meet to hear thy voice, of such is the king- * Presbyterian Church, Washington Square. DOM OF HEAVEN. Oh how changed art thou, since thou didst lay thy hand upon the little ones: now that thou sittest on high: and thy countenance is like the sun shining in his strength. Look not forth upon us in that glory which cast the beloved disciple at thy feet as dead. Let us see thy softened light in the holiness of thy disciples, and let us reflect that light upon the world. Help us in our families—in our schools—the elder and the younger together to receive and reflect thy healing light! Amidst the interests of this holy convocation; its earnest longings for a blessing on the young and on the old; its personal penitence and faith and holy resolution, the sun is descending towards the mountains of the west. The light is softer; the air is cooler; the hour is fitter for holy consecration, while as with one heart we say in silence, The living, the living they shall praise thee, as we do this day: the elder to the younger shall declare thy truth.... As we rise to depart, the evening shades are beginning to cover the earth: the sun is gilding the mountains and the rivers of the west with its setting glories—a sign of the soft influence of the Sabbath schools, as it shall fall over that mighty valley which is the object of our councils to-day. The night is coming over us and we are to be cast again upon the protection of the Almighty, the Lord and giver of life. Surely he is not withdrawn, for he is bringing forth the full moon to pour all night the reflected light of the sun upon the earth; and the stars, to show himself to feeble man as the Lord God of Hosts. We will both lay ourselves down and sleep in peace, for thou Lord only makest us to dwell in safety. Oh if we arise to see once more the light of the sun, and again and again arise from our nightly death; as we breathe the air and live upon the bounty of the Lord God of hosts, we will say by thy grace, The living, the living they shall praise thee: the elder to the younger shall declare thy truth. If we rise again, we will go forth in the spirit of the departed Raikes, and try, until from the Atlantic shore to the farthest regions of the west, the whole land shall be filled with the public and the domestic schools of piety: and the Sabbath and the week shall unite their power to instruct and bless the rising race. "Then shall our sons be as plants grown up in their youth, and our daughters as corner stones polished after the similitude of a palace. Happy is that people that is in such a case: yea, happy is that people whose God is the Lord! THE FIELD OF DEATH. Dated New York, April 6th, 1822. It requires no time for imagination to call up unseen realities, or to transport itself to past and future ages; and faith can rest upon things unseen, and upon the most distant futurity, as intently and as firmly, as sight, upon present things and passing events. It is profitable in the highest sense, amidst the cares and bustle and hopes of this world, to indulge, not a wild, wandering imagination; but a sacred and scriptural imagination, which calls up in authorised forms those unseen realities, which bear an awful relation to our present condition. Who that will do this can have his mind engrossed and enslaved by "things seen and temporal?" While he mingles with unseen and future realities, and dwells upon the hidden scenery which Revelation discovers, he feels the sacredness and the responsibilities of his passing hour. Born but yesterday, and living for a day, I need not bury my soul in the present and the visible. There are other realities, in which I am infinitely more concerned. I love to recall the imagery of ages past, which history and Scripture warrant, and to throw myself into the midst of that untried futurity of which inspiration pourtrays the instructive and awful scenery. Often when walking the streets of our great and busy city, do I turn away my mind from the passing scenes, and lose myself in the vivid conception of the unnoticed realities which have a present existence—of the revealed realities, which I conceive are yet to have existence on the very ground we now tread upon—and never with deeper interest than upon that "Field of Death," situated at the very confluence of the business and pleasure of our city.* From what a commanding position, and with what a ghastly aspect, does Death look down upon the great arena† of our city's business: how he seems with his hundred thousand victims to block up the very avenue of our wealth, our speculations, and our commerce; and what a chilling frown meets the expecting and the successful speculator, as he strives to pass by, unheeding and un- * "It is ascertained that more than 100,000 persons have been interred in Trinity Church-yard alone; and it has long been impossible to inter one, without disinterring another." † Wall-street. heeded! With what a ghastly aspect looks he on the ceaseless stream of parade and fashion and pride, along that brilliant thoroughfare, over which he presides!* With what a chilling frown he meets the glee of folly, the swell of vanity, the gaudy trappings of the gay, morning or evening, as they pass, by thousands! Did they see the frown of Death—did they see his heaps of long slain victims, piled rudely upon heaps—what an appalling thrill and dread would come upon the throng who give all their activity and earnestness, to the gettings of this fleeting hour! How are their eyes holden that they see Him not!—Was it for nought that He who directs the destinies of our citizens, whose Providence concurs with the gracious invitation of his word to lay hold on eternal life, should have placed on such a commanding eminence, Death’s crowded field! Oh, was it for nought that the parade, and fashion, and pride, which glory in the vanishing realities of the moment, must all pass the Field of Death!—that vanity must strut and vapor its fleeting hour, that giddy thoughtless folly must all day long tread on the border of the Field of Death! * Broadway. How silent is this crowded field! Was ever crowd so still? They utter no complaints, they teach no lesson, save with the silent eloquence of Death. Once they were moved by our motives, and mingled in the affairs of men:—but now how silent! The rattle of the crowded streets disturbs them not. The anxieties, and hopes, and desires, which hold the passing throng of living mortals in eager pursuit, and which mingle with the air from a multitude of tongues, as all day long the ceaseless procession of "a proud and gay and gain devoted city" passes on, disturb not the breathless silence of the Field of Death. If the sun shine, or the heavens blacken; if summer cover their habitation with the green grass and the flowering shrub, and gently wave the foliage of its elms and willows with its southern breeze; or if winter strip away the vegetation of the Field of Death, and overspread it with an icy covering, howling among the surrounding tenements of the living, around the Death environed church, and across the graves—this mighty congregation heed it not:—The Sabbath, with its symphony of prayer and praise, seeming to disenthral the Death-field of its terrors; the week day, with its jar, and noise, and confusion—are alike to them. Along the avenues of business and pleasure they cast no anxious looks. Who owns the real estate; who manages the banks; whose richly laden ships come safely into port; who rolls in splendor, and who rules the state—they ask not, care not. Gathered from the east and from the west, the children of two continents and many islands here sleep together. The rich and the poor, the honorable and the base, crowd each other in their narrow house—moulder and mix their earthly frames together. They lie unnoticing, but not unnoticed. He, who sustained their living frames, watches their sleeping dust. The fair and garnished habitation here lies in ruins. But the immortal tenants stay not in this Field of Death. It is sown with the bodies, not with the spirits of men. They live, but where? Oh, who can tell me where? Live they in the bliss of Heaven, or in the pains of Hell? Oh, could the Death-Field speak and tell the history of its hundred thousand—what mingled sounds might break upon our ear! What wailings might issue from the tombs of the gay youth, of the prosperous speculator, of the rich capitalist, whose souls were required of them in the moment of their vainest hopes! What songs of the youthful and aged, of the rich and poor, of the prosperous and afflicted, who were found watching and waiting for the coming of their Lord! We know not where they dwell. Yet we believe, that to every ruined habitation, there is in some untried region, a living but absent spirit. As many bodies moulder sown in earth, so many spirits live, awaiting the day of their re-union. Immortal spirits wait to re-occupy their long deserted habitations, and often cast a look through the distance to those mouldering ruins, once so precious, now so vile, and yet again to be restored by the power of God. When many ages more shall have rolled over the sleeping thousands of this crowded Death-field; when the vacant squares of our wide extended city shall be each a garden of the dead; when many millions shall have slept with us and with our fathers, and mingled quietly with the dust of the Island of our habitation; when, after many ages of the reign of Jesus, countless millions await a glorious resurrection, and Satan for a little season beguiles again the sons of men:— Then, on such a morning as I now behold, when the sun is mounting towards his midday height, tracing his track, trodden from the foundation of the world, and seen as he has been seen in the splendor of an unclouded morn by the countless millions of the human race—when busy crowds are careering their way along our avenues of business and pleasure, and their hearts are beating high with hopes of long life and large inheritance: Then, will the heavens thicken with a mighty cloud, and in the twinkling of an eye a trump will sound, arresting in the street and in the house, on the land and on the sea, the thousands of our city, and the millions of our world—and every eye from pole to pole shall in an instant turn to gaze upon the portentous signs of an abused or received Redeemer in the air: Then shall the hidden dust ascend—a great and living congregation from the Field of Death—a spectacle to an amazed city, mingling in the rising ruins of the whole family of man—a momentary prelude to the ascension of the living and the conflagration of the world. Reader, thou wilt shortly be a tenant in a Field of Death. In the day of thy soul’s reunion to its mouldered body, mayest thou hear thy Saviour say, “Enter thou into the joy of thy Lord.” NEW YORK AND THE AMERICAN BIBLE SOCIETY. MAY 12th, 1831. The following article was forwarded for the New York Observer of May 7th, but did not reach its destination. It is preserved in its original form, as a memorial of the spiritual blessings of the spring of 1831, and a grateful recognition of the signs which claim the faith and zeal of the revived church. Years have passed since the writer of this article has had the privilege of mingling with his fellow Christians at the Anniversary of the American Bible Society, around which so many other interesting convocations are arranged like clusters on the vine. Let me in imagination be seated under the shadow of it, and let its fruit be sweet unto my taste. I shall conceive the 12th of May, 1831, as ushered in upon the assembled Christians of the country, by a soft and gentle dawning, shining brighter and brighter, until the glowing and unclouded sun shall arise above the horizon, as in the most lovely mornings of that lovely month. That soft and gentle dawning will assure the city of our solemnities, that he is still travelling his wonted and unwearied course, and has reached already the shores of Newfoundland. In an hour, he will rise and shine upon the eastern boundary of the United States. In half an hour, he will gild the turrets of New York, greeting from Christian temples, the Everlasting Sign, the glowing image of the Sun of Righteousness:—Then he will come forth from his eastern chambers as a bridegroom; rejoicing as a strong man to run a race. In an hour, he will rise upon the lakes and rivers of the west; and from every dew-drop glistening in his rays, from every stem and leaf and flower in every hue of beauty, will be confirmed to that mighty valley, the ancient promise of the Almighty: Unto you that fear my name, shall the sun of righteousness arise with healing in his wings. In an hour, the Rocky mountains will reply to his rising beams: With everlasting kindness will I have mercy on thee, saith the Lord thy Redeemer.* In another hour he will reach the shores of the Pacific; revealing to the wild man, the love of God. The hill-tops on which the storm has been raging and the thunders rolling, will glow with the morning sun, and be vocal with the music of the birds, foreshowing the day-spring from on high, and * Isaiah liv. 8—10. the songs of Zion. In another hour, when the friends of the Bible will be assembling from every quarter of the city, speaking one to another the words of those that fear the Lord, and that think upon his name; while the record is making on high in characters of light, and the soft whisper to the soul is heard, They shall be mine when I make up my jewels; the sun will be rising upon the waste expanse of the Pacific, where to his morning rays, ocean, with all its waves, will reply in chorus, While the sun rises and the ocean roars, God will be gracious. In another hour, the day will begin to dawn upon the Christian natives of Oahu and Otaheite, as they go forth amid the palm groves to welcome the God of the morning to their hearts. At that sacred hour, when the Bible shall be opened and its promise read in the ear of the listening crowd, the EVERLASTING SIGN will be declaring its truth, in the morning dawn, on the Christian isles of the Pacific, and in the last rays of evening twilight upon the patient toils of those who sow in hope on the coast of Malabar; covering all America and Europe and Africa and western Asia, with its unchanging testimony to the oath and promise of the Lord: Yet giving its special witness, amid the lingering glories of the far spent day upon the mountains of Israel, and greeting with its special smiles the collected friends of the Bible, in the city of New York. It was not without reason that the American Bible Society, following the example of her illustrious Parent, chose for her annual convocation, that season when the Sun looks forth upon the northern hemisphere, with his vernal radiance and warmth: gladdening the valleys and the hills and the mountain tops: giving the seed time, and the promise of the harvest: everywhere expanding nature into beauty and production. The festival will return again, amidst the unfolding testimonies of the renovated earth. Here, the sun will be entering the pores of the mellowed soil, warming into life-giving death, seeds innumerable, that they may bring forth food for man. There, he will be training the tall wheat, how rapidly from its bed of snow! here, swelling the bud: there, unfolding the flower, or from the embryo fruit dropping the fragrant petals to be wafted in the breeze: everywhere carpeting the earth with grass, and adorning garden and hedge and field and forest, with flowers of every fragrance and color: everywhere receiving from the warmed and moistened earth the responding witness, My word shall not return to me void. On that festal day, while the sun is coming up and overspreading the whole breadth of our land with light, and fertility, and beauty, the mountains and the hills will break forth into singing, and all the trees of the field will clap their hands; while from earliest dawn the birds, in simple notes and oft repeated song, will say on this May morning, Let every thing that hath breath praise the Lord. The mind has fixed upon this morning and vernal scenery, at once beautiful and sublime; awakening the deepest awe and the most softened and gentle admiration; because it is so fitted to win the trembling heart of man to Him, who daily gives in all the habitations of men, the mildest and most powerful tokens of his love; and because the natural sun is the chosen sign and emblem of that grace which the Bible reveals for the salvation of the world. Can any thing in human thought or language be more attractive to the heart, than those scriptures, which represent the Saviour under the emblem of the natural sun? Unto you that fear my name, says Malachi, shall the sun of righteousness arise with healing in his wings. Zechariah prophesied of the day-spring from on high, to give light to them that sit in darkness and in the region and shadow of death; to guide our feet in the way of peace. Simeon, with the infant Saviour in his arms, said, Mine eyes have seen thy salvation, which thou hast prepared before the face of all people, a light to lighten the Gentiles, and the glory of thy people Israel. Peter commended the sure word of prophecy to be considered and pondered, until the day should dawn and the day-star arise in the heart; perhaps remembering the while, that morning of bitter weeping, when from the hall of Pilate, he went forth as with grief incurable:—how amidst the shadows of the moon, descending behind the city of the Great King, he went perhaps again over the brook Kidron, towards the Mount of Olives, dark, gloomy, and despairing;—how the cock-crowing through all Jerusalem cut him to the heart;—how, as he went up the Mount of Olives, crying in agony, Lord, to whom shall I go but unto thee; thou hast the words of eternal life; the serene day-star met his weeping eye, and the dawning of the day came in on that night of sorrow; and that look of the Saviour in the hall of Pilate, seemed changed into the balmy morning of the Sun of Righteousness, rising with healing in his beams! It must be, as in the case of the writer, that this lovely and glorious emblem of the grace of the Saviour will have new interest at the present season, from greater familiarity with that darkness and shadow of death in which the convinced sinner feels himself after the candles of his own righteousness are gone out. It must be that many will be present at the anniversary of the twelfth of May, whose privilege it has been, like his, to come in as a feeble day-star amidst the darkness, and to have witnessed the dawning of the Sun of Righteousness on the night of the soul, healing its agony. It must be that many, as they meet to do honor to the Bible, will say within themselves as they tread thy hallowed streets, Metropolis of Light, "Surely the day has dawned upon our cheered hearts; we have been healed by the gentle beams of the Sun of Righteousness." Doubtless those will be there, over whom many a cloud has passed; hiding for a season his cheering beams: whose eye has often drowsed or slept; yet who again and again have been joyful in his returned or discovered light, varying, yet shining brighter and brighter in the progress to the perfect day. On that hallowed morn the Christian multitude, joyful in the presence of their Lord, will be able to say: We have not followed cunningly devised fables, when we made known the power and coming of the Lord Jesus, but have been eye witnesses of his majesty. The day-star has arisen, the day has dawned, and the healing beams have shone upon our souls. Will not the believing, joyful throng, hear the voice which attends the day: His voice who goes forth as a bridegroom to win each family and neighborhood and town and city, our country and all nations, as his bride? Arise and shine, for the glory of the Lord is arisen upon thee. If there be any occasion when the American church may be supposed, in one universal assembly, to be sitting in the light, and listening to the voice of her heavenly bridegroom, it must be at the anniversary of that Society, which unites ALL who love the Lord, in circulating the book, which bears the name of its divine Author, and is called THE WORD OF GOD:—itself the medium of that light, of which He, the true light, is the fountain and the source; and which goes forth in silent majesty converting the soul. And if there be any anniversary, appropriate to such a call, it must be that which is to occur on the 12th of May, 1831. Amidst the beauties of this spring-time of the natural and spiritual year, she will hear the voice of her beloved: Rise up my love, my fair one, and come abroad. For lo, the winter is past, the rain is over and gone; the flowers appear on the earth; the time of the singing of birds is come, and the voice of the turtle is heard in our land. Arise, my love, my fair one, and come abroad. Arise and shine, for the glory of the Lord is risen upon thee. Never, perhaps, since the world began, did such a spring awake upon the earth; never such a May morning dawned and shone, as hastens to adorn the coming anniversary; never, before the light of the church had been cast upon the pagan world, and reflected back with increasing glory on her sons and daughters around her; and they were seen, as now, flying like a cloud, and as the doves to their windows. With what interest, with what a hearty covenant, with what fixed determination, yet with what trembling self-distrust, and glorying only in infirmity, may we suppose the Redeemer's call will be heard by that representative assembly; will be urged by those who shall be the organs to the public ear, and will be transmitted to the Green mountains and the lakes, to Massachusetts bay and Passamaquoddy, over the Alleghany ridge, along the rivers of the west and the Atlantic shore: Arise and shine, for the glory of the Lord is risen upon thee. But the Redeemer waits not the eloquence of man. Ere yet the tongue of the learned and devout wakes into harmony, the call will have gone over the whole land in the soft eloquence of light: in the truth of an everlasting sign. "Church of the living God, arise. Look to thy heavenly bridegroom with the eye of prayer: as meek, as earnest, as wakeful, as of old, from gloom and suffering, from the dungeon and the scaffold. Shine, humble as the morning, when the dark, colorless earth receives from the sun her renewed robe of beauty; yet, glorious as thy day, more holy, more useful than prophets and righteous men who lived in the dawning. Vaunt not thyself; but call thy kindred and thy neighbors and all nations to the glory of the Lord arisen on thee. Shine in the free light of day. Rush not to tower or mountain top. The light falls on the valley and the plain; is reflected from the lawn, from the field, from the garden, from stem and leaf and flower, as bright as from tower top or mountain brow. If any say, Lo here! lo there! believe them not; but in thy place and station receive and reflect the glory of the Lord arisen upon thee. Abide in the light. Hide not thyself again in the dungeon of declension, or in the chambers of indolence, singing or dreaming amidst thy chosen gloom:—The bridegroom has departed, and seeks not now to win all people as his bride. Shine by the word of God; enlightening the eyes, converting the soul. Wield the arts of men with the mightiest energies of nature, and throw the sun beams into every habitation of men." In harmony with the soft eloquence of light, with the truth of the everlasting sign, how will the assembly sit entranced, persuaded! How melted in holy sympathy! How earnest for the Saviour's glory and for man's salvation! How strong in holy resolution! How sincere the assurance in the felt presence of the all-seeing: Lord thou knowest all things, thou knowest that we love thee! The resolution will be accepted: the assurance acknowledged, as the sun descending towards the west, enlightens the paths of the dispersing crowds,—bearing its witness over the whole continent of America, over the Society and Sandwich islands: while the response will be sung by all the waves of the Atlantic and the Pacific, as they roll basking in the day. Let the assembly disperse in faith and holy resolution. As the day fades away, the moon and the stars will give forth over our land their silent witness the livelong night; while the dawning and glory of the morning will rise upon Japan and China and the mountains of Himalaya and the plains of the Ganges, striking the turrets of pagoda and mosque; and calling all the dark nations to worship the God of the morning, the Author of salvation. Let the assembly disperse in hope and holy resolution. As the evening twilight of the 12th of May shall fade away, the morning will begin to dawn upon the tops of Lebanon and Carmel and Zion! Once more on the heights of Israel, desolate and forsaken, the sign will be renewed, that God will fulfil to his people the promise of the new covenant, that all shall know the Lord from the * Jer. xxxi. 35. least to the greatest; and afflicted, chastened, broken-hearted Jerusalem be built again to the Lord from the tower of Hanemeel unto the gate of the corner: and the whole valley from the brook Kidron be holy unto the Lord! As the American Church rests in the arms of the Redeemer on the night of the 12th of May, the rising sun will strike the minarets of Jerusalem: and the desolate top of Zion:—a sign to the trespassing Mussulman, to the unbelieving Jew, and to thy heralds, word of God,—the sun for a light by day, a renewed and everlasting sign, of the recovery of the seed of Israel as life to the world. Go forth, thou oracle of the living God! Rod of the Saviour’s strength, go forth from Zion. Great sun of righteousness, shine by thy word, and enlighten, and convert our kindred, our neighbors, our country, and the world. Go make the nations thy willing people: in the beauties of holiness, from the womb of the morning, fresh, abundant in the dew of thy youth! On Lebanon, on Zion's hill, Arise and in thy glory shine; Ages have given, oh Lord fulfil Thine own, the everlasting sign! Arise, with healing in thy wings, And comfort Israel's broken heart; While Judah's daughter wakes and sings Thy kindness never can depart. Let Israel's glory move the night From desert and from wilderness; On Zion's top reveal its light, And all the Gentile nations bless. From darkness and from dawn, break forth Thou glorious Sun, converting word! From east to west, from south to north, O'er Christian realm, o'er savage horde. Thro' all our land, thy light reveal; And let its soften'd radiance glow, From mountain, plain, and deepest vale, On which thy beams of glory flow. Lighten the islands of Japan; O'er China's millions rise and shine: At Delhi, and at Ispahan; Around the cross of Constantine. From Egypt to the Caffree's land, From Greece, and Italy, and Spain, To Iceland and the Loffoden, From Finisterre to Astracan. From Alaska to Newfoundland, O'er Christian realm, o'er pagan horde, From Mexico to Magellan, Arise, thou all-converting word! The oath confirm'd by daily light: Confirm'd by every ocean's swell; By all the splendors of the night, Confirm'd for ages, now fulfil! THE BOX OPENED. Dated New York, Oct. 6, 1821. The public have been informed that a leaden box, containing medallions, coins, books, newspapers, &c. has been formally deposited in one of the four pillars, just erected at the great gateways of the Park in this city—deposited as a legacy to a distant posterity. Sublime and solemn act! How many ages may roll away before those solid pillars, tottering by the hand of time, shall be removed, and the legacy be found! When all our present citizens have long—long been buried in the earth; when all that is now great or honorable of our population shall be lost in the night of antiquity; when our never-dying spirits shall have been long happy in heaven or wretched in hell; when the city of New York, long a "city of the Lord," shall overspread the Island of Manhattan, and the millions of her population shall be of one heart and one mind; when the Bible shall be the dearest treasure of every individual, and the voice of prayer and praise, and the sacrifice of obedience shall ascend from every house; when "her people shall be all righteous," her "walls salvation, and her gates praise;" when the eye, surveying the crowds up and down our present Broadway, shall not see one who is not a friend and brother in Jesus Christ; when her thousand temples shall be filled with devout and joyful worshippers, and her Sabbaths a heaven below: When the forests of the west, subdued by the hand of Christian industry, shall have become a fruitful field, and an English and Christian people shall spread from the Atlantic to the Pacific—from Mexico to Labrador; when from a million temples, shall ascend the joyful homage of the American church "in spirit and in truth;" when the gospel shall have been preached in all nations, and received by all nations; when war shall have ceased under the mild reign of the PRINCE OF PEACE, and "every knee shall bow and every tongue confess"—when "One song employs all nations, and all cry, 'Worthy the Lamb, for he was slain for us:—'" When the church, having perhaps for centuries embraced the whole earth, remembers no more the reproach of her widowhood;—when the tradition of a sinful age has ceased, and only on the page of history can be learnt, the folly, and vice, and impiety of ancient times: THEN, since ruin is inscribed upon the strongest monuments; since massy columns cannot stand forever—THEN, if itself survive the wreck of time, will this long unknown memorial of ancient days be discovered by a generation so remote that they cannot trace back their line of ancestry to us; be inherited by a new and holy population: a memorial of the arts and sciences, the heroes and statesmen, not only, but of customs and follies, and vices, which have long passed from the memory and tradition of men. Methinks as the men of other times cluster around the spot where these pillars stand, and listen to some venerable and holy sage, while he describes the manners of the ancients from their bequeathed memorials, that regret, and shame, and astonishment will awhile trouble every bosom. "There," he may say, "stood the theatre, where the mouldered dead, as says the page of ancient story, trifled their short lives away, met the votaries of vice, and drowned their souls in perdition. Yonder stood the debtor's prison;—yonder the bridewell—receptacle of crime:—there stands the ancient hall of justice, now the seat of mercy. Here, on the Park, met and mingled on many a day of dissipation the vicious and profane. All along down Broadway were the lottery offices, the idler's hope, the country's ruinous tax. There poured an unceasing tide of dress, and fashion, and parade—of vanity and pride. Gain was the people's god. Strong locks and bars guarded every house, and were trafficked in every street. Drunkenness was an article of commerce, was bought and sold in every block—the buyer's and the seller's ruin. The sword and spear, the musket and the cannon, which history describes as weapons of murderous warfare, and as beaten centuries ago into ploughshares and pruning hooks, were sold in enormous quantities, and stored in mighty magazines.—Even here, often poured the parade of soldiery; here roared the thunder of arms, in mimic warfare, while music, made for heaven's praise, pealed insultingly in martial strains to heaven. In all the earth, pride and ambition filled the minds of men, and even tarnished the purity of believers. The daily news revealed thefts, and contentions, and murders, from the cottage to the throne; and piracies and man-stealing, and nation warring against nation.” Awful exhibitions of antiquity, to men whose hearts are love; when lust and hate and war are known no more, but lie forgotten like the barbarous rites of our own heathen ancestry. Yet they will dwell upon whatever of great and good they find in the memorial. When all that this world admires has ceased to be admired, the record of the Bible Society will consecrate the Daily Advertiser, which transmits the most glorious of all our institutions to the admiration of a distant posterity:—THE BIBLE SOCIETY, which sends down a stream of mercy, swelling as it flows, turning the desert into a fruitful field, the wilderness into a garden of the Lord. It was an oversight, in sending down the doings of our age to a posterity, who will love the Bible and the Saviour, not to add the religious publications of our city:—the Christian Herald, the Christian Journal, the Methodist Magazine and the Missionary Register, little known indeed to our present generation, but worthy to transmit along with the memorials of our arts and sciences, our follies and our vices, the general efforts which are making to bring to pass among all nations, the dominion of the Saviour of the world. Then might the assembled crowd which I have imagined gathered around the ancient dilapidated pillars at the Park, see that even in our iron age, there was a little of the fear of God, of the love of Jesus, of the faith of good things to come. Such an assembly our eyes never saw. An assembly of which the present dwellers of our city are unworthy to be the ancestry. What emotions of holy gratitude would fill their souls, while looking back upon sinful antiquity! And, assembled on a spot once the arena of folly and sin, would they not spontaneously pour forth a song of praise to the giver of all good. The writer of this article is far too feeble in genius, too low in holy feeling to conceive the strain. Yet he seems to hear thousands of voices, from ground once polluted, sending up to heaven a song of praise in a far higher, holier, more heavenly strain than this: MILLENNIAL HYMN. Christ 's our King: He reigns below: We yield ourselves to Thee alone: Redeem'd from sin, redeem'd from wo, We cleave entirely to thy throne. The locks and bars are all remov'd Throughout our city family: The prisons now are known no more: Love, is our bond of harmony. In all our streets, they love thy name, To Thee, in every house they bow; A thousand temples sound thy praise, Our Sabbaths are a heaven below. The sword and spear, and wrath and pride, The battle-field, the victor's mirth, A country's joy at groans and death, Are known no more in all the earth. The tribes of men are all the Lord's, Earth is his wide and fair abode; The sun in all his daily course, Shines only on the sons of God. I do not know that my sufferings were greater than the common lot. I was less afflicted than multitudes; far less, than those who tremble for years under the uplifted hand of persecution; or who pined in the dungeon, or died on the scaffold, or at the stake; or who bewailed their imprisoned and murdered friends. I was even less afflicted than multitudes, who sought in outward blessings their highest good: whose schemes were baffled; whose prospects were blasted; whose heart was overwhelmed, without a rock of refuge. Yet I was a sufferer, and out of great tribulation, came up hither. I was tried in the furnace that I might be brought forth as gold. I had my share of afflictions which were obvious to the public eye, but my most frequent and poignant sufferings were secret to myself. My heart alone knew its own bitterness, and no stranger was acquainted with my grief. God only was the observer of my sufferings. He chastened me as a father for my profit, that I might be a partaker of his holiness. He chastened me in secret, that I might be forced to tell him my sorrows; that I might not seek instead of Him, some friend helpless as myself, but might be meekly and quietly in subjection to the Father of spirits and live. I was a sufferer even at my best estate. As I arose the mount of prosperity, I was still in the region of cares and woes; of fears and sufferings. As I took my rank among the prosperous, I saw them beset with perplexities and difficulties, peculiar to themselves, and still exposed to the common lot of sickness, calamity, losses, and bereavement. I submitted to a discipline from which there was no escape. I even yielded, often cheerfully and with thankfulness, to the hand which chastened me amidst the profusion of its gifts. It was good for me that I found a thorough intermixture of bitterness in the sweetest cup of life. Thus my longing was increased for the waters of salvation; I was made to hunger and thirst after righteousness until I was filled. But I was often visited with cares and sufferings so perplexing, and so severe, that I was tempted to forget the blessings intermixed. I needed food, and raiment, and shelter—and they were given me as my heavenly Father saw best, but amidst uncertainties and difficulties which awoke sometimes the deepest anxiety. I was compelled to rise early, to sit up late, and eat the bread of carefulness, as if the body was to be sustained by the neglect of the soul; and I sighed for more leisure for meditation and prayer, even for a lonely wilderness and a hermit's cell, that I might give my thoughts to God. I had not yet learned, that the occasions of care, and anxiety, and labor, furnished the school for the recovery of my soul. It was here that I learned how much I needed the help of Him, who won me by his daily kindness amidst daily necessities, to make Him my everlasting portion. I felt that I could not live a day, or an hour, or a moment, without Him. "I cried unto the Lord in my trouble, and he delivered me out of my distresses." I even said—"There is none upon the earth that I desire besides thee." He supplied the wants which forced me to ask his kindness, and gave me infinitely more; a fixed and settled choice of Himself, as my portion and friend. As he stretched out His hand to feed and clothe me, I saw that He was able and willing to bestow more than I could ask or think; even to give me a perfect and unbroken fellowship with the Friend, to whom in my extremity I resorted so anxiously. In my prosperity, I forgot the Giver of my mercies, which I was learning to ascribe to my own industry and skill; which I was heedlessly and proudly taking to myself as other gods. How kindly did my heavenly Father threaten to strip me of every blessing, and make me cast over the future a look of helplessness and despair. How precious were the seasons, when I turned and asked of my Father in heaven my daily bread, and found myself hungering and thirsting after righteousness. I began with seeking the provisions of my body, and I ended in a fulness of blessings for the soul. I was abundantly satisfied with the fatness of his house; and drank of the river of his pleasures. I said within my heart, "with thee is the fountain of life; in thy light shall I see life. Oh continue thy loving kindness unto them that know thee, and thy righteousness unto the upright in heart." I had longed to be released from want, and care, and perplexity, and temptation, that I might have fellowship with God; but I learned by experience, that these were the means which I needed to incline my heart to draw near to Him. I needed yet severer discipline. I was vigorous in body and in mind, and was concentrating my energies upon a design which promised good to men and glory to God; yet overrating my skill, and forgetting how helpless I was—I was leaning to my own understanding, and setting a value upon my own services, when I was humbled under the mighty hand of God, that I might be exalted in due time. I sunk under bitter sickness. My strength of body, and vigor of mind, failed me in an hour. I felt as a worm and no man. At the divine rebuke, all my beauty consumed away as the moth. As I lay crushed beneath the hand of the Almighty, unable to lift my head from the pillow which his kindness had left me, I bowed submissive to his chastening hand. I bewailed the folly which made me honor my loaned powers; I prayed for restoring, forgiving, and upholding mercy. "Oh Lord, I am oppressed; undertake for me. What shall I say? He hath both spoken unto me, and himself hath done it. I shall go softly all my days in the bitterness of my soul." The Lord heard my groaning, and forgave me; and brought me up slowly from the gates of death, and helped me to say, "Thou hast in love to my soul delivered it from the pit of corruption: for thou hast cast all my sins behind thy back.—The Lord was ready to save me." As my strength returned, I received it as the gift of God. As my mind resumed its energies, I tried to employ all its faculties for Him; saying, "The living, the living, he shall praise thee as I do this day." I had maladies of the mind. I felt often trembling, and darkness, and horror, when all was well to the eye of my nearest and dearest friends. I could not tell always, how far they were produced by diseases of the body, when they belonged to the weakness and helplessness of a creature, or when they were the direct consequences of the errors of a sinful mind. Sometimes, I knew they were the secret workings of disease upon a frame always tending to death: holding a spirit not yet fully taught to make God its refuge. Sometimes they were the passions of a feeble spirit, struggling with inward sins and outward difficulties, still failing to trust in God its strength; and often, I knew they were the misgivings of an immortal soul which sought again and again a fulness from the dry fountains of the world. Blessed seasons, of trembling, and darkness, and horror! I came to God because I was broken and decayed in spirit—because I was craving in my inmost soul; and I sought Him and found Him as my portion forever. I said again and again, in the midst of light bursting over clouds of darkness and gloom, "Why art thou cast down, O my soul, and why art thou disquieted within me? Trust thou in God, for I shall yet praise Him, who is the health of my countenance and my God!" But I did not live alone. I was successively a member and a head of a family, and from childhood to life's end, had its varied opportunities; its tribulations as well as its mercies. I suffered from anxieties and fears, even when no real evils occurred. I trembled in expectation of accidents, which never befel me or mine; of diseases, of which I saw only the faintest symptoms; of deaths, which for years a kind Providence delayed. I cannot now call it folly to have been so sensitive. I see that I was placed not only in a state of suffering, but of anxiety and fear; that when no heavy calamities were upon me, I might seek the guidance and the aid and the cheering presence of my heavenly Father. A thousand times, when pride, and vain glory, and worldly satisfaction, were taking away my heart, I was affrighted back by my anxiety and fear. When I was not smitten with the rod of correction, I saw again and again the uplifted hand; the sight of which reminded me of my sins, and made me earnestly long to be a partaker of the holiness of God. But my social discipline was not confined to anxieties and fears. I was required to watch the sick bed of wife and children; sometimes held in the most anxious suspense, whether life or death would be the issue; sometimes feeling the most distressing certainty of coming calamity. I was compelled to resign the dearest and the best to Him who gave them; and to feel bereaved and alone. Still, I knew I was undergoing a discipline for heaven; that a Father’s hand was leading me on in a pilgrimage to glory. I learned, in the bitterness of my social anxieties and sorrows, still more tenderly to bewail my sins, to humble myself before God, and to cleave to Him as my everlasting portion. As I felt His sacred influence on my recovered soul, sorrow gave place to joy, and I was enabled to say, amidst the most tender recollections, "The Lord gave, and the Lord hath taken away, blessed be the name of the Lord." My tribulation wrought patience and experience, and a hope that never made me ashamed. Yet my severest trials were those which I brought directly upon myself. I desired conformity to God; I sought to be transformed into his image; to do all his will; but in the hour of temptation, how often did I fail; blinding my own eyes, forgetting my own vows, forsaking God my helper, and leaving my bosom so far the seat of evil passions, that I became like a troubled sea, when it cannot rest; tossed by the winds of pride, and lust, and anger, and discontent, which at sometimes I thought had given place forever to the gentle breathings of the Holy Spirit. Then it would seem that sin had regained all its power, and that the severest struggles were in vain. What bitterness did I suffer in those hours of trial, from the reluctance of my sinful mind to bear the cross which I had made for myself, and from the fear that I should deny my Lord, and bring upon myself swift destruction! Often, I seemed to stand on the very verge of ruin; to be just ready to seal my character in hopeless sinfulness. But thanks be to my Redeemer that he prayed for me, that my faith should not fail; that I was enabled to lift my heart to the Prince and Saviour, and to cry, "Help, Lord, for my foot slippeth." At the moment when I was ready to fall, I found the everlasting arms around me, holding me fast upon the rock of salvation; and higher raised towards its heavenly summit. Then the language of mercy to the church seemed as if spoken in soft accents to my ear. "In a little wrath I hid my face from thee for a moment, but with everlasting kindness will I have mercy on thee, saith the Lord my Redeemer." But success, even victories renewed again and again, exposed and hindered me. As I gained fresh triumphs, I was "exalted above measure." I lost sight of the weakness and sinfulness of my heart. I trusted in myself, and became again under the power of sin, until the Spirit set my sins in order before me; often, until He who led me through great tribulation, startled me by some alarm, blasted some high hope, or withdrew some idolized blessing, and I learned again that I was a worm and no man; that I had no skill in the heavenly warfare, and no hope of success in my conflict with inward corruption, but in the aid of my all-sufficient Redeemer. Then, I did not count myself to have apprehended, but "forgetting those things which are behind, and reaching forth unto those which were before, I pressed towards the mark for the prize of the high calling of God in Christ Jesus my Lord." Often, I thought that the last conflict was at hand. Often, did my guide and friend show before me the prospect of death, preparatory to that trying event. I needed, often to stand on the border of eternity, to convince me how naked, and poor, and blind, and miserable I was. Sometimes my pride and self-sufficiency were blasted in a moment. I heard, or thought I heard, "Thou fool, this night thy life will be required." Then the brightest visions of my tempted fancy vanished as a dream. Yet I knew the voice to be of warning, and not of judgment; the voice of conscience, and of the striving and quickening Spirit, whom I had grieved from my bosom. Thus was I restored, and with a fearful, trembling heart, came back from the shades of death, to walk before God in the light of the living; trusting in Almighty aid; strong in my weakness; and finding his grace sufficient for me. Oh how sweetly did I live, as I sung, in my soul, "The Lord preserveth the simple; I was brought low and he helped me. Return unto thy rest, oh my soul, for the Lord hath dealt bountifully with thee. I will walk before the Lord in the land of the living." Then the victory seemed to be won, and the last trial past; and I enjoyed such fellowship with God, that I seemed almost a possessor of the promised glory. Yet again and again I forsook my rock and my refuge, and vainly gloried in myself; and again and again was driven back by the rod of my heavenly Father. How often did I hear the soft and cheering whisper to my soul, "I have chosen thee in the furnace of affliction. I have tried thee, and thou shalt come forth as gold." Each suffering had its kind design; each made me more than before a partaker of his holiness; and darkened as my path often seemed, changing as was the light I saw, and wayward as I was, I was led onward in a path "shining brighter and brighter to the perfect day." Yet there was another tribulation needful, so deep, so growing, so enduring, as to throw the past into oblivion. Amidst life's uncertainties, it was so probable it would never come, that I thought to pass into rest without its darkness and horrors. But he who knew my heart, appointed for my profit the lengthened discipline of weak, fearful old age. I trembled as I felt its gradual approach, lest I should fail in the last long trial; lest I should be consumed in the last heat of the furnace of affliction. At length the days came in which I had no pleasure. I would have used my limbs, but they refused to support my frame, and seemed to have no other use but as the seats of pain. I would have looked out at the windows of my soul, but all was dark. I would have listened to the daughters of music, but the sweetest harmonies were lost upon my ear. I was afraid of that which was high; the almond tree flourished on my head; the grasshopper was a burden, and desire failed. I looked forward without hope; and could see nothing in my worldly prospect, but man going to his long home, and the mourners going about the streets. Soon, said I, the silver cord will be loosed; the golden bowl will be broken; the dust will return to the earth, as it was, and the spirit unto God who gave it. I had never before seen days in which there was no pleasure. I was never before so naked, helpless, and forlorn. Man could not help me. The kindest children could not give me the succor which I needed. Wealth could not buy what my desolate state required. I was blessed with comfortable abode, with kind attendants, with abundant supplies. I could have pampered an epicurean's appetite; but I had no desire for food. I could have clothed richly my withered and aching limbs, and laid my body on a bed of down; but I could not ease my pains, or rest my limbs, or quiet my trembling mind. After all my trials, I was unprepared for this; for these days of darkness, and gloom, and suffering, and hopelessness. I felt the risings of impatience; and even the terrors of unbelief. I felt that pride and self-confidence had mingled with my former peace and joy. I trembled, lest after all, I should be filled with my own devices, and perish helpless and alone! But oh, how short was the fear, repeated though it was, again and again. For I remembered the word, "As thy day is, so shalt thy strength be." "And even to your old age, I am He; and even to hoar hairs, I will carry you; I have made, and I will bear; even I, will carry and will deliver you." I cried out amidst the gloom, when it thickened around me, "Oh, Lord, thou art my guide from my youth; forsake me not when my strength faileth; cast me not off in the time of old age. Now when I am old and grey headed, oh God, forsake me not until I have showed thy strength unto this generation, and thy righteousness unto every one that is to come." I was sure that he heard my prayer; and that underneath were the everlasting arms. Out of weakness, I was made strong; as my outward man was perishing, my inward man was renewed day by day. I could hear in the inmost recesses of my soul the promises of my Redeemer; I could see Him interceding for me at the right hand of God; I could remember all the path of loving kindness from my youth. As I looked out of the windows, it was dark; but as I looked upon Jesus, it was light; and I felt myself more rapidly than before changing into his image, from glory to glory; and I could see nearer and still nearer, the city which hath no need of the sun, or of the moon, because the Lamb was the light thereof. At length the last trial came. I had passed fourscore years, and felt that the hour feared and longed for had arrived. I had learned the deceitfulness and wickedness of my heart. As I fixed my dying eyes on the infinitely pure, I was overcome with a sense of unworthiness. An awful foreboding came over me, that sin so obstinate might never be removed. For a moment I was tempted to despair. At that moment I cried anew, "Lord, to whom shall I go, thou hast the words of eternal life? Lord, save me, or I perish." I had not reached that certainty of my salvation which I once expected; but I had tried the Saviour so often, that I believed he would not now turn away my prayer, nor his mercy from me. As the last crisis came, all the events of my life were clustered before me; I was encouraged by all my former victories. All my resolutions revived, and were united in one act of universal consecration to God. All my prayers seemed concentrated in one act of faith. I prayed for a victory over sin with unutterable "groanings," and seemed to gain within my soul, the strength of the Almighty; as I beheld with unclouded eye, the Lord, my righteousness and strength. I cried with a faith which I felt to be absorbing, "Lord Jesus, receive my spirit." I knew not death. I felt merely transformation, and complete resemblance to my Lord. I was guided and borne by the angels. As I arose, I forgot the tribulations and the withered body from which I was released; I was so ravished with the pleasures of unsullied holiness, with the transforming face of God, and with the holy society of angels, and the just made perfect. I was bemoaned, and buried, and given up to worms, before recollection came in as the handmaid of my joy. Then I remembered the path of tribulation which led me to my rest; how, amidst my sorrows, I saw the Lamb of God who came to take away the sins of the world; how, out of the depths, I cried unto the Lord, until I could say, "There is forgiveness with thee that thou mayest be feared;"—until by faith in Him, I was turned from darkness unto light, and from the power of satan unto God;—until at length my purified spirit parted from its earthly tabernacle, and rose triumphant to its house not made with hands, eternal in the heavens. At length the period came for the Redeemer's kingdom to be closed. I was summoned to judgment—I descended fearless to the earth, the scene of my dangers and my victories. The trumpet sounded, and in a moment I found myself the tenant of a body—how like, and yet how unlike that which had been sown in corruption. I saw the lands of my nativity; the hill top where I had often gone to the burial of kindred and neighbors; the hills and valleys remaining as in the days of my childhood. I rose from the fields where I was sown in corruption. There my corruptible put on incorruption, and my mortal put on immortality. I ascended in the air, with inconceivable swiftness; yet soaring without a struggle, and seeming to have passed thousands of miles in an instant, I was at the right hand of my Judge. I saw his benign countenance, like the sun shining in his strength; and yet as softened in the light which he shed upon the redeemed, as "the bright and morning star." As I looked, the remembrance of passed sins came over me almost to overwhelming; and yet I was most happy in a sweet and calm conviction of unworthiness, while his light shone into all the once darkened corners of my soul. As I looked upon my Redeemer and Judge, I felt assured that darkness and night would be no more. I heard the invitation to my final rest. I had forgotten, as I heard them recalled, the services which I had rendered to my Lord. But He had preserved the record; and when our humbled minds failed to remember as services, what we had done amidst temptations and sins, He bade us look upon the redeemed. I had not noted till then, the scene around me. We seemed in open space, suspended in the regions of immensity; innumerable, shining in glorious bodies, like unto that glorious sun, upon which our whole attention had been fixed. No earthly beauty could be remembered comparable to this. Though we could see that some in childhood—some in youth—some at maturity, and some in old age, had been laid in the grave. All were beautiful; all were diverse; and there were countless gradations, as one star differeth from another star in glory. At the Saviour's bidding, I looked for the witnesses that I had ministered unto Him. I could recall the earthly looks of the glorified around me; visible through the splendors of their immortal bodies. I remembered, without confusion, and with perfect ease, the day, and the hour, and the place, where amidst sin and shame I met and aided them on earth. The board where the hungry was fed; the cup of water which I had given to the thirsty; the sick man’s bed, and the captive’s prison, were instantly before my imagination: with other forms of human wo, which, as a sufferer myself, I had learned to pity; and though a sufferer, had been able to relieve: above all, many whom I had warned of their spiritual dangers, and whose spiritual burdens I had been able to bear. The services, which I recalled, had been so mixed with sin, as to be remembered only amidst the tears of repentance; but then, I saw how my momentary and imperfect kindnesses had been blessed in comforting the afflicted; in cheering the Christian on his way; or in softening the stony heart, and bringing it to the Saviour. One, I saw, whose case broke upon me with a sudden overpowering. We had met as happy spirits disembodied; but had not recognised each other; for I met him whom I knew as a youth, with all the marks of aged saintship. But now—when each had recovered in the form of glory, his earthly body, from which not eternity can remove its distinguishing appearance—making heaven as various as beautiful—I turned, and his eye caught mine. It seemed kindled with the same fire, as when I saw his first espousals to Christ, only purer and lovelier. I had found him going astray from the instructions which his dying parents had left him. I warned, I besought in vain. He sickened, and I visited him—not knowing him yet as one of the Redeemer's brethren. I helped his feeble mind to reflect—his weak memory to recall the teaching of his tender mother. I watered his couch with my tears—his heart melted—his conscience awoke—his vows were made, and his anxious, trembling prayer was offered up, "God be merciful to me a sinner." I led him to Jesus; half doubting the while, whether I knew the way myself. I saw the commencement of his course, and we parted, and met not for years before I was called from the earth. When I knew him again, it was in his incorruptible body, on the morning of the resurrection. The Saviour seemed pointing to him, and saying, "inasmuch as thou didst it unto him, thou didst it unto me." I turned again, and saw father, and mother, and son, and daughter, and neighbor, and friend, and many a stranger, too, who in all my sorrows and sins, were the Saviour's ministers unto me; and who now received their reward, as if they had done their various service unto Him. Then, confused views rose upon my eye, since becoming more and more distinct and glorious, of the feeble kindness of such a worm as I, spreading its saving influence over the wide world, and descending down through many generations, and giving me a part in the joy of millions and millions, standing at the right hand of the Redeemer, and called [His brethren. Then, I could see all the streams of human kindness, bearing the love of God to the heart of man, descending and widening through all time, and uniting the redeemed throng with charity, the bond of perfectness. Then, I beheld, as in every face, the sweet expression of gratitude for mutual kindness, mingled with silent praise, to the infinite Redeemer, through the innumerable throng, and earth's little moments seemed to be the fountain of a peaceful eternity, flowing from never failing charity. We rose together, obedient to the word, "Come, ye blessed of my Father, inherit the kingdom prepared for you from the foundation of the world." We rose as a cloud, not tempest driven, but as we had seen on earth the clouds of the morning borne by the gentle breeze, and gilded by the rays of the morning sun, until we arrived at the city which "hath no need of the sun, neither of the moon to shine, for the glory of God doth lighten it, and the Lamb is the light thereof; and the nations of them that are saved, walk in the light of it; and the kings of the earth bring their glory and their honor into it." Death was swallowed up in victory. It seemed so as I rose triumphant, a redeemed and holy spirit, after the last earthly struggle, when I left my worn-out body on the death-bed. It was then a glorious triumph; for while my body was given to the worms and wasted into dust, I was still with the God of my living spirit; happy, without hindrance or defect. But my happiness was not complete until the resurrection; when resuming my body, which I used to find the occasion of sin, that it might be the means of holiness, I found it a temple of holiness and joy. My senses became so many aids to the knowledge of God. When corruption put on incorruption, and mortal put on immortality, and death was swallowed up in victory, oh what new glories burst upon our sight, so large as to invite our diligence forever. What new harmonies broke upon our ears, and dwelt upon our tongues, as we learned more and more to turn our knowledge into praise. As we entered, new arrived, hosts of angels burst forth in universal acclamation; "O death where is thy sting! oh grave where is thy victory!" Ten thousand times ten thousand, multitudes innumerable, replied in chorus—Thanks be to God which giveth us the victory through our Lord Jesus Christ! Through much tribulation, I have entered into rest, and my robes have been washed white in the blood of the Lamb, and I am before the throne of God, and serve him day and night in his temple, and he that sitteth on the throne dwelleth with me. I hunger no more, neither thirst any more, neither doth the sun light on me, nor any heat. For the Lamb which is in the midst of the throne, feedeth me, and leadeth me unto living fountains of waters, and God has wiped away all tears from my eyes. I am changed into his image. My bosom is like crystal, reflecting the perfect image of the sun. I retain my individuality, and hold in perfect remembrance, all the path of tribulation, which led me hither; and yet, as I look upon the Lord God, and reflect the beams of His glory, I feel that I am one with Him. I am still his servant; not permitted, not desiring to be an idle partaker of his happiness. I serve Him day and night, if earthly measure be taken to express my unceasing diligence; but I am never weary. My employments are my refreshment—while I toil, I rest. I am progressive. Hitherto I have no measure of duration, but the improvement of my faculties—higher happiness—greater knowledge, and increased resemblance to my God. I desire no other measure. I think of improvement, but not of time. I am wiser, holier, happier than I was. I shall be wiser, holier, happier than I now am. What other measure can I desire? It is enough that my progress has no boundary. I could be happy if I were alone, reflecting the glorious image of my God. Yet it befits my nature more, to unite with other spirits in love, service, adoration and praise. My capacities of happiness seem now, as if multiplied by all the myriads of heaven. I feel so one with the countless hosts of the Redeemed, and even with other hosts of happy spirits, that the happiness of all seems as if poured into my single bosom, while we sing; blessing, and honor, and glory, and power, be unto Him that sitteth on the throne, and to the Lamb forever! LONDON; A RETROSPECT OF THE BRITISH AND FOREIGN BIBLE SOCIETY. MAY 1, 1816. At the anniversary of May 1, 1816, the British and Foreign Bible Society shone forth in meridian glory. Then, the mists and clouds which obscured its rise and progress, seemed completely dispersed—dissolved in a serener air—distilled in dews, and dropped down in rain upon the earth, warmed and thriving in its beams. The storms and tempests which had been gathered over Europe, had cleared away—leaving the lands from which the Bible had been either withheld or discarded, watered and open to its rays. The more distant nations, clouded beneath their own dark exhalations, saw in that mid-day glory, a light piercing their gloom. No people was hidden from the heat thereof. I entered the hall at great Queen-street, at 10 o'clock; and sat partly in meditation, and partly in that Christian fellowship, which the occasion produced, until the hour appointed for the annual solemnities: joining in the burst of welcome, with which again and again, the more illustrious friends of the Bible were greeted. At twelve, the President, Lord Teignmouth, as humble as honored in that exaltation, opened the meeting, which I have ventured to characterize as the meridian of the British and Foreign Bible Society. At that moment, the sun shining from its mid-day height, poured its rays over the habitable world! Shedding the light of day over both continents—it cheered the western coast of America with the rising glories of the morning, and lingered in the evening twilight on the islands of Japan; thus designating London as the moral centre of light to the world. Even so the Bible Society, having emerged from the night of modern infidelity, pursued its course amidst the mists and clouds of the morning, and darkened by lowering and roaring tempests, over all the sky, until that day shone forth over the whole earth! At that moment, America was collecting her scattered rays into one great beam of light, sufficient to cover with glory the whole western hemisphere. Russia, and her northern and southern neighbors, seemed suddenly opened to the whole light of day; while it was seen gleaming far on distant Africa, and Asia, and the islands of the sea. There was no speech nor language where the voice was not heard. The symbol retired; and day and night, and night and day have succeeded each other, for half a generation, yet daily renewing at noon, the symbol of the meridian glory of the British and Foreign Bible Society: still shining, London, over thee, and flowing in streams of mercy over the whole earth. The condition of the world at the anniversary of 1816, was remarkably suited to fix it as the meridian era of the British and Foreign Bible Society; a condition more striking to the writer, because it had burst upon him suddenly with a noon-day glory. The events of the spring and summer of 1815, which brought the infidel drama to its close, and exalted the British and Foreign Bible Society to be the light of the world, broke upon him with an effect which they can scarce imagine, who received the successive items of intelligence by every new arrival from abroad. He had left the coast of Malabar in the month of October, just on the news of Napoleon's escape from Elba, arrival at Paris, and his rallying the armies of France against combined Europe; ready to embroil the world in the horrors of protracted war. In this state of suspense, when the conqueror was on the wing, the writer embarked for Great Britain, and remained ignorant of subsequent events until the middle of December, and within a few day's sail of the island of St. Helena. The reader may imagine with what interest we saw on the verge of the horizon, a ship, appearing now, not merely to break the solitude in which for weeks we had traversed the ocean, but to inform us of the fates of men, at one of the most remarkable crises of human history. He cannot imagine the effect upon our minds, when we received at once the detail of events, the battle of Waterloo, the fall of Napoleon, the pacification of Europe, and the imprisonment of the CONQUEROR OF THE WORLD, on that rock in the ocean to which we were hastening our way. A few day's sail brought us, with unimpaired interest, to the prison, where, as now appears, Napoleon was sent to die, by HIM who taketh the wise in their own craftiness, and subdueth the mighty by their own weapons. Even there, in the midst of the ocean, far from the Christian world, the mind was overcome, as that towering rock loomed higher and higher in the prospect, with the thought of Him who had judged among the nations and rebuked many people, that at length they might beat their swords into ploughshares, and their spears into pruning hooks, and learn war no more; and wrapt into the sympathy of the instructed nations, we seemed to hear and return the appropriate exhortation of that wonderful period, "Come ye and let us walk in the light of the Lord." On arriving in London, the last of February, 1816, the impression was deepened. The most striking wonders of that wonderful metropolis, were now the trophies of victory, turned into ensigns of peace among contending nations. The armor, the weapons, the chariot, the horses, the travelling bed and furniture, the varied regalia of him who had wasted Europe; were the objects of universal curiosity and admiration. The battle of Waterloo, itself, was made to pass in review,* before the tens of thousands of all nations, as if it were to be gazed at, as the last * Referring to the Panorama view exhibited in Lincolns-Inn-Fields. specimen of man butchering man on the tented field; the last sight of blood and battle, on which repentant man would fix his eye. There seemed at least such a pause in the conflicts of man, as at the birth of the Prince of Peace, when the temple of Janus was shut, and the nations were at rest; now so much the more promising, because contending nations acknowledged Him as Lord, and were open to receive the converting word. In the midst of these ensigns of returned peace, the spring of 1816, came clustered with its joyous anniversaries; succeeding each other through the month of April: all covered with new lustre by the glories of the times, and all under the golden motto, repeated by one of the most eloquent orators in those brilliant assemblies; AMICITIÆ SEMPITERNÆ—LET FRIENDSHIP REIGN FOREVER.* At length, May-day came, illustrious in the meridian glory of the British and Foreign Bible Society, casting its light over the whole earth. On that blessed day, faith seemed to see throngs of pacific angels hovering over the METROPOLIS of all nations, having the everlasting gospel to preach unto them that * Rev. Mr. Bunting, of the Methodist Church. dwell on the earth, and to every kindred, and tongue, and people; and to hear, instead of the noise of battle, soft harmonies of songsters innumerable, "GLORY TO GOD IN THE HIGHEST, ON EARTH PEACE AND GOOD WILL TO MEN." It is impossible to make this retrospect upon LONDON, as the centre of moral light to the world, without reflecting upon its original darkness; as sitting in its ancient shadow of death. How has the land of deepest gloom come at length to shed the noon-day light upon the world! When our Lord commanded his disciples, "Go ye into all the world, and preach the gospel to every creature," LONDON, now radiant with the word of God, signalized its May-day, by the Bealan fires, fearfully glaring over every town and village of Britain, Ireland, and even Gaul, between which, beasts and men ready to be sacrificed were compelled to pass. Mid-summer, too, was sacred to Belinus; and honored with the most splendid pageantry. "When the sun enters into Cancer," says Quintus, in his letter to his brother, M. T. Cicero, "it is the greatest festival of the god; and on all high mountains and eminences of the country, they light fires at the approach of that day, and make their wives, their children, and their cattle, to pass through the fire, or to present themselves before the fire in honor of the deity. Deep and profound is the silence of the multitude during this ceremony, until the appearance of the sun above the horizon, when with loud and continued exultations and songs of joy, they hail the utmost exaltation of that luminary, as the supreme triumph of the symbol of the god of their adoration." On the eve of the first of November, a more striking act of reverence was paid. Over all Britain, the fires in every house were extinguished; and every master of a family, sought the prosperity of the following year, by bringing home to his domestic hearth the consecrated fire! Bless the Lord, London, that the adored sun has become the symbol of that light which has arisen over thee, and pours its healing beams over the whole earth! How slow the rising of that light! How little welcomed were those healing and reviving beams! In times unknown, the word of life found its uncheered way among the untutored Britons; yet forced to hide itself on the heights and in the depths of the mountains of Cornwall and Wales. Then we see Gregory sending the heralds of the cross, that he might prepare these Angles, to be angels; and the gospel welcomed through all the Saxon kingdoms. Then, again, we see Egbert translating the gospel, and venerable Bede giving his countrymen the Bible and dying with the words of faith on his lips, "I will go to him from the flesh who formed me out of nothing; my soul desires to see Christ my King in his beauty:" and Alfred, too, that wise and Christian king, preparing the laws and diffusing the holy light under which England is blessed after near a thousand years! Yet as we look again, Britain is seen worshipping sun, and moon, and fountains, and rivers, and trees,* and amidst its later darkness, him who seated in the temple of God, shut up the Bible from the eyes of men. Amidst the gloom, the light was not extinguished. Through the distance of five hundred years, Wickliffe rises on our view, offering to his dark countrymen the light of life; and for so great a crime, dug from the grave, burned to ashes and scattered in the river; yet followed by Lord Cobham, a martyr to his zeal, hung in chains for spreading the knowledge of the Bible. * In the 11th century, Canute forbade this worship by law. But see! a distant light breaking over Europe! The chain of superstition, rusted and worn in ten thousand links, suddenly bursts at the stroke of the monk of Wittemberg; and England and her sister nations hasten to be free. Yet half Europe rivets again her iron bondage, and darkens her windows, to the spreading light. Even England doubts and hesitates, whether to come forth into the freedom and glory of the gospel. Now, she unfolds the word; now, she closes it and seals its hidden contents in the blood of her sons; and the prison, and the axe, and the stake join their cruel hands to drive the Bible from the shores of England. And when these did not avail, the council of the church and the power of foreign states covered the seas with the invincible armada, to destroy the friends of light, and cover England again with darkness and the shadow of death. Then treachery laid its train and prepared its secret thunders, that one instant might blast forever all the hopes and joys of England. Yet no weapon prospered against the word of God. The Spanish host was wasted by the tempest; and the gunpowder plot defeated by the misgivings of cruelty. "The bright occidental star," had not long set, when the morning star arose in the publication of the authorized version, which by means of the Bible Society covers all who speak the English tongue with light. . . . . Yet look again—reluctant England did not even then welcome the healing and reviving light; was not then ready to be its almoner to all nations. Scarce was the victory gained over cruel superstition, when studious and crafty philosophy prepared most skilfully its glittering weapons for a warfare against the Bible. Now, it pretended that human reason was too skilful in reading the book of nature to need the book of revelation, and now declared it impossible for God to confirm a message to mankind. Now, it could see no testimonies of history to establish the records of the Bible; and now blinded its eyes to rites and customs of Jews and Christians, which no wit could explain if those records were false. Now, it plunged into the archives of eastern nations, and came back from their hidden depths with dates and dynasties which put all scripture history to shame; and now dug through crust after crust around the crater of the volcano, until it exulted in the discovery of an under-soil, far older than that which arose out of chaos on the morning of creation. Oh, Britain, what a guilty part had thy sons, in the infidelity of Europe; in the vain attempt to quench the light of life, and cover the earth with darkness! Thy profound philosophers—thy sagacious historians—thy enterprising travellers; what pains did they take to stop the dawning day—the rising sun! Miserable men; monuments of folly, amidst your boasts of wisdom! How hopeless was your task, to stop his giant course! As ye wrought your little night, and involved yourselves, and your disciples, in a chosen darkness; little did ye think, how soon the word of God would arise in glory, over London, Britain, and the world. Little did ye think, as ye welcomed the skirts of the gathering clouds, how soon the tempest of infidelity would burst, and leave all Europe and the world, sitting in the clear light of day! Little did ye think, that when ye were laid in the darkness of your faithless graves, a light would break forth, which would disperse the fictions of your toil; and even quench the lights, which the advocates of the Bible had set on high amidst the gloom of infidelity. Little did ye think, while ye dreamed of darkness, and clouds, and gloom, about to overshadow your European paradise, a light would issue from the kraals of the Hottentots—from the dark isles of the Pacific, so bright as to cover all Europe, and the world with light. Little thought ye, that from the fancied midnight in which ye were covering your own native isle, a light would break forth from ignorance and want, which would cheer the Bible onward to its meridian glory; that the cry GIVE US THE BIBLE, from the mountains of Wales, would rally its friends to the work of spreading it abroad over all nations; and that amidst the boasts of victory and the wailings of your defeat, it would shine forth as the sun in meridian glory over LONDON AND THE WORLD. It was thus, that we saw LONDON, May-day, 1816, the centre of a world at rest, and sat down amidst assembled throngs, to rejoice in the triumph of the word of God. The scene returns again as we write, radiant with its former splendor; and London is before us, as we saw it in those gala days of Europe, or as on that May-day, basking in the LIGHT. As the vision returns, the wonders of that wonderful metropolis, borrow their interest from their relations to the remembered and growing glory. We revisit the tower, and seem to hear the splash of the oars which bear a lover of the Bible to its silent chambers; and from those silent chambers, to his last sufferings, and testimony; we listen to his prayers; we see his blood, as seeds of increase to the word of God. Or amidst the trophies of ancient victory over the foes of the Bible—the relics of the invincible armada—instruments of torture for the protestants of England, we pause to read the inscription, unwritten, yet glowing in fancy's eye, "No weapon formed against thee shall prosper." Or surveying the pageant kings of many generations, in their mute parade,* the Williams, Edwards, Henrys, Charleses, Jameses, Georges; we listen to their silent eloquence, "All the glory of man is as the flower of grass. . . . . . . But the word of the Lord endureth forever." . . . . . . We retrace the sepulchral courts of Westminster Abbey; the mausoleum of the gifted and honored dead; but it is only to ask, amidst the glory in which ye lived and died, did ye give heed to the sure word of prophecy, and follow its dawning and increasing light, until ye were led to * In effigy, in armor, and on horse-back. the realms of eternal glory, in which your earthly fame is all forgotten; and even now, are ye bending from your thrones, with bursts of joy, as ye see the word of God triumphant. We see again, St. Paul's, rearing its emblematic ball and cross; now, symbols of the sun of righteousness, arisen on high, over London and the world.* As we enter that magnificent edifice, we forget again its architectural grandeur, lost in the moral sublimity of the Apostle of humanity, standing in marble before us! The outstretched arm, extending the key, to unlock the prison door—the intrepid step—the fearless look, softened into the sweetness and gentleness of an angel of mercy, mark the immortal Howard;—the harbinger of the age of the spread of the gospel: showing with what energy, and courage, and kindness, the next generation must go forth, proclaiming the acceptable year of the Lord:—healing the broken hearted—preaching deliverance to the captives, and opening the prison to them that are bound. We reascend, that lofty * The symbol, and the name of that splendid cathedral, may be said to have been welcomed by the See of London, when the excellent Bishop Porteus, led the way of the English Bishops in the patronage of the British and Foreign Bible Society. tower, and see again, London, itself a world! sitting beside its liquid pathway; long the track of commerce and of war: now, how remarkably of the word of God; bordered by her surrounding villages, in all the beauty of green England; and to fancy's eye, by the growing glories of the summer, over all the fertile lands of Britain; the memorial of that prophecy which closed the report of 1816: So shall my word be, that goeth forth out of my mouth; it shall not return unto me void, but it shall accomplish that which I please, and it shall prosper in that whereto I sent it. It was not England alone, which was concerned in the anniversary of 1816. . . . All Europe seemed that day, to have trodden back her wandering paths of infidelity, and to have come to do homage to the Bible, and to God. Europe, for her pride and her atheism, had been cut down to the stump, until three times seven times, had passed over her. She had been cast out, desolate and forlorn, forsaken of her rejected God and Saviour, until chastened by her sufferings and disgusted with her pride, she came back, acknowledging the Author of the Bible as the most high God; before whom all the inhabitants of the earth are reputed as nothing: who doeth according to his will, in the armies of heaven; and among the inhabitants of the earth. Did we not see Europe, that day, with her reason restored—her honor and brightness returned—honoring and extolling the King of Heaven, all whose works are truth, and his ways judgment—and those who walk in pride, he is able to abase? Did we not hear the decree unto all people, nations, and languages: How great are his signs, and how mighty are his wonders. His kingdom is an everlasting kingdom, and his dominion is from generation to generation! The Chancellor of the British Exchequer, may be said to have been the HERALD of smitten and returning Europe, when he announced to the British and Foreign Bible Society on its first succeeding anniversary, the two-fold work of the sovereigns of Europe and their people,—THE HOLY ALLIANCE, and the PATRONAGE OF THE BIBLE SOCIETIES, by the Emperor Alexander, and the associate kings. The holy alliance was signed at PARIS, September 26, 1815; and significantly published at St. Petersburg, Dec. 25th; "the day of the birth of our Saviour:" having received, meanwhile, the accession of Holland, and the informal, but "entire concurrence," of the Prince Regent of Great Britain. Its terms were most striking, when considered as issuing from Paris; so lately the metropolis of atheism and deism. The three powers take for their sole guide, the precepts which the holy religion of our Saviour teaches. They declare that there is no other Sovereign than Him, to whom alone power really belongs, our divine Saviour, the Word of the Most High, the Word of Life. Thus failed the watch-word of Voltaire, ecrasez l'infame, crush the wretch! Thus ended the decree at Paris, 1793, which abolished the gospel! Thus ended the law and the inscription, there is no God. After the lapse of fifteen years, our views of this great public act, of returning Europe, are not changed.* Notwithstanding the sneers which have been cast upon it, the Holy Alliance will never be blotted from the records of * The Christian Observer, at the time, styled it a "solemn recognition of the Supreme Authority of the great Sovereign of earth and heaven." heaven—will never lose its importance in the annals of the world. Be it as some have supposed, a piece of state chicanery—the offspring of the hypocrisy of courts—a plan of the princes, to fix the nations in slavery; what was it but Europe in the person of her hollow-hearted sovereigns, sick of the vanities of her former confidence, humbling herself under the mighty hand of God? The more just supposition is, that the princes were influenced by the discipline of Providence; and that they, with their subjects, were awed into an acknowledgment of the Scriptures, and of God: were bowed to the authority, under the auspices of which they had triumphed over the atheistical empire. No wonder that the conquerors, were awe-struck at their victories! when they found themselves triumphant on the spot where Christianity was threatened with extermination; at the fountain of that atheism, which was to have overspread the globe! No wonder that they were awed to acknowledge THE WORD OF THE MOST HIGH—THE WORD OF LIFE; and to welcome its light as the only blessing to Europe and the world! Inconsistency has ensued;—but the act is not blotted from the records of heaven; and will never lose its importance in the annals of the world. In one sense, no doubt, THE HOLY ALLIANCE was premature. The heart of Europe was not yet ready to obey the dictates of its conscience; was not yet ready to adopt a treaty, which broke upon it like a sudden day-spring from on high; the very dawning of the millennial morn. Europe has not been ready to follow her princes—her inconsistent and backsliding princes, have not been ready to lead their people, in paying the vows which their mouth had spoken in their trouble, and renewed in the first grateful moments of deliverance; yet, on the records of heaven, and in the annals of mankind, THE HOLY ALLIANCE will remain, as a great preliminary movement, fore-showing the final submission of the nations, when kings shall be nursing fathers, and queens nursing mothers, in the church,—the kingdom of the KING OF KINGS! It may be, that the inconsistent leader of that Christian treaty, was sincere in his effort to guide allied Europe to the God of the Bible, and that being forgiven for his defection from the cause he had so earnestly and successfully served, he was taken away from the evil to come. Yet, even now, that treaty is required at the hand of the nations—at the hand of the princes, who have forgotten the God of their deliverance; and flourishing in the pride of their hearts, are faithless to the bond by which they or their predecessors bound the nations to Him, to whom alone power really belongs, our divine Saviour, the Word of God. Alas, neither Europe nor her princes have humbled their heart, though they knew all this; but have lifted up themselves against the Lord of Heaven, and have not glorified the God in whose hand their breath is, and whose are all their ways. Even now the handwriting of affliction and judgment, burns in characters of fire, on the walls of the palace of the Czar; whether amidst his exultations over Poland, or his fears lest the arch of empire should fall, or his plans of foreign war to widen its base and compact its parts. Nay, the writing is seen on all the high places, and in all the streets of Europe, whether for forewarning and recovery, or for overthrow and ruin, who can interpret? War has returned with its havoc, or its threatening, upon the lands which were solemnly bound to the Redeemer, by the Holy Alliance: while He that rules among men, stands ready to arrest and exceed the havoc of war, by the overflowing pestilence. Let not the nations be deceived, by the pause in the judgments of the Almighty: nor riot in the joys of a fancied escape from the overwhelming scourge. He who rules among men, and to whom allied Europe did homage half a generation ago, our divine Saviour—the Word of the Most High, the Word of Life can now, or when he will, gird his sword upon his thigh, and ride forth, King of Kings and Lord of Lords. He can array the passions of courts and nations, which they will not restrain, and let them loose as the instruments of mutual slaughter; or he can lead on the pestilence:—the cholera, at which the world stands aghast, or other pestilence from the store-house of his judgments; a scourge over Christian Europe, as overwhelming as its opportunity has been great, and its vows public and distinct. Who does not perceive the signs of the times? Who does not see that He to whom alone power really belongs, is looking down upon the nations, who have taken counsel against the Lord, and against his anointed, to cast away the cords of 1815: yet still, while war for a moment grounds its arms,* and pestilence stays its havoc, saying in accents of utmost kindness, "kiss the Son lest He be angry, and ye perish from the way. . . . . . Blessed are all them that put their trust in Him." Yes, and over France; whose desecrated capital welcomed the princes, when they paid in her streets, the homage of allied Europe, unto God; and whose approving voice was given to the Holy Alliance, the voice of warning soundeth loud; mingled still with the sweet accents of mercy. Can she escape the contest and the trial if it come, when all Europe shall rush together for carnage and destruction, with a conflict of passion more universal and more deadly, than ever before she broke her returning vows? or escape the pestilence which can dash her glory like a potter's vessel? And England; that happy nation, which amidst the storms, of infidelity welcomed and speeded the word of life rising over the nations; and when the reason of maddened Europe was restored, was found sitting under its meridian glory, shining over London, Bri- * March 9, 1832, in view of intelligence to January 21st. tain and the world, is not excepted from the warning, while she hears still in sweetest, loudest chorus, the accents of mercy. For while the sounds of foreign war, and internal contentions call her forth to mingle in the wasting conflicts of men, and pestilence stands smiting her uttermost border, the kindness which she ministers to Europe and the world, is re-echoed to her heart, from many a family, and village, and island, and nation, joyous in the word of life published by her sons. And America, far from the strifes of men; with the wide Atlantic, as her cordon, against the march of the cholera. Is not she forewarned? Is not she invited to the Redeemer? Let England and America, allied in the great work of saving mankind, accept the forewarning: and as they see the ruin which threatens to involve the nations, let them hear the voice, "Come out from among them, that ye be not partakers of their plagues;" and renew their counsels, and increase their efforts for spreading the word of life over the world. But amidst the forebodings of the storm, let us hope for something better than infatuation and destruction among the afflicted and threatened nations. Let us hope that northern and southern Europe will note the footsteps of their abused Redeemer; that they will be forewarned by the first skirmishes of the great battle, which is yet to purify the world:—and whether amidst the thunder and the storm, or the intervening calm, will hear the assurance, Blessed are all they that put their trust in Him. Sure, conscience will awake, and the melted heart flow forth in all the palaces of Europe, in all her cities, and villages, and families, and the golden promise of cleaving to the Lord, will come forth again out of the furnace of affliction; and the banner of another holy alliance, will be unfurled, and the nations of Europe and the world, sit down together in peace and holiness, under the meridian light of the word of God, shining over London, Britain, Europe and the globe! THE GANGES, AND THE MISSISSIPPI. It is impossible to cast one's eye upon the map of the world, without perceiving a striking resemblance between the two great rivers of India and the United States; and those wide and fertile districts, through which they pursue their long pathway to the ocean. The resemblance prepares the mind to dwell with intenser interest upon contrasted views, which force themselves upon it; and upon the prospects which faith opens to the eye, when those well watered regions shall be equally glorious under the reign of the Redeemer. The Ganges, that king of Asiatic rivers, flowing from the actual heights of the world—the mountain tops of Himalaya, claims to itself the oriental title of heaven descended Gunga; and a divine origin from the forehead of Mahesa, "When Gunga from his brow, by heavenly fingers press'd, Sprang radiant, and descending, graced the caverns of the west." As she flows through the most fertile regions of the earth, the means of food and raiment, and the channel of intercourse to bordering millions, she passes as a goddess of beneficence, scattering her gifts in rich profusion, and receiving by her side, and in her sacred bosom, the vows and the reverence of her worshippers; from age to age: even now, the homage and the offering and the victim, amidst the light and improvements of the modern world. Alas! that in far distant times, the gifts of the Creator, the bounties of Providence, should have been so abused by the perverseness of man; that when they knew God, they refused to glorify Him as God, and worshipped and served the creature, rather than the Creator. Alas! that along this mighty river, these ever fertile plains, kindness did not win the ancients to their Maker;—that when the fountains of posterity were opened, and the streams were enlarged, on which millions after millions, were to flow in successive ages, they were polluted so heedlessly; that light was turned to darkness, wisdom to folly, religion to idolatry, as the inheritance of unnumbered immortals; passing their pilgrimage and opportunity, worshippers, of the rivers, and the trees, and all animated nature, and men, and shapeless idols, until bewildered fancy could imagine three hundred and thirty millions of gods! That branch of the Ganges, most familiarly known to Europeans, is the Hoogley; which passing Calcutta, the capital of British India, empties itself into the bay of Bengal, we may say between two remarkable Teert'hu St'hannu—places of salvation, to which the Hindoos have been accustomed to resort, that returning, or ne'er returning, they might bear witness that the people hold a perpetual lie in their right hand. Strange, that for ages, Gunga should have been adored as heaven descended, and been visited in ceaseless pilgrimage—and been born to be worshipped over all India; when she empties herself unceasingly, as between the jaws of some nether hell; as the only proper outlet to a stream, whose bosom swells from its uttermost heights, with the dying victims of superstition, from hoary age to the first breath of infancy! The grand landmark by which this mouth of the Ganges is approached, is the temple of Juggunath, on the coast of Orissa; the temple of the ruler of the world: how fearful a symbol of the cruelty of that religion, which has consecrated Gunga; whose waters without much stretch of fancy, may be said to wash its base as on its western bank; how fearfully showing, in its annual groans, and death—and in the whitening bones of men, covering the country round, for fifty miles, what havoc man makes of himself, when for the creature he forsakes the Creator. On leaving the temple of Juggunath, the voyager, losing sight of land, sails eastward for the pilot ground, bordering the sand heads, formed by the influx of the mouths of the Ganges. On receiving a pilot, the dangerous passage is accomplished in a few hours, and the Hoogley is entered, washing on its eastern shore, the celebrated island of Saugor, the other place of salvation, to which we have referred as an emblem and symbol of that cruel religion, which Gunga patronizes from the foot of the mountains of Himalaya to the bay of Bengal—a fearful spot, guarded by the fiercest tygers and alligators, and sending up from deep jungles exhalations of death; yet frequented twice a year by vast crowds, when many sacrifice themselves, and many a mother casts her helpless infant into the alligator’s jaws, or leaves it the tyger’s prey. Unhappy land! How strong the current of thy sin! How increasing the stream as it has flowed from age to age:—thy passing generations, refusing still to worship the uncreated BREMH, ever lingering in the borders of thy perverse mythology: refusing still to worship the author of thy annual rains, gathering over heavens of brass; who makes thy seasons fruitful, and fills thee with food and gladness: refusing still to learn the lesson which tradition mingled with thy false religion, and choosing but the dregs of the patriarchal stream—refusing too the gospel; of which remnants, few and feeble, prove thee a rejecter of the good news published to all nations. Yet, India, there are healing streams for thee. There is a heaven descended river, which will yet flow through thy plains, and will drop down in rain, and distil in dew with the waters of salvation; each drop of which, like the fabled Gunga, abounds with life, and is the precursor of the ever-flowing stream; and as it flows amidst thy realms of death, it will raise thy people with a power so great and so benign, as amidst thy gathered, listening crowds, was never fancied of the sons of Sūgūrū.* Thou hast become wearied in 60,000 sons of Suguru were restored from their ashes, according to the legend, by the descent of Gunga. the greatness of thy way; long lost in the mazes of the wisdom of the East,—in self-confidence and pride. Even now, thou hast begun to receive from the fools of the West, the streams of salvation. From Britain, from Europe, from America, from thy sister valley, thou shalt receive the full stream of the gospel, and thy 100,000,000, and thy 400,000,000, shall rise in a spiritual and immortal life. But we turn to view the river—the rivers of the West, for ages watering and fattening their fertile valley, to be settled in the fulness of times, when the lessons of all ages should be ripened on the human mind, and its growing millions might be only blessed and a blessing: Flowing in reserve: until the arts of men, struggling upwards, through the hindrances opposed by sin and Satan, should be ready to bless the valley of the West, covering its plains with every boon of life; traversing, with ease, its impetuous rivers, and binding it, with its Eastern and Western neighbors, and the world, in useful and harmonious intercourse, giving and receiving all that can adorn and comfort the home of man; * Mr. Ward supposes that all Eastern Asia, including China, is under the influence of Hindoosim. —until the institutions of our pilgrim fathers, and the wisdom of Washington and his compatriots, and the British constitution, like the oak of Britain, spreading its branches, and strengthening its roots, beneath the storms which break off its rotten, useless limbs, should have given the opportunity of sheltering the happy valley under the protecting shade of a just, beneficent, and enduring government; wide enough to cover the mountains and the shores from sea to sea;—until the powers of the common mind should be understood, and education should have descended with her offer of knowledge into all the walks of life; and science should have become habited in the people's mother tongue, and the school, and the academy, and the lyceum, and the press, should be ready for the welcome of its growing population. Flowing in reserve: until that boon was ready for a universal reception, which, received in part, had thus prepared the conveniences, and comforts, and growing knowledge for an earthly paradise;—until the idolatry of the ancient and modern world should have been seen by fearful exhibitions, as the offspring and the parent of vice and misery; and superstition had shown what guilt and folly she could clothe in the garments of Christianity; what malice, and misery, and blood she could lead on under the banner of the gospel; and the nations, some in repentance, and some in disgust, had cast her chains, that man might reject all human lords from the temple of his worship, and come to the word of life, as the fountain of salvation; and infidelity, the daughter of superstition, had broken her promise to mankind;—until it was proved, by the ripened experience of all times, that neither idolatry, nor superstition, nor infidelity, have the promise of the life that now is, or that which is to come. And, at length, free from the priestcraft of superstition, and free from the priestcraft of infidelity, fearless of the papal frown, or the wicked man’s sneer, the gathering people might wait on the message from heaven, and THE BIBLE, confirmed by the growing history of all times, first meet a universal welcome in the valley of the West, and a Christian people, the influx of all Christian nations, be cemented as one, by believing and obeying the word of life. Flowing in reserve: until amidst the glories of the age of propagation, the gathering crowds from all Christian lands might know their privilege to welcome to their heart the healing light, and to speed it through the world; to receive and to convey the waters of salvation: and all the paths of emigration, eastern, western, northern, southern, might seem none other than the way of holiness, covered with the ransomed of the Lord; while forth from every avenue, blessings should flow abroad upon the nations. . . . . . Until, may we not say, Faith, matured by Providence for ages, and revived and cheered by success in all lands, had gained strength for the most difficult achievement, in the recovery of two races of men, lingering on the borders of the happy valley, almost until now, in hopeless despondency; until the light reflected from the islands of the Pacific, from Caffraria and Guinea, and from Brai-nerd and Eliot, might shine upon the dark, ill-boding clouds, in the blest assurance, The things that are impossible with men are possible with God; inviting to that heavenly skill and power, which may fix the remnant children of the Forest, in such goodly gardens, as shall make them lovely and beloved, in the future history of our country. Nay, until that cloud, which even now lowers with the bursting storm, might seem glowing to the eye of faith with the most cheering of all encouragements, The things that are impossible with men, are possible with God; when every feeling of despondency might be changed into prayer, and the hearts of tens of thousands ask wisdom of Him who will not upbraid when they ask His guidance, in recovering a state of society,* in which master and servant are alike unhappy; until, as of old over the tabernacles of Israel, a guiding pillar shall be seen. It does not become the writer to presume to foresee whither *The writer believes that the great question of the welfare of the servile population of the South will never be fairly met by the mingled benevolence of the country, until at the North, we regard their condition in the same light, as we do any other unhappy state of society which modern benevolence attempts to relieve; that, for instance, of the idolaters of India,—instead of urging the irritating charge of a crime. Viewing slave-holding as a crime, it were difficult to decide which were the most guilty, the masters, who by birth-right, hold the peasantry in a servile state; or the slaves, who by the same tenure, hold the masters in a state of mastership; a bondage more severe and more fearful than their own. As a state of society, generated like all other existing evils, in the errors of former times, it is bad; a considerate master sees it to be so, without casting the blame upon the slave, who holds him in the unhappy condition of owner; as a considerate slave may be supposed to do, without casting blame upon the master, who holds him in the condition of servility. It certainly becomes us at the North, who, without our care or payment, are placed in neither predicament, instead of blaming either, or calling for a disruption of society, ruinous to both parties, to unite with all the kindness and wisdom of the South in those measures, which may produce a gradual change, to the mutual advantage of all. Enough has been seen of the progress of society, under wise Christian applications at the present day to show that gradual, means neither sluggish nor slow. As to crime, they will be guilty, whether at the South or North, slave holders or spectators, who refuse to apply the principles of this age of benevolence to an unhappy state of society; who waste their time and talents either in indolence or reproaches; and refuse to labor in hope, even against hope, that He, who has released Polynesia from a worse bondage, will guide a Christian people in the attempt to bless their own servile population. that pillar will guide the steps of a thoughtful and praying people; how the triumph will be gained over one of the most difficult problems of human society. Yet, as we look, we cannot fail to see in no distant prospect, Africa—desolated, afflicted Africa, released from the oppression of the slave-dealer and his allies, her cruel and avaricious princes, and from the worship of gods and devils;—and on the Senegal, and the Gambia, and the Mesurado, and the coast of the gulf of Guinea, inward to the Niger, and the Mountains of the Moon, a people renovated and blessed by their returning sons;—the children returning to their mother’s bosom, with the word of life! And at home, that wisdom which Britain was reluctant to learn on the shores of the Ganges; gaining by instruction, and kindness a sway over a virtuous and religious people, such as can never be held over ignorant and vicious minds; and in ways, which prayer is yet to learn of the infinitely wise, blessing the African race more rapidly than a growing population can be removed from our shores; so that while one division of the stormy cloud passes on, to drop down over Africa the saving dew and rain of the Word of God, the other may become the adorning of our own horizon. Flowing in reserve: that the happy valley might have its settlement in the morning of earth's millenium; that one spot on earth might form its habits at such an auspicious era, that at the last coming of the Redeemer, no rod of iron might need to be stretched forth over it, no potter's vessel to be dashed in pieces, no flesh of captains to be given to the fowls of the air, but blessed, and a blessing might welcome his coming; that one spot on earth might be settled under such happy influences, as might sanctify its beginning and its progress; and its tide of people, swelling from age to age, might flow into the ocean of immortal bliss, welcomed as they come to the bosoms of the patriarchs, who opened the fountains and enlarged the streams of a posterity so glorious;—who led the way first to the rivers and the valley of the West, and thence to the paradise above, while angels, who rejoice over one redeemed, sound their welcome in chorus of rejoicings. Here Pittsburgh, Cincinnati, Louisville, St. Louis, Natchez, New Orleans, fulfilling their promised numbers in their growing history, rise upon our view, centres of Christian influence, more hallowed than the famous cities of the Ganges; and the Ohio, the Mississippi, the Missouri, the Arkansas, flowing by other hallowed cities yet to rise, and by the well tilled and fruitful farms and gardens of a virtuous people, a multitude won by the kindness of the Creator to his worship; while the matron states of the East, and the daughters of the farthest West, shall be bound to the central valley by the cords of filial and maternal love; and the Union, from sea to sea, be secured by that divine covenant which subdues the heart of man; by that oath which calls to witness the sun for a light by day, and the moon for a light by night, shining over it, and all the waves of the Atlantic and the Pacific roaring on its shores. Yet fancy is no fool—no alien to reason and truth; and in the glowing prospect sees that there is an option; that only by practising the lessons of ages; by yielding to the principles displayed in the history of centuries; by co-operating with the means which divine Providence offers to the times; by believing that it is possible with God to direct and aid the work which he demands of men, can the valley of the West possess its privilege. If the prospect shall arise and become a speedy and glorious reality, it will be because the settlers of the West, at such a time as never rolled on before, and in such favorable circumstances as never existed before, are such a studious, watchful, active, prayerful race, as never, on sea or river, laid the foundations of empire before. If it shall rise, the East will continue and increase their efforts for the West; and the West will welcome all the aid which the East will or can bestow; yet multiplying it a thousand fold; as their peerless valley, receiving the head waters of the rivers which enrich and adorn it, contributes ten thousand streamlets from its own bosom to increase them as they flow, and distills the dew, and drops down the rain on every field and plant; thus forming a bond of union as gentle and as everlasting as binds the Ohio to the Alleghany and Monongahela, and that double river to the mountains and fountains of the East. If this glowing prospect shall become a reality, it will be because opportunity is met by a spirit of self-improvement, such as never gave impulse to infant state before; showing in every village and hamlet, amidst the humblest walks and works, such wisdom and refinement as were wont to be shown only on the high places and leisure grounds of society. It will be because science, education, philosophy, whether coming to the high or low, has laid aside her self-sufficiency, nor ventures to teach her disciples with their eye withdrawn from the Infinitely Wise... because, with her demon cast out, Philosophy sits, with her train of disciples, at the feet of Jesus in her right mind, preserving thus the valley of the West from entering such a maze of folly and ruin as was opened by the boastful, baneful wisdom of the East: because the growing people look upward to the Prince and Saviour, asking and receiving that gift of the Spirit, which even now is poured forth, claiming the universal welcome of the valley of the West, because they join the company which publishes the gospel—blessed, by blessing others. Happy land! Wilt thou be blessed and a blessing? Ye emigrants from the Atlantic shore, descending the rivers and the lakes: ye people of all Christian lands,—will ye enter on your inheritance, harvesting the ripened wisdom of ages? Free from the bondage of superstition, and the licentiousness of irreligion, will ye inherit your fertile valley, in the liberty of the gospel? Or shall we look again—unwilling—down the vista of a people, false to their opportunity; traitors to posterity?... See those clouds of smoke, once ready to vanish away, thickened again and covering the whole valley with one dark cloud, underneath which the liquid fire parts from its harmless companions, that it may be poured into the bosoms of a vicious people, and speed them on in folly, debauchery, and violence! See those hapless families wasted with poverty amidst the riches of the valley of the West; those hamlets, villages, continuing late at the cup in contention and blood, amidst the beauties of those moonlight evenings; and breaking even the chains of debauchery to desecrate those balmy mornings with violence and wrong! . . . Hark!—hear those oaths, mingled with loud laughter, the mockery of happiness! Hear those yells of anger, more fearful than the ancient war whoop! those loud voices, within that inn, amidst that liquid fire; the horsewhip—the bludgeon—the pistol! See genteel malice—murder in the garb of honor—the fashion of the times! See those freemen at the polls, symbolizing the liberty of their inheritance, by ribaldry and oaths, and staggering, and bruises, and blood, and eyeless, ghastly countenances! See that fearful panorama of a valley too fertile—opportunity too favorable; of posterity cursed by the folly and the guilt of an ancestry self-wise; posterity, willingly polluted, until the gangrene spreading from the vitals shall have destroyed the whole body, and the republic shall lie a loathsome carcass, the abhorring of all lands—from which the ransomed empires of the world shall turn to worship the Lord on a new earth, and under new heavens! Alas! the decision is at hand. The valley of the West, reserved for settlement in the last, best days, will not come to its crisis by the slow process of less favored times. Nay, the nations of the earth with it, on the heights of opportunity, will with it hasten to the crisis. The hopes of the world will not linger much longer on the sin and folly of man. The lesson of ages is nearly complete; and the favored valley must welcome it to her heart, or be hastened to her ruin. If she will be faithless to her opportunity, and enter the mazes of superstition or irreligion, then will human passions hasten with more speed and more acrimony, than was ever known before, to the havoc of men, without waiting for the slow progress from licentiousness to despotism, and from despotism to revolution, in their partial, ceaseless round of desolation; but rousing themselves in all the subdivisions of society, will fill the land with violence and wrong; or, since there is a power which, as of old, knows how to arrest the violence of men, human passions may be outstripped by the pestilence walking in darkness and raging at noon-day; as has been only faintly shadowed on the banks of the Ganges; sweeping the valley of the Mississippi, and the border heights of the Rocky and the Alleghany mountains, and the Atlantic and the Pacific shores, until a remnant, great or small, shall welcome Him who ruleth over men! But we turn with horror from the sight; nay, with hope, that the last great valley, reserved for settlement until now, will need neither human violence, nor judicial pestilence, to cleanse it for the millenial reign of the Son of man; no breaking with a rod of iron; no dashing in pieces like a potter’s vessel; no giving of the flesh of captains to the fowls of the air: that one spot on earth, reserved for settlement in the happiest times, shall so early and so fully yield to the Redeemer, that only in mercy shall it see the signs of his coming, until with unbroken harmony the Mississippi shall say to the Ganges, and the Ganges to the Mississippi, the Western to the Eastern, and the Eastern to the Western world, "Come ye, and let us walk in the light of the Lord." Deacidified using the Bookkeeper process. Neutralizing agent: Magnesium Oxide Treatment Date: April 2005 Preservation Technologies A WORLD LEADER IN PAPER PRESERVATION 111 Thomson Park Drive Cranberry Township, PA 16066 (724) 779-2111 12 20 LIBRARY OF CONGRESS 0 013 985 545 0
Economic Value of Low Flow Data Jens Chr. Refsgaard and Eggert Hansen Technical University of Denmark, Copenhagen By means of the Bayesian decision approach the economic value of low flow data is considered with special reference to the design of treatment plants for typical Danish conditions. For a specific case the worth of primary data in form of direct observations of low streamflows has been investigated in terms of an expected opportunity loss (EOL). Indirect information concerning the low flow properties of interest may be obtained by means of a secondary data set in combination with a regression model. In this study is suggested a heuristic method for the economic evaluation of a secondary set of data taking into account the uncertainty embedded in the regression model. Introduction Low streamflows are of great importance. One of the reasons being that very often the streams are utilized as recipients for effluents from waste water treatment plants. As a certain minimum of dilution is required the critical period of the year is the time when low flows occur. Consequently low streamflows are crucial in the design of treatment plants discharging their effluents into streams. Ideally in this design the magnitude as well as the duration of low flows should be considered. Very often in practice, however, only the magnitude of low flows is taken into account. In Denmark e.g. the median value of the minimum daily discharge during each year is taken as one of the basic design quantities for treatment plants. Thus it is accepted that cases of dilution less than a certain critical value are experienced on the average every second year. For at particular stream under investigation primary data in terms of direct observations of streamflow are often available only for a short period of time on the basis of which it is impossible to get more than a rather crude estimate of the relevant design factors. Thus the engineer, who has to design the treatment plant, is faced with a general problem of taking a particular action and making decisions under uncertainty. In this situation the engineer evidently has to consider the possibility of improving the basis of his decision either by postponing the project to collect additional primary data or by transmitting further information about the relevant design factor from a secondary set of hydrological data e.g. by means of a mathematical model. The Bayesian decision approach is a rational statistical method for choosing and evaluating design alternatives when the »true« state of nature (the design factors) is unknown. Furthermore such an analysis makes it possible to evaluate in monetary terms the worth of primary data as well as secondary data. The effect of uncertainty is taken into account through the use of probability density functions for the hydrological quantities treated as uncertain. This type of decision theory focuses on the decision to be made and not on the hydrological parameters as the final results. Within the field of hydrology the Bayesian approach has only been applied recently, but the interest is increasing. Besides a comprehensive study by Davis (1971), Davis et al. (1972) applied the Bayesian approach to a flood levee design problem. Jacobi (1974, 1975) considered the problem of economic value of sediment load data and optimum record length, respectively. Wood et al. (1974) applied the Bayesian approach to extreme hydrologic events while Vicéns et al. (1974) applied the Bayesian approach to hydrological time series modelling. Theoretical Background The statistical decision approach According to Jacobi (1974) the different steps in the Bayesian decision procedure can be outlined as follows: A) Define the decision to be made and identify the alternatives. B) Form the goal function including the selection of variable describing »the state of nature«. C) Derive stochastic properties of state variables. D) Select best alternative by i) calculating the expected value of the goal function for each alternative, and ii) choosing the alternative which minimizes the expected value of the goal function. E) Evaluate uncertainties and find the worth of additional data by i) determining the expected opportunity loss, $EOL$ (due to uncertainty), and ii) finding the reduction in $EOL$ either by a) collecting more primary data by postponing the project, or b) transmitting information from a secondary set of data by a regression model. F) Find the net worth of additional information in monetary terms as the decrease in $EOL$ minus the cost of obtaining this information. The present study deals with the design of a wastewater treatment plant discharging its effluent into a stream. Thus the set of decision alternatives to be considered are the possible low flows of the stream $M^{alt}$ to be used for the design. The hydrologic parameters treated as uncertain, the state variables, are the mean $\mu$ and the variance $\sigma^2$ of the annual minimum daily discharge $Y$. The goal function $G(M^{alt}\|\mu, \sigma^2)$ is a penalty function. It indicates the excess cost that has to be paid because of either a realized overdesign or underdesign during the lifetime of the plant. We return in more detail to the construction of the goal function later on. The uncertainty about »the state of nature« is reflected by assigning a probability density function to the state variables $\mu$ and $\sigma^2$. Assuming that the annual minimum of daily discharge, $Y$, is a random stochastic variable following a normal distribution with parameters $\mu$ and $\sigma^2$, the joint a posteriori distribution of $\mu$ and $\sigma^2$ may be obtained by assuming a diffuse a priori knowledge. Given a set $y_i$ of independent primary observations of $Y$ the distribution is found to be $$f(\mu, \sigma^2 \| n, \bar{y}, s^2) = \left\{ \sqrt{\frac{1}{2\pi \sigma^2/n}} \exp\left(-\frac{1}{2} \frac{(\mu - \bar{y})^2}{\sigma^2/n}\right) \right\} \left\{ \frac{(n-1)^{n-3/2}}{\Gamma(n-3/2)} \frac{1}{s^2} \left(\frac{s^2}{\sigma^2}\right)^{n-1/2} \exp\left(-\frac{n-1}{2} \frac{s^2}{\sigma^2}\right) \right\}$$ (1) where $\bar{y}$ and $s^2$ are the sample statistics $$\bar{y} = \frac{1}{n} \sum_{i=1}^{n} y_i ; \quad s^2 = \frac{1}{n-1} \sum_{i=1}^{n} (y_i - \bar{y})^2$$ (2) A decision is made by choosing the alternative $M^$, that minimizes the expected value of the goal function $$R(M^) = \min_{M^{alt}} \int \int G(M^{alt}\|\mu, \sigma^2) f(\mu, \sigma^2 \| n, \bar{y}, s^2) d\mu d\sigma^2$$ (3) $R(M^)$ is called the Bayes Risk and $M^$ the Bayes Solution. If the »true« state variables $\mu_t$ and $\sigma_t^2$ were known the alternative chosen $M_t$ would be the one that minimizes the goal function $$G(M_t\|\mu_t, \sigma_t^2) = \min_{M^{alt}} \{G(M^{alt}\|\mu_t, \sigma_t^2)\}$$ (4) Having used $M^$ instead of $M_t$ an opportunity loss ($OL$) is the result because of the nonoptimal choice. The expected opportunity loss ($EOL$) is calculated as $$EOL = \int \int [G(M^\|\mu, \sigma^2) - G(M^t\|\mu, \sigma^2)] f(\mu, \sigma^2 \| n, \bar{y}, s^2) d\mu d\sigma^2$$ (5) where $M'$ is the design alternative that minimizes the goal function for given values of $\mu$ and $\sigma^2$. It is seen that $EOL$ is a function of the available data sample. Additional information from secondary data So far, having performed step $A$ through $E.i$, primary data have been our only concern. As previously mentioned direct observations of low flows for a particular stream under investigation are very often sparse. It is not unusual that data for only a couple of years are available when the decision has to be made on the design of the treatment plant. One way to overcome the problem is by augmenting the primary data by means of a secondary set of hydrological data. The secondary set of data may be low flow data from a nearby stream for which direct observations of streamflow exists for a much longer period, or it may be calculated streamflows based on a rainfall-runoff model in combination with longer series of meteorological observations. A linear regression model in combination with the secondary set of data may be utilized for lengthening the primary set of data and thereby transmitting information. Immediately the following question turns up: What is the worth of the secondary set of data? Matalas and Jacobs (1964) treated the question by comparing the estimates of the mean and the variance based on the short and the lengthened series, respectively. They showed that if the correlation coefficient, which measures the strength of the linear regression, exceeds about 0.5, then the estimates of the mean and the variance based on the lengthened series are better than the estimates based on the short series. For a specific problem related to the design of sediment control reservoirs Jacobi (1974) applied the Bayesian approach in an approximate calculation of the economic worth of a secondary data set of water discharges. The approximations made were, however, rather crude. From the foregoing section it is recalled that an important feature in making economic evaluations is the probability distribution $f(\mu, \sigma^2 \| n_1, \bar{y}_1, s_1^2)$ of the state parameters $\mu$ and $\sigma^2$ given a set of sample statistics based on a series of independent primary observations. From a strictly Bayesian point of view we ought to derive the revised distribution of $\mu$ and $\sigma^2$ based on the primary plus the secondary sets of data in order to take into consideration the experienced gain in information concerning the state parameters $\mu$ and $\sigma^2$. A revision without some unacceptable simplifying assumptions turns out to give raise to insurmountable calculations why this approach has to be abandoned. In the present study a plausible approximate method for the calculation of the economic value of secondary data is suggested. The method is based on a heuristical combination of the Bayesian approach and the previously mentioned results obtained by Matalas and Jacobs (1964). In order to account for the suggested approximate method for the revision of the distribution the observed set of primary and secondary data is considered. Thus the primary data exist for $n_1$ years while the secondary data besides the same $n_1$ years are at hand for some other $n_2$ years. Utilizing a linear regression model based on the $n_1$ concurrent values the primary data are supplemented by $n_2$ calculated values according to the linear regression equation. $$\hat{y}_i = \bar{y}_1 + b (x_i - \bar{x}_1) ; \quad i = n_1 + 1, \ldots, n_1 + n_2$$ (6) where the sample statistics are defined as follows $$\bar{x}_1 = \frac{1}{n_1} \sum_{i=1}^{n_1} x_i \quad s^2_{x_1} = \frac{1}{n_1-1} \sum_{i=1}^{n_1} (x_i - \bar{x}_1)^2$$ $$\bar{y}_1 = \frac{1}{n_1} \sum_{i=1}^{n_1} y_i \quad s^2_{y_1} = \frac{1}{n_1-1} \sum_{i=1}^{n_1} (y_i - \bar{y}_1)^2$$ $$b = \frac{\sum_{i=1}^{n_1} (x_i - \bar{x}_1)(y_i - \bar{y}_1)}{(x_i - \bar{x}_1)^2} \quad r = b \frac{s_{x_1}}{s_{y_1}}$$ $b$ and $r$ are estimates of the regression coefficient and the correlation coefficient $\rho$, respectively. For the primary data series lengthened by the $n_2$ calculated values obtained by Eq. (6) Matalas and Jacobs (1964) derived the following revised estimates for the state parameters $\mu$ and $\sigma^2$ $$\bar{y}_{1+2} \equiv \bar{y}_1 + \frac{n_2}{n_1+n_2} b (\bar{x}_2 - \bar{x}_1)$$ (7) $$s^2_{y_{1+2}} = \frac{1}{n_1+n_2-1} \left[ (n_1-1)s^2_{y_1} + (n_2-1)b^2s^2_{x_2} + \right.$$ $$\left. + \frac{n_1n_2}{n_1+n_2} b^2 (\bar{x}_2 - \bar{x}_1)^2 + (n_2-1)\alpha^2 (1-r^2)s^2_{y_1} \right]$$ (8) where $$\alpha^2 = \frac{n_2(n_1-4)(n_1-1)}{(n_2-1)(n_1-3)(n_1-2)} \quad \bar{x}_2 = \frac{1}{n_2} \sum_{i=n_1+1}^{n_2} x_{n_1+i}$$ $$s^2_{x_2} \equiv \frac{1}{n_2-1} \sum_{i=1}^{n_2} (x_{n_1+i} - \bar{x}_2)^2$$ Furthermore they derived expressions for the variance of the estimates given by Eqs. (7) and (8). \[ \text{var}\{\bar{y}_{1+2}\} = \frac{\sigma^2}{n_1} \left[ 1 - \frac{n_2}{n_1 + n_2} (\rho^2 - \frac{1-\rho^2}{n_1-3}) \right] \tag{9} \] \[ \text{var}\{s^2_{\bar{y}_{1+2}}\} = \frac{2\sigma^4}{n_1-1} + \frac{n_2\sigma^4}{(n_1+n_2-1)^2} (A\rho^4+B\rho^2+C) \tag{10} \] $ A, B, $ and $ C $ are constants given by rather lengthy expressions of $ n_1 $ and $ n_2 $ (Matalas and Jacobs. 1964), $ \rho $ is the linear correlation coefficient. By comparing the variances given by Eqs. (9) and (10) with the corresponding expressions for the variances of estimates obtained on the basis of $ n $ independent observations \[ \text{var}\{\bar{y}\} = \frac{\sigma^2}{n} \quad \text{var}\{s^2\} = \frac{2\sigma^4}{n-1} \tag{11} \] we may define an equivalent number of independent observations with respect to estimation of the mean \[ n_{em} = n_1 + n_{2em} = n_1 \left[ 1 - \frac{n_2}{n_1 + n_2} (\rho^2 - \frac{1-\rho^2}{n_1-3}) \right]^{-1} \tag{12} \] and with respect to estimation of the variance \[ n_{ev} = n_1 + n_{2ev} = 1 + 2\left[ \frac{2}{n_1-1} + \frac{n_2}{(n_1+n_2-1)^2} (A\rho^4+B\rho^2+C) \right]^{-1} \tag{13} \] Firstly we realize that the equivalent number of independent observations with respect to the mean and the variance are in general different functions of $ n_1, n_2 $ and $ \rho $. Secondly we realize that only when the right hand side of Eq. (12) is greater than $ n_1 $ the lengthened series contains more information about the mean than does the short series. Similarly we realize from Eq. (13) that only when the right hand side of this equation is greater than $ n_1 $ the lengthened series contains more information about the variance than does the short series. The first immediate reaction in relation to the desired revision of the probability density function $ f(\mu, \sigma^2 \| n_1, \bar{y}_1, s^2) $ due to the incorporation of $ n_2 $ secondary observations could be to substitute for $ \bar{y}_1 $ and $ s^2_1 $ the corresponding revised expressions given by Eqs. (7) and (8). At the same time $ n_1 $ should be substituted by an equivalent number of primary observations that contains as much information as $ n_1 $ primary observations plus $ n_2 $ secondary observations. As described above this last question has, however, no simple and unique answer. Thus $ n_2 $ secondary observations contain as much information about the mean as $ n_{2em} $ (Eq. (12)) primary observations while they contain as much information about the variance as $ n_{2ev} $ (Eq. (13)) primary observations where in general $ n_{2em} \neq n_{2ev} $. In the following is suggested a heruristically based approach to circumvent the problem of how to make relevant substitutions for $ n_1 $ in the probability distribution $ f(\mu, \sigma^2 \| n_1, \bar{y}_1, s^2_1) $. We observe that the probability distribution given by Eq. (1) can be interpreted as a product of two independent distributions. The terms in the first bracket is the wellknown Normal distribution for the sample mean with parameters $(\mu, \sigma^2/n)$. Except for a factor $(n-3)/(n-1)$ close to unity the terms in the second bracket is the well-known Gamma distribution for the sample variance with parameters $((n-1)/2, 2\sigma^2/(n-1))$. These observations in combination with the possibility of calculating the equivalent number of independent primary observations corresponding to the sample mean $n_{em}$ as well as the sample variance $n_{ev}$, (Eqs. (12) and (13)) contain the clue to the problem. By substituting $n_{em}$ for $n$ in the first bracket of Eq. (1) and $n_{ev}$ in the second bracket we take approximately account of the different amount of information contained in a secondary data set of length $n_2$ with respect to the sample mean and the sample variance of the primary observations. Thus in summary it is suggested in this study to arrive at a probability density function of the state variable $\mu$ and $\sigma^2$ given a set of $n_1$ primary data and $n_1 + n_2$ secondary data by substituting for $\bar{y}_1$ and $s^2_1$ in $f(\mu, \sigma^2 \| n_1, y_1, s^2_1)$ the improved expressions given by Eqs. (7) and (8) and at the same time substituting for $n_1$ different expressions (Eqs. (12) and (13)) dependent on the place of occurrence of $n_1$. Hereby the following revised density function is obtained \[ g(\mu, \sigma^2 \| x_1, \ldots, x_{n_1+n_2}, y_1, \ldots, y_{n_1}) = \\ \sqrt{\frac{1}{2\pi \sigma^2 / n_{em}}} \exp\left(-\frac{1}{2} \frac{(\mu - \bar{y}_{1+2})^2}{\sigma^2 / n_{em}}\right) \frac{n_{ev}^{-1}}{\Gamma\left(\frac{n_{ev}-3}{2}\right)} \frac{1}{s^2_{y_{1+2}}} \left(\frac{s^2_{y_{1+2}}}{\sigma^2}\right)^{\frac{n_{ev}-1}{2}} \\ \cdot \exp\left(-\frac{n_{ev}-1}{2} \frac{s^2_{y_{1+2}}}{\sigma^2}\right) \] (14) Having obtained this distribution it is possible to perform all the Bayesian risk calculations described in the foregoing section. By substituting $g()$ for $f()$ we may for instance by means of Eq. (5) calculate $EOL(n_1, n_2, \rho)$ corresponding to the situation with $n_1$ primary data and $n_1 + n_2$ secondary data, where the primary and secondary data are linearly correlated with a correlation coefficient $\rho$. Having obtained $EOL(n_1, n_2, \rho)$ an equivalent number of primary observations $n_e$ may be defined as shown in Fig. 1. Thus we may conclude that for the particular problem considered $n_2$ secondary observations are as valuable as $n_{2e} = n_e - n_1$ primary observations. It has to be emphasized that because the weighing is performed in economic terms the calculations depend on the applied goal function why the obtained results are not universal but intimately related to the specific case. Case Study. Assumption and Results Probability distribution of low flows As previously mentioned the economic value of low flow data is considered in the present study with special reference to the design of treatment plants for typical Danish conditions. Table 1 \| River \| Time period \| Median minimum (l/s) \| Mean $\bar{y}$ (l/s) \| Standard deviation $s_y$ (l/s) \| $s_y/\bar{y}$ \| \|------------------------\|-------------\|----------------------\|----------------------\|-------------------------------\|---------------\| \| Ringsted Å \| 1950-59 \| 100 \| 113 \| 50 \| 0.44 \| \| Middle Zealand \| \| \| \| \| \| \| Halleby Å \| 1932-59 \| 143 \| 162 \| 64 \| 0.40 \| \| Western Zealand \| \| \| \| \| \| \| Ryom Å \| 1933-59 \| 180 \| 178 \| 75 \| 0.42 \| \| Eastern Jutland \| \| \| \| \| \| \| Uggerby Å \| 1931-59 \| 347 \| 358 \| 118 \| 0.35 \| \| Northern Jutland \| \| \| \| \| \| \| Arup Å \| 1936-59 \| 525 \| 533 \| 123 \| 0.23 \| \| Jutland \| \| \| \| \| \| \| Guden Å at Egeballe \| 1931-55 \| 705 \| 725 \| 108 \| 0.15 \| \| Middle Jutland \| \| \| \| \| \| Low flow data for six Danish streams have been selected and their distributions examined. By plotting on normal probability paper the distributions have been found to be approximately normal. An example is shown in Fig. 2. Some main characteristics are given in Table 1. ![Graph](image) Fig. 2. Plotting of annual minimum streamflows for Uggerby Å 1931-59. The goal function The goal function is constructed on the basis of the diagram in Fig. 3, showing the cost of a wastewater treatment plant as a function of the design streamflow, the amount of wastewater, and the degree of purification. The diagram is prepared by civil engineer Jan Hassing, Cowiconsult and it is constructed under the following assumptions: i mechanical-biological treatment, ii one person equivalent (p.e.) is the same as 250 l/day with a $BOD_5$ concentration of 200 mg/l. The maximum hourly flow into the plant is assumed to be 1/12 of one day’s flow, iii the critical output concentration of $BOD_5$ from the plant is determined by mass balance calculations on the basis of the design discharge of the stream in combination with a critical $BOD_5$ concentration of 2 mg/l in the stream just downstream the outlet, as recommended by the Agency of Environmental Protection (1974). Fig. 3. Cost of sewage treatment plant (construction + capitalized working expenses). Interest rate $i = 15\%$ and design lifetime $L = 30$ years. prices in 1974 D.kr. (Prepared by civil engineer Jan Hassing, Cowiconsult). The case considered in the present study is a community corresponding to 10.000 p.e. From the diagram in Fig. 3 a cost function $COST(M, M_{alt})$ is constructed, showing the extra cost of the plant designed for a median minimum streamflow value $M_{alt}$, while the realized median minimum within the design lifetime, $L$, turns out to be $M$. As stated earlier the annual minimum $Y$ follows a normal distribution with parameters $\mu$ and $\sigma^2$. The distribution of the median $M$ of $L$ values of $Y$, $\phi(M\|\mu, \sigma^2)$, is shown by Gumbel (1958) to be approximately normal, with mean $\mu$ and variance $1.56 \sigma^2 / L$. The goal function is the expected value of the cost function, given the state parameters $\mu$ and $\sigma^2$. The functional form is $$G(M_{alt}\|\mu, \sigma^2) = \int_0^\infty COST(M, M_{alt}) \cdot \phi(M\|\mu, \sigma^2) dM$$ \hspace{1cm} (15) By the construction of the cost function two cases have to be considered, namely underdesign and overdesign. Case 1: Underdesign, $M_{alt} > M$ In this case more pollution is made than society can accept. Therefore the cost ought to be connected with the consequences of this extra pollution. This is, however, a very difficult problem (a study in itself) because tangibles as well as intangibles are involved. Instead $COST(M,M_{alt})$ is chosen somewhat arbitrarily as the cost by extending the plant to a capacity large enough to meet the requirements of society. These extra costs are said to be the difference between the costs of a plant designed for $M_{alt}$ and a plant designed for $M$, plus an overhead of 20%. Case 2: Overdesign, $M_{alt} < M$ In this case a bigger capital investment than necessary is made. Therefore $COST(M,M_{alt})$ is the difference between the costs of a plant designed for $M$ and a plant designed for $M_{alt}$. In this way $COST(M, M_{alt})$ is found as the curves shown in Fig.4. For computational purposes the cost function is wanted in analytical form so the cost function has been approximated by linear expressions as indicated in Fig.4. ![Graph showing cost function](image) Fig. 4. The cost function cost $(M,M_{alt})$ showing the cost of either a realized overdesign or underdesign: (1) cost curve as computed from Fig. 3. (2) approximated cost curve used in the computations. Results The computations of Bayes Risk, Bayes Solution and $EOL$ (Eqs. (1) - (5)) have been made on a digital computer. The necessary computer programs have been placed at our disposal by civil engineer Sven Jacobi, Ph.D. The following values of the sample statistics have been used \[ \begin{align} \bar{y} &= 300 \text{ l/s} \\ s^2 &= (60 \text{ l/s})^2, (105 \text{ l/s})^2, (150 \text{ l/s})^2 \\ n_1 \varepsilon [6,50] \text{ years} \end{align} \] In Eq.(16) the range of ratio $ s/\bar{y} $ is chosen to be 0.2-0.5 in accordance with typical Danish conditions (see table 1). The results of the EOL computations are shown in Fig.5, where the importance of the sample length and the sample variance are seen. Fig. 5. EOL as a function of the length, $ n_1 $, of the primary set of data for different ratios between the sample mean ($ \bar{y} = 300 \text{ l/s} $) and the sample variance, $ s^2 $. By means of the approach outlined in the foregoing section similar computations have been made including secondary sets of data with the sample statistics \[ \begin{align} n_1 &= 10 \text{ years}; \quad n_2 = 50 \text{ years} \\ \bar{y}_1 &= \bar{x}_1 = \bar{x}_2 = 300 \text{ l/s} \\ s^2_{y_1} &= s^2_{x_1} = s^2_{x_2} = (105 \text{ l/s})^2 \\ \rho \varepsilon [0.30,1.00] \end{align} \] The sample mean and the sample variance have without loss in generality been assumed to be identical in the primary and the secondary series. By use of the formulas derived in the theoretical section it is now possible to calculate the equivalent number of primary observations, $n_{2e}$, corresponding to $n_2$ secondary observations. $n_{2e}$ is seen to be a function of $n_1$, $n_2$, and $\rho$. As an example Fig. 6 shows $n_{2em}$, $n_{2ev}$, and $n_{2e}$ as functions of $\rho$ for given values of $n_1$, and $n_2$. It is seen that the equivalent length of a secondary set of data with respect to the mean, $n_{2em}$, is much greater than the equivalent length with respect to the variance, $n_{2ev}$. Besides it is realized that by the economic weighing $n_{2em}$ is dominating relative to $n_{2ev}$. This last result is in good agreement with the study of Jacobi (1974) who found that information about the mean was more important than information about the variance. ![Graph showing $n_{2em}$, $n_{2ev}$, and $n_{2e}$ as functions of $\rho$](image) Fig. 6. Equivalent length of a secondary set of data as a function of the correlation $\rho$ between primary and secondary data. Using Figs. 5 and 6 it is now possible to find the worth of a secondary set of data $B(n_1, n_2, \rho)$ as a decrease in $EOL$. Thus for the specific case considered in this study corresponding to $n_1 = 10$ years of available primary observations supplemented by $n_2 = 50$ years of secondary observations it is observed that the worth of the secondary data depends heavily on the value of the linear correlation coefficient $\rho$, between primary and secondary data. The cost of obtaining the secondary information depends on which procedure one applies. Utilizing for instance a numerical rainfall-runoff model like the one developed by Nielsen and Hansen (1973) computations of daily values of streamflow may be performed after calibration of the model on the basis of input of daily values of precipitation and temperature together with monthly values of potential evapotranspiration. The cost of collecting the necessary meteorological observations and performing the computer calculations has to be compared with the worth of the obtained information as given in Table 2. We realize that for the case considered it is economically sound to spend up to D.kr. 10,000 - 70,000 in obtaining secondary information depending on the correlation coefficient $\rho$. \| $\rho$ \| 0.50 \| 0.60 \| 0.70 \| 0.80 \| 0.90 \| \|--------\|------\|------\|------\|------\|------\| \| $B(10,50,\rho)$ (D.kr.) \| 10.200 \| 19.400 \| 34.100 \| 51.200 \| 70.800 \| In order to get an idea of the potential economic benefits of transmitting information on low flows by means of for instance a rainfall-runoff model let us consider a not too unusual situation where primary observations of minimum discharges are only available for $n_1 = 3$ years. If a treatment plant for 10,000 p.e. is built on the basis solely of this information of the low-flow conditions in the stream an $EOL$ of approximately D.kr. 500,000 may result. Now unfortunately it is not possible to apply the technique of transmitting information from a secondary set of data by means of a linear regression model. This is due to the fact that a regression model based on as few as $n_1 = 3$ values contains by far too much uncertainty to transform any real information. Of course this does not mean that secondary data calculated by means of a rainfall-runoff model are without any value. If the applied model from experience with other streams is known to simulate low-streamflow values in an appropriate way it might be possible to transmit from previous meteorological observations an information equivalent to for instance 7 years of primary low-streamflow values. Utilizing again Fig. 5 we realize that the resulting $EOL$ has decreased to approximately D.kr. 100,000. Thus a total decrease of approximately D.kr. 400,000 is experienced by utilizing a rainfall-runoff model in this case. This result indicates that the economic value of secondary data under certain circumstances may have a value of an order of magnitude larger than that given in Table 2. Summary and Conclusions Nonoptimal decisions due to lack of basic data are frequently occurring within the field of water resources management. In the present study low streamflows and their importance for the design of treatment plants have been considered and evaluated in economic terms by means of the Bayesian decision approach. For a specific case the worth of primary data in form of direct observations of low streamflows have been investigated in terms of an expected opportunity loss (EOL) as a function of the number of primary observations. Indirect information concerning the low-flow properties of interest may be obtained by means of a secondary set of data in combination with a regression model. In this study is suggested a heuristic method for the economic evaluation of a secondary set of data taking into account the uncertainty embedded in the regression model. It is shown that corresponding to a situation with $n_1$ primary observations and $n_1 + n_2$ secondary observations an equivalent number of primary observations $n_{2e}$ may be defined so that the $n_{2e}$ primary data give raise to the same decrease in EOL as the $n_2$ secondary observations. For typical Danish conditions it is shown that the economic value of secondary information concerning low-streamflow values as obtainable by means of a rainfall-runoff model lies in a very wide range of D.kr. 10,000 - 400,000 depending upon the available number of primary observations and the functioning of the rainfall-runoff model. If for instance 10 years of primary observations are available a rather moderately operating rainfall-runoff model may produce calculated secondary low-flow values worth approximately D.kr. 20,000. If on the other hand only 3 years of primary data are available a good rainfall-runoff model may produce calculated low-flow values worth approximately D.kr. 200,000. References Davis, D. R. (1971) Decision making under uncertainty in systems hydrology. Tech. Report No. 2, Hydrology and water resources interdisciplinary program, University of Arizona, Tucson, Arizona. Davis, D. R., Kisiel, C. C., and Duckstein, L. (1972) Bayesian decision theory applied to design in hydrology. Water Resources Research, 8, No. 1, pp. 33-41. Gumbel, E. J. (1958) Statistics of extremes. Colombia University Press, New York and London. Jacobi, S. (1974) Economic worth of sediment load data in a statistical decision framework. Ph. D. dissertation, Civil Engineering Department, Colorado State University, Fort Collins, Colorado. Jacobi, S. (1975) Economic optimum record length. Nordic Hydrology, 6, pp. 28-42. Matalas, N. C. and Jacobs, B. (1964) A correlation procedure for augmenting hydrologic data. U.S. Geological Survey, Professionel paper 434-E, Washington. Nielsen, S. A. and Hansen, E. (1973) Numerical simulation of the rainfall-runoff process on a daily basis. Nordic Hydrology, 4, pp. 171-190. Vejledende bestemmelser for udlædning af spildevand. Vejledning nr. 6/1974. København (Recommendations for discharge of wastewater. Agency of Environmental Protection). Vicéns, G. J., Rodriguez-Iturbe, J., and Schaake jr., J. C. (1974) A Bayesian approach to hydrologic time series modelling. M.I.T. Ralph M. Parsons Laboratory for water resources and hydrodynamics. Report No. 181, Cambridge, Mass. Wood, E. F., Rodriguez-Iturbe, J., and Schaake jr., J. C. (1974) The methodology of Bayesian inference and decision making applied to extreme hydrologic events. M.I.T. Ralph M. Parsons Laboratory for water resources and hydrodynamics. Report No. 178, Cambridge, Mass. Received: 12 December, 1975 Address: Institute of hydrodynamics and hydraulic engineering ISVA Technical University of Denmark Bldg. 115, DK-2800-Lyngby, Denmark
Memo To: ALL PROBATION OFFICERS From: Sandra M. Brulo Chief Juvenile Probation Officer Date: February 20, 2003 By order of Judge Ciavarella ......... Youth on probation are to be violated and detained for any violation of the terms of their probation. Examples: zero tolerance, not attending school, not attending appointments, curfew. When in doubt ask me or your supervisor. Thank you. SMB/wam \| Orientation Training \| \| \|----------------------\|---\| \| Post Orientation - Supervisor observes P.O., and may use checklist to measure P.O.'s competency in Core Job Tasks \| \| \| - Intake Assessment \| \| \| - Disposition Investigation \| \| \| - Probation Supervision \| \| \| - Home Visit \| \| \| - Community Visit \| \| \| - Court Hearing \| \| \| Modules - Trainee completes following trainings within X years: \| \| \|---------------------------------------------------------------\|---\| \| - Phase I Staff Safety – 9 hrs. \| \| \| - Educational Law/Working with Schools – 12 hrs. \| \| \| - Adolescent Substance Abuse: Trends, Intervention and Ref. Skills – 9/12 hrs. \| \| \| - Assessment, Case Planning, Interviewing & Counseling Skill Building – 15 hrs. \| \| \| - Cultural Competence – 9 hrs. \| \| \| - Introduction to Group Counseling Techniques – 9 hrs \| \| To: All Probation Staff From: Sandra M. Brulo Chief Juvenile Probation Officer Date: November 19, 2003 Judge Ciavarella has agreed to reinstate a graduated sanction policy for youth as well as new standards for first time offenders. While I have outlined some basics, most of the cases will be reviewed individually to determine what consequences will be imposed for probation violators. First time offenders who are coming through intake will be reviewed every Wednesday at a staffing to make a determination as to what sanctions will be imposed. I ask you to please ask if you have any questions. GRADUATED SANCTIONS POLICIES LUZERNE COUNTY JUVENILE PROBATION OFFICE I. First Time Offenders- Youth at the Intake and Detention Center Level 1. Intake—Cases will be discussed each Wednesday following the Detention Meeting. Please note—Youth requiring drug and alcohol rehabilitation will not necessarily be required to initially attend Camp Adams but rather go directly to drug and alcohol rehabilitation. II Releases from Drug and Alcohol Rehabilitation Intensive drug and alcohol treatment plus other conditions of probation. III Youth Currently on Probation: Formal/Consent Decree 1. Drug and Alcohol Violations Exception: heroine, cocaine A. First Time Detention - 72 hours House Arrest with monitor Intensive Drug and Alcohol treatment Extension of Probation by 2 months B. Second Time Detention - 72 hours House Arrest with monitor Intensive drug and alcohol treatment 4 weekends at Camp Adams (males) Enrollment in Saturday Aftercare Program (females) Extension of Probation by 2 additional months. C. Third Time Residential Placement 2. General Technical Probation Violations A. First Time House Arrest with monitor Extension of Probation by 2 months B. Second Time Detention - 72 hours House Arrest with monitor Extension of Probation by 3 additional months C. Third Time Residential Placement IV Violation of Informal Probation A. Youth will return to Court for hearing on original charges B. Upon adjudication, youth will spend 72 hrs in Detention C. A new Probation Plan will be developed at a Wednesday Staff Meeting. Waiver of Right to Counsel I am aware I have the right to counsel in the juvenile matter before the Court. I have consulted and been advised by a responsible adult who is aware of the fifth and sixth amendment rights guaranteed to me by the United States Constitution. I AM WAIVING MY RIGHT TO COUNSEL. _________________________________________ _______________________ Name of Juvenile Date _________________________________________ _______________________ Name of Advising Adult Date Relationship to Juvenile: ________________________________ I have no conflict of interest in the juvenile matter before the Court and have advised the juvenile to waive counsel without bias. _________________________________________ _______________________ Name of Advising Adult Date The Court has reviewed the waiver of counsel statement with the juvenile and is permitting the case to proceed to hearing. BY THE COURT, _________________________________________ _______________________ J. WAIVER OF COUNSEL (Understanding the Consequences) 1. Does the defendant and parent/guardian understand that he or she has the right to be represented by counsel, and the right to have free counsel appointed if he or she is indigent? 2. Does the defendant and parent/guardian understand the nature of the charges against him or her and the elements of each of those charges? 3. Is the defendant and parent/guardian aware of the permissible range of dispositions (probation, consent decree, residential placement, detention, certification) and/or fines for the offenses charged? 4. Does the defendant and parent/guardian understand that if he or she waives the right to counsel, he or she will still be bound by all the normal rules of procedure and that counsel would be more familiar with these rules? 5. Does the defendant and parent/guardian understand that there are possible defenses to these charges of which counsel might be aware, and if these defenses are not raised at trial they may be lost permanently? 6. Does the defendant and parent/guardian understand that, in addition to these defenses, the defendant has many rights that, if not timely asserted, may be lost permanently; and that if errors occur and are not timely objected to, or the objections are not otherwise timely raised by the defendant, these errors may be lost permanently? _________________________ ________________________________ Date Defendant _________________________ ________________________________ Date Parent / Guardian JUVENILE PROBATION OFFICE and JUVENILE COURT 11TH JUDICIAL DISTRICT PENNSYLVANIA COURT OF COMMON PLEAS COUNTY OF LUZERNE WRITTEN WAIVER OF COUNSEL We, ____________________________, have been told that we have the right to have a lawyer represent the juvenile defendant, and the right to have a free lawyer appointed to represent the juvenile defendant without cost if we cannot afford to pay a lawyer. We give up these rights and knowingly and voluntarily choose to proceed without the assistance of an attorney. We hereby acknowledge that we have read and signed the Waiver of Counsel document attached hereto and made a part hereof. We further acknowledge that we understand the rights as set forth in the signed Waiver of Counsel. ______________________________ ________________________________ Date Defendant ______________________________ ________________________________ Date Parent / Guardian I have determined that the defendant has made a knowing, voluntary and intelligent waiver of this right to counsel. ______________________________ P.J. All Luzerne County Juvenile Probation Intake Officers will abide by all Pre-Adjudicatory Procedures, Commencement of Proceedings, Detention, Transfer, Adjudication, Disposition and Post-Disposition Reviews rules outlined in the Pennsylvania Supreme Court Rules of Juvenile Probation and Juvenile Act with the understanding that certain parts of the Juvenile Act are superceded by the Supreme Court Rules of Juvenile Probation. Additional Duties of a Juvenile Division Intake Officer are as follows: 1. Intake Officer is responsible for learning and understanding the Pennsylvania Crime Codes. 2. Intake Officer is responsible for learning Medical Necessity Criteria that qualifies a youth for Medical Assistance. 3. Intake Officer, during the interview, will ask the youth if he/she admits to the charge(s) and then document the response. 4. Intake Officer will comment in writing to any questions with (YES) response on the Problem Severity Index. It is the duty of the intake officer to expound on any question in the interview process that could lead to more useful knowledge of case. 5. Intake Officer will prepare in writing recommendations to the courts containing the following information: 1) Name of client; 2) DOB of client; 3) Date of recommendation; 4) Order for Forensic Psychological, and Psychiatric when necessary; 5) Brief description of recommendation (body of recommendation must contain): Why you made the current recommendation; What underlying issues are presently occurring if any exists (define briefly); What alternatives could work for child and family; Suggestions for placements (names of facilities) that could meet the needs of the youth if placed by the Courts. 6. All recommendations must be signed by Intake Officer and a Supervisor in the Juvenile Probation Department. 7. All Intake Officers must learn and be able to perform all duties associated with the intake department, i.e., Acceptance of allegations and submission to the Clerk of Courts; Assigning court dates to the allegations; Completing record checks; Assigning cases to an intake officer; Completing Informal; Scheduling and conducting detention hearings; Preparing court materials and conducting juvenile court hearings, as well as all other duties that fall under the intake department. VIOLATION OF PROBATION / PROCEDURE EFFECTIVE IMMEDIATELY Procedure after a Violation of Probation: 1. Probation Officer is responsible to make sure that clients on their caseload who violate HIS/HER probation must be contacted face to face twice a week after their violation hearing for a period of six weeks. 2. ONE FACE TO FACE CONTACT MUST OCCUR WITH CHILD AND PARENT PRESENT; THIS CONTACT MUST OCCUR OUTSIDE NORMAL BUSINESS HOURS; AND THIS MUST OCCUR FOR A SIX WEEK PERIOD. AFTER SIX WEEKS THE CASE MAY BE REDUCED BACK TO ONE TIME PER WEEK VISIT, AND ALL COMMUNITY BASED STANDARDS WILL APPLY. A SUPERVISOR MUST SIGN OFF ON THE REDUCTION TO MAKE IT VALID. ALL MEETINGS DURING THIS SIX-WEEK PERIOD MUST BE DOCUMENTED IN THE COMPUTER SYSTEM. 3. IN ORDER TO ACCOMPLISH THIS PROCEDURE A PROBATION OFFICER CAN REARRANGE THEIR SCHEDULE, HOWEVER, ANY CHANGE TO SCHEDULES NEED TO BE PREAPPROVED ONE-WEEK IN ADVANCE. The request must be written and sent to Executive Secretary via email for placement on the weekly roster. 4. Allowable schedule changes can be any period in which a probation officer can work a full day’s compliment of 7 hours. This also includes working on a Saturday as well as working on a Sunday. Saturday and Sunday work if approved will require that an officer work a full 7-hour compliment, and he or she take one day off during the next week. A supervisor will be available during those office times. Luzerne County Probation Services (Juvenile Division) Policy and Procedure Intake Department All Luzerne County Juvenile Probation Intake Officers will abide by all Pre-Adjudicatory Procedures, Commencement of Proceedings, Detention, Transfer, Adjudication, Disposition and Post-Disposition Reviews rules outlined in the Pennsylvania Supreme Court Rules of Juvenile Probation and Juvenile Act with the understanding that certain parts of the Juvenile Act are superceded by the Supreme Court Rules of Juvenile Probation. Additional Duties of a Juvenile Division Intake Officer are as follows: 1. Intake Officer is responsible for learning and understanding the Pennsylvania Crime Codes. 2. Intake Officer is responsible for learning Medical Necessity Criteria that qualifies a youth for Medical Assistance. 3. Intake Officer, during the interview, will ask the youth if he/she admits to the charge(s) and then document the response. 4. At the conclusion of the Intake Process, the Intake Officer shall meet with the Deputy Director of Forensic Programs to review the intake information and, if required, make an appropriate recommendation to the court. If necessary, the Deputy Director of Forensic Programs will meet with the child and the child’s parents either during or subsequent to the intake process. If after a recommendation is made and the recommended placement cannot be accomplished, the case shall be returned to the Director of Forensic Programs for further recommendation. Any PSI with a substance abuse indicator will be referred to the Substance Abuse Specialty Court for review and recommendation. The Specialty Courts review is to be considered part of the intake process and any Specialty Courts recommendation will be presented to the court for its dispositional consideration. The Deputy Director of Forensic Programs will review and approve all after care plans for juveniles being discharged from a residential treatment facility. The Deputy Director of Forensic Programs will be responsible for the training of intake officers and any additional training required. The above Policy and Procedure is hereby effective on February 26th, 2008 and will be followed by the Juvenile Division of the Luzerne County Probation Services Department. Honorable President Judge & Juvenile Court Judge Director of Probation Services 2/26/08 The above Policy and Procedure is hereby effective on February ______ 2008 and will be followed by the Juvenile Division of the Luzerne County Probation Services Department. Honorable President Judge & Juvenile Court Judge ______________________________ Director of Probation Services ______________________________ Deputy Chief of Administration March 26, 2009 All Luzerne County Police Chiefs & Pennsylvania State Police Commanders RE: JUVENILE DETENTION Dear Sirs: Please be advised that Juvenile Probation will utilize a Detention Assessment Tool and the following Detention Standards prior to admitting a juvenile to secure a detention: DETENTION: Detention may be authorized for juveniles who have committed any of the following offenses: - Murder 18 Pa.C.S.A. § 2502 / Voluntary Manslaughter 18 Pa.C.S.A. § 2503 / Involuntary Manslaughter 18 Pa.C.S.A. § 2504; - Rape 18 Pa.C.S.A. § 3121 / IDSI 18 Pa.C.S.A. § 3123 / Aggravated Indecent Assault 18 Pa.C.S.A. § 3125; - Robbery 18 Pa.C.S.A. § 3701 / Robbery of a motor vehicle 18 Pa.C.S.A. § 3702; - Aggravated Assault 18 Pa.C.S.A. § 2702; - Kidnapping 18 Pa.C.S.A. § 2901; - Arson 18 Pa.C.S.A. § 3301; - Burglary of a structure adapted for overnight accommodation 18 Pa.C.S.A. § 3502; - Terroristic Threats 18 Pa.C.S.A. § 2706; - Stalking 18 Pa.C.S.A. § 2709.1; - Risking Catastrophe 18 Pa.C.S.A. § 3302; - Riot 18 Pa.C.S.A. § 5501; - Drug Felonies 34 P.S. § 780-113(a)(30); - Retaliation against victims/witness 18 Pa.C.S.A. § 4952; - Any Offense utilizing a deadly weapon/firearm/explosive; - Warrant from another jurisdiction. Any of the felonies against persons or felonies involving a deadly weapon shall result in automatic detention if it is deemed necessary. Other factors to consider when judging whether or not detention is warranted: - Whether the juvenile is a repeat felony offender; - If the juvenile is on probation at the time of the new violation; - If the juvenile has a history of failure to appear (if known at the time); - If the juvenile has previously escaped or run away from home or a facility; - Parent/guardian refusal to provide appropriate supervision; - Juvenile poses a risk to the community; - Juvenile has a history of violence in the home; - Juvenile has a serious substance abuse problem. DETENTION ALTERNATIVES: Officers should seek alternatives when warranted prior to requesting secure detention: - Contact Children and Youth if the child has no supervision and request that Children and Youth provide shelter care for the juvenile (they have a 24hr on call person); - Home Detention may be considered with the request that within 24 hours electronic monitoring be established in the home; - Request Juvenile Probation to expedite court scheduling; - Explore options for alternative supervision with relatives of the juvenile; - Worse case scenario, the officers may detain the juvenile in their station cells as long as the juvenile is in a cell by themselves or with their other juvenile co-defendants (i.e. no adults) and the juvenile is removed from the cell as soon as other suitable arrangements may be made. If the juvenile is intoxicated at the time of arrest, Juvenile Probation must refuse to detain due to Detention's policy. If secure detention is warranted the juvenile may be admitted once they are medically cleared through a hospital. FYI: If the detention Center in Pittston Twp. is at capacity, the officer will have to transport the detained juvenile to an alternative location. DIRECT FILING TO CRIMINAL COURT: A juvenile over the age of fifteen (15) at the time of the offense may be charged as an adult if they are charged with the following crimes and a deadly weapon is utilized (FYI: a BB gun is considered a deadly weapon): - Voluntary manslaughter 18 Pa.C.S.A. § 2503; - Aggravated assault 18 Pa.C.S.A. § 2702 sections (a)(1) or (a)(2) (only); - Rape 18 Pa.C.S.A. § 3121 /IDSI 18 Pa.C.S.A. § 3123 /Aggravated Indecent Assault 18 Pa.C.S.A. § 3125; - Kidnapping 18 Pa.C.S.A. § 2901; - Robbery 18 Pa.C.S.A. § 3701 sections (a)(1)(i)/(a)(1)(ii)/(a)(1)(iii) (only); - Robbery of a motor vehicle 18 Pa.C.S.A. § 3702; - Attempt 18 Pa.C.S.A. § 901/conspiracy 18 Pa.C.S.A. § 903/solicitation 18 Pa.C.S.A. § 902 of any of above. If a juvenile over the age of fifteen (15) at the time of the offense is charged with any of the above listed offenses and was previously adjudicated of any of the above offenses, he/she may be charged as an adult (aggravated assault is not included in this provision). A juvenile previously convicted of any offense graded above a summary in any jurisdiction should be charged as an adult for any offense they commit regardless of the grading (once an adult, always an adult). If you have any questions, please don’t hesitate to contact us at (570) 825-1674. Thank you for your attention to this matter. Very truly yours, BY: ________________________________ JACQUELINE MUSTO CARROLL District Attorney _______________________________ STEFANIE L. POLLOCK Assistant District Attorney JMC:sl Dear J.J., I have a weekly telephone conference call with Kidspace and discuss all referrals to and discharges from their program. This week, we discussed [redacted] and the referral from Luzerne JPO to their RTC for dual diagnosis children. I questioned what were the discharge recommendations from St. Michael's (where he was for approximately 12 months) and what are the recommendations by the BH-MCO. I was advised that the recommendations by CCBHO was for a residential D&A facility, licensed by the PA Department of Health and funded by the BH-MCO. I have no reason to disagree with this recommendation and therefore I am questioning why CCBHO is not making the proper referrals and reimbursing for the service and why Luzerne County JPO is making the referral and offering to pay. OCYF funds should not be used when treatment services are medically necessary. That is the responsibility of the BH-MCO. Please advise and thanks. Richard J. Gold Deputy Secretary Office of Children, Youth and Families Pennsylvania Department of Public Welfare Rm. 131, Health & Welfare Bldg, Commonwealth Ave. & Forster Street Harrisburg, PA 17105-2675 717-787-4758 Fax: 717-787-0414 firstname.lastname@example.org Dear Mr. Johnson, If the Problem Severity Index Tool indicates the need for a psychiatric evaluation, then a referral should be made to the BH-MCO in Luzerne County to work with the county on providing such an evaluation as soon as possible and within the time constructs outlined in the Juvenile Act. Such an evaluation will be face-to-face and not only based on the PSI assessment tool and the record. It is the responsibility of the BH-MCO to arrange for the evaluation and for all 3 systems (probation, MH and child welfare) to work together in an integrated manner. OCYF will not reimburse the county for psychiatric evaluations if the child is MA eligible since the BH-MCO has already been funded for such services. Richard Gold -----Original Message----- From: JOHNSON, JOHN [mailto:email@example.com] Sent: Thu 1/24/2008 12:04 PM To: Gold, Richard J Cc: SABA, LARRY Subject: Re: Dear Richard, In regards to your questions about [redacted] case, my response is as follows: Juvenile Probation does not make any recommendations to the courts about placement of youth, or recommend the need for Psychiatric evaluations. The Deputy Director of Forensics Program (Sandra M. Bruolo, M.P.A., L.S.W.) makes these recommendations to the courts based on her expertise as a clinician in the realm of mental health. She reviews the Problem Severity Index Tool completed by the JPO Intake staff who interviews the child and parent. Juvenile probation makes the referral once the recommendation has been made by Ms. Bruolo. I would ask that you refer your questions to Ms. Bruolo; she can be reached by phone or email (570) 825 1728, email firstname.lastname@example.org Thank You, JJ John E. Johnson, M.S. Deputy Chief of Administration (570) 825-1850 *mailto:email@example.com
The following Assistant Professors have been selected for UGC sponsored 79th Orientation Programme organized by UGC-Human Resource Development Centre commencing from 17/07/2018 to 13/08/2018. The Selected candidates are requested to report for joining positively at Patna College, Patna University, Patna 800005 on 17/07/2018 at 9:30 am along with necessary documents as mentioned under terms and conditions attached. The course will be conducted at the UGC-Human Resource Development Centre, Department of Education Building Campus, Dariyapur, Baripath, Patna, Patna University, Patna, from 17/07/2018 to 13/08/2018. Note: Selected Participants are requested to go through the Terms and Conditions attached minutely and report for the Course only when they are fully eligible for the Course. As per UGC Letter no F.no.1-13/2017 (HRDC) dated 04.06.2018, There shall be no payment of Dearness Allowance to participants of any courses. However Travelling allowance will continue as per UGC Guidelines. \| Sr. No \| Name Of Participants & Designation \| Department & Colleges \| Telephone No. \| \|--------\|-----------------------------------\|-----------------------\|---------------\| \| 1 \| Dr. Subalal Paswan, Assistant Professor \| Department Of Psychology, R.P.S. College, Chakeyaj, Mahnar, Vaishali \| 7070174529/9931860451 \| \| 2 \| Mr. Shyamanand Shandilya, Assistant Professor \| Department Of Maithili, J.N. College, Nehra, Darbhanga-847233 \| 9534338272 \| \| 3 \| Mr. Pramod Kumar Paswan Assistant Professor \| Department of Maithili U.P. College, PUSA(Samastipur) 848125 \| 9534882994 \| \| 4 \| Mr. Abhay Kumar , Assistant Professor \| Department Of Mathematics S.D. College, Kaler, Arwal-824127 \| 9555219241 \| \| 5 \| Dr. Shailja Singh Assistant Professor \| Department of Chemistry Patna Science college, Patna 800005 \| 9999249844 \| \| 6 \| Dr. Manish Kumar Verma, Assistant Professor \| Department Of Physics, Magadh Mahila College, Patna-800001 \| 9650844970 \| \| 7 \| Dr. Vikash Kumar Assistant Professor \| Department Of Psychology, R.C. College, Sakra, Dholi, Muzafferpur843105 \| 8102596563 \| \| \| Name and Title \| Department and College \| Contact Number \| \|---\|--------------------------------------\|-------------------------------------------------------------\|-------------------------\| \| 8 \| Dr. Bhavprita Kumari, Assistant Professor \| Department Of Psychology, S.B. College, Ara-802301, Bhojpur \| 9199424712 \| \| 9 \| Dr. Ashish Kumar, Assistant Professor \| Department Of English, Sri Ugaratara B.M. Sanskrit College, Mahishi, Saharsa-852216 \| 9466554818 \| \| 10\| Dr. Chakradhar Thakur, Assistant Professor \| Department Of Maithili, B.D. College, Patna \| 9631201629/0612-2280927 \| \| 11\| Mr. Sachchida Nand Kumar, Assistant Professor \| Department Of Chemistry, S.P.M. College, Udantpuri-803101, Nalanda \| 9939692677 \| \| 12\| Dr. Kumari Seema, Assistant Professor \| Department Of Chemistry, J.D. Women's College, Patna 800023 \| 9709601875 \| \| 13\| Mr. Ajay Kumar, Assistant Professor \| Department Of Mathematics, M.D. College, Naubatpur, Patna \| 9631900767/947068085 \| \| 14\| Dr. Shabnam Azmi, Assistant Professor \| Department Of Psychology, J.D. Women's College, 800023 \| 7277774324 \| \| 15\| Dr. Kalpana Singh, Assistant Professor \| Department of Philosophy, Dr. S.K.Sinha Women's College, Motihari \| 9523081472 \| \| 16\| Mr. Subrata Kumar Das, Assistant Professor \| Department of English, V.S.J College, Rajnagar, Madhubani-847235 \| 09441376438 \| \| 17\| Dr. Madhu Kumari Gupta, Assistant Professor \| Department of Chemistry, Magadh Mahila College, Patna-800001 \| 9470002618 \| \| 18\| Mrs. Maridula Sahay, Assistant Professor \| Department of Mathematics, Ganga Devi Mahila College, Lohia Nagar, Kankar Bagh, Patna 800020 \| 9304754133 \| \| 19\| Dr. Sudhir Kumar Jha, Assistant Professor \| Department of Maithili, Patna University, Patna-800005 \| 9661819662 \| \| No. \| Name and Title \| Department and Location \| Contact Number \| \|-----\|--------------------------------------\|--------------------------------------------------------------\|------------------\| \| 20 \| Dr. Jeebachh Ram, Assistant Professor \| Department of Maithili, Patna University, Patna-800005 \| 9525949639 \| \| 21 \| Dr. Sudhanshu Shekhar, Assistant Professor \| Department of Philosophy, T.P. College, Madhepura-852113 \| 9934629245 \| \| 22 \| Chandan Chandra Chunna, Assistant Professor \| Department of Economics, Hari Singh College, Havelli, Kharagpur, Munger811213 \| 9308312826 \| \| 23 \| Ram Nagina Rajak, Assistant Professor \| Department of Education, R. K. College, Madhubani \| 9470837307 \| \| 24 \| Dr. Renu Choudhary, Assistant Professor \| Department of Social & Social Anthropology, A.N. Sinha Institute of Social Studies, Patna 800001 \| 9431279357 \| \| 25 \| Dr. Aviral Pandey, Assistant Professor \| Department of Agriculture Economics, A.N. Sinha Institute of Social Studies, Patna 800001 \| 8987100796 \| \| 26 \| Dr. Sandhya Rani Mahapatro, Assistant Professor \| A.N. Sinha Institute of Social Studies, Patna 800001 \| 9470738204 \| \| 27 \| Dr. Dinesh Kumar Majhi, Assistant Professor \| Department of Mathematics, Patna College, Patna-800005 \| 7352547860 \| \| 28 \| Mr. Avinash Kumar, Assistant Professor \| Department of Mathematics, Patna College, Patna-800005 \| 8439540892 \| \| 29 \| Mrs. Kiran Kumari, Assistant Professor \| Department of Philosophy, A.P.S.M. College, Barauni-851112, Begusarai \| 8877596210 \| \| 30 \| Dr. Amrita Prasad, Assistant Professor \| Department of Chemistry, Magadh Mahila College, Patna 800001 \| 9955748253 \| \| No. \| Name and Title \| Department \| Address \| Contact \| \|-----\|----------------\|------------\|---------\|---------\| \| 31 \| Mrs. Deepshikha Choudhary, Assistant Professor \| Department of Geography, M.S.M. Samta College, Jandaha, Vaishali-844505 \| 9472060887/06227231101 \| \| 32 \| Dr. Pratyaksha Raj, Assistant Professor \| Department Of Philodophy, Ramesh Jha Mahila College, Near Thana Chawk, Saharsa-852201 \| \| 33 \| Dr. Reshma Sultana, Assistant Professor \| Department Of Philosophy, Vaishali Mahila College, Hajipur-844101 \| \| 34 \| Dr. Rajesh Kumar, Assistant Professor \| Department Of Chemistry, Ramdayalu Singh (R.D.S) College, Muzafferpur-842002 \| \| 35 \| Mr. Manikesh Kumar, Assistant Professor \| Department Of Psychology, Dr. R.M.L.S. College, Muzafferpur, Bihar \| \| 36 \| Dr. Veena, Assistant Professor \| Department Of Psychology, L.N. College, Bhagwanpur, Vaishali-844114 \| \| 37 \| Dr. Dharam Prakash Paswan, Assistant Professor \| Department Of Economics, L.N. College, Bhagwanpur, Vaishali-844114 \| \| 38 \| Dr. Shashi Bhushan Kumar, Assistant Professor \| Department Of Philosophy, L.N. College, Bhagwanpur, Vaishali-844114 \| \| 39 \| Dr. Neeraj Kumar, Assistant Professor \| Department Of Education (B.Ed), A.M. College, Katari Hill Road, Bodh Gaya \| \| 40 \| Dr. Bandana Kumari, Assistant Professor \| Department Of Philosophy, B.R.M. College, Munger \| \| 41 \| Dr. Javed Iqbal, Assistant Professor \| Department Of Zoology, Jamshedpur Workers College, Mango, Jamshedpur-831012 \| \| 42 \| Mrs. Anupam Kumari, Assistant Professor \| Department Of Education, Fatima Degree College, Gonputa, Phulwari Sharif, Patna 801505 \| \| No. \| Name and Title \| Department and Address \| \|-----\|----------------\|------------------------\| \| 43 \| Mr. Abhishek Anand, Assistant Professor \| Department Of Economics, Darshan Sah College, Katihar-854105 \| \| 44 \| Dr. Sovan Chakraborty, Assistant Professor \| Department Of English, Patna Science College, Patna 800005 \| \| 45 \| Dr. Jitesh Pati Tripathi, Assistant Professor \| Department Of Mathematics, R. N. College Hajipur, Vaishali-844101 \| \| 46 \| Mrs. Sanyukta, Assistant Professor \| Department Of Psychology, College Of Commerce, Arts & Science, Kankerbagh, Patna-800020 \| \| 47 \| Dr. Sheerin Masroor, Assistant Professor \| Department Of Chemistry, A.N. College, Boring Road, Patna 800013 \| \| 48 \| Dr. Amrita Chakraborty, Assistant Professor \| Department Of Chemistry, A.N. College, Boring Road, Patna 800013 \| \| 49 \| Mr. Abhishek Dutta Assistant Professor \| Department of Economics, A.N. College, Krishna Puri, Boring Road, Patna-800013 \| \| 50 \| Dr. Kaushal Kishore Assistant Professor \| Department Of Sociology & Social Anthropology, A.N. Sinha Institute Of Social Studies, Patna800001 \| \| 51 \| Mrs. Jagriti Anand Assistant Professor \| P.G Department Of Economics, Veer Kunwar Singh University, Ara 802301 \| \| 52 \| Dr. Vandana Maurya Assistant Professor \| Department Of Psychology College Of Commerce, Arts & Science, Kankerbagh Patna 800020 \| \| 53 \| Dr. Nisha Kumari Assistant Professor \| Department Of Chemistry, A.N. College, Boring Road, Patna800013 \| \| 54 \| Dr. Santwana Kumari, Assistant Professor \| Department Of Psychology S.M. College, Bhagalpur-812002 \| \| \| Name and Designation \| Department \| \| \|---\|----------------------\|------------\|---\| \| 55 \| Dr. Manoj Prabhakar Assistant Professor \| Department Of Economics, Patna University, Patna 800005 \| \| \| 56 \| Dr. Jyotish Kumar Assistant Professor \| Department of Physics, A.N. College, Boring Road, Patna 800013 \| \| \| 57 Draft Wanting \| Vijya Kumari, Assistant Professor \| Department of Economics, T.P. College, Madhepura-852113 \| \| \| 58 Draft Wanting \| Arun Kumar, Assistant Professor \| Department of Maithili, T.P. College, Madhepura-852113 \| \| Document Wanting Joining / Appointment letter (or other documents) are not attached with the Application Form and must be produced before joining/attending the Orientation Programme. Draft Wanting Those who have not sent Bank Draft Directly to Director, HRDC, may bring the same at the time of joining the course. Important: As the selected probable participants have been consented to join the course on telephone, they are directed to come with their relieving letter and three recent photographs from their respective institutions to join the course on 17th July, 2018 at 9:30 am positively. Memo No : UGC-HRDC/PAT/40 /____ 1. Copy forwarded to all candidates mentioned above and Principals of the concerned institutions / HODs concerned / Registrars of the Universities concerned / Director of Higher Education concerned for information and necessary action. 2. Registrar, Patna University / Secretary to the Vice Chancellor, Patna University for information. 3. Joint Secretary, UGC HRDC, Bureau, University Grants Commission New Delhi for information. The Principals / HODs concerned are requested to relieve the selected participants immediately to enable them to join the aforesaid Course. Professor Chandrama Singh $ Eligibility, Target Group and Duration Teachers working in universities and colleges that are included under Section 2(f) of the UGC Act, even though they may not yet be fit to be included under Section 12 (B), may be invited to participate in the Orientation Programmes and Refresher Programmes. The teachers of colleges that do not yet come within the purview of Section 12(B), but have been affiliated to a university for at least two years, will be permitted to participate in the programmes/courses. However, they won't be paid Dearness Allowances for attending the course. For the Orientation Programmes, newly appointed Assistant Professors within two years of continuous service and all those teachers who require orientation for getting a higher grade will be allowed. For the Refresher Courses, participation in the Orientation Programme is a prerequisite for admission. However, in exceptional cases a teacher may be allowed to complete a Refresher Course prior to completion of an Orientation Programme. Also, there should be a minimum gap of one year between two courses, though it may be relaxed if adequate number of participants is not available or it is essential for the teacher to fulfill eligibility conditions for career advancement. The Orientation Programme will be of four weeks duration, with minimum of 24 working days (excluding Sundays) and 144 contact hours (six hours a day). The refresher course will be of three weeks duration, with minimum of 18 working days (excluding Sundays) and 108 contact hours (six hours a day Part time/ Ad hoc/temporary/contract teachers who have been teaching for at least three academic sessions in an institution which has been affiliated to a University for at least two years may be permitted to participate in the Orientation Programme/ Refresher Course to enhance their skills. Teacher participants may be paid up to a limit of A/ C III railway fare (all trains) or A/C deluxe bus fare on production of tickets. For admission to Refresher Course, participation in Orientation Programme is a prerequisite. Thereafter, with a gap of one year they can opt for Refresher Course. Also there should be a minimum gap of one year between two Courses. If the candidate has been working as reader/or not undergone Orientation Programme/ or has attended Refresher Courses during the last one year to the present, he/she should not report for this programme. The Selected candidates are requested to report for join positively at Patna College, Patna University, Patna on the date and time as mentioned in the selection letter for registration along-with (1) A letter from the Principal/HOD relieving him/her on deputation/duty leave to attend the said programme on the format mentioned below and (2) three recent passport sized photographs. No candidate will be allowed to attend the programme if he/she does not attend the inaugural lecture of the said programme. As this is a contact based programme, no participants will be allowed to leave during the operation of the programme. No certificate is to be issued to any participant unless adequate numbers of hours are completed by the concerned participant. CERTIFICATE FOR CLAIMING TA To The Director UGC-Human Resource Development Centre Patna University, Patna Subject: Certificate for Claiming TA Dear Sir, This is to certify that our college/institution/organization (Name of the Institution) ............... is affiliated to ........... University since .............. and recognized under Section 2(f) & included under Section 12(B) of UGC Act 1956, vide letter No ............ dated ............ received from University Grants Commission, New Delhi. The above information provided is true to our knowledge, and if found to be otherwise, the concerned teacher Mr. / Mrs. / Ms. / Dr. ........................................ who is attending the Course /Programmes organized by UGC-HRDC, Patna University will not be entitled for his/her TA as per the new UGC Guidelines for HRDCs. Yours sincerely, Date: Place: Principal/Head of Institution (Signature with seal) Note: As per XII Plan of UGC, Teachers working in universities and colleges that are included under Section 2(f) of the UGC Act, even though they may not yet be fit to participate in the Orientation Programmes and Refresher Programmes. The teachers come within the purview of Section 12(B), but have at least two years, will be permitted to participate in the programmes/courses. However, they won’t be paid any allowances for attending these courses. The Vice Chancellors/ Registrars All Central Universities/ State Universities receiving grant-in-aid Deemed to be Universities receiving grant-in-aid from UGC/Government Inter University Centres of UGC Sub: Whether regular teachers participating in RCs/OPs be treated ‘On Duty’ – Clarification regarding. Sir, This is to clarify that the regular teachers from the colleges/universities participating in the Refresher Courses/Orientation Programs etc. organized by UGC-HRDCs may be treated ‘ON DUTY’. Further, it is specifically conveyed that the period for which these regular teachers are participating in RCs/OPs etc. at UGC-HRDC’s will be treated as ‘ON DUTY’ and not as on duty leave. With regards, Yours faithfully (Dev Swarup) To, The Director’s (All 66 HRDCs) Subject: Dearness Allowance to the Participants - regarding. Sir/Madam, I am directed to refer on the subject cited above and to convey the decision of the Commission for information and necessary action in this regard at your end. \| Dearness Allowance \| No \| \|--------------------\|----\| \| Travelling Allowance \| To continue \|
Trading Private Range Counting over Big IoT Data Zhipeng Cai Department of Computer Science Georgia State University Atlanta, United States firstname.lastname@example.org Zaobo He Department of Computer Science and Software Engineering Miami University Oxford, United States email@example.com Abstract—Data privacy arises as one of the most important concerns, facing the pervasive commoditization of big data statistic analysis in Internet of Things (IoT). Current solutions are incapable to thoroughly solve the privacy issues on data pricing and guarantee the utility of statistic outputs. Therefore, this paper studies the problem of trading private statistical results for IoT data, by considering three factors. Specifically, a novel framework for trading range counting results is proposed. The framework applies a sampling-based method to generate approximated counting results, which are further perturbed for privacy concerns and then released. The results are theoretically proved to achieve unbiased, bounded variances, and strengthened privacy guarantees under different privacy levels. Moreover, a pricing approach is proposed for the traded results, which is proved to be immune against arbitrage attacks. The framework is evaluated by estimating the air pollution levels with different ranges on 2014 CityPulse Smart City datasets. The analysis and evaluation results demonstrate that our framework greatly reduces the error of range counting approximation; and the optimal perturbation approach enables that the private counting satisfies the specified approximation degree while providing strong privacy guarantee. Index Terms—Differential Privacy, Range counting, Pricing I. INTRODUCTION The Internet of Things (IoT) has been regarded as a new paradigm of big data platform. For example, smart city applications have been deployed to timely monitor, analyze and response upon volumes of physical data. As a fundamental data analyzing operation, range counting aggregation acts as a critical component for these applications. For instance, data analyzers compute range counting over massive particulate matter level, traffic volume or weather data to monitor pollution levels. These aggregates are not only valuable to data owners, but also attractive to other communities with business purposes. However, data in IoT are collected in a distributed manner and strongly correlated with users’ sensitive status, aggravating the cost and privacy concerns for data analyzing operations. Therefore, this paper proposes a novel framework for range counting aggregation, which jointly considers utility, cost, privacy preservation, and charging for derived answers. Actually, many information platforms have emerged to facilitate such data circulation from raw data to data consumers (i.e., service requesters). In particular, these platforms usually prefer to trade statistics of raw data, like the range counting aggregation results, to data requesters. As shown in an FTC’s survey on several data brokers [1], Axiom, as an essential broker, collects personal information from more than 700 million users, and sells aggregation statistical information to big companies such as Oracle, Microsoft, AT&T, etc. Although attractive, conducting and trading range counting aggregation in IoT brings two major challenges, i.e., the concerns on resource consumption and privacy disclosure, and the design of pricing mechanisms. The first challenge lies in the performance concern and the privacy concern. If all IoT data are collected to compute the exact range counting aggregation, considerable communication and computation overhead will be incurred [2] [3]. However, in many cases, approximate range counting results with less overhead are actually sufficient enough for data customers to perform data analysis [4], [5], [6]. Meanwhile, data privacy is another serious concern, which obstacles the wide deployment and adoption of smart devices. The underlying reason is that smart devices collect, understand and interact with a user in a pervasive and intimate way [7]. Thus, aggregation results released to requesters should avoid considerable leakage of sensitive information. Although individual efforts have been made for each concern respectively, it remains unsettled to compute privacy-preserved approximate range counting aggregation efficiently. The second challenge is to establish effective pricing mechanisms for trading approximate range counting aggregation results. This challenge arises from potential arbitrage opportunities in trading procedures. Generally, data consumers are usually allowed to specify their own expected approximation degrees, upon which data brokers compute an approximate aggregation result and perturb it for privacy preservation. In this case, a smaller approximation degree intuitively leads to a higher price. However, with a poorly designed pricing mechanism, malicious consumers may circumvent to pay the desired price of a query. These consumers turn to buy multiple cheaper results with high variance, and reduce the variance by averaging the returned results. Then this sophisticated trading practice is an arbitrage attack when the total price of high-variance aggregation is less than a single one with low-variance. As far as we know, the arbitrage attack has not been investigated for pricing mechanisms in IoT. To mitigate the gaps, we propose a sampling-based algorithm for privacy-preserved approximate range counting aggregation. The algorithm presents an unbiased estimator for range... counting with bounded variance. To derive privacy-preserved range counting, the state-of-the-art approach Differential Privacy [8] is adopted, which allows unlimited reasoning power and background knowledge of adversaries. To meet such a requirement, our algorithm introduces an additional noise to the original approximate result. However, putting differential privacy into approximate range counting remains a challenging problem, since the two-stage compositive approximation should still meet the requirement of a customer. To address the above issues, we formulate an optimization problem that takes approximation degree as input, and outputs an optimal noise-adding mechanism, such that the derived range counting result satisfies the specified approximation degree and privacy preservation can be optimized. Specifically, our formulation traverses all valid intermediate approximation aggregations, and calculates the minimized differential privacy budget for them, while the final result still guarantees the input approximation degree. We theoretically prove that the derived range counting result meets the specified approximation degree while providing the strongest privacy preservation. Finally, to develop a pricing mechanism to avoid arbitrage, a sufficient and necessary condition for all arbitrage-avoiding pricing functions is proposed, with the information regarding how fast arbitrage-avoiding pricing functions can decrease with the approximation degree. Our key contributions are summarized as follows. - We propose a privacy-preserved approximate range counting aggregation algorithm, in which an unbiased estimator with bounded variance for range counting aggregation is presented. - An optimization problem is formulated to achieve the strongest differential privacy, while satisfying the approximation degree specified by data customers. A solution is provided accordingly. - An arbitrage-avoiding pricing mechanism is proposed to eliminate arbitrage attacks. A set of pricing functions can be constructed based on our identified critical condition to guarantee justice of trading. - We extensively evaluate the performance of our approach based on real-world dataset, i.e., the CityPulse Smart City Datasets. The rest of this paper is organized as follows. Section II introduces our system and adversary models, together with some preliminary knowledge on differential privacy and our problem definition. Section III presents the sampling based range counting aggregation algorithm and the optimization mechanism for providing the strongest differential privacy. The pricing mechanism is introduced in Section IV. Section V illustrates our evaluation results. Section VI discusses the related works, and Section VII concludes the paper. II. Problem Formulation This section first presents our system model and adversary model. Then necessary preliminary knowledge on differential privacy and the concept of Arbitrage Avoiding is introduced. At the end of this section, we present the problem definition of differentially private $(\alpha, \delta)$-range counting. A. System Model As shown in Fig. 1, there are three major entities in our system model including IoT networks, data brokers and data consumers. IoT networks consist of large scales of smart devices, which collect data generated by sensing modules or other input channels. Denote $D$ as the global dataset collected by all smart devices in IoT networks. Instead of transferring the entire $D$ to the base station, each smart device only sends a sample of its locally collected data to the base station. This will significantly reduce the communication cost of data transmission. Then a sample $S$ of $D$ is stored in the base station, which opens the data access API to data brokers. In this paper, we consider range counting queries on datasets collected by smart devices, and the definition of range counting is as below. Definition 2.1: Range Counting. Given range parameters $l$ and $u$ ($l \leq u$) together with dataset $D$, the range counting of $D$ with lower bound $l$ and upper bound $u$ is $\gamma(l, u, D) = \|\{x \| l \leq x \leq u, x \in D\}\|$. Computing exact range counting from scratch is expensive in terms of real-time communication in IoT networks. In many scenarios, an approximate range counting with acceptable accuracy suffices to meet customers’ requirements. Definition 2.2 presents the notion of $(\alpha, \delta)$-range counting, which parameterizes range counting with accuracy parameters specified by customers. Definition 2.2: $(\alpha, \delta)$-Range Counting. Given $0 \leq \alpha \leq 1$ and $0 \leq \delta \leq 1$, for any range parameters $l$ and $u$ such that $l \leq u$, the $(\alpha, \delta)$-range counting of dataset $D$, denoted as $\hat{\gamma}(l, u, D)$, satisfies that $\Pr[\|\hat{\gamma}(l, u, D) - \gamma(l, u, D)\| \leq \alpha\|D\|] \geq \delta$. Data customers send $(\alpha, \delta)$-range counting requests denoted by $\Lambda(\alpha, \delta)$ to a data broker. The data broker may access $S$ to response these requests. However, the sensitive information may still be inferred by adversaries with background knowledge, even if approximate aggregates, rather than the raw dataset, are released to data customers. Thus, the IoT network entrusts the protection of data privacy to the data broker. The data broker first accesses $S$ to compute a $(\alpha', \delta')$-range counting, where $\alpha' \leq \alpha$ and $\delta' \geq \delta$. Then the data broker employs the standard notion of differential privacy, and adds carefully controlled noise to the $(\alpha', \delta')$-range counting. Subsequently, the $(\alpha', \delta')$-range counting and the noise jointly composite an $(\alpha, \delta)$-range counting. Finally, the data broker responses a customer with the composited $(\alpha, \delta)$-range counting, and charges the customer with price $\pi(\alpha, \delta)$. B. Adversary Model The adversaries in this paper are closestfied or malicious customers, who look for arbitrage opportunities against the trading pricing designed by a data broker. For example, an adversary is interested in an aggregate result with low variance. However, instead of making full payment, the adversary turns to purchase several aggregates with high variance at a cheaper price. Then the adversary can reduce the variance by averaging the returned aggregates. A benefit-concerned data broker would like to rule out such arbitrage behaviors. Thus, the pricing mechanism should provide the property of arbitrage avoiding, which is defined as follows [9]: Definition 2.3: Arbitrage Avoiding. A pricing function $\pi(\alpha, \delta)$ is arbitrage avoiding if $\forall m \geq 1$, $\{\Lambda(\alpha_1, \delta_1), \ldots, \Lambda(\alpha_m, \delta_m)\} \mapsto \Lambda(\alpha, \delta)$ implies: $$\pi(\alpha, \delta) \leq \sum_{j=1}^{m} \pi(\alpha_j, \delta_j),$$ (1) where $\{\alpha_i, \delta_j \| i, j \in \{1, \ldots, m\}, \alpha_i > \alpha, \delta_j < \delta\}; \mapsto$ is an operation that composites these $m$ range counting results to a result with $(\alpha, \delta)$-approximation. Definition 2.3 implies that for a desirable $\pi(\alpha, \delta)$, $\pi(\alpha, \delta) \leq \sum_{j=1}^{m} \pi(\alpha_j, \delta_j)$. Then adversaries cannot obtain aggregation service $\Lambda(\alpha, \delta)$ with a lower price through buying multiple aggregates with diverse higher variances and averaging aggregates at a cheaper price than $\Lambda(\alpha, \delta)$. ### C. Differential Privacy Differential privacy is a well accepted standard notion for protecting sensitive information in statistical aggregates. The formal definition of differential privacy is given as follows: Definition 2.4: $\epsilon$-Differential Privacy. A randomized algorithm $G$ satisfies $\epsilon$-Differential Privacy ($\epsilon$-DP) if and only if for any two neighboring datasets $D$ and $D'$ that differ in only one item and for any possible output $O$ of $G$, the following condition holds: $$\Pr[G(\gamma(D)) = O] \leq e^\epsilon \cdot \Pr[G(\gamma(D')) = O].$$ The Laplace mechanism [10], introduced as below, is a standard approach to achieve differential privacy. Laplace Mechanism. For a function $\gamma : \mathbb{D} \rightarrow \mathbb{R}^d$, Laplace mechanism derives the following result: $$G(D) = \gamma(D) + \text{Lap} \left( \frac{\Delta \gamma}{\epsilon} \right)^d,$$ where $$\Delta \gamma = \max_{D \simeq D'} \\|\gamma(D) - \gamma(D')\\|_1,$$ and $$\Pr[\text{Lap}(\eta) = x] = \frac{1}{2\eta} e^{-\|x\|/\eta}$$ to achieve $\epsilon$-differential privacy. Given the privacy budget $\epsilon$, the amount of noise is also denoted as $\text{Lap}(\epsilon)$ for abbreviation. ### D. Problem Definition The problem of computing differentially private $(\alpha, \delta)$-range counting is defined as follows: Input: 1) Data set $D$; 2) Range parameters $l$ and $u$ ($l \leq u$), accuracy parameters $\alpha$ and $\delta$ ($0 \leq \alpha \leq 1$ and $0 \leq \delta \leq 1$). Output: 1) Differentially private $(\alpha, \delta)$-range counting with the minimum privacy budget. In addition to the above problem definition, this paper studies the design of arbitrage avoiding pricing mechanisms for trading $(\alpha, \delta)$-range counting aggregation. ### III. $(\alpha, \delta)$-DIFFERENTIALLY PRIVATE RANGE COUNTING In this section, we propose a sampling based algorithm to provide $(\alpha, \delta)$-differentially private range counting. Section III-A presents our RankCounting Estimator to answer $(\alpha, \delta)$-range counting based on samples. Then we define an optimization problem and present a solution in section III-B to achieve the optimal differential privacy under the constraint that $(\alpha, \delta)$-range counting is guaranteed. #### A. Sampling Based $(\alpha, \delta)$-Range Counting In this section, we handle the first part of our work, namely the sampling-based $(\alpha, \delta)$-range counting. High accuracy and low communication cost are essential to the performance of the entire system. To achieve this, we propose an estimator namely RankCounting for answering $(\alpha, \delta)$-range counting aggregations. We assume the network is organized in a flat model, in which each node communicates with the base station directly. Note that algorithms on flat models can be easily extended to a general tree model. After samples are collected from underlying nodes, they will be used to answer future range counting aggregations if the required accuracy can be satisfied. Otherwise, the base station will inform the underlying nodes to collect more samples from the network. A straightforward estimation (denoted as BasicCounting) to the range counting is $\hat{\gamma}_B(l, u, S) = \frac{\|\{x \| x \in S, l \leq x \leq u\}\|}{p}$. This estimator is unbiased and its variance is $\frac{\|\{x \| x \in D_k, l \leq x \leq u\}\|(1-p)}{p}$, which may grow to $\frac{\|D\|(1-p)}{p}$ when a large range is queried. This in turn increases the communication cost of sample transmission since more samples should be drawn to guarantee query accuracy. To reduce communication cost, we leverage the rank of sampled data elements to present the RankCounting estimator. Let $S_i$ be the set of sample drawn by node $i$, and $D_i$ denote the set of data collected by node $i$, $i = 1, \ldots, k$. Let $S = \bigcup_i^k S_i$, $D = \bigcup_i^k D_i$ be the global set of sample and data respectively. Let $n_i$ be the number of data collected at node $i$, and $n$ denote the total number of data collected at $k$ nodes. Let $fst$ and $lst$ denote the first and last data collected at node $i$, $fst \leq lst$, respectively. Given the lower and upper ranges $l, u$, our estimator first computes $\hat{\gamma}(l, u, i)$ using $S_i$, namely the range counting with parameter $(l, u)$ at node $i$, and then obtains the range counting at $S$, denoted as $\hat{\gamma}(l, u, S)$. The RankCounting Estimator. Each node $i$ first independently samples each of its data with a certain probability $p$ (to be determined later). For each sampled data $x$, it computes $r(x, i)$, the local rank of $x$ at node $i$, i.e., the rank of $x$ in $D_i$. Finally, node $i$ sends all the sampled data and corresponding ranks to the base station. Similarly, if the existing samples are unable to satisfy the query accuracy requirement, more samples should be drawn and their ranks are also transferred to the base station. Given query parameters $l$ and $u$, we denote $r(l, i)$ as the smallest rank of any element in $D_i$ whose value is no smaller than $l$, and $r(u, i)$ is defined as the largest rank of any element no larger than $u$. Note that $l$ and $u$ may not exist, in most cases, in the sampled data $S_i$, so neither the broker nor the RankCounting estimator is able to obtain the ranks $r(l)$ and $r(u)$. Nevertheless, the concept of $r(l, i)$ and $r(u, i)$ are only employed in the accuracy analysis of the proposed estimator, and they are not involved in estimation calculation at all. Let $p(x, i)$ denote the predecessor of $x$ in the sampled data $S_i$, i.e., the largest sampled data no larger than $x$. Likewise, let $s(x, i)$ denote the successor of $x$ in the sampled data from node $i$, i.e., the smallest value larger than $x$. It is worth noting that $p(x, i)$ and $s(x, i)$ may not exist. To distinguish different cases with regards to the existence or non-existence of $p(l, i)$ and $s(u, i)$, we make the following denotations, together with their probabilities below: - $\omega_p$: $p(l, i)$ exists, $\Pr[\omega_p] = 1 - (1 - p)^{r(l)}$; - $\overline{\omega}_p$: $p(l, i)$ does not exist, $\Pr[\overline{\omega}_p] = (1 - p)^{r(l)}$; - $\omega_s$: $s(u, i)$ exists, $\Pr[\omega_s] = 1 - (1 - p)^{n_i - r(u)}$; - $\overline{\omega}_s$: $s(u, i)$ does not exist, $\Pr[\overline{\omega}_s] = (1 - p)^{n_i - r(u)}$. Based on the above notations, we carry out our estimator RankCounting, and RankCounting estimates $\hat{\gamma}(l, u, i)$ as below. $$\hat{\gamma}(l, u, i) = \begin{cases} \gamma(p(l, i), s(u, i), i) - \frac{2}{p}, & \text{if } \omega_p, \omega_s; \\ \gamma(p(l, i), lst, i) - \frac{1}{p}, & \text{if } \omega_p, \overline{\omega}_s; \\ \gamma(fst, s(u, i), i) - \frac{1}{p}, & \text{if } \overline{\omega}_p, \omega_s; \\ \gamma(fst, lst, i) & \text{else}. \end{cases}$$ In the calculation of $\hat{\gamma}(l, u, i)$, ranks of certain samples are employed to improve estimation accuracy. Given query parameters $l, u$ and the collected samples in $S_i$, RankCounting determines which of the four cases should be adopted. Then the corresponding calculation is carried out, and in this process terms $\gamma(p(l, i), s(u, i), i)$, $\gamma(p(l, i), lst, i)$, $\gamma(fst, s(u, i), i)$, and $\gamma(fst, lst, i)$ can be exactly calculated with the ranks of $p(l, i)$, $s(u, i)$, $fst$ and $lst$. Here, $p(l, i)$ and $s(u, i)$ are in $S_i$, so RankCounting can obtain their ranks. The ranks of $fst$ and $lst$ are simply 1 and $n_i$. Based on $\hat{\gamma}(l, u, i)$, RankCounting estimates $\hat{\gamma}(l, u, D)$ as $$\hat{\gamma}(l, u, S) = \sum_{i=1}^{k} \hat{\gamma}(l, u, i). \quad (2)$$ Next, we illustrate the high accuracy of RankCounting with Theorem 3.1, which shows that RankCounting can use $\hat{\gamma}(l, u, i)$ to accurately estimate $\gamma(l, u, i)$. Theorem 3.1: For any $l$ and $u$, $\hat{\gamma}(l, u, i)$ is an unbiased estimation of $\gamma(l, u, i)$ with variance $\text{Var}[\hat{\gamma}(l, u, i)] \leq \frac{k}{p^2}$. Proof. Denote $\Psi = \hat{\gamma}(l, u, i) - \gamma(l, u, i)$, then $\Psi$ could be formulated under different cases as follows, $$\Psi = \begin{cases} \gamma(p(l, i), s(u, i), i) - \gamma(l, u, i) - \frac{2}{p}, & \text{if } \omega_p, \omega_s; \\ \gamma(p(l, i), lst, i) - \gamma(l, u, i) - \frac{1}{p}, & \text{if } \omega_p, \overline{\omega}_s; \\ \gamma(fst, s(u, i), i) - \gamma(l, u, i) - \frac{1}{p}, & \text{if } \overline{\omega}_p, \omega_s; \\ \gamma(fst, lst, i) - \gamma(l, u, i), & \text{else}. \end{cases}$$ The proof first shows that $E(\Psi) = 0$ and then illustrates the bounded variance of $\hat{\gamma}(l, u, i)$. We observe that $\gamma(p(l, i), s(u, i), i)$, $\gamma(p(l, i), lst, i)$, $\gamma(fst, s(u, i), i)$ and $\gamma(fst, lst, i)$ are all no smaller than $\gamma(l, u, i)$, since they contain additional data elements compared to the query range $(l, u)$. The number of additional data elements can be studied in four cases according to the existence or non-existence of $p(l, i)$ and $s(u, i)$. $\gamma(p(l, i), s(u, i), i) - \gamma(l, u, i)$ represents the number of additional data elements located in range intervals $(p(l, i), l)$ and $(u, s(u, i))$, when both $p(l, i)$ and $s(u, i)$ exist. $\gamma(p(l, i), lst, i) - \gamma(l, u, i)$ represents the number of additional data elements located in $(p(l, i), l)$ and $(u, lst)$, when only $p(l, i)$ exists. Likewise, $\gamma(fst, s(u, i), i) - \gamma(l, u, i)$ represents the number of additional data elements located in $(fst, l)$ and $(u, s(u, i))$, when only $s(u, i)$ exists. $\gamma(l, u, i) - \gamma(fst, lst, i)$ represents the number of additional data elements located in $(fst, l)$ and $(u, lst)$, when neither $p(l, i)$ or $s(u, i)$ exists. We introduce term $C[i]^{\pm}_{\pm} = \|\{x \| x \in D_k, s \leq x \leq e\}\|$ to denote the number of data elements from $D_k$ located in the range interval $(s, e)$. For example, $C[l]^{\pm}_{\pm}$ and $C[u]^{\pm}_{\pm}$ represent the number of data elements located in $(p(l, i), l)$ and $(u, g(u, i))$, when $p(l, i)$ and $g(u, i)$ exist, respectively. For a given $j$ satisfying $1 \leq j \leq r(l)$, the probability of $C[i]_{p(i)}^l = j$ is $p(1 - p)^{j-1}$, when $C[i]_{p(i)}^l$ exists. Similarly, for a given $j$ satisfying $1 \leq j \leq n_i - r(u)$, the probability of $C[i]_{u}^{g(u)} = j$ is $p(1 - p)^{j-1}$, when $C[i]_{u}^{g(u)}$ exists. According to the definition of $C$, we have $C[i]_{fst}^l = r(l, i)$ and $C[i]_{u}^{lst} = n_i - r(u, i)$. Note that for given query parameters $l$ and $u$, $C[i]_{p(l,i)}^l$, $C[i]_{u}^{g(u,i)}$, $C[i]_{fst}^l$ and $C[i]_{u}^{lst}$ can be regarded as random variables. $C[i]_{fst}^l$ is independent of $C[i]_{p(l,i)}^l$ and $C[i]_{u}^{lst}$, while $C[i]_{u}^{g(u,i)}$ is independent of $C[i]_{p(l,i)}^l$ and $C[i]_{fst}^l$. The notation of $\Psi$ could be formulated using $C$ as follows: \[ \Psi = \begin{cases} C[i]_{f(l,i)}^l + C[i]_{p(l,i)}^l - \frac{2}{p}, & \text{if } \omega_p, \omega_s; \\ C[i]_{f(l,i)}^l + C[i]_{u}^{lst} - \frac{1}{p}, & \text{if } \omega_p, \overline{\omega}_s; \\ C[i]_{fst}^l + C[i]_{u}^{g(u,i)} - \frac{1}{p}; & \text{if } \overline{\omega}_p, \omega_s; \\ C[i]_{fst}^l + C[i]_{u}^{lst}. & \text{else} \end{cases} \] In the following analysis, we omit the identifier $i$ of node in $r(l, i), C[i]_{p(l)}^l, p(l, i)$ and $g(u, i)$, and use $r(\cdot), C(\cdot), p(\cdot)$ and $g(\cdot)$ respectively for abbreviation when focusing on node $i$. The expectation of $\Psi$ can be computed as shown below: \[ E(\Psi) = \sum_{1 \leq m \leq l \leq r(l)} (m + n - \frac{2}{p}) \Pr[C_{p(l)}^l = m, C_u^{(u)} = n] + \sum_{1 \leq m \leq r(l)} (m + n_i - r(u) - \frac{1}{p}) \Pr[C_{p(l)}^l = m, \overline{\omega}_s] + \sum_{1 \leq m \leq n_i - r(u)} (r(l) + n - \frac{1}{p}) \Pr[C_u^{(u)} = m, \overline{\omega}_p] + (r(l) + n_i - r(u)) \Pr[\overline{\omega}_p, \overline{\omega}_s] = \Pr[\omega_s] E[C_{p(l)}^l - \frac{1}{p}] + \Pr[\omega_p] E[C_u^{(u)} - \frac{1}{p}] + \Pr[\omega_s] E[C_{p(l)}^l - \frac{1}{p}] + \Pr[\omega_p] (n_i - r(u)) \Pr[\overline{\omega}_s] + \Pr[\omega_s] r(l) \Pr[\overline{\omega}_p] + \Pr[\overline{\omega}_p] E[C_u^{(u)} - \frac{1}{p}] + (r(l) + n_i - r(u)) \Pr[\overline{\omega}_s] \Pr[\overline{\omega}_s] = E[C_{p(l)}^l] - \frac{1}{p} \Pr[\omega_p] + E[C_u^{(u)}] - \frac{1}{p} \Pr[\omega_s] + r(l) \Pr[\overline{\omega}_p] + (n_i - r(u)) \Pr[\overline{\omega}_s] = \sum_{j=1}^{r(l)} jp(1-p)^{j-1} - \frac{1}{p}(1-(1-p)^{r(l)}) + \sum_{j=1}^{n_i-r(u)} jp(1-p)^{j-1} - \frac{1}{p}(1-(1-p)^{n_i-r(u)}) + (1-p)^{r(l)}r(l) + (1-p)^{n_i-r(u)}(n_i-r(u)) = 0 \] So we have $E[\hat{\gamma}(l, u, i)] = \gamma(l, u, i)$, and the RankCounting estimator produces an unbiased estimation to $\gamma(l, u, i)$. Given query parameters $l$ and $u$, the exact query result $\gamma(l, u, i)$ is a constant, and it indicates that $\text{Var}[\hat{\gamma}(l, u, i)] = \text{Var}[\Psi]$ since $\Psi = \hat{\gamma}(l, u, i) - \gamma(l, u, i)$. Next, we investigate the variance of $\hat{\gamma}(l, u, i)$ by calculating the variance of $\Psi$. \[ \text{Var}[\hat{\gamma}(l, u, i)] = \text{Var}[\Psi] = E[\Psi^2] - E[\Psi]^2 = E[\Psi^2] = \sum_{1 \leq m \leq l \leq r(l)} (m + n - \frac{2}{p})^2 \Pr[C_{p(l)}^l = m, C_u^{(u)} = n] + \sum_{1 \leq m \leq r(l)} (m + n_i - r(u) - \frac{1}{p})^2 \Pr[C_{p(l)}^l = m, \overline{\omega}_s] + \sum_{1 \leq m \leq n_i - r(u)} (r(l) + n - \frac{1}{p})^2 \Pr[C_u^{(u)} = m, \overline{\omega}_p] + (r(l) + n_i - r(u))^2 \Pr[\overline{\omega}_p, \overline{\omega}_s] < \sum_{1 \leq m \leq l \leq r(l)} (2m^2 + 2n^2 + \frac{4}{p^2}) \Pr[C_{p(l)}^l = m, C_u^{(u)} = n] - \sum_{1 \leq m \leq l \leq r(l)} \frac{2(m+n)}{p} \Pr[C_{p(l)}^l = m, C_u^{(u)} = n] + \sum_{1 \leq m \leq r(l)} (2m^2 + 2(n_i - r(u))^2 + \frac{1}{p^2}) \Pr[C_{p(l)}^l = m, \overline{\omega}_s] - \sum_{1 \leq m \leq r(l)} \frac{2(m+n_i-r(u))}{p} \Pr[C_{p(l)}^l = m, \overline{\omega}_s] + \sum_{1 \leq m \leq n_i - r(u)} (2r^2(l) + 2n^2 + \frac{1}{p^2}) \Pr[C_u^{(u)} = m, \overline{\omega}_p] - \sum_{1 \leq m \leq n_i - r(u)} \frac{2(r(l)+n)}{p} \Pr[C_u^{(u)} = m, \overline{\omega}_p] + (2r^2(l) + (n_i - r(u))^2) \Pr[\overline{\omega}_p, \overline{\omega}_s] < 2E[(C_{p(l)}^l)^2] + 2E[(C_u^{(u)})^2] + 2Pr[\overline{\omega}_p] r^2(l) + 2Pr[\overline{\omega}_s] (n_i - r(u))^2 + \frac{4}{p^2} - \frac{2}{p} E[C_{p(l)}^l] + E[C_u^{(u)}]) = \frac{4}{p^2} + 2 \sum_{j=1}^{r(l)} j^2 p(1-p)^{j-1} + 2 \sum_{j=1}^{n_i-r(u)} j^2 p(1-p)^{j-1} + (1-p)^{r(l)}r(l)^2 + (1-p)^{n_i-r(u)}(n_i-r(u))^2 - 2 \left( \sum_{j=1}^{r(l)} jp(1-p)^{j-1} + \sum_{j=1}^{n_i-r(u)} jp(1-p)^{j-1} \right) \leq \frac{8}{p^2} \] Compared with BasicCounting whose variance is bounded by $\frac{(D)(1-p)}{p}$, our estimator does provide advantage to improve the communication cost. The total number of samples drawn in the system is expected to be $\|S\| = \|D\|p$. In general cases, only a small fraction of the raw data will be sampled, so we have $1 - p > 0.5$. If $\|S\| = \|D\|p > 16k$ and $\frac{8}{p^2} < \frac{\|D\|(1-p)}{p}$, it indicates that the proposed estimator provides smaller variance and incurs smaller communication cost. In contrast, if $\|S\| = \|D\|p \leq 8k$, it means that the average number of samples transferred by each node is no larger than 16. In this case, a node could pack the samples into an ordinary heartbeat message to the broker, and no more communication cost is incurred either. Since the global range counting aggregation is the sum of local range counting aggregations, which is produced independently, we conclude that $\hat{\gamma}(l, u, S)$ is an unbiased estimator of $\gamma(l, u, D)$ with bounded variance in Theorem 3.2. Theorem 3.2: For any $l$ and $u$, $\hat{\gamma}(l, u, S) = \sum_{i=1}^{N} \hat{\gamma}(l, u, i)$ is an unbiased estimation of $\gamma(l, u, D)$ with variance $\text{Var}[\hat{\gamma}(l, u, i)] \leq \frac{8k}{p^2}$, where $k$ is the number of nodes. Therefore, by setting $p = \sqrt{8k/\alpha n}$, the variance will be $(\alpha n)^2$. According to Chebyshev’s inequality, this means that $\hat{\gamma}(l, u, S)$ approximates $\gamma(l, u, D)$ within an additive error of $\alpha n$ with constant probability. We can make this constant probability arbitrarily close to 1 by enlarging $p$ by appropriate constant factors. The total communication overhead in this case is $\sqrt{8k}/\alpha$, since this is the expected number of samples to be transferred. Furthermore, according to Chebyshev’s inequality, we can extend the above approximation, which is within additive error of $\alpha n$ with constant probability, to the case with a certain probability guarantee (say $\delta$): Theorem 3.3: Given query parameters $l$ and $u$, the number of nodes $k$, for any $0 < \alpha < 1$ and $0 < \delta < 1$, if the sampling probability $p$ in the RankCounting estimator satisfies that $p \geq \frac{\sqrt{2k}}{\alpha n} \frac{2}{\sqrt{1-\delta}}$, then $\hat{\gamma}(l, u, S)$ is an $(\alpha, \delta)$-range counting. Proof. The RankCounting estimator provides an unbiased estimation $\hat{\gamma}(l, u, S)$ and its variance is no larger than $\frac{8k}{p^2}$. Combined with the Chebyshev’s inequality, we have $$\Pr[\|\hat{\gamma}(l, u, S) - \gamma(l, u, D)\| \leq \alpha n]$$ $$= \Pr[\|\hat{\gamma}(l, u, S) - \mathbb{E}[\hat{\gamma}(l, u, D)]\| \leq \alpha n]$$ $$\geq 1 - \frac{\text{Var}[\hat{\gamma}(l, u, S)]}{(\alpha n)^2} = 1 - \frac{\frac{8k}{p^2}}{(\alpha n)^2} \geq 1 - \frac{(\frac{\sqrt{2k}}{\alpha n} \frac{2}{\sqrt{1-\delta}})^2}{(\alpha n)^2} = \delta.$$ To this end, we can see that $\hat{\gamma}(l, u, S)$ satisfies the requirement of $(\alpha, \delta)$-Range counting. □ ### B. Differentially Private Approximate Range Counting To keep sensitive data in $D$ private, we adopt a two-phase approach to response to $(\alpha, \delta)$-range counting in a private manner, and we formulate an optimization problem aiming at achieving the optimal differential privacy, while taking accuracy requirements as constraints. Given query parameters $l$, $u$, $\alpha$ and $\delta$, a data broker firstly chooses a pair of $(\alpha', \delta')$, and computes $(\alpha', \delta')$-range counting $\hat{\gamma}(l, u, S)$. Then the Laplacian mechanism is employed with privacy budget $\epsilon$, and finally $\gamma^(l, u, S) = \hat{\gamma}(l, u, S) + \text{Lap}(\epsilon)$ is returned to the querying data customer. Note that $\gamma^(l, u, S)$ should satisfy the $(\alpha, \delta)$-range counting accuracy. We can get $\alpha' < \alpha$ and $\delta' > \delta$, since $\hat{\gamma}(l, u, S)$ must be more accurate than $\gamma^(l, u, S)$, otherwise, it will violate the accuracy requirement after adding Laplacian noise $\text{Lap}(\epsilon)$. For each pair of $(\alpha', \delta')$, the data broker should include as many samples as possible in the computation of $\hat{\gamma}(l, u, S)$. Then in the next step a larger search space of $\epsilon$ will be obtained under the final $(\alpha, \delta)$ accuracy requirement. For a given pair of $(\alpha', \delta')$, the data broker calculates the smallest $\epsilon$ making $\gamma^(l, u, S)$ a $(\alpha, \delta)$-range counting. After traversing all pairs of $(\alpha', \delta')$, the data broker is able to achieve the optimal differential privacy, e.g., the smallest $\epsilon$. A problem in the above process is how to calculate the accuracy of $\gamma^(l, u, S) = \hat{\gamma}(l, u, S) + \text{Lap}(\epsilon)$, given $\hat{\gamma}(l, u, S)$ is an $(\alpha', \delta')$-range counting and Laplacian noise is determined by $\epsilon$. This problem is equivalent to calculate the probability $\Pr[\|\hat{\gamma}(l, u, S) - \gamma(l, u, D)\| \leq \alpha n]$. It is known that $\Pr[\|\hat{\gamma}(l, u, S) - \gamma(l, u, D)\| \leq \alpha' n] \geq \delta'$. For a specified $\epsilon$, suppose $\Pr[\|\text{Lap}(\epsilon)\| \leq (\alpha - \alpha')n] \geq \tau$, then we have $$\Pr[\|\gamma^(l, u, S) - \gamma(l, u, D)\| \leq \alpha n]$$ $$\geq \Pr[\|\hat{\gamma}(l, u, S) - \gamma(l, u, D)\| + \|\text{Lap}(\epsilon)\| \leq \alpha n]$$ $$\geq \Pr[\|\hat{\gamma}(l, u, S) - \gamma(l, u, D)\| \leq \alpha' n]\Pr[\|\text{Lap}(\epsilon)\| \leq (\alpha - \alpha')n]$$ $$= \delta \tau.$$ From the above inequality, it is seen that if $\Pr[\|\text{Lap}(\epsilon)\| \leq (\alpha - \alpha')n] = \tau \leq \frac{\delta}{\delta'}$, then we will derive that $\Pr[\|\gamma^(l, u, S) - \gamma(l, u, D)\| \leq \alpha n] \geq \delta$ holds. It is worth mentioning that random variables $X = \hat{\gamma}(l, u, S) - \gamma(l, u, D)$ and $Y = \text{Lap}(\epsilon)$ are independent, and if $\tau < \frac{\delta}{\delta'}$, then we cannot guarantee that $\Pr[\|\gamma^(l, u, S) - \gamma(l, u, D)\| \leq \alpha n] \geq \delta$ holds. After injecting a Laplacian noise scaled at $\text{Lap}(\epsilon)$, not only the inaccuracy is composited by sampling and Laplacian noise, but also the final privacy degree. In other words, the differential privacy budget achieved by our two-phase approach denoted as $\epsilon'$ is determined, in a collaborative manner, by sampling probability $p$ and privacy budget $\epsilon$. Lemma 3.4 generalized from [11] shows the relationship of $\epsilon'$, $p$ and $\epsilon$. Lemma 3.4: If function $\phi(\cdot)$ is $\epsilon$-differentially private, and function $S(\cdot)$ returns independent random samples with probability $0 \leq p \leq 1$, then $\phi(S(\cdot))$ is $\epsilon'$-differential private, where $\epsilon' = \ln(1-p+p\cdot e^\epsilon)$. Proof. Let $D$ and $D'$ be any pair of neighboring datasets, assuming $D = D' \cup \{i\}$. Let $o$ be any output of $\phi(S(\cdot))$. $$\Pr[\phi(S(D)) = o]$$ $$= \sum_{Z \subseteq S(D')} p \Pr[S(D') = Z] \Pr[\phi(Z \cup \{i\}) = o]$$ $$+ \sum_{Z \subseteq S(D')} (1 - p) \Pr[S(D') = Z] \Pr[\phi(Z) = o]$$ $$\leq (pe^\epsilon + 1 - p) \sum_{Z \subseteq S(D')} \Pr[S(D') = Z] \Pr[\phi(Z) = o]$$ $$= (pe^\epsilon + 1 - p) \Pr[\phi(S(D')) = o].$$ Similarly, we can get that $$\Pr[\phi(S(D')) = o] \leq (pe^\epsilon + 1 - p) \Pr[\phi(S(D)) = o],$$ and $\phi(S(\cdot))$ is $\epsilon'$-differential private, with $\epsilon' = \ln(1-p+p\cdot e^\epsilon)$. □ Now we can formulate the optimization problem of achieving the optimal differential privacy (with the smallest privacy budget) under the $(\alpha, \delta)$-range counting accuracy requirement. In the following, we present the formulation: \[ \begin{align} \min & \quad e' = \ln(1 + p(e' - 1)) \\ \text{s.t.} & \quad \frac{\sqrt{2k}}{\alpha' n} \cdot \frac{2}{\sqrt{1 - \delta'}} \leq p \\ & \quad \alpha' \leq \alpha \\ & \quad \delta \leq \delta' \\ & \quad \Pr[\|\text{Lap}(e)\| \leq (\alpha - \alpha')n] \geq \frac{\delta}{\delta'} \\ & \quad \epsilon \geq \frac{\delta}{\delta'} \end{align} \] Given $\alpha, \delta$ and $p$ as inputs, the optimization problem in (3) can be constructed, taking the optimized $\epsilon$ (together with intermediate $\alpha', \delta'$) as output, which is used to conduct an $(\alpha, \delta)$-range counting. The search space of (3) contains all the feasible solutions for $(\alpha, \delta)$-range counting conducted by the two-phase approach. Next, we show how to compute the optimal solution to (3) and how to conduct a differentially private $(\alpha, \delta)$-range counting based on the optimal solution. For a fixed $\alpha'$, to avoid repeated sampling in continuous queries, a data broker uses all the existing samples (collected with $p$) at the base station to compute an $(\alpha', \delta')$-range counting. Then, $\delta'$ can be obtained by setting $\frac{\sqrt{2k}}{\alpha' n} \cdot \frac{2}{\sqrt{1 - \delta'}} = p$. According to (3), $\alpha, \delta, p, k$ and $n$ are all constants, and the data broker can obtain a minimum $\epsilon$ for the fixed $\alpha'$ with the following constraint: \[ \Pr[\|\text{Lap}(e)\| \leq (\alpha - \alpha')n] = 1 - e^{-\frac{(\alpha - \alpha')n\epsilon}{\Delta \hat{\gamma}}} \leq \frac{\delta}{\delta'}. \] Then we know $\epsilon \geq \frac{\Delta \hat{\gamma}}{(\alpha - \alpha')n} \ln \frac{\delta'}{\delta - \delta}$, and the optimal differential privacy is achieved by setting $\epsilon = \frac{\Delta \hat{\gamma}}{(\alpha - \alpha')n} \ln \frac{\delta'}{\delta - \delta}$. $\Delta \hat{\gamma}$ is the sensitivity of $\hat{\gamma}(l, u, i)$. In the worst case, $\Delta \hat{\gamma}$ grows to $n$; with an extremely small probability, and adopting $\Delta \hat{\gamma} = n$, will totally destroy the aggregation utility. A fair solution is to use the expectation of $\Delta \hat{\gamma}$, which is $\frac{1}{q}$ in the general cases. By traversing $\alpha'$ in $[0, \alpha]$, an optimal solution consisting of $\alpha', \delta'$ and $\epsilon$ can be found. Although the searching range of $\alpha'$ is continuous, we can approximate it to a discrete domain with arbitrarily small intervals. Given an optimal solution consisting of $\alpha', \delta'$ and $\epsilon$, the data broker carries out the following two steps. First, an $(\alpha', \delta')$-range counting is computed as $x' = \sum_{i=1}^{k} \hat{\gamma}(l, u, i)$. Second, Laplacian noise $\text{Lap}(\epsilon)$ is added on $x'$ and the final result is $x'' = \sum_{i=1}^{k} \hat{\gamma}(l, u, i) + \text{Lap}(\epsilon)$, which is an $\epsilon'$-differentially private $(\alpha, \delta)$-range counting. IV. PRICING MECHANISM This section discusses the existence of arbitrage attacks for pricing mechanisms, and how the pricing mechanism should be designed to thwart such attacks. Initiatively, $\pi(\alpha, \delta)$ should monotonically decrease with the approximation degree $\alpha$ and increase with $\delta$, to achieve a negative correlation with the variance. However, a carelessly designed $\pi(\alpha, \delta)$ could still be vulnerable under the negative correlation. Example 4.1: A data consumer wants to buy a range counting aggregation service $\Lambda(\alpha, \delta)$ with low aggregation variance. Therefore, she will intuitively specify a small value of $\alpha$ and a large value of $\delta$, and a higher price will be charged. However, she may also circumvent to pay the full price, and turns to buy multiple cheaper services of the same range counting with higher variances, denoted as $\{\Lambda(\alpha_i, \delta_i)\}_{i \in \{1, \ldots, m\}}$, $\alpha_i > \alpha, \delta_i < \delta$. We use $V(\alpha_i, \delta_i)$ to indicate the variance derived from the parameter pair $(\alpha_i, \delta_i)$, respectively. Afterwards, the data consumer estimates the result according to Formula (4). \[ \begin{align} \{\Lambda(\alpha_1, \delta_1), \ldots, \Lambda(\alpha_m, \delta_m)\} \\ \longmapsto \gamma(.) = \frac{1}{m} \sum_{i=1}^{m} \hat{\gamma}_i(.) \\ \longmapsto V(.) = \frac{1}{m^2} \sum_{i=1}^{m} V(\alpha_i, \delta_i), \end{align} \] In other words, the data consumer obtains a final range counting aggregation by computing the average of $m$ noisy answers, with an accumulated variance $\frac{1}{m^2} \sum_{i=1}^{m} V(\alpha_i, \delta_i)$ potentially lower than $V(\alpha, \delta)$. If the pricing function $\pi(\alpha, \delta)$ is arbitrage avoiding, the following conditional statement must hold: \[ \frac{1}{m^2} \left\{ \sum_{i=1}^{m} V(\alpha_i, \delta_i) \right\} \leq V(\alpha, \delta) \Rightarrow \sum_{i,j=1}^{m} \pi(\alpha_i, \delta_i) \geq \pi(\alpha, \delta). \] With the above conditional statement, we first use Lemma 4.1 to show the equivalence property of arbitrage-avoiding pricing function $\pi(\alpha, \delta)$ according to the variance. Lemma 4.1: For an arbitrary-avoiding pricing function $\pi(\alpha, \delta)$, there must be another function $\psi(V(\alpha, \delta)) = \pi(\alpha, \delta)$, when $\pi(\alpha, \delta)$ is arbitrage free. Proof. Assume there are two arbitrary sets of parameters $\alpha, \delta$ and $\alpha', \delta'$ with $V(\alpha, \delta) = V(\alpha', \delta')$. Then we must have $\pi(\alpha, \delta) = \pi(\alpha', \delta')$, i.e., $\psi(V(\alpha, \delta)) = \psi(V(\alpha', \delta'))$, indicating the price is uniquely determined by the variance. Otherwise, we have $\pi(\alpha, \delta) \neq \pi(\alpha', \delta')$, which means $\pi(\alpha, \delta)$ is not uniquely determined by $V(\alpha, \delta)$. We prove the lemma by contradiction. Assume $\pi(\alpha, \delta) > \pi(\alpha', \delta')$, then an arbitrage attack exists, as $V(\alpha', \delta') = V(\alpha, \delta)$, and $\pi(\alpha', \delta') < \pi(\alpha, \delta)$. This contradicts the assumption that $\pi(\alpha, \delta)$ is arbitrary-free. The proof is finished. $\square$ With the above equivalent expression, the following theorem shows how to address an arbitrage attack. Theorem 4.2: Any pricing function $\pi(\alpha, \delta)$ is arbitrage-avoiding if and only if the following properties hold: 1) $\pi(\alpha, \delta) = \psi(V(\alpha, \delta))$ 2) $\forall \alpha = \alpha_0, \delta = \delta_0, \Delta \delta \geq 0$, the relative difference of $\pi(.)$ and $V(.)$ follows \[ \frac{\pi(\alpha_0, \delta_0 + \Delta \delta) - \pi(\alpha_0, \delta_0)}{\pi(\alpha_0, \delta_0 + \Delta \delta)} \geq \frac{V(\alpha_0, \delta_0) - V(\alpha_0, \delta_0 + \Delta \delta)}{V(\alpha_0, \delta_0)} \] 3) $ \forall \alpha = \alpha_0, \delta = \delta_0, \Delta \alpha \geq 0 $, the relative difference of $ \pi(\cdot, \cdot) $ and $ V(\cdot) $ follows \[ \frac{\pi(\alpha_0, \delta_0) - \pi(\alpha_0 + \Delta \alpha, \delta_0)}{\pi(\alpha_0, \delta_0)} \leq \frac{V(\alpha_0 + \Delta \alpha, \delta_0) - V(\alpha_0, \delta_0)}{V(\alpha_0 + \Delta \alpha, \delta_0)}. \] Proof. Part 1: Sufficiency. We first prove the sufficiency of the properties, where the three listed properties can guarantee the pricing function to be arbitrage-avoiding. Assume there are two groups of parameters: $ \{(\alpha, \delta)\} $ and $ \{(\alpha_1, \delta_1), (\alpha_2, \delta_2), \cdots, (\alpha_m, \delta_m)\} $. Data consumers could either request for range counting once with parameters $ \{(\alpha, \delta)\} $, or request multiple times with $ \{(\alpha_2, \delta_2), \cdots, (\alpha_m, \delta_m)\} $. In the latter strategy, data consumers will apply the average value of all results as the final conclusion, with the variance equals $ Var(\frac{1}{m} \sum_{i=1}^{m} R_i) $, where $ R_i $ indicates the result for the $ i $th query. To prove pricing function $ \pi(\alpha, \delta) $ is arbitrage-avoiding, we need to prove that $ \pi(\alpha, \delta) \leq \sum_{i=1}^{m} \pi(\alpha_i, \delta_i) $ when $ Var(\frac{1}{m} \sum_{i=1}^{m} R_i) \leq V(\alpha, \delta) $. Firstly, we have \[ Var(\frac{1}{m} \sum_{i=1}^{m} R_i) = \frac{1}{m^2} \sum_{i=1}^{m} V(\alpha_i, \delta_i) = \frac{1}{m^2} \sum_{i=1}^{m} k_i \cdot V(\alpha, \delta). \] (5) We also have \[ \sum_{i=1}^{m} \pi(\alpha_i, \delta_i) = \sum_{i=1}^{m} l_i \cdot \pi(\alpha, \delta). \] (6) For an arbitrary set $ \{\alpha_i, \delta_i\} $, we introduce an intermediate set $ \{\alpha, \delta\} $, and estimate its relative change on variance and price compared to $ (\alpha, \delta) $. According to Property 2), we can accumulate the difference and \[ \frac{\pi(\alpha_i, \delta_i)}{\pi(\alpha, \delta)} \geq \frac{V(\alpha, \delta)}{V(\alpha_i, \delta_i)}. \] (7) We can further estimate the change according to Property 3) by accumulating the difference, and achieve the following inequality: \[ \frac{\pi(\alpha_i, \delta_i)}{\pi(\alpha, \delta)} \geq \frac{V(\alpha, \delta)}{V(\alpha_i, \delta_i)}. \] (8) Then we can combine the two inequalities, and the following conclusion holds: \[ \frac{\pi(\alpha_i, \delta_i)}{\pi(\alpha, \delta)} \geq \frac{V(\alpha, \delta)}{V(\alpha_i, \delta_i)}, \] (9) which means $ l_i \geq \frac{1}{k_i} $. As $ \frac{1}{m^2} \sum_{i=1}^{m} k_i \cdot V(\alpha, \delta) \leq V(\alpha, \delta) $, we have $ \sum_{i=1}^{m} k_i \leq m^2 $. Assume $ \overline{k} = \frac{\sum_{i=1}^{m} k_i}{m} $. Then we have $ \overline{k} \leq m $. The conclusion can be further extended: \[ \overline{k} \geq \frac{1}{m}. \] (10) The result in Inequality (10) can be combined with the following fact: \[ \sum_{i=1}^{m} \frac{1}{k_i} \geq \sum_{i=1}^{m} \frac{1}{\overline{k}}, \] (11) and guarantees \[ \sum_{i=1}^{m} \frac{1}{k_i} \geq 1. \] (12) Finally, we combine the results: \[ \sum_{i=1}^{m} l_i \geq \sum_{i=1}^{m} \frac{1}{k_i} \geq 1. \] (13) This is the same as the conclusion that \[ \sum_{i=1}^{m} \pi(\alpha_i, \delta_i) \geq \pi(\alpha, \delta), \] (14) and the proof of sufficiency is completed. Part 2: Necessity. Now we prove the necessity of the properties. Assume pricing function $ \pi(\alpha, \delta) $ is arbitrage-avoiding. We achieve the conclusion by contradiction. The necessity of the first property is proved in Lemma 4.1. As for the second property, assume there is a pair of $ \alpha' $ and $ \delta' $, where \[ \frac{\pi(\alpha', \delta' + \Delta \delta) - \pi(\alpha', \delta')}{\pi(\alpha', \delta' + \Delta \delta)} < \frac{V(\alpha', \delta') - V(\alpha', \delta' + \Delta \delta)}{V(\alpha', \delta')} \] (15) for some $ \Delta \delta > 0 $. Assume $ V(\alpha', \delta') = k \cdot V(\alpha', \delta + \Delta \delta) $, and $ \pi(\alpha', \delta') = l \cdot \pi(\alpha', \delta + \Delta \delta) $. Then there must be a pair of integers $ m_1 $ and $ m_2 $, such that data consumers can either buy range counting queries $ m_1 $ times with $ V(\alpha', \delta') $, or $ m_2 $ times with $ V(\alpha', \delta + \Delta \delta) $, while their general variances are identical. In this case, the total cost for the first strategy is smaller than that of the second one, as $ l \leq \frac{1}{k} $ according to Inequality (15). Then it contradicts the fact that $ \pi(\alpha, \delta) $ is arbitrage-avoiding, as we can achieve the same query result and variance with a lower cost. The proof of the third property is similar with that of Property 2), thus omitted here. Generally, any arbitrage-avoiding pricing function $ \pi(\alpha, \delta) $ must guarantee all the properties, and the necessity is proved. $ \square $ ### V. EXPERIMENTS This section presents the evaluation results of the proposed method. Both the results for approximate range counting aggregation, and the tradeoff between privacy preservation and utility are investigated. Datasets. The evaluation employs a real-world dataset, the pollution records in CityPulse Smart City Datasets [12]. The data are contributed by sensors with exact locations along the rode sides. Each record contains 5 air quality indexes: ozone, particulate matter, carbon monoxide, sulfur dioxide, and nitrogen dioxide. The dataset has 17568 records collected from 0:05am, 8/1/2014 to 0:00am, 10/1/2014. A. Effectiveness of the Sampling-Based Algorithm We evaluate the performance of the proposed sampling-based algorithm for answering approximate range counting queries. The first evaluation validates the impact of sampling probability $p$ on the accuracy of the sampling algorithm. In this evaluation, the maximum relative error of the sampling algorithm is calculated, while $p$ increases from 0.0173 to 0.4048. Fig. 2 shows that the maximum relative error is high with the maximum value 27% when the sampling probability is less than 0.12. Furthermore, with less data being preserved for querying, the accuracy oscillates considerably. Querying accuracy is very high if there are more than 15% data which are preserved in the samples; and accuracies remain stable. From Fig. 2, we further conclude that the proposed algorithm has high aggregation accuracy, i.e., the maximum relative error can be bounded to 3% if there are more than 5% data which are preserved in the samples. The second evaluation studies the impact of $\alpha$ and $\delta$ on the accuracy of the achieved results. The accuracy is computed while $\alpha$ and $\delta$ increase from 0.08 to 0.8. Fig. 3 shows that the maximum relative error remains stable when $\delta$ is larger than 0.3. Furthermore, the maximum relative error is less than 0.019 when $\delta$ is larger than 0.3. When a small $\delta$ is given, e.g., $\delta < 0.3$, the maximum relative error shocks significantly. The third evaluation investigates the impact of data size on the sampling probability. In this evaluation, $\alpha$ and $\delta$ are set to be 0.055 and 0.5, respectively. The sampling probability is computed while the size of data increases from 10% to 100% of the original dataset. As shown in Fig. 4, our algorithm is suitable for big data range counting aggregation, as it largely reduces computation and communication overhead. When data size is very large, the sampling probability can converge to a stable state with less data collected. Therefore, our sampling probability can appropriately balance aggregation accuracy and overhead. B. Privacy-Utility Tradeoff We also evaluate the performance of the differential privacy mechanism for realizing privacy-utility tradeoff. The first evaluation investigates the impact of privacy budget $\epsilon$ on the accuracy of the range counting aggregation. The accuracy is computed while $\alpha$ and $\delta$ increase from 0.08 to 0.8. Furthermore, $\epsilon$ increases from 0.01 to 8, and $p = 0.4$. As shown in Fig. 5, with the decreasing of the privacy requirements, aggregation accuracy can be improved, indicating less privacy concerns lead to better utilities. Moreover, if a strong privacy guarantee is required with $\epsilon = 0.1$, our algorithm can still bound the relative error under 8% for all these 5 datasets, showing the high stability of the proposed framework. The second evaluation investigates the impact of sampling probability on privacy preservation under different privacy budgets. Fig. 6 shows that querying accuracy is low when $p < 0.15$. With the increasing of sampling probability from 0.0173 to 0.25, querying accuracy is improved as more samples are collected. The above observations indicate that the global sensitivity of $\hat{\gamma}(b_1, b_u, S)$ satisfies $GS(\hat{\gamma}(b_1, b_u, S)) \propto 1/p$, and a larger $p$ means smaller volume of differential privacy noise. VI. LITERATURE REVIEW Approximate Range Counting Aggregation. Sampling-based algorithms have been proposed for approximate data aggregation in many areas [13] [14] [15], such as data stream, tradition database systems, P2P networks, etc. However, these works are not designed for range counting queries for big data in IoT, indicating that no guarantees on performance are provided for this kind of queries. The sampling-based algorithms have also been applied for long-term queries via continuous data collection. Considering the high correlation in temporal and spatial dimensions, the work in [16] proposes a distributed approximate aggregation algorithm which can considerably reduce aggregation overhead. The work in [17] proposes some algorithms to realize the tradeoff between aggregation overhead and aggregation accuracy in order to prolong network lifetime through allocating tolerable error bounds to each sensor node in a network. Some indexing structures are proposed in [18] to conduct spacial online sampling and data aggregation on big temporal-spatial and spatial datasets. As a sampling-based algorithm, the structures proposed in [18] have good performance for dynamic datasets. However, these works mainly focus on the reduction of long-term bandwidth consumption. They have no promise on the performance of the one sample multiple queries discussed in this paper. Furthermore, they neglect the privacy issues underlying the collected data. Differentially Private Data Aggregation. There are many state-of-the-art paradigms investigating differentially private data aggregation [19]. Especially, the work in [20] studies the problem of hierarchical decomposition with differential privacy guarantee based on spatial decomposition trees. The paradigm proposed in [20] can efficiently answer differentially private range counting by eliminating the dependence of querying sensitivity on the maximum height of the decomposition tree. The work in [21] investigates the problem of releasing the degree distribution of a graph under differential privacy. To reduce Laplace noise volume, the work in [21] transfers high-dimensional graph data to low dimensional data based on a graph projection approach. However, these studies mainly focus on reducing the scale of noise introduced by privacy preservation, while they fail to demonstrate an unbiased estimator and cannot be applied directly for data in distributed environment. Trading Private Aggregation. For the trade of data analyzing results, current research mainly focuses on arbitrage avoiding in querying-based pricing mechanisms [22] [23]. The work in [9] proposes some arbitrage avoiding pricing mechanisms for arbitrary formats of queries. However, these conclusions cannot be extended to the range counting aggregation, and no guarantees on privacy and utility tradeoff are given. VII. CONCLUSION This paper investigates the problem of trading approximate range counting for big IoT data. The objective is to derive query answers with optimal differential privacy guarantee, while the answers satisfy the specified approximation degree. A two-phase sampling-based approach with bounded variance is proposed, including an unbiased estimator based on sampled data and a perturbation mechanism designed for optimal differential privacy. The paper further studies the pricing mechanism for trading counting results, facing the arbitrage attacks from cunning consumers. As a result, the critical condition to establish a proper pricing function is established. All the proposed methods are validated towards real-world datasets. ACKNOWLEDGMENT This work is partly supported by the National Science Foundation (NSF) under grant NOs. 1252292, 1741277, 1829674, and 1704287. REFERENCES [1] “Data brokers: A call for transparency and accountability: A report of the federal trade commission.” https://www.ftc.gov/reports/data-brokers-call-transparency-accountability-report-federal-trade-commission-may-2014, 2014. [2] S. Madden, M. I. Franklin, J. M Hellerstein, and W. Hong, “The design of an acquisitional query processor for sensor networks,” in Proceedings of the 2003 ACM SIGMOD International Conference on Management of Data, ser. SIGMOD ’03, 2003. [3] ——, “Tango: a tiny aggregation service for ad-hoc sensor networks,” ACM SIGOPS Operating Systems Review, vol. 36, 2002. [4] J. Li and S. Cheng, “(ε, δ)-approximate aggregation algorithms in dynamic sensor networks,” IEEE Transactions on Parallel and Distributed Systems, vol. 23, no. 3, pp. 385–396, 2012. [5] J. Li, S. Cheng, Z. Cai, J. Yu, C. Wang, and Y. Li, “Approximate holistic aggregation in wireless sensor networks,” ACM Transactions on Sensor Networks (TOSN), vol. 13, no. 2, p. 11, 2017. [6] Z. He, Z. Cai, S. Cheng, and X. Wang, “Approximate aggregation for tracking quantiles and range countings in wireless sensor networks,” Theoretical Computer Science, vol. 607, pp. 381–390, 2015. [7] “Invasion of the internet-of-things data snatchers,” https://www.aclu.org/blog/speech/invasion-data-snatchers-big-data-and-internet-things-means-surveillance-everything, accessed: 2010-09-30. [8] C. Dwork, “Differential privacy,” in Automata, Languages and Programming, 2006, pp. 1–12. [9] B.-R. Lin and D. Kifer, “On arbitrage-free pricing for general data queries,” Proceedings of the VLDB Endowment, vol. 7, no. 9, pp. 757–768, 2014. [10] C. Dwork, F. McSherry, K. Nissim, and A. Smith, “Calibrating noise to sensitivity in private data analysis,” in Theory of cryptography conference. Springer, 2006, pp. 265–284. [11] S. P. Kasiviswanathan, H. K. Lee, K. Nissim, S. Raskhodnikova, and A. Smith, “What can we learn privately?” SIAM Journal on Computing, vol. 40, no. 3, pp. 793–826, 2011. [12] “Citypops: Urban city datasets,” http://iot.ee.surrey.ac.uk:8080/datasets.html, accessed: 2014. [13] H. Harb, A. Mahkoui, D. Lajmiyani, and A. Jaber, “A distance-based data aggregation technique for periodic sensor networks,” ACM Transactions on Sensor Networks (TOSN), vol. 13, no. 4, p. 32, 2017. [14] S. Cheng, J. Li, and Z. Cai, “O (ε)-approximation to physical world by wireless sensor networks,” in INFOCOM, 2013 Proceedings IEEE. IEEE, 2013, pp. 3087–3092. [15] S. Cheng, Z. Cai, and J. Li, “Approximate sensory data collection: A survey,” Sensors, vol. 17, no. 3, p. 564, 2017. [16] A. Bouli, S. Ganeriwal, and M. B. Srivastava, “Aggregation in sensor networks: Energy-accuracy trade-off,” Ad hoc networks, vol. 1, no. 2-3, pp. 317–331, 2003. [17] X. Tang and J. Xu, “Extending network lifetime for precision-constrained data aggregation in wireless sensor networks,” in Proceedings IEEE INFOCOM 2006, 25TH IEEE International Conference on Computer Communications, April 2006, pp. 1–12. [18] X. Wang, R. Christensen, P. Li, and K. Yi, “Spatial online sampling and aggregation,” Proceedings of the VLDB Endowment, vol. 9, no. 3, pp. 84–95, 2015. [19] R. Chen, B. Fung, B. C. Desai, and N. M. Sosou, “Differentially private transit data publication: a case study on the medical transportation system,” in Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2012, pp. 221–231. [20] J. Zhang, X. Xiao, and X. Xie, “Privtree: A differentially private algorithm for hierarchical decompositions,” in Proceedings of the 2016 International Conference on Management of Data, 2016, pp. 155–170. [21] W.-Y. Dang, J. Li, and M. Lyu, “Publishing differentially private distribution tables,” in Proceedings of the 2016 International Conference on Management of Data. ACM, 2016, pp. 123–138. [22] S. Deep and P. Kourlis, “Qirana: A framework for scalable query pricing,” in Proceedings of the 2017 ACM International Conference on Management of Data. ACM, 2017, pp. 699–713. [23] Y. Ning, Z. Zhao, F. Wu, S. Yang, T. Gao, and G. Chen, “Unlocking the value of privacy: Trading aggregate statistics over private correlated data,” in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. ACM, 2018, pp. 2031–2040.
Trust Broker: A Defense Against Identity Theft From Online Transactions Michael George Edvalson Brigham Young University - Provo Follow this and additional works at: https://scholarsarchive.byu.edu/etd Part of the Computer Sciences Commons BYU ScholarsArchive Citation Edvalson, Michael George, "Trust Broker: A Defense Against Identity Theft From Online Transactions" (2005). Theses and Dissertations. 348. https://scholarsarchive.byu.edu/etd/348 This Thesis is brought to you for free and open access by BYU ScholarsArchive. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of BYU ScholarsArchive. For more information, please contact email@example.com, firstname.lastname@example.org. TRUSTBROKER: A DEFENSE AGAINST IDENTITY THEFT FROM ONLINE TRANSACTIONS by Michael G. Edvalson A thesis submitted to the faculty of Brigham Young University in partial fulfillment of the requirements for the degree of Master of Science Department of Computer Science Brigham Young University April 2006 Copyright © 2006 Michael G. Edvalson All Rights Reserved BRIGHAM YOUNG UNIVERSITY GRADUATE COMMITTEE APPROVAL of a thesis submitted by Michael G. Edvalson This thesis has been read by each member of the following graduate committee and by majority vote has been found to be satisfactory. _____________________________ ________________________________ Date Kent E. Seamons, Chair _____________________________ ________________________________ Date Charles D. Knutson _____________________________ ________________________________ Date Sean C. Warnick As chair of the candidate’s graduate committee, I have read the thesis of Michael G. Edvalson in its final form and have found that (1) its format, citations, and bibliographical style are consistent and acceptable and fulfill university and department style requirements; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the graduate committee and is ready for submission to the university library. Date Kent E. Seamons Chair, Graduate Committee Accepted for the Department Parris K. Egbert Graduate Coordinator Accepted for the College Thomas W. Sederberg Associate Dean, College of Physical and Mathematical Sciences The proliferation of online services over the years has encouraged more and more people to participate in Internet activities. Many web sites request personal and sensitive information needed to deliver the desired service. Unfortunately, it is difficult to distinguish the sites that can be trusted to protect such information from those that cannot. Many attempts to make the Internet easier to use introduce new security and privacy problems. On the other hand, most attempts at creating a safe online environment produce systems that are cryptic and hard to use. The TrustBroker system is based on a specialized online repository that safely stores user information and helps the user determine which sites can be trusted with their sensitive information. Also, the repository facilitates the transfer of the user’s information. The overall effect of the system is to inspire greater confidence in online participation among users who desire to protect their personal information. ACKNOWLEDGMENTS This thesis would not have been possible without the support from my wife Brandi who held down the fort during the many hours I spent at school. Thanks also goes to my daughter Amanda who always gave me a reason to smile after a long day in the lab. Sincere thanks goes to Dr. Kent Seamons for his deep insight and support as my advisor and employer. Lastly, I greatly appreciate the ideas and contributions offered by the fellows in the Internet Security Research Lab. This research was supported by funding from the National Science Foundation under grant no. CCR-0325951 and prime cooperative agreement no. IIS-0331707, and The Regents of the University of California. # Table of Contents 1 Introduction 1 1.1 Establishing Trust and Transferring Information 2 1.2 Introduction to TrustBroker 3 1.3 Thesis Overview 4 2 Related Work 5 2.1 Transport Layer Security 5 2.2 Identity Management 5 2.2.1 Microsoft Passport 7 2.2.2 Liberty Alliance 8 2.3 Rating Systems 9 2.4 Reputation Systems 10 2.5 Attribute-based Systems 12 3 System Requirements and Design 15 3.1 Requirements Analysis 15 3.1.1 TrustBroker Repository 15 3.1.2 Client-side Module 16 3.1.3 Web Server 16 3.2 Design Analysis 16 3.2.1 TrustBroker Token 17 3.2.2 Protocol 21 3.2.3 Data Obfuscation ........................................ 28 3.2.4 Repository Policy Agglomeration ....................... 31 4 Implementation .............................................. 35 4.1 Client-side Module ......................................... 36 4.1.1 User Interface ..................................... 36 4.1.2 Token Management .................................. 42 4.2 TrustBroker Repository ................................... 43 4.2.1 Web Interface for Users ............................ 44 4.2.2 Remote Interface for Web Servers .................. 44 4.3 Web Server ............................................... 48 4.3.1 Page-enabling ..................................... 48 4.3.2 Handling the Token ................................ 48 4.3.3 Acquiring the User’s Information from the Repository .... 48 5 Threat Analysis ............................................. 51 5.1 TrustBroker Token ........................................ 51 5.1.1 Impersonation ..................................... 51 5.1.2 Replay ............................................. 54 5.1.3 Modification ...................................... 54 5.2 Pre-existing Threats ..................................... 55 6 Conclusion .................................................. 59 7 Future Work ................................................ 61 References ..................................................... 62 List of Figures 3.1 Elements of a TrustBroker token. . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.2 Channels of communication in a typical TrustBroker session. . . . . . . . . . . . 21 4.1 Overview of a TrustBroker transaction. . . . . . . . . . . . . . . . . . . . . . . . 35 4.2 TrustBroker tool menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.3 Client module About dialog box. . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.4 TrustBroker toolbar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.5 Client module Settings dialog box. . . . . . . . . . . . . . . . . . . . . . . . . . . 40 4.6 TrustBroker settings as part of web browser preferences. . . . . . . . . . . . . . 40 4.7 Confirmation dialog box when browser initially loads. . . . . . . . . . . . . . . . 41 4.8 Sending a token to the client. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.9 Notifying the client that a site is token-enabled. . . . . . . . . . . . . . . . . . . 43 4.10 Protocol messages between web server and TrustBroker repository . . . . . . . 45 5.1 Impersonation attack. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 5.2 The encrypted domain element of TrustBroker token. . . . . . . . . . . . . . . . 53 5.3 The token hash element of TrustBroker token. . . . . . . . . . . . . . . . . . . . 55 LIST OF FIGURES Chapter 1 — Introduction The remarkable growth in the number of online services over the last few years has encouraged more people to turn to the Internet for finding information and purchasing goods. As additional sites become available, people are more able to find services that satisfy their interests. Many of these services desire to create a relationship with their users, introducing the need to acquire personal information. At the same time, people are wary of giving out sensitive information in fear that it might be abused. In the past, closed systems have been the predominant form of information control. A closed system is an environment in which proposed services are filtered by a gatekeeper before being offered to the members of the community. In this type of environment, the growth and viability of eligible services are controlled and maintained by a centralized authority. This allows one to predict the quality and trustworthiness of the service based on what is known about the gatekeeper’s policies. However, this also allows for censorship and restrictions on the amount and types of services allowed. As an example, consider a highly specialized bookstore that seeks a strong reputation for its technical literature. This reputation helps customers predict the quality of any book found in the store. However, it is developed according to the subjective judgement of the owner of the bookstore by selling only the books he considers of worth, while refusing to sell those that are deemed less valuable. The Internet, on the other hand, is an open system. There is no gatekeeper that filters content or services. Rather, anyone can publish anything or offer any service. Such an environment facilitates a proliferation of published opinions, ideas, information, and services. However, because everyone (including malicious entities) has a voice, determining what and who is believable is a difficult challenge. Indeed, it can be difficult to distinguish sites that can be trusted to protect a person’s information from those that cannot. Identity theft from participating in online activities is a significant problem. Phishing attacks are increasingly more sophisticated. The incessant demand for new and fresh spam targets elevates the incentive for online businesses to sell personal information while turning a blind eye to the consequences. As the power, flexibility, and availability of the Internet continue to expand, so does the threat of identity theft and other abuses of personal information. Thus, a user’s willingness to participate in new and exciting opportunities on the Internet is dampened by the requirement to proceed cautiously, sometimes suspiciously. 1.1 Establishing Trust and Transferring Information Two motivating examples demonstrate common viewpoints of the same problem faced by the average user while online. Suppose a potential e-commerce customer, George, is shopping for a particular digital camera online. While browsing, George discovers a web site that appears to offer the camera for a better price than any other site. Despite the assurance from his web browser that his session with the remote server is secure, he may be somewhat suspicious that this site may not adequately protect his personal information. In this case, helping George gain confidence that the server will protect his personal information may be the single most important factor in facilitating the desired transaction between them. Secondly, suppose Janice is an avid participant of a popular Internet auction site. She receives an email from the auction site indicating that they are completing a review of all accounts. The email instructs her to click on the supplied link and login with her normal username and password to verify that her account is still active. She clicks the link, which takes her to the very familiar login page of the auction site. She is on her own to determine if the site she is looking at is the authentic login site of the auction or if it is the front of a sophisticated phishing attack. In addition to the stress of determining the credibility of the remote server in each of these scenarios, a second problem exists. If a user decides to interact with the remote site, he is required to enter personal information. The repeated entry of this information can be monotonous and time consuming, which may further discourage would-be participants. 1.2 Introduction to TrustBroker The solution to both of these problems can be accomplished using TrustBroker: a service that comprises 1) an online repository of user information, and 2) a protocol that serves to establish trust between remote entities and automatically transfer user information between them. The online repository aids the user by establishing trust on his behalf. Once trust has been established between the web server and the repository, the repository discloses the user’s requested information. If trust cannot be established, no information is revealed. This process provides simultaneous solutions to the two problems previously defined. First, the user determines if he can trust the unfamiliar server by noticing if the web server is able to acquire any of his personal information. Second, the user’s information is automatically sent to the remote server, eliminating the need for the user to provide it manually. 1.3 Thesis Overview The remainder of this thesis is outlined as follows. Chapter 2 enumerates the related work. Chapter 3 explains the challenges and possible solutions associated with the design of each of the elements of a successful TrustBroker architecture. Chapter 4 delineates a proof-of-concept implementation along with a detailed description of what approaches were utilized and why. Chapter 5 comprises a thorough security analysis of the TrustBroker system, including its strengths and weaknesses. Chapters 6 and 7 include the conclusions of this thesis and possible avenues for future work. Chapter 2 — Related Work A significant amount of research has been devoted to the problem of information privacy and usability on the Internet. This chapter is devoted to introducing the current state of research in this field as well as what is currently available in the consumer market. 2.1 Transport Layer Security Transport Layer Security (TLS) is a widely used protocol that performs two important tasks with regard to security and privacy. First, it offers a mechanism by which each participating entity can verify the identity of the other. This protects against malicious impersonation attempts. Second, it establishes a secure connection between the two parties over which sensitive information can safely flow. These two tasks are critical to a secure online environment. However, the Public Key Infrastructure (PKI) upon which TLS depends is not perfect. There are numerous ways to circumvent the security features of a PKI [?]. In addition, while TLS can ensure that the line of communication is secure, it cannot guarantee that either entity will protect information divulged by the other. 2.2 Identity Management Many ideas have surfaced to remedy the problem of managing user identity on the Internet in an attempt to make the Internet easier to use. One method is to use a local repository, such as a data and password manager. This repository stores a user’s personal information and passwords in an encrypted file on the local computer and supplies the correct username, password, and other data to a given site as directed by the user. This enables the use of multiple strong (i.e., hard to remember) passwords because the user is not required to remember them. However, a number of drawbacks have been discovered in such systems. First, the keys to all accounts and passwords are generally protected by a single password, so if the file can be downloaded and cracked off-line by a malicious person, unauthorized access to all accounts and information is granted. This is especially dangerous given the tendency for people to select weak passwords. Second, since that information resides on only one computer, a person will not be able to access their online accounts from any other computer. Third, the loss of this password file renders all accounts inaccessible. Another method of managing a person’s online identity is to federate a set of websites so that the username/password that grants access to any of them also grants access to the rest. Federated Identity Management is a broad field of research that incorporates a diverse set of motivations and objectives. The common theme, however, is to unify a group of distinct security domains so that the identity of a particular user may be recognized from one domain to another with little or no user intervention [?]. The research space is immense, due mostly to the huge diversity of needs and usage models that exist in large institutions today. One example is the centralization of directory and account services at the University of Pittsburgh [?]. Another example is federated access control for databases [?, ?]. A third example is the desire for Internet-wide single sign-on access control for every Internet user [?]. Though the requirements of these systems are diverse, they share a fundamental ideology in that they mostly rely upon a centralized store of user information within a federated community. Two significant players in this field are Microsoft’s Passport [?] and the Liberty Alliance’s Federated Identity System [?]. Despite strong similarities between their ultimate objectives, these two groups have approached the solution in remarkably different ways. Together, they serve as contrasting case studies regarding the problems associated with federated identity systems and different ways to solve them. A short description of these two systems follows, including a brief analysis of their respective benefits and drawbacks. 2.2.1 Microsoft Passport Microsoft Passport [?] is built around a central user information repository. Each person that wishes to participate must create a relationship (a user account) with this repository. Each business that wants to participate must establish a business relationship with Passport and implement the proprietary technology that interfaces with Passport’s central user information store. When a user visits a site that is Passport enabled, he need only enter a username and password to authenticate. This username and password combination is universal to all systems that are Passport enabled. With this information, the site can retrieve other information about the person from the central Passport repository. The key benefit of this solution is that the user has to remember only a single username and password. Also, the user’s information is centrally located. Changing pieces of information is a simple process that can be done from any computer with an Internet connection and web browser. Some potential drawbacks exist, however. First, each person who wishes to participate must rely on the privacy practices of Passport to protect his information from theft and wrongful distribution. There are no alternative repositories. Thus, if Passport proves less secure with user information than a person deems reasonable or necessary, that person only has the option to stop participating. Second, having all the user information in one place makes a very inviting target for attackers. Third, not every online company can feasibly participate in Passport because of economic issues. According to Microsoft, participation in Passport requires a yearly $10,000(US) fee in addition to a periodic $1,500(US) compliance fee per URL that is Passport enabled [?]. This is in addition to any consulting/development costs needed to implement the technologies required by Passport. Clearly, this eliminates many small- to medium-sized businesses from participating. In the end, this prohibits the wide-scale adoption of this technology. 2.2.2 Liberty Alliance The Liberty Alliance [?] was also formed to solve the problem of online identity management. Their primary goal is to enable a federated identity management system between business partners and consumers. This means that after authenticating to one site, a user may visit any of the site’s federated partners without re-authenticating—his identity follows him to each site. A problem with this solution is that a user’s identity federation and single sign-on capabilities are constrained by whether the sites that he visits have chosen to participate in a federation. If they have not, or if they are not allowed to participate in a particular group, the user cannot enjoy these benefits. Additionally, a site may only belong to one federation. Thus, if one interesting site belongs to one federated group and another interesting site belongs to a different group, the user cannot enjoy the benefits of a federation that includes both sites. The potential problems become apparent as this structure is logically applied to the real world. Consider the situation in which a site wishes to federate with more than one group. For example, an online clothing store may wish to federate with a bank that is in one federated group. It may also wish to federate with other online stores owned by the same clothing company. Though, the Liberty Alliance has mentioned the intent to resolve this, their specifications currently prohibit a site from joining multiple federations. Additionally, this model requires that the user already have an account with a site before he can participate in the federated system. In the common situation where a user is discovering new sites, such as with George in the introductory scenario, this forces the user to decide if he can trust the new server with his personal information before he can take advantage of any features of the federated system. Because of this, it may also discourage the active search for new Internet content and services. Another issue arises from a business perspective. Since groups are formed among sites in order to establish single sign-on capabilities between them, the barriers to entry for a company may become very large. In some cases, groups could restrict entry to only those who meet very specific and perhaps strict criteria. For example, it may be possible that a group of large corporations decide that membership in their group requires a hefty fee. In this case, small- to medium-sized businesses are effectively banned from participating in that group. Another example of failure might occur when a user desires to check airplane ticket prices from multiple competing websites. These companies, in competition with each other, may not desire to federate. Again, the system breaks and the user is not able to benefit from single sign-on technology. In any of these scenarios, the core ideology of federated identity systems may become frustrated. Thus, for the average Internet user, the details of the Liberty Alliance specifications may restrict the desired user experience and undermine the real intent of federated identity systems. 2.3 Rating Systems There has been a significant amount of research regarding the problem of establishing trust in an untrustworthy environment like the Internet [?, ?, ?, ?]. Some methods prescribe a rating system where users who view the content are allowed to rate it. Thus, the more a site is viewed and rated, the more accurate the rating will be with regard to the general consensus of the population. Amazon.com is one enterprise that uses such a rating system. However, since it is often a driven minority who are the most vocal, it becomes important to rate the raters [?]. Such a rating system becomes recursive as now you must rate the raters of the raters, and so on [?]. Some claim that this process should stop after the first recursion, with the raters of the raters; it becomes unmanageable to continue [?]. However, the level of reviewers at which the recursion stops cannot be held accountable for their actions because there is no further reviewing process. An approach taken by Slashdot.org is to periodically select groups of people at random from an established section of the general population to perform the ratings [?]. While this allows for a much more manageable rating system, it still may not convey the true value of the site with regard to the general population. Those selected to perform the ratings might be biased towards the feelings of those who chose them as raters. Another fundamental hurdle with these types of systems is that immediate trust in new content or entities cannot be established. As mentioned, the rating of the site only becomes accurate over time as more and more users take time to view and rate the content. The Internet contains too much information to be rated piece by piece. The implementation of such a system on the Internet is infeasible. 2.4 Reputation Systems A reputation system is loosely defined as a connected graph in which every node develops a reputation among its neighbors. While a fully connected graph would be ideal, the staggering number of nodes needed to represent the entire Internet prohibits such a graph. Thus, each node in such a large graph is usually connected to a manageable number of neighbors. Essentially, the reputation of a single node can be discovered by searching the graph until the node’s neighbors are found and queried. When a particular node encounters an unknown entity, it may ask closely trusted neighbors if it should trust the new node [?]. The aggregated recommendation from its trusted neighbors becomes the new trust policy towards the new node, and a new connection in the graph is made. The proliferation of research in this field focuses primarily on the problems of identifying the attributes that can be used to compute trust and then transferring the meaning of that trust from one node to another. Several approaches have been taken by researchers [?]. Among them, the common premise is that if a node can be trusted, all of the nodes that it recommends can be trusted, although possibly with a greater degree of caution. Further, all of the newly trusted nodes’ recommendations should also be trusted. When trusted neighbor nodes disagree about whether an unfamiliar node can be trusted, an algorithm may be used to determine the likelihood that the new node can be trusted based on information from as many trusted nodes as possible. Thus, a reputation graph is established where any node can ask about the trustworthiness of any other node. The major benefit of this type of system is that it is scalable. As more nodes are added to the graph, the overall graph gets bigger, but only those nodes that are directly connected to the new node are affected by the increase. Most reputation-based proposals acknowledge frustrating flaws. Primarily, the addition of a new node to the graph is a difficult task. It is difficult to create an initial trust classification that is accurate. In other words, it is difficult to bootstrap the system and difficult to add nodes in a reliable manner. Another significant problem is that trust is context-sensitive. A person that is trusted to give advice about car repair may not necessarily be trusted to give sound financial advice, and vice versa. Both may be trusted, but only within their spheres of expertise. Thus, asking whether a node in the reputation graph can be trusted is meaningless unless a context for the trust is also provided. Thus, it is not enough to maintain a single trust rating for each neighbor. Each node must maintain a trust rating for each neighbor with regard to every context that may be of interest. This point necessarily breaks a reputation network into many layers, or many reputation networks, each layer representing a different context and intertwined with other layers. Therefore, a globally accepted list and definition of contexts must be maintained, as well as a mechanism for adding new contexts. Delimiting the entire list of possible contexts alone is likely an insurmountable task, not to mention the incredible task required of each node of maintaining its own location in many concurrent reputation systems. Lastly, most algorithms for defining and transferring reputation-based trust are easy targets for deceptive nodes. In cases where some nodes may knowingly report false information, such as reporting that a node is trustworthy when it is not, or reporting that it is not trustworthy when it is, the reliability of information is severely damaging. Compounding the problem, there is usually no way to determine if a node is purposefully being deceptive, so the validity of the trust inference of nodes on the network may become highly questionable. This undermines the entire purpose of such systems [?]. 2.5 Attribute-based Systems Attribute-based systems for developing trust are relatively new. A fundamental observation that led to their inception is that identity-based systems are rigid, inflexible, and unscalable. When considering an open and dynamic environment such as the Internet, identity-based trust establishing systems are not a good fit. The reasoning follows: “Instead of trying to figure out who they are, let’s care about what they are.” Thus, if two entities are able to assert certain attributes about each other, a level of trust can be established even though their identities remain a mystery. Trust negotiation [?] is an example of an attribute-based trust establishment model. The core process entails the iterative exchange of access control policies and credentials that attempt to satisfy those policies. These credentials usually take the form of certificates that are signed by a trusted third party and assert some attribute about the holder. Once the policies that govern access to a resource have been satisfied, the resource is granted to the requesting party. In this way, strangers are able to develop some level of mutual trust. Attribute-based systems seem to play well in open and dynamic environments, and appear to be scalable and flexible. These attributes make them suitable for authentication and trust establishment models in open systems such as the Internet. However, there are some obstacles to overcome. The policies that govern access to resources are generally very complex and difficult to create correctly. In addition to forming correct access control policies for sensitive data, credentials need to be obtained, managed, and protected. The loss, theft, or transfer of credentials is quite hazardous, and creates a quagmire for the management and distribution of credentials. In essence, these systems are hard to use. CHAPTER 2. RELATED WORK Chapter 3 — System Requirements and Design The exploration of the TrustBroker system design is divided into two sections: 1) the requirements analysis and 2) the design. The requirements analysis section provides a detailed exploration of what each component needs to accomplish to make TrustBroker successful. The design section analyzes alternative technological approaches that satisfy these requirements, and discusses the ramifications of design decisions on an actual implementation. 3.1 Requirements Analysis This section discusses the overriding design goals and challenges of TrustBroker. It is important to first mention that while a privately managed, single-user repository is certainly a viable player in the TrustBroker system, this research focuses on the establishment of a public, multi-user repository. The arguments made hereafter are directed towards the multi-user paradigm. User-friendly security is the fundamental principle of TrustBroker. While information security is essential, it must not compromise usability, and vice versa. The TrustBroker architecture has three basic components: repository, client-side module, and web server. Each of these is discussed in the following sections. 3.1.1 TrustBroker Repository The TrustBroker repository must provide two remote interfaces. The first interface allows a user to register and manage personal information and privacy settings. In addition to these fundamental services, there may be ancillary services available to a user, such as options to view and manage a subscription or learn details about how the repository protects his personal information. Because usability is a fundamental design goal, this interface must be clear, simple to use, and comprehensible. to humans. Remote web servers may connect in order to request information about a person through the second interface exposed by the repository. Through this interface the repository establishes trust with the remote server and handles incoming requests for personal information. The repository must support fine-grained access control for each piece of user information. It must guarantee that it will release information only to trustworthy servers. 3.1.2 Client-side Module The client module resides on the client computer and provides a user interface to TrustBroker. The interface must be easy to use, yet leave the user in control over the disclosure of personal information. Thus, the design challenge is two-fold: 1) discover an appropriate way to grant maximum control to the user over the distribution of information; and 2) refrain from being intrusive or annoying when adhering to protective security practices. Additionally, the client module must operate correctly behind a firewall or from a private network. 3.1.3 Web Server A web server interacts with the TrustBroker client module and repository by notifying the client module that it is TrustBroker-enabled and requesting personal information from a TrustBroker repository. When necessary, it must prove to the repository that it can be trusted to protect that information. 3.2 Design Analysis The previous section outlined the TrustBroker requirements. This section presents design alternatives that satisfy the requirements, along with the pros and cons of each option. 3.2.1 TrustBroker Token A significant design issue is to develop a strategy by which the repository can guarantee that the correct person’s information is being distributed, and only to trustworthy servers. It must be able to guarantee that the client seeking services from the web server is who he says he is and that the server can be trusted to protect any disclosed information. To accomplish this, the repository must issue to the client some sort of immutable token, which can be passed to the web server instead of the requested personal information. The web server passes this token to the repository to indicate who’s information is being requested. Because this token is responsible to help guarantee the integrity of the system, the design and format of the token is critical. The token must provide these assurances: 1. The server must know which TrustBroker repository is representing the client. 2. The TrustBroker repository must be assured that the client making the request owns the token. 3. The token must provide proof of data integrity because it is transferred through a public, untrusted environment. Any violations of these demands will negate the usefulness of TrustBroker. The token contains 6 elements: the domain of the TrustBroker repository; the port on which the repository is listening; a user identifier; an expiration timestamp; a hash of the token (either keyed or signed); and an encrypted element that the client code adds (Figure 3.1). These elements are described below. TrustBroker Domain. The first element of the token is the DNS domain name of the TrustBroker repository. When the web server receives a token from the client, it uses the domain name to contact the appropriate repository. TrustBroker Port. Instead of setting a fixed port for the TrustBroker repository, a repository may choose to listen on any port. This helps reduce the threat of a widespread Denial of Service attack against TrustBroker. Therefore, the second element of the token is the port on the TrustBroker domain to which the web server should connect when requesting data. User Identifier. Some user identifier must be embedded in the token to identify the user to the TrustBroker repository. The following are options that satisfy this requirement: Actual User Identifier. This user identifier could be the actual user identifier (such as the unique login name) of the user on the repository. However, this may provide in clear text unnecessary information to the web server and anyone else who gains access to the token. Encrypted Form of Actual User Identifier. The next alternative is an encrypted form of the client’s user identifier as defined by the repository. This allows only the TrustBroker repository to determine the owner of the token. Random Number The user identifier in the token could also be a random number associated with the user during the lifespan of the token. When the repository generates the token, it also generates a random number that is unique among all the users known to the repository. This number is temporarily associated with the client so that when the web server sends the token back to the repository, the TrustBroker repository can determine the owner of the token. This is a compelling option because it stores absolutely no information about the user, either in clear text or encrypted. Also, it removes the demand for an additional encryption/decryption from the TrustBroker repository. Lastly, this ID would change to a new random number with each token. This reduces the chance that it could be used to obtain unauthorized access to the client’s sensitive information because every time a token is issued the identifier from the previous token is forgotten and therefore invalid. The only overhead required would be to ensure the random number is unique to all other currently issued tokens. Token Authenticator. A clear principle of TrustBroker’s security model is to ensure that the contents of the token cannot be altered after it is created by the repository. Therefore, one element of the token is a value that can be used by the repository to determine if the token’s contents have been modified. One way to accomplish this is to sign a hash of the token’s contents with the private key of the TrustBroker repository. This would allow any remote entity to verify the authenticity of the hash through decryption with the repository’s public key. Also, it would prevent anyone from spoofing the original hash by changing the contents of the token and recreating the hash, unless the private key of the repository was compromised. The second form of protection is to use a hash-based message authentication code (HMAC). The repository maintains a secret key used when creating the token. When a request for information is received along with a token, the repository uses the secret key to verify the integrity of the token. This computation is very fast when compared to a database access or public key computation. This also means that no remote entity (e.g., the web server) would be able to independently check the integrity of the token. More about the security aspect of this element of the token are discussed in Chapter 5. Web Server Identifier. In order to generate the last element of the token, the repository generates a shared symmetric key and sends it to the user along with the token. The client uses this key to encrypt the web server’s domain name using a strong symmetric-key cipher, such as the Advanced Encryption Standard (AES). This encrypted value is appended to the token. When the token is directed back to the repository from the web server, the repository attempts to decrypt this element using the key that is shared with the client. If the decrypted string does not match the domain of the web server as perceived by the repository, the token is deemed invalid. This prevents anyone who may obtain access to the token from impersonating the client. Again, more about the security aspect of this element will be discussed in Chapter 5. Expiration Timestamp. When important digital documents travel encrypted or signed through public domains, they are often accompanied by an expiration timestamp. This is employed as an automatic invalidation mechanism so that the creator can ensure that the public is in possession of the latest version. Adding a timestamp to the token is a precautionary measure to ensure that the token cannot be stored for a long period of time while attacks against it or the repository are levied. It limits the window of possibility of a successful attack to the lifespan of the token. 3.2.2 Protocol TrustBroker utilizes three communication channels (Figure 3.2). The first channel is for the user to request a token from the TrustBroker repository. Over the second channel the client sends his token and a request for a resource to the web server, and the web server responds to this request. The third channel occurs when the web server contacts the TrustBroker repository to request the user’s information. To protect against eavesdroppers, all channels operate over secure communication. links. The following sections describe these three communication channels. 18.104.22.168 Client – TrustBroker Repository Communication It has already been stated that a web interface needs to exist for the user to login and obtain a TrustBroker token. However, by forfeiting certain security and design goals, this does not need to be the case. Historically, the client computer has been quite vulnerable to attack, largely through the myriad of spyware that exists. To nullify this angle of attack on the TrustBroker system, a significant design goal is to remove all sensitive data from the client computer. However, if this were not important, or the threat could be dealt with another way, the username and password to the TrustBroker repository account could be stored on the client computer. The pair could either be stored in a password protected file or as browser settings. In this case, a mechanism could be built in which the client would simply click a button to retrieve a TrustBroker token. The user would not need to navigate to a login page or enter any information; a click of the button would suffice. A slight twist to this model is to store the username and password on an external device, such as a flash drive. This way, the username and password do not reside on the computer but are available when the device is connected. This eliminates part of the risk because the data is not always available for exploitation. Instead the data on the device becomes a target only when it is connected to the computer, which might be infrequent. In this way, the benefits of single-click token acquisition might be realized with reduced security risk. This research strives to minimize the risk of attack on the client computer. One way to satisfy this goal is to store all sensitive TrustBroker-related information in a secure online repository. Assuming that a web interface is used to obtain a token from the TrustBroker repository, the next design decision is to determine the mode of transport. One possibility would be for the client module to open a socket and attempt communication with the TrustBroker repository in a side band after the client has authenticated. The question then arises: who should attempt to make the connection? The client could try to connect to the TrustBroker repository or the repository could connect to the client. Allowing the repository to connect to the client may not be a good idea for several reasons. First, it opens an unnecessary socket on the client to the outside world, which is poor security practice and may make the client a viable target for attack. Secondly, there may be a firewall or private router that might block the connection attempt. Therefore, this does not seem like an appropriate approach. However, if the client attempted a second connection to the TrustBroker repository to request the token after the client had authenticated over the first connection, these problems are largely fixed. Indeed, this is a viable option, albeit somewhat complex. The last possible option would be to transport the token within the same HTTP session as the login. The following is a list of mechanisms through which the token might be sent, along with an explanation of their strengths and drawbacks. Browser Cookies. The token could be sent using the standard and well-used cookie protocol. Advantage. The advantage of this approach is that browsers already support cookies. This reduces the amount of coding needed to create the client and web server modules. Disadvantage. The cookie mechanism is designed to allow a web server to place a key-value pair on the client computer that will be made available only to the server that generated it when the browser returns. The transfer of cookies is transparent to the user. Since the design goals of TrustBroker are to transfer tokens to sites that did not issue them and to grant the user control of the distribution of the token, the cookie mechanism seems ill-suited for the task of transporting the token. Special header tags. A second way to transport the token is in special tags inside the HEAD portion of the document. The HEAD section is designed to include meta information about the page and is not displayed to the user. The token could be placed inside special TRUSTBROKER tags, or inside the standard META tags. Advantage. The major advantage of this approach is its simplicity. Also, it requires the client-side module to scan only the HEAD section of a web page, thus reducing the time required to discover a token. Disadvantage. No significant drawbacks to this approach are apparent. ### 22.214.171.124 Client - Web Server Communication When sensitive information is needed to complete a transaction, the client sends the token instead. There are two main approaches to accomplishing this. Either the web server could actively collect the token from the browser, or the web server could passively notify the browser that it is enabled to receive TrustBroker tokens. Allowing a web server to snatch a token without the user’s knowledge violates one of the basic design principles, namely that the ultimate control of the token should be governed by the user. This would require the user to authorize or reject every token request from the web server. However, another design goal is to remain unobtrusive to the user. Thus, allowing the web server to actively collect the token does not permit the client module to simultaneously adhere to both of the fundamental goals. of TrustBroker: usability and security. One succeeds at the expense of the other. The second approach is to create a notification mechanism that only informs the browser that the web server is capable of handling a TrustBroker token. The browser notifies the user in any number of ways, including methods that are unobtrusive and not restricted to yes/no dialog boxes. The token would be sent only when the user actively requests it. This approach seems to fulfill both of TrustBroker’s fundamental design principles: the user retains full control over the distribution of the token and yet is not constantly bothered by obtrusive dialog boxes. Given a passive notification method, the design of the notifier becomes important. The web server would need to embed the notifier into the returned web page and the client module would need to be actively looking for it. The following is a list of possible notification methods, with their advantages and drawbacks. Hidden form field. One way to notify the browser that the server could accept a token would be to generate a form that includes a special hidden field. This hidden field would not have an initial value, but would have a standardized name that indicates to the client module that it should include the token as the value of that hidden field when the user clicks the submit button. Advantage. One of the advantages of this approach is that it would require the client module to only scan web forms for the special hidden field. Since many pages do not have forms, the scanning would usually be very fast. Another nice aspect of this approach is that the client module would simply have to insert the value into the form. The basic functionality of web forms in browsers would do the rest and the token would be transmitted via the post data. Disadvantage. The biggest drawback to this approach is that the user may miss clues from the client module that his token is being requested. A malicious server could place a form with this special hidden field in every page it serves, trying to phish for any tokens it can acquire. The only way to stop this would be to generate increasingly obtrusive notification mechanisms. This interferes with the usability principle of the system. HTML Elements. Like receiving the token, the notification mechanism could be passively embedded in the HEAD section of a web page, in either special TRUSTBROKER tags or the standard META tags. Advantage. A strong advantage to this approach is that the client code could scan pages, notice when a web server is token-enabled, and notify the user in an unobtrusive fashion. No active response is necessary, and if the user ignores or misses a hint from the browser, there is no risk of accidentally sending his token. This allows the user to fully benefit from TrustBroker without being annoyed by obtrusive security measures. Disadvantage. There are no apparent disadvantages to this mechanism. ### 126.96.36.199 Web Server - TrustBroker Repository Communication The communication channel between the web server and the TrustBroker repository is the simplest because it involves no humans. Once this communication channel opens, the user has received a token and has willingly granted the web server authorization to use it in order to acquire his information. Even though the human usability of the system is largely irrelevant at this point because no human is involved in this transaction, there remain significant technical design challenges. The main challenge is the communication protocol used between the web server and the repository. It is not limited to HTTP as other communication channels are; any data transport protocol could be used. This raises the question of whether a proprietary protocol should be created specifically for this communication or if an existing, standardized protocol should be utilized. For the ease of implementation as well as for propriety’s sake, the responsible answer seems to be that if a standardized communication protocol exists that could satisfy the demands of a TrustBroker interaction, it should be used. While many approaches may suffice, messages formatted in XML and sent directly over the secure communication channel would provide a simple, flexible mechanism to transfer information. Another decision regarding the communication protocol between the web server and the repository is the delineation of messages that are required to complete a transaction. There are three classes of required messages: an information request message, the required trust negotiation messages, and an information response message. The final design element of this part of the protocol is the contextual language that is spoken. For example, when a web server requires the first name of the client, it must somehow encode this into the request so the repository can correctly extract the desired piece of data. Researchers have made strides in defining naming standards for commonly used pieces of information on the web [?]. One such set of standards is defined by the Electronic Commerce Markup Language [?], which standardizes the names of pieces of information that are commonly used in online commercial transactions. As the demand for transferring other kinds of information (such as medical and scholastic) increases, more standards will need to be defined and used. TrustBroker will need to adapt to these new standards as they are developed. Additionally, there may be cases where a user desires to add non-standardized fields to his repository of information, such as a username and password for a particular site. This would be highly beneficial against phishing attacks. An email that asks a user to login via a link in the email is suspicious in nature, and may be a sophisticated phishing attack. Under normal circumstances, a careful user would simply ignore the email or at least visit the site through normal methods (i.e., browser bookmark, typing the URL into the address bar, etc.). However, if the username and password were stored in the repository, the user could safely click the link from the web page. Instead of manually entering the login information, the user would initiate the TrustBroker system and let the remote site attempt to acquire the information from the repository. By allowing the repository to establish trust for him, a user could be protected against phishing attacks. 3.2.3 Data Obfuscation Since a typical TrustBroker repository stores sensitive information about many people, it is a rich target for attack. If an attacker compromises the database and gains access to the information, significant damage could be done. One method of mitigating this risk is to obfuscate the data stored in the database so that a successful attack on the repository’s database yields only encrypted, and therefore useless, data. Thus, the challenge is to grant only authorized people access to the data in a clear text form. 188.8.131.52 Protection from External Attacks Protection against external attacks could be implemented by storing the data in an encrypted form. When a new user registers their data with the repository, the data is sent in clear text (but over a TLS connection to prevent eavesdropping). Once the repository has the data, it generates a strong key, encrypts the data in the database, and sends the key back to the client. The repository does not store this The key is appended to the token when it is sent to a web server. Thus, if the server is able to acquire information from the repository, it can decrypt it using the key provided by the client. If it cannot gain access to the data, it is left with the key, but without access to the encrypted information. With this system, the data cannot be acquired directly from the repository’s database in clear text form. The key from the client is also necessary. This raises the required level of attack sophistication, as now an attacker must obtain unauthorized access to the information and acquire the appropriate user’s key. In addition, if an attacker (namely the web server, as it is often granted access to the token) was able to acquire the key from a token, and gain unauthorized access to the repository’s database, it could still obtain information for only a single user. The information of other users would remain secure because it is encrypted with unique keys. Of course, once the web server has a user’s key, it is no longer secret. To mitigate the risk caused by repeatedly distributing the key, the user would periodically renew the encrypted data with a new key. The downside of this approach is that the loss of the key on the client side means the data cannot be retrieved. This would require the client to re-register his information with the repository. Despite the fact that the key is being released to some web servers, it is still prudent to minimize its distribution. A second downside to this approach is that malicious web servers that are able to acquire the key may readily distribute it to other servers. Also, the fact that the client stores the key locally means that it is now an avenue for attack. While there are drawbacks to this approach, its goal is to protect personal information from direct attacks on the database and attacks on the repository service. To protect only against attacks on the database, any number of commercially encrypted databases could be employed. 184.108.40.206 Protection from the Repository Another way to encrypt the user’s data in the repository is for the client to generate its own key, and send only encrypted data to the repository when the client first registers. Thus, the repository never sees the data in clear text. The key would be sent in the token so the web server could decrypt the information, as in the previous example. In addition to the problems of the previous example, the drawback to this solution is that it begins to encroach upon the basic principle of usability. The user would need to be more heavily involved with the generation of the key and the encryption and transmission of the data. 220.127.116.11 Protecting Against External and Internal Threats To protect against both external and internal threats, this idea of data obfuscation could be carried one step further. As in the previous section, the data would be sent to the repository encrypted with a key generated and stored by the client. However, in this case, the client would not append this key to the token. Instead, when the server obtains the information from the repository, it would send the data back to the client for decryption. The client would decrypt the information with its secret key and return the data to the web server in clear text. This solution solves the key distribution problem as noted in previous solutions. However, the complexity of the communication is significantly increased. A further study of the benefits and risks of attempting such modifications to the proposed TrustBroker system is left to future work. 3.2.4 Repository Policy Agglomeration Each TrustBroker repository may allow separate pieces of a user’s information to be protected by individual policies. For example, it is likely that a person would want the repository to protect his credit card information more strictly than his first name. In this case, the policy protecting the release of his credit card information would require a higher level of trust than the policy governing the access to his first name. Thus, separate policies might be applied to disparate pieces of information. Therein lies a challenge for the TrustBroker repository to assimilate the information being requested and unlock only the pieces that the web server is authorized to see. Several options are available. First, the repository may send all of the policies that govern all of the requested data to the web server. The web server would look through the policies to determine which ones it thinks it can satisfy. This approach reduces the number of iterations needed to establish trust. However, it also requires the web server to employ more resources during this portion of the transaction because it must logically analyze the set of policies received from the repository. Another approach is to employ a linear method, wherein the policy that governs each piece of data is sent to the web server in turn. This allows the policy analysis to remain on the repository, but may cause significant delays in the transaction due to the number of rounds that may be required. In some instances, it may make sense to assume that policies can be ordered. In other words, the statement “the credential that satisfies policy A offers greater privacy assurance than the credential that satisfies policy B” may be possible for every policy available. When this is the case, several approaches seem reasonable. The first would be to challenge the web server with the protecting policies in descending order starting from the most stringent. If the web server could satisfy the most stringent policy, it could be inferred that it could satisfy the less stringent ones. Therefore, the trust negotiation session could stop and the web server would be authorized to see all the data requested. If it cannot satisfy that policy, the next easiest policy would be attempted. This would continue until the web server is able to satisfy the policy, at which time the trust negotiation would stop and the corresponding data would be delivered. The possible benefit is that only one round of trust negotiation would be necessary for servers that were able to satisfy the most stringent policy. However, the drawback would be that the trust negotiation session with web servers that possess no satisfactory credentials might take a long time. Thus, a second method would be the exact opposite: start the trust negotiation with the least strict policy, then iterate through consecutively stricter policies until the web server fails to satisfy it. The repository could safely assume that the web server wouldn’t be able to satisfy any others and could terminate the session and return the data tied to the policies that had been satisfied. The potential benefits and drawbacks are the opposite of the previous method. Detecting web servers that are not very trustworthy might occur very quickly. However, the trust negotiation session with web servers that are highly qualified to protect user information may take a long time. A happy medium might be to begin in the middle of the policies, with regard to their strictness. If the web server could satisfy that policy, stricter policies could be challenged. If not, less strict policies could be challenged until the web server finally satisfied one. This way, the worst-case time it would take to determine the level of privacy assurance that a web server could guarantee would be half the worst-case time of either of the previous two alternatives. A slight improvement might be made by performing the equivalent of a binary search through the policies. This would allow the repository to establish trust in logarithmic time. CHAPTER 3. SYSTEM REQUIREMENTS AND DESIGN Chapter 4 — Implementation In order to establish the practicality of the TrustBroker system, a proof-of-concept implementation was developed. This chapter presents the implementation’s details, especially as they relate to the design principles mentioned in the previous chapter. The explanation of the implementation is divided into the three core pieces of a TrustBroker system: the client-side module, the TrustBroker repository, and the web server. Although several trust establishment protocols are feasible for the web server and TrustBroker repository to establish trust, the implementation uses attribute-based trust negotiation as its underlying trust establishment mechanism. While the TrustBroker design calls for TLS sessions between each entity, they are not necessary to establish the feasibility of a commercial TrustBroker system. Therefore, they were not included in the implementation. ![Diagram](image) Figure 4.1: Overview of a TrustBroker transaction. The implementation supports the TrustBroker system architecture shown in Figure 4.1. First, the client logs in to the TrustBroker repository (1) and is issued a token (2). When the client requests a resource from a web server, the token is also sent (3). The web server forwards the token to the TrustBroker repository and requests the necessary information to complete the transaction (4). TrustBroker verifies whether the web server satisfies the policies governing the client’s information (5). If the web server satisfies the required policies, it is granted access to the requested data (6). The server asks the client for confirmation of the data and grants access to the resource (7). 4.1 Client-side Module The client module is an extension to the Mozilla Firefox web browser and provides two key functions. First, the user interacts with the system via a graphical interface that consists of a browser toolbar and a submenu added to the default browser tool menu. Second, the client module manages the acquisition, storage, and transmission of a token. 4.1.1 User Interface All of the client module functionality can be accessed via the tool menu (Figure 4.2). It includes the ability to request, send, delete, and view the contents of the token; adjust user settings and preferences; and see details about the installed TrustBroker client module. The first two items allow a user to request and send a token, respectively. The third item on the menu (“Delete Token”) allows the user to delete the current token. This option is valuable in situations where the client may want to explicitly remove the token from memory, such as when using a public computer. The “View Token” item allows the user to view the contents of the token, which may help the user Figure 4.2: TrustBroker tool menu. Figure 4.3: Client module About dialog box. determine if his current token has expired, but is provided mostly for development and debugging purposes. The fifth item allows the user to adjust user settings and preferences, which will be discussed shortly. The last item causes an informational dialog box to be displayed that has detailed information about the client module, such as the author, its version number, and where to go for updates (Figure 4.3). The second main graphical element of the client-side module is an optional toolbar, which duplicates functionality found in the tool menu (Figure 4.4). Its purpose is to provide quick access to the two most common user actions: acquiring the token and sending it. The “Acquire Token” button is used to request a token from the TrustBroker repository. When clicked, the browser navigates to the login page of the user’s repository. If the user already has a valid token, a dialog box will appear asking the user to confirm that he would like to revoke the current token and acquire a new one. This dialog box is necessary because the circumstances in which a new token is desirable over the current valid one are rare and it is likely that the user clicked the button by mistake or as a mechanism to check if he currently has a valid token. There is another aspect of the “Acquire Token” option that increases the system usability. Before the browser navigates to the login page of the user’s TrustBroker repository, it stores the current URL. After acquiring the token, the client module navigates the browser back to this URL. This allows the user to navigate to any page before acquiring a token, and he will be returned to that page directly after receiving it. Since most sites do not require personal information, sending a TrustBroker token to each server is undesirable. Because of this, the “Send Token” option is disabled by default (Figure 4.4(a)). However, when a site is TrustBroker-enabled (a description of how a site becomes TrustBroker-enabled is found later in this chapter), the “Send Token” button is enabled (Figure 4.4(b)). This allows the user to send his token instead of manually submitting personal information. Thus, when the user expects a site to be TrustBroker enabled, such as when filling out a web form, he may verify it by checking the status of the button. If enabled, and if he desires, he may click the “Send Token” button to send the token instead of entering data in the form. This method of transferring the token is both secure and easy. It grants the user strict control over when the token is sent, while remaining unobtrusive during normal Internet use. The fifth item on the menu opens a dialog box that allows the user to view and modify settings that determine the behavior of the client module (Figure 4.5). These settings are stored as part of Firefox’s browser settings and are loaded and stored like other user preferences. Figure 4.6 shows the TrustBroker-related settings displayed alongside other user-configurable settings. The `trustbroker.enabled` property allows the user to enable/disable the client module. When disabled, the client module becomes dormant, hides the toolbar, and does not respond to any TrustBroker-related event. Figure 4.5: Client module Settings dialog box. Figure 4.6: TrustBroker settings as part of web browser preferences. If the `trustbroker.acquireTokenOnLoad` property is set to true, the browser will immediately navigate to the TrustBroker repository’s login page when it loads. However, if the `trustbroker.promptAcquireOnLoad` property is true, the user will be asked to confirm his desire to obtain the token when the browser loads (Figure 4.7). At this point, the user will be able to cancel the automatic request for a token if he desires. The `trustbroker.showToolbar` property allows the user to determine if the toolbar should be visible. If the toolbar is hidden, the user can still interface with the TrustBroker system as before through the tool menu. The `trustbroker.tblogin` property stores the URL of the TrustBroker repository’s login page. When the user clicks the “Acquire Token” button on the toolbar or tool menu item, the client module directs the browser to the specified URL. If no URL is specified, a dialog box will request that the user enter a URL. Again, these settings are modified by the user from the settings dialog (Figure 4.5). These graphical elements of the client module combine to create the user interface to TrustBroker. It allows the user to remain in control of the token, yet is unobtrusive in that the user is not bothered by repeated pop-up boxes or security 4.1.2 Token Management The second aspect of the client-side module consists of its interface with the TrustBroker system. This entails being able to request, interpret, and send the token when needed. Each of these tasks will be explained in detail. 18.104.22.168 Requesting and Interpreting a Token A token request is always initiated by the user clicking the “Acquire Token” button on the toolbar (Figure 4.4) or by selecting the menu item from the tool menu (Figure 4.2). The only exception to this is when the user specifically asks the browser to request a token when it is first loaded (Figure 4.5). In any case, when this request is made the client-side module temporarily stores the current URL, including the query string if it exists, and navigates the browser to the TrustBroker login page as specified by the URL in the settings panel (Figure 4.5). From that point on, the client-side module scans the HEAD element of every page visited, looking for a token sent from the repository. After discovering the token, the module redirects the browser to the stored URL. The repository sends a token to the client by including a special META tag in the HEAD element of any web page (Figure 4.8). The token consists of a string concatenation of the five required elements listed in Section 3.2.2 delimited with the ‘@’ character. The secret key is the 128-bit key shared between the repository and the client, which is used to generate the sixth element of the token. When receiving this token, the module checks to make sure the domain sending the token is the same as the domain of the user’s repository as specified in the settings panel. This prohibits any arbitrary website from replacing the token. 22.214.171.124 Sending the Token The client module allows the token to be sent to any website that is token-enabled. In order to be token-enabled, a website must include a special META tag in the HEAD element of the web page (Figure 4.9). When the module detects this special tag, it enables the “Send Token” button on the toolbar (Figure 4.4) as well as the “Send Token” option in the tool menu (Figure 4.2). If a user desires to send the token, he simply clicks either of these two controls. The module encrypts the domain of the website, appends this encrypted value to the token, and sends the token to the site indicated in the value portion of the META tag attribute as a query string. After sending the token, the module waits for a response from the web server. The response consists of a URL that points to the location where the client may continue the transaction process or be notified of any problems that may have occurred. 4.2 TrustBroker Repository There are two pieces of the TrustBroker repository’s implementation. First, people use a web interface to register their information and request tokens. Second, web servers request/receive client information and perform trust negotiation through a remote server interface. 4.2.1 Web Interface for Users This portion of the implementation is straightforward. It consists of a few web pages specially created to support the TrustBroker system. The first is a login page. This page has a simple username and password form that allows a person to login and request a TrustBroker token. Also on the page is a link to the registration page where a user may register his information with the TrustBroker repository. The login attempt is sent through form data to the second web page, the login confirmation page. This page authenticates the username and password. If they are valid, it also generates a token and shared key unique to that user and token. The token and key are placed inside the special META tag in the HEAD element of the page that is sent back to the client (Figure 4.8). When the user desires to register, a form is provided so the user can enter all the personal information he wants to store and allows him to assign each piece of data a protection level. Each protection level is translated into a policy that governs access to the data. Thus, the higher the protection that a user demands on a piece of information, the stricter the policy is. This mechanism allows a user to determine how strictly each piece of information is protected. The current implementation stores the user’s information in clear text in the database. 4.2.2 Remote Interface for Web Servers The TrustBroker repository’s second interface consists of a simple TCP socket. This socket listens on an arbitrary port, chosen by the repository. This port is included in the token, to ensure the web server is aware of how to establish the connection. Through this interface, a series of messages are exchanged that lead up to and include the transfer of user information. The messages are formatted in XML for <?xml version="1.0" encoding="UTF-8"?> <trustBroker type="clientHelloMessage"> <token> TrustBrokerDomain@TrustBrokerPort@UserId@ExpirationDate@HMACOfToken@EncryptedWebServerDomain </token> <ECMLData type="Ecom.ShipTo.Postal.Name.First"></ECMLData> . . . <helomessage ...> ... standard trust negotiation client hello message ... </helomessage> </trustBroker> (a) Initial Request for Information / Trust Negotiation Client Hello Message <?xml version="1.0" encoding="UTF-8"?> <trustBroker type="serverHelloMessage"> <helomessage ...> ... standard trust negotiation server hello message ... </helomessage> </trustBroker> (b) Initial Request for Information / Trust Negotiation Server Hello Message <?xml version="1.0" encoding="UTF-8"?> <trustBroker type="tnTermination"> <ECMLData type="Ecom.ShipTo.Postal.Name.First">Michael</ECMLData> <... standard trust negotiation termination message ... </trustBroker> (c) Transfer of User Information / Trust Negotiation Termination Message Figure 4.10: Protocol messages between web server and TrustBroker repository easy parsing. There are four types of messages that are used to carry out this communication: Initial Request for Information / Client Trust Negotiation Hello. After receiving a token from the client, the web server will send this message to the repository to request the client’s information. The contents of this message include the token, a list of desired data items in Electronic Commerce Markup Language (ECML) notation, and the contents of a standard Trust Negotiation Client Hello message (Figure 4.10(a)). Server Trust Negotiation Hello. Upon receipt of the client hello message, the repository evaluates the validity of the token according to the following steps. 1. It computes the HMAC of the received token using the secret key created when the user obtained the token and compares it to the original HMAC generated at the same time. Since the secret key is required to produce a correct HMAC, a matching comparison guarantees the token’s authenticity (as long as the repository has safely protected the key). This check protects against modification attacks. 2. It decrypts the signed domain name with the shared key associated with the user indicated in the token and compares that name with the domain name of the server that is requesting the information. As long as DNS remains reliable, this domain name cannot be spoofed because TLS guarantees domain name authenticity. This comparison protects against impersonation attacks. After establishing the validity of the token, the repository evaluates the requested data items and determines all of the policies that govern their disclosure. In the event that the information is guarded by separate, but hierarchical policies (meaning they can be ordered in the level of protection they offer), the server determines which policy to try first and integrates that into the standard Trust Negotiation Server Hello message (4.10(b)), which is then sent to the web server. If the token is not valid, the server sends the Trust Negotiation Termination message indicating that it is not prepared to offer any information to the web server. Trust Negotiation Continuation. After the hello messages are exchanged, trust negotiation ensues. The repository determines how much to trust the web server according to the policies the server satisfies during trust negotiation. Trust negotiation is completed when the repository can satisfactorily determine how much to trust the web server. Trust Negotiation Termination / Transfer of User Information. Trust negotiation can be terminated at any time by either side. The repository terminates the negotiation when it discovers the level of trust that may be applied to the web server. The web server may terminate the negotiation if it becomes too lengthy, resource-intensive, or for any other reason. When either side decides to terminate trust negotiation, it sends this message to the other party. The Trust Negotiation Termination message contains the standard trust negotiation termination information. If the TrustBroker repository is terminating the communication, the requested data that has been unlocked via trust negotiation (Figure 4.10(c)) is also included. Requested data fields that are protected by policies that were not satisfied by the web server are included but left blank, indicating that the repository either does not have the requested data, or refuses to disclose it. The web server may notify the client if any necessary information could not be acquired or if any errors occurred. 4.3 Web Server There are three basic requirements that a web server needs to satisfy in order to interface with the TrustBroker system. First, it must notify the client module that a page is TrustBroker-enabled. Second, it must scan incoming resource requests for TrustBroker tokens. Third, it must be able to negotiate trust with a web server to acquire a user’s information. 4.3.1 Page-enabling As mentioned previously, the mechanism for enabling a web page is to include a special META tag in the HEAD element of the page (Figure 4.9). This tag indicates to the client module that the current site is capable of handling a TrustBroker token instead of any requested information, and provides the URL where the token should be sent. 4.3.2 Handling the Token When processing a request, the token-enabled page checks to see which kind of data has been supplied. If standard form data has been sent, the transaction proceeds as normal. If the request includes a TrustBroker token instead, the server uses this token to request the desired information from the TrustBroker repository. 4.3.3 Acquiring the User’s Information from the Repository If a resource request includes a TrustBroker token in place of the expected form data, the web server must use the token to acquire the user’s information from the TrustBroker repository. To do this, it must make the appropriate request and establish itself to the repository as a credible server. Trust negotiation is used as the method of establishing this credibility. The current implementation of TrustBuilder, a package developed by the Internet Security Research lab at Brigham Young University to perform core trust negotiation tasks, is written in the Java language. To leverage the benefits of a homogenous solution, the web pages that handle TrustBroker tokens are also written in Java, in the form of JavaServer Pages (JSP). Thus, the Java packages that are required to support trust negotiation (including TrustBuilder) are loaded directly into the environment where the JSPs are compiled and executed (Apache’s Tomcat, in this case). When a token-enabled web page discovers a token in the request, it parses the token and makes a TCP connection to the specified TrustBroker repository on the specified port. From there, it sends the Initial Request for Information / Client Trust Negotiation Hello message (Figure 4.10(a)) to the repository, thus initiating the communication between the web server and the repository. The repository challenges the credibility of the server by asking it to satisfy the security policies governing the requested data. If the repository determines that it can trust the web server, it releases the desired information. Once the Trust Negotiation Termination / Transfer of User Information message (Figure 4.10(c)) has been received, the JSP closes the connection to the repository and analyzes the client’s information it received. At that point, it may ask the client to confirm the data or provide more information. CHAPTER 4. IMPLEMENTATION Chapter 5 — Threat Analysis In the search for usability, the security of a system must not be compromised in any way. Thus, a thorough security analysis of TrustBroker is imperative. In most cases, a security model cannot formally be proven correct. Instead, it is analyzed with respect to the known attacks and vulnerabilities of other systems as applied to the new system. The longer it withstands this scrutiny, the more secure it is considered. This chapter presents a threat analysis of TrustBroker. The client, web server, and TrustBroker repository are each analyzed for possible infractions in correct security principles. There are numerous well-known security threats that are not addressed by TrustBroker. Although some of these attacks are analyzed, the focus of this chapter is on threats introduced by TrustBroker. 5.1 TrustBroker Token The token issued by the TrustBroker repository is a target for attack. It is a key element to TrustBroker’s security, and is the focus of the threat analysis. There are three classes of attack that could be levied against the token: impersonation, replay, and modification. 5.1.1 Impersonation An impersonation attack occurs when an attacker convincingly assumes the identity or characteristics of another person. Consider the steps of the impersonation attack illustrated in Figure 5.1. The client requests a resource from a malicious web server posing as a legitimate service provider, and sends his TrustBroker token instead of personal data (A). The normal behavior of a legitimate web server would be to send this token to the TrustBroker repository and request the client’s information (B). Instead, the malicious web server requests a resource from a legitimate web server and attempts to impersonate the client by sending the client’s token (C). The legitimate web server sends the token to the designated TrustBroker repository and requests the client’s information (D). The TrustBroker repository authenticates the legitimate web server and releases the information (E). The malicious web server gains access to the client’s information and is granted services from the legitimate web server using the client’s identity (F). However, the TrustBroker system design prevents this type of attack. Consider the sixth element of a TrustBroker token (Figure 5.2). It consists of the domain of the intended token recipient encrypted with a secret key shared only between the TrustBroker repository and the client. When a TrustBroker repository receives a token and request for information it decrypts this element and compares it to the domain of the web server making the request. If they do not match, the request is Figure 5.2: The sixth element of the TrustBroker token is the domain of the intended token recipient encrypted with the secret key shared between the client and TrustBroker repository. denied. A third party attempting to impersonate a client will not be able to generate the correct value because it does not have access to the shared secret key. As long as the secret key is protected, only the client will be able to correctly generate this element. Therefore, impersonation attacks are not possible. 5.1.2 Replay A replay attack is successful if the token is captured by a malicious entity and successfully re-used without the consent of the client. Since all TrustBroker-related communication occurs over a secure channel, the only substantial threat is a malicious, but trusted, web server. If the web server is malicious, but cannot satisfy the policies governing access to the client’s data, the token is of little value anyway and no replay attack is possible. If the web server is able to satisfy the policies that govern access to the client’s personal information, it is unlikely to be malicious because the credentials necessary for unlocking the user’s information would not be granted to a server with a history of malicious behavior. 5.1.3 Modification A modification attack is successful if a malicious entity is able to modify the contents of a TrustBroker token such that a different person’s information is accessed. This attack could be attempted by either a malicious client or web server. The fifth field (Figure 5.3) of the TrustBroker token prevents such an attack. This authenticator field contains either a digital signature or an HMAC (i.e., keyed hash) that only the repository is able to generate. The repository can detect when a token has been fabricated or modified by recreating this authenticator from the contents of the token and checking its validity. 5.2 Pre-existing Threats There are numerous well-known attacks that can be levied against many systems, including TrustBroker. This section briefly discusses some of them that are most relevant to TrustBroker. Defective software. Experience shows that the design of a secure system can be rendered ineffective through faulty implementation. Faulty software on any of the entities involved could be exploited to break the TrustBroker system, either by gaining access to a user’s information or by obtaining and abusing an otherwise valid token. This is especially dangerous on the client computer and the repository. The greatest protection against such problems is to use reputable software and keep up-to-date on software patches. Still, there is no way to guarantee that the software is free from error. Spyware and other malicious software. Malicious software running on any of the three entities could be dangerous. The client could give up the token, secret shared key, username and password of the user with a simple keystroke logger or Trojan horse. The web server and repository could both disclose the sensitive information of the client. However, the implementation of standard security practices (such as firewall, antivirus, and intrusion detection solutions) minimizes this threat. Malicious Insider. The problem of trusted, yet malicious, employees is well-known and difficult to solve. Often the measures taken to solve this problem collide head-on with the privacy and productivity of the employee. A malicious employee on the web server or the repository could gain access to user information with the intention of abusing it. With regard to TrustBroker, a possible solution to the malicious insider is to obfuscate the data on the repository, as discussed in Section 3.2.3. Such an effort would help ensure that the repository does not provide information to attackers, either external or internal. Public Key Infrastructure. Numerous vulnerabilities exist in any Public Key Infrastructure (PKI) [?]. To the extent that TrustBroker relies on a PKI, problems such as private key compromise and faulty certificate generation might reduce the security of TrustBroker. This could become a problem with regard to the TLS sessions that exist to protect the transfer of information from eavesdroppers. It could also become a problem when an attribute-based form of trust establishment (such as trust negotiation) is used to build trust between entities. In either case, a server might be able to impersonate the identity and/or attributes of another server if the PKI model is compromised. Distributed Denial of Service. Distributed Denial of Service (DDoS) attacks are among the most common form of successful attack against online entities. They certainly are a viable form of attack in the TrustBroker system. They could target a particular website (as the majority do now) but a potentially more debilitating attack would be against a specific repository. If the repository became inoperable, the advantages of TrustBroker would become moot. The success of website-based DDoS attacks is largely due to the fact that web servers listen on a standard port (port 80). Since the TrustBroker system allows the repository to select any port to listen on, this threat is greatly reduced. CHAPTER 5. THREAT ANALYSIS Chapter 6 — Conclusion As the Internet continues to grow in the number of available services, more people are participating in online activities. Unfortunately, the Internet can be a hostile environment, leaving the average user without help in protecting sensitive information. In many cases, the choice is to risk the possible abuse of this information or simply not participate. Similarly, phishing attacks are becoming more sophisticated, sometimes fooling even savvy users. Additionally, the transfer of information is often a tedious task. Unlike physical stores where a user is required to divulge minimal information when making a purchase, online stores often require the user to fill out extensive forms that contain information about identity, physical location, and preferences. TrustBroker is a system that seeks to alleviate these problems simultaneously. It is a system that utilizes a trust establishment mechanism to help the user gain confidence in the trustworthiness of a remote server, and automatically transfers the user’s required information once trust has been established. TrustBroker allows a user to obtain a token from a trusted online repository that stores his personal information. When online services request his personal information to complete a desired transaction, he submits his token instead. The server attempts to exchange this token for his information from the online repository. Before releasing the information to the server, the repository will attempt to establish trust that the server will not abuse this information. Once this trust has been established, the user’s information is released to the server. The average Internet user should have greater confidence while participating in online transactions involving TrustBroker. A user stores his sensitive information in a TrustBroker repository, allowing him to benefit from the trust establishment mechanism without being required to understand its details. Also, he is generally not required to submit any information to the server, as it can acquire the information it needs from the repository if it demonstrates that it is trustworthy. This research attempts to simultaneously support the seemingly contradicting design goals of security and usability. It presents design alternatives of a successful TrustBroker system, and analyzes the pros and cons associated with each approach. The implementation presented in this research demonstrates the practicality of TrustBroker. It uses trust negotiation as the mechanism to establish trust, and a browser extension that allows the user to interface with the system. TrustBroker remains unobtrusive while protecting the information of the user and allow the user to remain in control of each transaction. TrustBroker bolsters the necessary confidence to participate in new and exciting online activities by establishing trust on their behalf, and automatically transferring the necessary information when required. Additionally, this is all done with minimal user intervention, which allows the average user to participate without being required to learn the details of difficult and complex trust establishment mechanisms. Chapter 7 — Future Work TrustBroker establishes a framework for extensive future research. Much of this research has been mentioned in this paper. However, there remains significant areas of research that would serve to expand and strengthen TrustBroker. The current design and implementation of TrustBroker utilizes trust negotiation as the trust establishment mechanism because it works well in open systems, such as the Internet. However, the trust establishment mechanism is independent of the TrustBroker framework, and may be interchanged with other trust establishment mechanisms without affecting the rest of the system. An interesting exploration would be to consider the benefits and drawbacks of employing other trust establishment mechanisms. Rating and reputation systems are promising prospects. Two other research topics that are currently being pursued in the Internet Security Research Lab at Brigham Young University could be useful extensions to TrustBroker: digital receipts and context-sensitive trust negotiation. Digital receipts are the digital equivalent of a receipt obtained as a record of transaction from a physical store and serve as proof of a transaction conducted online. Considering that web servers contact TrustBroker repositories directly when requesting a user’s information, digital receipts could strengthen TrustBroker greatly. In the current system, a user never knows for sure what parts of his information are requested by a web server. Also, he never knows for sure what parts of his information were released by the repository. It may be that the web server requests and is granted access to much more of his information than is required to complete the transaction. Digital receipts could provide an audit mechanism to notify the user whenever his information was requested and/or transferred, in addition to which server requested the information, and what credentials were used to satisfy the access control policies. This would allow the user to be aware of how his information is being distributed. Second, the idea of context in trust negotiation is becoming a significant concern. The problems lie in a malicious remote entity’s attempt to either phish for another’s credentials or create a DoS attack. The first attack could be levied by submitting irrelevant policies in the attempt to collect information about which credentials the entity owns. The second attack could be effected by indefinitely continuing rounds of negotiation with irrelevant policies and credentials, thus consuming the entity’s resources. Either attack could be avoided if the server understands the context of the request for information and matches it with the context of the policies and credentials. If a server receives policies or credentials that are out of context, they would simply be ignored. Although this might lead to a termination of the trust negotiation, it would help identify errors in policies and protect against malicious attempts to gather information or cause a DDoS situation. Applying context to the trust negotiation that ensues between the web server and repository would strengthen TrustBroker. It would help ensure that the repository divulges only the information relevant to the current transaction. [1] D. Eastlake and T. Goldstein. ECML v1.1: Field specifications for e-commerce. IETF Request for Comments: 3106, April 2001. [2] Carl Ellison and Bruce Schneier. Ten risks of pki: What you’re not being told about public key infrastructure. Computer Security Journal, XVI(1), 2000. [3] Jennifer Golbeck and James Hendler. Inferring reputation on the semantic web. In Proceedings of the 13th WWW Conference, New York, NY USA, 2004. ACM Press. [4] A. Josang. The right type of trust for distributed systems. In Proceedings of the 1996 New Security Paradigms Workshop, New York, NY, USA, 1996. ACM Press. [5] Michael Koch. Global identity management to boost personalization. In Proceedings of the Ninth Research Symposium on Emerging Electronic Markets, Basel, Switzerland, 2002. Institute for Business Economics, University of Applied Sciences Basel. [6] Karl Krukow and Mogens Nielsen. Towards a formal notion of trust. In Proceedings of the Fifth ACM SIGPLAN International Conference on Principles and Practices of Declarative Programming, New York, NY, USA, 2003. ACM Press. [7] Microsoft. Microsoft .net passport for businesses. http://www.microsoft.com/net/services/passport/business.asp, March 2003. [8] Rob Murawski. Centralized directory services and accounts management project. In Proceedings of the 28th SIG UCCS Conference on User Services, New York, NY, USA, 2000. ACM Press. [9] Birgit Pfitzmann and Michael Waidner. Federated identity-management protocols. In Proceedings of the 11th Cambridge International Workshop on Security Protocols, Berlin, Germany, 2004. Springer-Verlag. [10] Liberty Alliance Project. Liberty alliance architecture overview version 1.1. http://www.projectliberty.org/specs/liberty-architecture-overview-v1.1.pdf, January 2003. [11] R. Guha. Open rating systems. http://tap.stanford.edu/wot.pdf, 2003. [12] Matthew Richardson, Rakesh Agrawal, and Pedro Domingos. Trust management on the semantic web. In Proceedings of the Second International Semantic Web Conference, Berlin, Germany, 2003. Springer-Verlag. [13] Slashdot. Slashdot meta-moderation. http://slashdot.org/faq/metamod.shtml, May 2003. [14] Slashdot. Slashdot comments and moderation. http://slashdot.org/faq/commod.shtml, June 2004. [15] Kerry Taylor and James Murty. Implementing role based access control for federated information systems on the web. In C. Johnson, P. Montague, and C. Steketee, editors, Conferences in Research and Practice in Information Technology, Vol. 21, Adelaide, Australia, 2003. Australasian Information Security Workshop 2003, Australian Computer Society, Inc. [16] Marjorie Templeton, Herbert Henley, Edward Maros, and Darrel J. Van Buer. Interviso: Dealing with the complexity of federated database access. In The VLDB Journal - The International Journal on Very Large Data Bases, Volume 4, Issue 2, pages 287–318, Secaucus, NJ, USA, April 1995. Springer-Verlag New York, Inc. [17] W3C. The semantic web. http://www.w3.org/2001/sw/, July 2004. [18] W. Winsborough, K. Seamons, and V. Jones. Negotiating disclosure of sensitive credentials. In Proceedings of the Second Conference on Security in Communication Networks, Amalfi, Italy, 1999.
By-Laws Of Local 194 International Federation Of Professional And Technical Engineers AFL/CIO, CLC Article I- Name 1. This organization shall be known as the New Jersey Turnpike Employee’s Union, Local 194, International Federation of Professional and Technical Engineers, AFL/CIO. Article II- Jurisdiction 2. The jurisdiction of this Local shall include all (non-supervisory) employees of the New Jersey Turnpike Authority, and any subsequent group(s) organized by the Local. 2.1. The Local shall consist of two Divisions: Regular, which covers all full-time employees and Auxiliary, which covers all part-time employees. In the event the Local organizes private sector workers, a third Division will be established in for that unit. Article III- Objectives 3. The objectives of this Local shall be found in the International Constitution and shall unite into one labor organization all workers eligible for membership, to secure improved wages, hours, working conditions, and other economic advantage for the members through collective action, through advancement of our standing in the community, and in the labor movement, to provide educational advancement and training for the officers, employees and members, to safeguard and promote the principle of free collective bargaining, the rights of workers and the security and welfare of all the people, and to protect and preserve the Union in the performance of its obligations. Article IV- Eligibility 4. Eligibility to membership in this Local shall be all (non-supervisory) current and former employees and associates of the New Jersey Turnpike Authority, and any subsequent group(s) organized by the Local. Article V- Officers 5. The Officers of this Local shall consist of a President, First Vice-President, Second Vice-President, Third Vice-President, Secretary, Secretary-Treasurer, Sergeant-at-Arms. These Officers together with the Tolls, Maintenance, and Office Grievance Chairpersons shall constitute the Executive Board of this Local. 5.1. Officers and all Executive Board members may be removed from office as provided for under Article 26 of these By-laws, except the appointed positions may be removed by the President with the advice and consent of the Elected Officers of the Executive Board. 5.1a All members of this local holding an elected or appointed position are required to attend all Membership Meetings scheduled by the Local, such as General Membership, Executive Council, Executive Board and Chapter Meetings unless excused by the (Local Union Excuse Committee) President for good cause. Failure of any elected or appointed official to comply with the above attendance rules may result in automatic removal from their respective office or position and they shall not be permitted to run or hold any position for the balance of the term (except as delegate to the Convention). Any Member removed from office under this provision may appeal to the General Membership in January or June providing there is a quorum of 10% of the Membership. Article VI- President 6. It shall be the duty of the President to preside at meetings of the Local, of the Executive Board, and of the Executive Council; to preserve order therein and enforce the International Constitution and the Local By-Laws; to see that all Officers perform their respective duties and to appoint committees. The President shall also have the right to serve on all committees by virtue of his/her office. 6.1. The President shall decide all questions of order subject to an appeal of the Membership; shall have the right to vote in the election of Officers; shall cast the deciding vote when a tie occurs in any question; shall announce the results of all votes and enforce all fines and penalties; and shall have the power to call special meetings when he/she deems it necessary or when requested in writing by 20% of the Membership for a Membership Meeting, 20% of the Executive Council for the Executive Council Meeting, or three (3) elected Officers of the Local for an Executive Board Meeting. 6.2. The President, or in his/her absence the Vice-President, together with the Secretary-Treasurer, shall sign all checks on bank accounts and, together with the Secretary-Treasurer, be the only representatives of this Local to handle and make payments from the funds of this Local. The President shall have the authority to interpret these By-Laws and to decide questions of law thereunder between meetings of the Executive Board and Executive Council. 6.3 In those years where either a President and/or Secretary-Treasurer are not re-elected, no checks shall be disbursed the following year until the first Friday until the newly-elected officers are installed. 4. The President shall have supervision of the conduct and affairs of the Union, its Officers, and employees, including without limitation, the right to appoint Members to the Union to chair meetings and preserve order therein. The President shall have the Authority to interpret these By-Laws and to decide questions of law thereunder between meetings of the Executive Board and Executive Council. Article VII- Vice-Presidents 7. It shall be the duty of the First Vice-President to preside at Union Meetings in the absence of the President, and of the Second Vice-President to preside in the absence of both the President and First Vice-President and the Third Vice-President to preside in the absence of all three. The First Vice-President, Second Vice-President and Third Vice-President shall perform such other duties and render such assistance as may be directed by the President. 8. The Secretary-Treasurer shall perform all duties imposed upon him/her by the International Constitution. He/she shall make a quarterly report of the financial standing of the Local and shall keep itemized records showing the source of monies received and to whom monies were disbursed; he/she shall keep records, vouchers, books, and accounts to verify the correctness of any report. 8.1. The Secretary-Treasurer shall upon request make available a copy for inspection of any annual report to any Member. Upon request made in writing setting forth good and sufficient cause for requesting examination of any books, records, and accounts necessary to verify any report filed in compliance with any statutory requirement. He/She shall make arrangements to have the records desired available for inspection at a convenient time and place. In the event that he/she believes that just cause does not exist for the request made, the Member shall have the right to appeal the matter within ten (10) days in writing to the Executive Board, mailed to the Union’s current address. 8.2. The Secretary-Treasurer shall receive all monies paid the Local and place same in the bank in the name of the Local except when order or directed otherwise by the executive Board. 8.3. The Secretary-Treasurer has no authority to proceed with the writing of a check until the required signed voucher has been received, and neither the President or the Secretary-Treasurer should sign a check which is not accompanied by a properly executed, signed voucher. It is the duty of the Secretary-Treasurer to question any out of the ordinary expense and to make sure all receipts tally with the total on the voucher. The only exception to this rule requiring a signed voucher are to pay the bills, obligations, and indebtedness of this Local Union. Under no circumstances shall a check be made out to “Cash” or any Officer or Representative of this local other than expenses mandated by these By-Laws. No Officer of this Local shall have the authority to issue credit cards to employees of this Union. 8.4. Whenever a Secretary-Treasurer’s term of office shall expire or otherwise be terminated, he/she shall give to the successor all papers, documents, records, vouchers, reports, money and other union property that may have been entrusted to him/her by virtue of their office. Such records, etc. shall be retained for a period of six (6) years. 8.5 The Secretary-Treasurer shall, at the end of each calendar year, submit all necessary financial records to an auditor selected by the Executive Board for that year, and said auditor shall be required to return such records together with an audit and financial statement. Article IX- Secretary 9. It shall be the duty of the Secretary to attend meetings of the Local, the Executive Board, and the executive Council, to keep minutes of the proceedings (which need not be verbatim), to keep a record of the names of the Members comprising each committee, to handle the correspondence of the Local where required, and to maintain an up-to-date list of the names and addresses of the Membership of the Local, said list not to be opened to inspection by any Members except by a bonafide candidate for office in connection with an election under these By-Laws. Article X- Sergeant-At-Arms 10. The Sergeant-at-Arms shall attend all meetings of the Local, the Executive Board and the Executive Council, and shall assist the Presiding Officer in maintaining order. He/She shall perform such duties and render such assistance as may be directed by the Presiding Officer. The Sergeant-at-Arms will take a roll call vote at all Executive Council, and shall have the authority to challenge or inspect any Member’s eligibility to attend any meeting of the Local. 10.1 Grievance Chairpersons shall attend all meetings of the Local, the Executive Board, and the Executive Council. Each Grievance Chairperson shall be responsible for the pursuit of grievances and the defense of disciplinary actions in his/her group; Tolls, Maintenance, or Office, and each may assist the other. Grievance Chairpersons shall be responsible for the co-ordination of all union activities between the Local and the Stewards. Grievance Chairpersons shall perform such other duties and render such assistance as may be directed by the President. Grievance Chairpersons will be appointed for one (1) year and reappointed on a yearly basis with the advice and consent of the Executive Board. Article XI- Grievance Chairpersons 11. Grievance Chairpersons shall attend all meetings of the Local, the Executive Board, and the Executive Council. Each Grievance Chairperson shall be responsible for the pursuit of grievances and the defense of disciplinary actions in his/her group: Tolls, Maintenance, and Office, and each may assist the other. Grievance Chairpersons shall be responsible for the coordination of activities between the Local and the Stewards. Grievance Chairpersons shall perform such duties and render such assistance as may be directed by the President. Grievance Chairpersons will be appointed for one (1) year and re-appointed on a yearly basis with the advice and consent of the Executive Board. Each Grievance Chairperson will keep a detailed record of their activities and turn a copy into the Union President. Grievance Chairpersons will only be appointed if he/she is a member in good standing. Article XII- Chapter Officers 12. The duties of Chapter Officers shall be consistent with the duties of each office enumerated and defined above except that said responsibilities shall apply only to the specific Chapter in which each Officer is elected. XIII- Executive Board 13. Except as may be otherwise provided in these By-Laws, the Local Executive Board is authorized and empowered to manage, invest, expend, contribute, use and acquire Local Union funds and property in the pursuit and accomplishment of the objectives set forth in the Constitution of the International Union and these By-Laws. 13.1 Union Funds are held in sacred trust for the benefit of the Membership. The Membership is entitled to assurance that Union funds are not dissipated and are spent for proper purposes. The Membership is also entitled to reasonably informed as to how Union funds are invested and used. Local 194 shall not permit any of the Union’s funds to be invested in a manner which results in personal profit or advantage or any Officer or Representative of the Local. The Local shall not make loans to its Officers, Representatives Employees, or Members of their families for advancing the private business of such person(s). 13.2 Any disbursements of money for any reason other than monies necessary to pay bills, obligations, and indebtedness of the Local must be approved by the Executive Board. 13.3 The Executive Board is hereby empowered, in addition to other such powers as are set forth herein or conferred by law, to: a. Provide for the salaries, allowances, direct and indirect disbursements, expenses, and reimbursement of expenses of Officers, agents and employees; b. Provide for the employment and payment of attorneys, accountants, and other such special or expert services as may be required by the organization; c. On behalf of the Local, its Officers, and employees or members, to initiate, defend, compromise, settle, arbitrate, or release or to pay the expenses and costs of any legal proceedings, desirable to protect, preserve, or advance the interest of the organization; d. Transact all business and manage and direct the affairs of the Local between Executive Council meetings and membership meetings; e. Be the Negotiating Committee on all Agreements between the Union and the Employer except those involving Auxiliary Members; f. Have full participation through discussion and vote on any agreements that affect the membership; g. No member of the Executive Board, consultant or business manager can serve on the Labor Relations Committee due to the appearance of conflict of interest. Article XIV-Executive Council 14. Except as may otherwise be provided in these By-Laws, the Local Executive Council shall do all acts, subject to the ratification of the membership whether or not expressly authorized herein, which the Council may deem necessary or proper for the protection of the property of the Local and for the benefit for the organization and members. 14.1 The Executive Council shall pass on all agreements between the Local and the Employer except those involving Auxiliary members prior to recommendation to the Membership. 14.2 The Executive Council shall have the power by majority vote, to over-rule any act of the president or other Local Officers, or Chapter, or Chapter Officers or the Executive Board, except those involving Auxiliary members, subject to appeal to the membership; 14.3 The Executive Council shall meet at least twice a year, and not more than sixth (6) months from the date of the last meeting with a two (2) weeks grace period, at the call of the President, or at the request, in writing, of 20% of its members. Article XV- Expense Reimbursement 15. Recognizing that the Officers of this Local must participate in cultural, civic, political, fraternal and educational activities in addition to their specific duties provided in the Constitution and By-Laws, that such activities benefit the organization and the members; that the time spent in such activities is unpredictable; such officers and employees may be granted allowances in such amount (daily, weekly, or monthly) as the Executive Board may determine. 15.1 On all assignments which require travel outside the normal work area, Officers and employees may receive a travel and expense allowance in such amount as is reasonable and necessary. 15.2 In addition to the allowances set forth above, all Officers and employees may be reimbursed for all expenses incurred in connection with their activities. 15.3 When a representative of the organization is engaged in activities in the interest of or for the benefit of the organization and its members, he/she shall be reimbursed for expenses incurred provided he/she presents the Secretary-Treasurer a satisfactory voucher for such expenses. Article XVI- Use of Automobile 16. The Local shall provide its representatives with automobiles or in lieu thereof, shall pay mileage allowance in such amount or at such rate as shall be determined by the Executive Board. In such instances where the Local provides an automobile, it is recognized that such Officers or employees are responsible for its safekeeping at all times. Accordingly, for the convenience of the Union and as partial compensation for such additional responsibilities, such Officers and employees shall be permitted private use of such car on a ‘round-the-clock’ basis, including such private use when the car is not required on Union business. Article XVII- Nominations and Elections 17. Nominations of Local Officers shall be held by mail with nomination forms to be mailed to each member at their last known address on or before the 15th of September in election years. Nomination forms shall be returnable on or before September 30th. Beginning with the election of Officers in 2006, elected Executive Board Officers will serve three (3) year terms. 17.1 A nomination for President shall also contain nominations for First Vice-President, Second Vice-President, and Third Vice-President. They will be placed in such order as the nominator wishes them to serve. A member can be a candidate for only one office. 17.2 The Nomination Form and Ballot shall, after the word “President” read: “/Delegate to the (-) IFPTE Convention” And after the word “Secretary-Treasurer” read: “/Delegate to the (-) IFPTE Convention.” In parenthesis will be placed the number of the next upcoming Convention. For example, the 2012 will have the “57th” where the parenthesis are. Each subsequent election will contain the number of the subsequent IFPTE Convention. 17.3 Persons being nominated for office must sign their nomination form. Any person signing to more than one office shall be declared not to be a candidate for any office. Persons nominated shall have until October 3rd to withdraw as a candidate. Should a person or two persons withdraw from a grouping of President and Vice-Presidents, the remaining candidate(s) in the same grouping shall have until October 5th to submit the name(s) of a substitute or substitutes who must be otherwise qualified according to these By-Laws. 17.4 All mail ballot elections as mandated by the By-Laws shall be conducted by a certified, independent outside agency or firm such as the Honest Ballot Association. 17.5 Official ballots shall be mailed to each member at the last known address no later than October 15th in election years and shall be returnable on or before October 31st. There shall be a period of fifteen (15) calendar days from the date of the mailing to the date of the return. No one has the Authority to collect or mail ballots for any other member of the Local. 17.6 A postage-paid return-addressed envelope shall accompany each Official Ballot. 17.7 Notice of Rules for Nomination must accompany each nomination form and Notice of Rules of Election must accompany each Official Ballot. In addition, Nomination of Officers must be announced in the Local 194 newspaper. Voting shall be by Secret Ballot and members shall not be required to mark or identify, nor shall the Official Ballot contain any indentifying marks, and any such ballots shall be void. 17.8 Every member in good standing, by payment of his/her dues, and who has been in such continuous good standing for each month in the nine (9) months period immediately prior to nominations, shall be eligible to hold office if that person is otherwise qualified under the International Constitution and these By-Laws. Every member in good standing as of October 1st of the Election year shall be eligible and entitled to vote. 17.9 Persons who leave the Bargaining Unit for a period in excess of six (6) months or longer shall not be eligible to hold any office until their return to the Bargaining Unit for a period of three (3) years. 17.10 Newly-elected Officers shall take office on the first Friday of January following their election. 17.11 Chapter Officers shall be nominated and elected at regular Chapter Meetings in November of election years, in the following order: President, Vice-President, Secretary, and Sergeant-at-Arms. 17.12 Newly-elected Chapter Officers shall be installed in advance of the January Executive Council meeting following their election. 17.13 In the event of a vacancy occurring in the office of President, the First Vice-President shall succeed, the Second Vice-President shall become the First Vice-President, and the Third Vice-President shall become the Second Vice-President. In the event of a vacancy occurring in the office of First Vice-President, the Second Vice-President shall succeed to First Vice-President, and the Third Vice-President shall succeed to Second Vice-President. In the event of a vacancy occurring in the office of Second Vice-President the Third Vice-President shall succeed. In the event of a vacancy occurring in the office of Third Vice-President, Secretary-Treasurer, Secretary, or Sergeant-at-Arms, such vacancy shall be filled by nomination at the next General Membership Meeting followed by a mail ballot election. A notice of vacancy containing the date, time, and location of the meeting must be published in advance of the meeting. 17.14 In the event of a vacancy, the President may make an interim appointment in order to carry out the duties and responsibilities of the vacant office but in no such event shall such appointee have a vote on the Executive Board, Executive Council, or Negotiating Committee by reason of such appointment. 17.15 Auxiliary Division (Part-time) members shall nominate and elect an Auxiliary Area Representative from the Auxiliary Division at each Chapter of the Operating Unit. The Auxiliary Area Representatives, along with the Union President and Auxiliary Division (Part-time) Vice-President will serve as the Negotiating Committee for the Auxiliary Unit. 17.16 Any member convicted of misrepresenting returns, altering, mutilating, or destroying deposited ballots, voting fraudulently or of intimidating others by threats or otherwise interfering with a member in the exercise of his/her rights to cast a ballot in a Local Union election shall be punished in accordance with the Trial procedures outlined in the International Constitution and these By-Laws. The member so convicted shall be disqualified from either appointed or elected office within the jurisdiction of this Local. Article XVIII- DELEGATES 18. Delegates to Convention shall be elected at the first General Membership Meeting following the election of Officers. 18.1 The President shall be the First Delegate and the Secretary-Treasurer shall be the Second Delegate. The Business Manager, if elected at the General Membership Meeting, shall be the Third Delegate. 18.2 Six (6) additional Delegates and three (3) Alternates shall be elected. The Office unit member receiving the highest number of votes will be the fourth Delegate, or Third Delegate if the Business Manager is not elected. The member receiving the highest number of votes will be the Fifth Delegate, and the member receiving the second highest amount of votes will be the Sixth Delegate, etc. 18.3 The number of Delegates shall be determined by the number of Delegates who will attend the Conventions. Attendance at any Convention will be in rank order according to the number of votes each Delegate received. 18.4 Protests and Appeals concerning the rulings of eligibility of voters and the conduct and validity of the election shall be filed within four (4) days with the President who, or his designee, shall decide such protests or appeals within (7) days. Decisions of the President shall be binding, subject to appeal as provided in the International Constitution. 18.5 When there are more than (2) two candidates for an office, the candidate receiving the most votes shall be declared elected. There shall be no write-in candidates for Officers. 18.6 The Secretary-Treasurer shall be responsible for reviewing the qualifications of a candidate as to his/her current membership status and continuous membership. 18.7 All nominations and election records shall be preserved for a period of at least one (1) year. Article XIX- DUES 19. The regular dues of this organization shall be determined by the Membership in a Secret Ballot vote mailed to each member at his/her last known address at least seven (7) days prior to the date such ballot is returnable. A postage-free return-addressed envelope shall be enclosed with the ballot. In no event shall the dues be less than the minimum established by the International Constitution. 19.1 Any change in the rate of dues must be by a majority of the votes cast. Any mark on a Ballot which may identify the voter, shall void the ballot. Intention to have dues increase referendum must be announced by the Executive Board. A special notice may be substituted in lieu of notice, through the Local Union newspaper. 19.2 Members who are former employees of the New Jersey Turnpike Authority must continue to pay the same monthly rate in dues as they would have if they were still an employee of the New Jersey Turnpike Authority, unless they belong to a newly established Private Sector Division. In that case, they pay the same percentage of salary each pay period as all other Members. If the dues are increased or decreased by a vote of the Membership, all Members dues will be increased by the same percentage. Article XX- INITIATION FEE 20. Initiation fees for membership in this organization shall be determined by vote of the Membership at the regular meetings of the Chapters or at a Membership meeting statewide, such vote to be ayes and nays, showing of hands for or against, standing for or against, or by Secret Ballot. XXI- MEETINGS 21. The six (6) Chapters for members shall consist of the North, North Central, Central, South Central, South, and Office. 21.1 Regular meetings shall be held in each Area and in the Woodbridge Area for the Office at such times and in such places as determined by the Area or Office Membership. Meetings will be held in March, May, September, and November, except those months in which General Membership Meetings are held, Chapter meetings may be suspended. 21.2 Full-time members cannot vote on issues particular to the Auxiliary Division. Auxiliary Division members cannot vote on issues particular to the Full-time members. 21.3 A minimum of two (2) General Membership meetings will be held in the months of January and June of each year. 21.4 Members in attendance at meetings shall have the right to express their views, arguments or opinions upon any business properly brought before the meeting. Subject to these By-Laws and the rules and regulations adopted pertaining to the conduct of meetings, but no member in exercising such rights shall evade or avoid his/her responsibility to the organization as an institution or engage in or advocate any conduct that would interfere in the Union’s performance of its obligations, or engage in any conduct personally which is unruly or boisterous. 21.5 Upon the failure of the President to call any special meeting within a reasonable time after a petition, properly containing an adequate number of signatures, has been submitted therefore, a majority of the Executive Board may call such meeting by action of a duly-called Executive Board meeting. 21.6 There shall be no quorum requirements for membership meetings. 21.7 No alcohol shall be permitted at Union Meetings. 21.8 Every member has the democratic right to speak out freely when recognized by the Chairperson. Everyone has the equal right with everyone else to have their say in the operation of the Union. Article XXII- Membership 22. There shall be two types of Member: ‘Regular’ and ‘Auxiliary’. A Regular member is one who is scheduled to work on a full-time basis. An Auxiliary member is one who is scheduled to work on a part-time basis. Limits on hours of work shall be indicated in the Agreements between Local 194 and the Employer. In the event the Local organizes private sector workers, a third Division will be established for that unit. 22.1 An applicant shall be considered a member when he/she submits an application for membership together with a Dues Deduction Authorization form and, when required, a initiation fee, and same has been accepted by the Union, such acceptance to be indicated by issuance of a current Membership Card. 22.2 Each member shall have full freedom of speech and the right to participate in the democratic decisions of the Union, subject to the reasonable rules and regulations. Each member shall have the right to criticize freely the policies and personalities of the Union Officials, however, that does not include the right to undermine the Union as an institution. The Local shall maintain adequate safeguards so that all its operations shall be conducted in a democratic procedure shall ever be permitted under any circumstances. 22.3 A member shall lose good standing next following withdrawal of Dues Deduction Authorization, by suspension or expulsion from Membership after appropriate proceedings consistent with the By-Laws and the International Constitution, or by terminating employment. 22.4 No member shall engage in dual unionism or espouse dual unionism or disaffiliation in the course of any meeting, shall not slander or libel the Union, its Members, or its Officers or Representatives, shall not be party to any activity to secure disestablishment of the Union as the collective negotiating agent, or interfere with any business representative or officer in the performance of his/her duties. 22.5 No member shall be permitted at any assembly or meeting of other members to engage in any conduct herein before described. 22.6 No member who is in an intoxicated condition (such condition to be determined by the Presiding Officer) shall be permitted to attend or participate in a union meeting and no member shall use intemperate, profane or abusive language during the course of a meeting, under penalty of being ejected and otherwise subjected to disciplinary proceedings. 22.7 Members shall process their grievances through the Union channels and be represented through the Union in disciplinary procedures. Failure to comply with this provision shall constitute conduct contrary to the best interests of the Local as defined in Article XVI of the International Constitution. Article XXIII- STEWARDS 23. Stewards are not Officers and shall have only such duties as the Executive Board may direct. Stewards shall be selected by vote of the group whom they will represent. 23.1 Stewards shall serve a term of one (1) year beginning each February and from year to year thereafter unless a request is submitted to the Executive Board for a new election. Such requests shall be in writing, signed by at least forty percent (40%) in the group, on or before January 15th of each year. 23.2 Alternates shall be appointed by the Stewards in each group. Article XXIV- AUXILIARY DIVISION 24. The Auxiliary Division shall house those Part-time Members not otherwise covered under the ‘Regular’ Membership provisions of these By-Laws. 24.1 Each Part-time work location should attempt to be represented by a Steward through election of the Auxiliary Members at their respective locations. Such Steward shall serve as long as he/she remains at the location. If a request for a change in Steward is received, signed by at last thirty percent (30%) of the Members, a nomination and election will take place for a new Steward, however the incumbent Steward may be a candidate and the person receiving the largest number of votes will be Steward. 24.2 In the event of a vacancy of Auxiliary Area Representative, the Alternate Auxiliary Area Representative will move up to Auxiliary Area Representative. Elections to fill the vacancy of Alternate Auxiliary Area Representative will be at the next Chapter meeting. Terms of office shall be three years. 24.3 All Auxiliary Division General Membership meetings will be held at a central location during the week, and not on weekends. Article XXV- CONFLICTS OF INTEREST 25. Any person who represents the Local and its Members, whether elected or appointed, has a sacred trust to serve the best interest of the members and their families. Therefore, every Officer and Representative must avoid any outside transaction which ever gives the appearance of a conflict of interest. The special fiduciary nature of union office requires the highest loyalty to the duties of Office. 25.1 The Mailing list of Local 194 is a valuable asset. In order to protect the interest of our entire membership, Union Officers, and Representatives shall not, under any circumstances, turn over a Union mailing list to any outsider for use in the promotion or sale of any goods or services that benefit an individual or a private concern. Mailing lists are to be used only to promote the necessary, legitimate functions of the Union. It is improper for an Officer or Representative of the Local to permit the use of the mailing list by any Third Party to promote the sale of furniture, appliances, automobiles, insurance, eyeglasses or any other item. However, the Union can promote vendors who offer professional services and discounts to the Members and may conduct a mailing containing a vendor’s solicitation if the vendor pays the Union the entire cost to the Union for the mailing. No Officer or Representative shall accept “kickbacks” under the table, payments, valuable gifts, or any personal payment of any kind from any vendor doing business with the Union other than the regular pay and benefits for work performed. XXVI- HEARINGS 26. Each member of this Union shall have the right to fair treatment in the application of union rules and law in accordance with the International Constitution and these By-Laws. In application, of all rules and procedures relating to the union discipline the essential requirements of due process of law-notice, hearing, and judgment based upon the evidence-shall be observed without, however, requiring technical formality followed in courts of law. 26.1 Members shall have the right to be represented by any member in good standing but no lawyers shall be permitted to appear on behalf of members in internal union proceedings. 26.2 Members of the Local may be charged only with offenses specified in the International Constitution and may be tried in accordance with procedures there specified. Any action by a member which cause embarrassment or results in the disciplining of another member by the employer, or causes a member ridicule by other members or workers shall constitute conduct contrary to the best interest of the Local as defined in Article XVI of the International Constitution. Article XXVII- INTERNATIONAL CONSTITUTION 27. The Local Union acknowledges that the Constitution of the International supersedes any provision of these By-Laws, which are inconsistent with such Constitution. If any provision of these By-Laws shall be declared invalid or inoperative by any competent authority, such provision is suspended and the Executive Board shall submit to the membership a substitute in its place. If any such provision is so declared, the remainder of these By-Laws shall not be affected thereby, but shall continue to be in full force and effect. Article XXVIII- AMENDMENTS 28. Any proposed By-Law Amendment(s) to these By-Laws shall be submitted in writing to the Local signed by a minimum of fifteen members in good standing. Such Amendments shall be promulgated in the local union publication next following their receipt by the Local and shall be voted upon at the next ensuing membership meeting statewide. Note: These By-Laws originally took effect January 1, 1969. This edition contains all revisions up to and including those voted on by the Membership at the January 29, 2011 General Membership meeting.
FRANÇOIS MORAIN JORGE OLIVOS Speeding up the computations on an elliptic curve using addition-subtraction chains Informatique théorique et applications, tome 24, n°6 (1990), p. 531-543. <http://www.numdam.org/item?id=ITA_1990__24_6_531_0> SPEEDING UP THE COMPUTATIONS ON AN ELLIPTIC CURVE USING ADDITION-SUBTRACTION CHAINS () by François MORAIN (1) Jorge OLIVOS (2) Communicated by P. FLAJOLET Abstract. — We show how to compute $x^k$ using multiplications and divisions. We use this method in the context of elliptic curves for which a law exists with the property that division has the same cost as multiplication. Our best algorithm is 11.11% faster than the ordinary binary algorithm and speeds up accordingly the factorization and primality testing algorithms using elliptic curves. Résumé. — Le but de cet article est de montrer comment calculer $x^k$ en utilisant des multiplications et des divisions. Nous utilisons cette méthode pour effectuer des calculs sur les courbes elliptiques, pour lesquelles la division a le même coût que la multiplication. Notre meilleur algorithme est 11,11 % plus rapide que la méthode binaire ordinaire et cela permet d'accélérer en conséquence les algorithmes de primalité et de factorisation qui utilisent les courbes elliptiques. 1. INTRODUCTION Recent algorithms used in primality testing and integer factorization make use of elliptic curves defined over finite fields or Artinian rings (cf. section 2). One can define over these sets an abelian law. As a consequence, one can transpose over the corresponding groups all the classical algorithms that were designed over $\mathbb{Z}/N\mathbb{Z}$. In particular, one has the analogue of the $p-1$ factorization algorithm of Pollard [29, 5, 20, 22], the Fermat-like primality testing algorithms [1, 14, 21, 26] and the public key cryptosystems based on R.S.A. [30, 17, 19]. The basic operation performed on an elliptic curve is the computation of the analogue of $x^k \mod N$ in the case where we work over () Received January 1990, accepted February 1990. (1) Institut National de Recherche en Informatique et en Automatique (I.N.R.I.A.), domaine de Voluceau, B.P. n° 105, 78153 Le Chesnay Cedex, France. On leave from the French Department of Defense, Délégation Générale pour l'Armement. (2) Departamento de Ciencias de la Computación, Universidad de Chile, Casilla 2777. Santiago, Chile, Proyecto Fondecyt 90-1121. Whatever the model of computation may be, the number of elementary operations (understood as being the product of two large integers modulo an integer $N$) is very high. Thus, reducing the number of such operations is a primary goal when implementing these algorithms. One can look first for expressions of the law that minimizes the number of operations [9]. Another idea is to reduce the number of multiplications needed to reach $x^k$. One way of handling this problem is to introduce the concept of addition chain [18] for exponents. An addition-chain for the exponent $k$ is given as an $(r+1)$-tuple $(k_0, \ldots, k_r)$ of positive integers such that $$k_0 = 1, \quad k_r = k,$$ and $$\forall i \in 1 \ldots r, \quad \exists a(i), \quad b(i) < i, \quad k_i = k_{a(i)} + k_{b(i)}.$$ Thus, if we have an addition chain for $k$ with length $r$, we can compute $x^k$ with $r$ multiplications. Many results are known [4, 11, 18, 28, 31] and some good algorithms exist, among which the binary algorithm, recalled in section 3, and some of its variations (see [18] and the implementation of the $2^m$ method in [10]). The authors of these papers have also studied briefly the so-called addition-subtraction chains, defined as in (1) but with $$\forall i \in 1 \ldots r, \quad \exists a(i), \quad b(i) \leq i, \quad k_i = \pm k_{a(i)} \pm k_{b(i)}.$$ This idea corresponds to the evaluation of $x^k$ by multiplications and divisions. In the case of integers, division is a costly operation and this idea does not seem to have implemented. The situation dramatically changes when we try to compute on an elliptic curve, because in this case, division is replaced by multiplication by the inverse and that inverse is available at no cost (cf. section 2). We are going to describe two algorithms that use addition-subtraction chains to compute $x^k$, the second one being an optimized version of the first one discovered by aid of a computer program. The expected gain is about 8.33% for the first one, and 11.11% for the second one. In section 2, we list some results about elliptic curves. Section 3 describes the binary algorithm. The new algorithms are presented in sections 4 and 5. We analyze the cost of these algorithms in section 6. In section 7, we compare the implementation of our algorithms to that of the $2^m$ method. 2. THE LAW ON AN ELLIPTIC CURVE Let $ K $ be a field of characteristic prime to 6. An elliptic curve $ E $ over $ K $ is a nonsingular algebraic projective curve of genus 1. It can be shown [7, 32] that $ E $ is isomorphic a curve of equation \[ y^2 z = x^3 + axz^2 + bz^3, \] (4) with $ a $ and $ b $ in $ K $. We note $ E(K) $ the set of points of coordinates $(x : y : z)$ which satisfy (4) with $ z = 1 $, together with the point at infinity $ O_E = (0 : 1 : 0) $. We may define on any $ E(K) $ an abelian law, but we shall introduce it only in the case where $ K = R $. This law, noted additively, plays the role of multiplication in our earlier discussion of addition chains. Addition chains are thus used to compute $ kM $ on a curve, where $ M $ is a point on $ E $. Over $ R $, let us consider the curve of equation \[ y^2 = x^3 + ax + b. \] (5) An example of such a curve is represented in figure 1 (assuming the discriminant $ \Delta = 4a^3 + 27b^2 $ to be negative). Let $ M_1 $ and $ M_2 $ be two points on the curve. We want to associate with them a third point $ M_3 $ lying on the curve that we call the sum of $ M_1 $ and $ M_2 $. Let $ D $ be the line $ M_1M_2 $ (if $ M_1 = M_2 $, $ D $ is the tangent line). One can see that $ D $ intersects $ E $ at only one other point $ P $. The point $ M_3 = M_1 + M_2 $ is then taken to be the reflexion of $P$ along the $x$-axis. The neutral element of this law is $O_E$. The opposite of a point $M$ of coordinates $(x : y : 1)$ is $-M$ whose coordinates are $(x : -y : 1)$. Thus taking the inverse of a point is essentially free. It is easy to make this process effective. We only list the results. First, $M + O_E = O_E + M = M$. When $M_2 = -M_1$ (i.e. $x_1 = x_2$ and $y_1 = -y_2$), then $M_3 = O_E$. Otherwise, the coordinates of $M_3 = (x_3 : y_3 : 1)$ are $$x_3 = \lambda^2 - x_1 - x_2,$$ $$y_3 = \lambda (x_3 - x_1) - y_1,$$ where $$\lambda = \begin{cases} (y_2 - y_1)(x_2 - x_1)^{-1} & \text{if } x_1 \neq x_2 \\ (3x_1^2 + a)(y_1 + y_2)^{-1} & \text{otherwise}. \end{cases}$$ One shows that the preceding equations are valid over any field $K$ and that they properly define the addition on $E(K)$. If $K$ is not a field, but just an Artinian ring, one can also define a law on $E(K)$ (cf. [3]). However, when working over $\mathbb{Z}/N\mathbb{Z}$, $N$ not prime, we do as if we were working over a field and use the preceding equations. The rationale behind this is that if we cannot invert an element, then we have a factor of $N$, which is the goal we usually want to reach. The reason why elliptic curves are so attractive is that two different curves provide two different laws, contrary to the case where we work in $\mathbb{Z}/N\mathbb{Z}$, where we only have multiplication. In the case of integer factorization, different curves define distinct factorization algorithms. It should be noted that adding two different points on a curve (over $\mathbb{Z}/N\mathbb{Z}$) requires three modular multiplications (i.e. multiplication of two integers followed by a reduction modulo $N$) and one gcd, and that doubling a point on the curve costs one more modular multiplication. We will come back to this problem later. All the algorithms that use elliptic curves over finite fields require the computation of the point $kM$ on $E$, where $k$ is a large integer and $M$ a point of the curve. These computations are to be performed as fast as possible, so this motivates the study that follows. 3. THE BINARY ALGORITHM AND SOME GENERALIZATIONS We just describe the algorithm. For validity and explanation, see [18]. procedure BINEXP $(P, M, k)$ $(^* P := kM, k = \sum_{i=0}^{n} k_i 2^i, 2^n \leq k < 2^{n+1} )$ - $Q := M;$ - if $k_0 = 1$ then $P := M$ else $P := O_E;$ - for $i = 1, \ldots, n$ - $Q := 2Q;$ - if $k_i = 1$ then $P := P + Q$ - end. Following [18], we introduce \[ \lambda(k) = [\log_2 k], \] \[ \nu(k) = \text{Card} \left\{ i \mid 0 \leq i \leq \lambda(k), k_i = 1 \right\}. \] Let $\mathcal{C}_{2P}$ be the cost of a “doubling” (i.e. evaluating $P + P$) and $\mathcal{C}_{P+Q}$ the cost of an “addition” (i.e. evaluating $P + Q$ with $P \neq Q$), supposed independent of $P$ and $Q$. Then the cost of BINEXP is \[ \mathcal{C}_{\text{BINEXP}}(k) = \lambda(k) \mathcal{C}_{2P} + (\nu(k) - k_0) \mathcal{C}_{P+Q}. \] The $2^m$-ary algorithm (see [18]) consists of working with base $2^m$, rather than base 2. The cost of this algorithm is roughly (see [10]) \[ \mathcal{C}_{m-\text{ary}}(k) = 2^{m-1} + \lambda(k) \mathcal{C}_{2P} + (\nu_m(k) - k_0) \mathcal{C}_{P+Q}, \] where $\nu_m(k)$ denotes the number of non-zero digits of $k$ in base $2^m$. 4. THE FIRST ALGORITHM The idea comes from the observation that long chains of 1’s in the exponent $c$ are better treated by division. For instance, we have \[ x^{15} = \frac{x^{16}}{x} = \frac{(((x^2)^2)^2)^2}{x}, \] which is more economical than the standard binary algorithm. In other words, one replaces a block of at least two 1’s by a block of 0’s and a division. If we imagine that we compute with exponents whose binary digits are 0, +1 and −1, then in terms of this extented representation, the algorithm corresponds to the transformation \[ 1^a \mapsto 10^{a-1} - 1. \] We now describe the first algorithm used to compute $ kM $ using two basic operations + and −. This idea is to construct two integers $ k_- $ and $ k_+ $ such that $ k = k_+ - k_- $, but for which the evaluation of $ k_- M $ and $ k_+ M $ require less operations than that of $ k M $. We represent the algorithm by the automaton of figure 2a. It is easy to deduce from this the following recursive procedure for the computation of $ k M $. procedure ADDSUBCHAIN-A (P, M, k)* $ Q := M; $ $ P := O_k; $ \{the result is contained in $ P \}\ TREAT0 (\( k $); end. procedure TREAT0 ($ k $) \{ invariant : $ R = P + k Q \}\ if \( k = 0 $ then return $(P)$ else if $ k $ is even then $ Q := 2Q; $ TREAT0 $(\lfloor k/2 \rfloor)$; else TREAT1 $(\lfloor k/2 \rfloor)$; end. procedure TREAT1 ($ k $) \{invariant : $ R = P + (2k + 1)Q \}\ if \( k = 0 $ then return $(P + Q)$ else if $ k $ is even then $ P := P + Q; $ $ Q := 4Q; $ TREAT0 $(\lfloor k/2 \rfloor)$; else $ P := P + Q; $ $ Q := 4Q; $ TREAT11 $(\lfloor k/2 \rfloor)$; end. procedure TREAT11 ($ k $) \{invariant : $ R = P + (k + 1)Q \}\ if \( k = 0 $ then return $(P + Q)$ else if $ k $ is even then $ P := P + Q; $ $ Q := 2Q; $ TREAT0 $(\lfloor k/2 \rfloor)$; else $ Q := 2Q; $ TREAT11 $(\lfloor k/2 \rfloor)$; end. Example: Using the binary method, $ k = 1101001110111 $ is calculated by 12 doublings and 8 additions; in total 20 operations. Using ADDSUBCHAIN-A, we compute \[ \begin{array}{ll} k & 1101001110111 \\ k_- & 100000010001 \\ \hline k+ & 10001010001000 \end{array} \] Figure 2 a. — Finite Automaton, version A. Figure 2 b. — Finite Automaton, version B. for which there are 13 doublings, 5 additions and 1 subtraction; in total 19 operations. The following is an optimized iterative from of this algorithm. procedure ADDSUBCHAIN (M, k); \{ k = k_n \ldots k_0 \} ● b := 0, P := O_k, Q := M; ● for i := 0 \ldots n if k_i = 0 then if b \neq 0 then P := P + Q; if b = 1 then Q := 2 Q; b := 0; endif Q := 2 Q; else if b = 0 then b := 1; else if b = 1 then P := P - Q; Q := 2 Q; b := 11; endif Q := 2 Q; ● P := P + Q; ● end 5. A SECOND ALGORITHM The idea is now to extend the preceding idea to the case where there are isolated 0's in the binary representation of the exponent. In this case, an isolated 0 inside a block of 1's only contributes one extra division. Using the rule of algorithm A, we have first \[ 1^a 0 1^b \mapsto 1 0^{a-1} - 1 1 0^{b-1} 1. \] But since $-2 + 1 = -1$, we can pile up the transformation $-1 1 \mapsto 0 - 1$, whence the rule describing algorithm B \[ 1^a 0 1^b \mapsto 1 0^a - 1 0^{b-1} - 1. \] This process is represented by a suitable modification of automaton A to produce automaton B: we introduce state 110 that takes this into account. We will make the analysis on this automaton (see section 6), but the actual program can be made simpler. This follows from the remark that the arcs that leave state 110 are the same as those that leave state 1. We can now build procedure ADDSUBCHAIN-B, that is the same as ADDSUBCHAIN-A except for procedure TREAT11. procedure TREAT1! k {invariant: $R = P + (k + 1) \ Q$} if $k = 0$ then return $(P + Q)$ else if $k$ is even then TREAT1! $(k/2)$; else $Q := 2 \ Q$ TREAT1! $(k/2)$; end. Exemple: With the same value of $k$, we find \[ \begin{array}{c\|c} k & 1101001110111 \\ k_+ & 100000001001 \\ \hline k_- & 10001010000000 \end{array} \] thus requiring only 13 doublings, 4 additions, and 1 subtraction; in total 18 operations. 6. ANALYSIS OF THE ALGORITHMS We recall that the expected cost of the binary method is $3/2n + O(1)$ and that of the $2^m$ method is $n(1 + (1 - 2^{-m})/m) + O(1)$ when all the operations have the same cost (see [15]). In order to analyze the algorithm (both versions), we will count the number of operations required to calculate $k$, assuming that $\mathcal{C}_{2P} = \mathcal{C}_{P+Q} = 1$ for the sake of simplicity in the first version and the real cost in the second version, i.e. $\mathcal{C}_{2P} = K + 4$ and $\mathcal{C}_{P+Q} = K + 3$ (see section 2 and below). Our analysis is based on the automata shown in figures 1 and 2. We will use an approach based on grammatical specification (see e.g. [16]). The associated bit strings belong to the language $L = \{0, 1\}^* 1$ where we have inverted the binary representation of a number greater than or equal to 1. The idea is to associate with each string in $L$ a commutative polynomial in variables $z$ and $u$ that represents the relevant parameters: the number of bits ($z$) and the calculation cost of the string ($u$). For instance, in the binary case \[011001 \mapsto z^6 u^7.\] The corresponding production rules (see fig. 2a) are the following \[ \begin{align} A & \rightarrow T_0 \\ T_0 & \rightarrow 0 \ T_0 + 1 \ T_1 \\ T_1 & \rightarrow 0 \ T_0 + 1 \ T_{11} + \varepsilon \\ T_{11} & \rightarrow 0 \ T_0 + 1 \ T_{11} + \varepsilon \end{align} \] Usual techniques can be used to obtain the generating function \[ A(z, u) = \sum_{n,m} a_{n,m} z^n u^m \] where $ a_{n,m} (= [z^n u^m] A(z, u)) $ is the number of strings with $ n $ bits (only $ n-1 $ of which are really involved here because the last is always 1), and with cost $ m $ (in number of operations). We only need to solve the following system of equations \[ \begin{align} A &= T_0 \\ T_0 &= zu T_0 + z T_1 \\ T_1 &= zu^3 T_0 + zu^3 T_{11} + u \\ T_{11} &= zu^2 T_0 + zu T_{11} + u \end{align} \] where each new bit has an associated $ z $ and $ u $ whose exponent is equal to the cost given on the corresponding arcs of the automata (fig. 2a). $ A(z, u) $ has been obtained by solving this system using MAPLE [8]. Since we are interested in the expected cost, we compute \[ a(z) = \left( \frac{\partial A}{\partial u} \right) \bigg\|_{u=1} = \frac{z^2 (2 + z + z^2)}{(1 - 2z)} + \frac{(z + 2z^2)}{(1 - 2z)} \] from this we deduce \[ \frac{1}{2^{n-1}} [z^n] a(z) = [z^n] 2a(z/2) = \frac{11}{8} n + \frac{1}{8}. \] This compares favorably with the cost of the binary algorithm which is $ \frac{3}{2} n + O(1) $. The relative saving in the number of additions is 25% [(1/2-3/8)/(1/2)], and there is an overall relative saving of 8.33% if all operations are considered. In a similar manner for version B (recall that program ADDSUBCHAIN-B does not correspond precisely to fig. 2b), the production rules are \[ \begin{align} B &\rightarrow T_0 \\ T_0 &\rightarrow 0 T_0 + 1 T_1 \\ T_1 &\rightarrow 0 T_0 + 1 T_{11} + \varepsilon \\ T_{11} &\rightarrow 0 T_{110} + 1 T_{11} + \varepsilon \\ T_{110} &\rightarrow 0 T_0 + 1 T_{11} \end{align} \] Introducing the respective costs for elliptic curves, the corresponding equations are \[ \begin{align} B &= T_0 \\ T_0 &= z u^{K+4} T_0 + z T_1 \\ T_1 &= z u^{3K+11} T_0 + z u^{3K+11} T_{11} + u^{K+3} \\ T_{11} &= z T_{110} + z u^{K+4} T_{11} + u^{K+3} \\ T_{110} &= z u^{3K+11} T_0 + z u^{3K+11} T_{11} \end{align} \] From which we found, using MAPLE, the excepted cost \[ [z^n] 2 b(z/2) = \frac{4K+15}{3} n + \frac{4K+9}{9} + O(2^{-n}). \] With version B we achieve a relative saving of 33% with respect to additions, and 11.11% if all operations are considered. We must note that version B of the algorithm was discovered with the help of $ \Lambda \gamma \Omega $ [12], a powerful system designed to perform automatic analysis of a broad class of algorithms, developed at I.N.R.I.A. We now give the results for elliptic curves when we have $ \mathcal{C}_{2p} = K+4 $ and $ \mathcal{C}_{p+Q} = K+3 $, where $ K $ is the cost of a gcd over the integers, the unit being the time of a multiplication modulo $ N $. For example, in [5], the author takes $ K=30 $. P. Zimmermann kindly computed the costs of the algorithms given this assumption, using $ \Lambda \gamma \Omega $. Here are the results Binary algorithm: $(3K+11)/2n + O(1)$ $2^m$-ary algorithm: $(K+4+(K+3)(1-2^{-m})/m)n + O(1)$ Algorithm A: $(11K+41)/8n + O(1)$ Algorithm B: $(4K+15)/3n + O(1)$. Our algorithms require no extra-storage, contrary to the $2^m$ method, which needs to store $2 \times 2^{m-1}$ $n$-bits integers (if we work over $ \mathbb{Z}/N\mathbb{Z} $ with $ N $ an $ n $-bit integer). 7. IMPLEMENTATION AND CONCLUSIONS The first author has used the second algorithm in his implementation of the so called Atkin’s test [26] for primality testing. The overall gain is about 3% in time for 100-digit numbers and 2.7% for 300-digit numbers. It should be possible to combine the idea of addition-subtraction chains with that of the $2^m$-ary algorithm. As for now, it is not clear how we could do that, the problem being that our algorithm must keep track of what happened a few bits before. The second author would like to express his gratitude to I.N.R.I.A. for an invited visit during which his work on the subject was done. Many thanks are due to P. Flajolet. This work has also benefitted from financial assistance from the French-Chilean cooperation program whose help is gratefully acknowledged. Both authors would like to acknowledge the help of P. Zimmermann with the $\Lambda_\gamma\Omega$ system and that of P. Flajolet who very carefully read the first version of the paper and made valuable remarks. REFERENCES 1. A. O. L. Atkin, Manuscript. 2. F. Bergeron, J. Berstel, S. Brlek and C. Duboc, Addition Chains using Continued Fractions, Journal of Algorithms, 10, 3, 1989, pp. 403-412. 3. W. Bosma, Primality Testing using Elliptic Curves, Report 85-12, Math. Instituut, Universiteit van Amsterdam. 4. A. Brauer, On Addition Chains, Bull. Amer. Math. Soc., 45, 1939, pp. 736-739. 5. R. P. Brent, Some Integer Factorization Algorithms using Elliptic Curves, Research Report CMA-R32-85, The Australian National University, Canberra, 1985. 6. J. Brillhart, D. H. Lehmer, J. L. Selfridge, B. Tuckerman and S. S. Wagstaff, Jr., Factorizations of $b^e \pm 1$, $b = 2, 3, 5, 6, 7, 10, 11, 12$ up to High Powers, Contemporary Math., A. M. S., 1983, . 7. J. W. S. Cassels, Diophantine Equations with Special References to Elliptic Curves, J. London Math. Soc., 1966, , pp. 193-291. 8. B. W. Char, K. O. Geddes, G. H. Gonnet and S. M. Watt, MAPLE, Reference Manual, Fourth Edition, Symbolic Computation Group, Department of Computer Science, University of Waterloo, 1985. 9. D. V. Chudnovsky and G. V. Chudnovsky, Sequences of Numbers Generated by Addition in Formal Groups and New primality and Factorization Tests, Research report RC 11262, I.B.M., Yorktown Heights, 1985. 10. H. Cohen and A. K. Lenstra, Implementation of a New Primality Test, Math. Comp., 1987, , 177, pp. 103-121. 11. P. Erdos, Remarks on Number Theory III: On Addition Chains, Acta Arithmetica, 1960, , pp. 77-81. 12. P. Flajolet, B. Salvy and P. Zimmermann, Lambda-Upsilon-Omega: An assistant algorithms analyzer. In Applied Algebra, Algebraic Algorithms and Error-Correcting Codes (1989), T. Mora, Ed., Lecture Notes in Comp. Sci., 357, pp. 201-212. (Proceedings AAECC'6, Rome, July 1988). 13. P. Flajolet, B. Salvy and P. Zimmermann, Lambda-Upsilon-Omega: The 1989 Cookbook, Research Report 1073, Institut National de Recherche en Informatique et en Automatique, August 1989, 116 pages. 14. S. Goldwasser and J. Kilian, Almost all Primes can be quickly Certified. Proc. 18th A.C.M. Symp. on the Theory of Compt., Berkeley, 1986, pp. 316-329. 15. G. H. Gonnet, Handbook of Algorithms and Data Structures, Addison-Wesley, 1984. 16. D. H. Greene, Labelled Formal Languages and Their Uses, Technical Report STAN-CS-83-982, Stanford University, 1983. 17. B. S. Kaliski Jr., A Pseudo-Random Bit Generator Based on Elliptic Logarithms, Proc. Crypto 86, pp. 13-1, 13-21. 18. D. E. Knuth, Seminumerical Algorithms, The Art of Computer Programming, T. II, Addison-Wesley. 19. N. Koblitz, Elliptic curve cryptosystems. Math. Comp., 1987, 48, 177, pp. 203-209. 20. H. W. Lenstra Jr., Factoring with Elliptic Curves, Report 86-18, Math. Inst., Univ. Amsterdam, 1986. 21. H. W. Lenstra Jr., Elliptic Curves and Number Theoretic Algorithms, Report 86-19, Math. Inst., Univ. Amsterdam, 1986. 22. H. W. Lenstra Jr., Factoring integers with elliptic curves. Annals of Math., 1987, 126, pp. 649-673. 23. D. P. McCarthy, The Optimal Algorithm to Evaluate $x^n$ using Elementary Multiplication Methods, Math. Comp., 1977, 31, 137, pp. 251-256. 24. D. P. McCarthy, Effect to Improved Multiplication Efficiency on Exponentiation Algorithms Derived from Addition Chains, Math. Comp., 1986, 46, 174, pp. 603-608. 25. P. L. Montgomery, Modular Multiplication without Trial Division, Math. Comp., 1985, 44, 170, pp. 519-521. 26. F. Morain, Implementation of the Atkin-Goldwasser-Kilian test. I.N.R.I.A. Research, Report 911, 1988. 27. F. Morain and J. Olivos, Un algoritmo de Evaluacion de Potencia utilizando Cadenas de Suma y Resta, Proc. XIV Conference Latinoamericana de Informatica (C.L.E.I., Expodata), Buenos Aires, September 1988. 28. J. Olivos, On Vectorial Additions Chains. J. of Algorithms, 1981, 2, pp. 13-21. 29. J.-M. Pollard, Theorems on factorization and primality testing. Proc. Cambridge Phil. Soc., 1974, 76, pp. 521-528. 30. R. L. Rivest, A. Shamir and L. Adleman, A Method for Obtaining Digital Signatures and Public-Key Cryptosystems, Comm. of the A.C.M., 1978, 21, 2, pp. 120-126. 31. A. Schönhage, A Lower Bound for the Length of Addition Chains, Theor. Comput. Science, 1975, 1, 1, pp. 1-12. 32. J. T. Tate, The Arithmetic of Elliptic Curves, Inventiones Math., 1974, 23, pp. 179-206.
Special R & D around the world NEW RAYTHEON MAGNETRONS FOR A WIDE RANGE OF APPLICATIONS Designed for C-band systems requiring tunability, the RK-7156 magnetron has a minimum peak power output rating of 250 kilowatts over a frequency range of 5,450 to 5,825 megacycles. Applications include a flight-tested, revolutionary airborne weather radar system. The RK-7156 is in quantity production. For ground-based and airborne radar systems, the RK-7529 magnetron provides a 2.0 microsecond pulse of 3.5 megawatts minimum peak power over 2,700 to 2,850 Mc. This liquid-cooled tube is interchangeable with other fixed-frequency S-band tubes operating at similar power levels. X-band magnetron for airborne search radar provides one megawatt minimum peak power and 875 watts average power within a frequency range of 9,340 to 9,440 Mc. Designated QK-624, this pulsed-type tube is liquid cooled and should give at least 1,000 hours of reliable service. A one kilowatt beacon magnetron, the RK-7578 weighs only 14 ozs., yet will withstand vibrations of 15 G's at 20 to 2,000 cycles and shock up to 100 G's. It is mechanically tunable and covers the 5,400 to 5,900 Mc range. Developed to withstand extreme environmental conditions, the RK-7449 magnetron is a lightweight, compact tube with a minimum peak power output of 45 kilowatts at the operating frequency of 24 kc. The RK-7449 is required to withstand repeated shocks of 50G. Stable operation is guaranteed at vibration frequencies up to 2,000 c.p.s. with 30G applied. Issue at a Glance Business Microelectronics Moving Fast. Report on a lively evolution........36 Air Agency Plans for '60. A look at FAA's electronics spending....40 X-rays for Communications? An exciting study for the Air Force...45 Hams Seek Facsimile Gear. Military Affiliate Radio System's needs..48 Radiation Used in Brain Study. Recent academic developments....53 Shoptalk .........................4 25 Most Active Stocks.......21 Business This Week...........11 Market Research ...........26 Washington Outlook .........14 Current Figures ...........26 Financial Roundup ..........21 Meetings Ahead ...........55 Engineering International flavor of progress in electronics is symbolized by flags from England, France, Japan, Australia, Israel, Italy, Sweden and Switzerland. See p 75................................................COVER Electronics Research & Development Around the World. Highlights of progress in other nations. An ELECTRONICS Special Report .................................................................75 Three-Element Semiconductor Materials. Developing new ternary compounds......................By J. H. Wernick and R. Wolfe 103 Wire Gage Provides Continuous Measurement. Circuit measures wire thickness non-destructively..............By K. H. Jaensch 109 Circuit Modifications for Boosting Counter Speed. Technique accelerates electromagnetic counters by 300 percent......By R. Ives 112 Electronic Tonometer for Glaucoma Diagnosis. Device measures abnormal eye pressure..............By R. S. Mackay and E. Marg 115 Character Generator for Digital Computers. High-speed display of alphanumeric characters......................By E. D. Jones 117 Departments Research and Development. Solid-State Electronic Tuning........122 Components and Materials. Small Beam Switch Extends Applications .................................................................126 Production Techniques. Wheel in Oven Makes 150 C Tests ......130 On the Market..............134 Plants and People...........192 New Books ..................182 News of Reps..............195 Literature of the Week.....188 Comment .....................196 Index to Advertisers......204 FREQUENCY STANDARDS PRECISION FORK UNIT TYPE 50 Size 1" dia. x 3¾" H.* Wght., 4 oz. Frequencies: 240 to 1000 cycles Accuracies: Type 50 (±.02% at —65° to 85°C) Type R50 (±.002% at 15° to 35°C) Double triode and 5 pigtail parts required Input, Tube heater voltage and B voltage Output, approx. 5V into 200,000 ohms 3¾" high 400 - 1000 cy. FREQUENCY STANDARD TYPE 50L Size 3¾" x 4½" x 5½" High Weight, 2 lbs. Frequencies: 50, 60, 75 or 100 cycles Accuracies: Type 50L (±.02% at —65° to 85°C) Type R50L (±.002% at 15° to 35°C) Output, 3V into 200,000 ohms Input, 150 to 300V, B (6V at .6 amps.) PRECISION FORK UNIT TYPE 2003 Size 1½" dia. x 4½" H. Wght. 8 oz. Frequencies: 200 to 4000 cycles Accuracies: Type 2003 (±.02% at —65° to 85°C) Type R2003 (±.002% at 15° to 35°C) Type W2003 (±.005% at —65° to 85°C) Double triode and 5 pigtail parts required Input and output same as Type 50, above 3½" high 400 to 500 cy. optional FREQUENCY STANDARD TYPE 2005 Size, 8" x 8" x 7¼" High Weight, 14 lbs. Frequencies: 50 to 400 cycles (Specify) Accuracy: ±.001% from 20° to 30°C Output, 10 Watts at 115 Volts Input, 115V. (50 to 400 cycles) FREQUENCY STANDARD TYPE 2007-6 NEW TRANSISTORIZED, Silicon Type Size 1¾" dia. x 3¼" H. Wght. 7 ozs. Frequencies: 400 — 500 or 1000 cycles Accuracies: 2007-6 (± .02% at —50° to +85°C) R2007-6 (±.002% at +15° to +35°C) W2007-6 (±.005% at —55° to +125°C) Input: 10 to 30 Volts, D.C., at 6 ma. Output: Multitap, 75 to 100,000 ohms FREQUENCY STANDARD TYPE 2121A Size 8¾" x 19" panel Weight, 25 lbs. Output: 115V 60 cycles, 10 Watt Accuracy: ±.001% from 20° to 30°C Input, 115V (50 to 400 cycles) FREQUENCY STANDARD TYPE 2001-2 Size 3¾" x 4½" x 6" H., Wght. 26 oz. Frequencies: 200 to 3000 cycles Accuracy: ±.001% at 20° to 30°C Output: 5V. at 250,000 ohms Input: Heater voltage, 6.3 - 12 - 28 B voltage, 100 to 300 V., at 5 to 10 ma. ACCESSORY UNITS for TYPE 2001-2 L—For low frequencies multi-vibrator type, 40-200 cy. D—For low frequencies counter type, 40-200 cy. H—For high freqs, up to 20 KC. M—Power Amplifier, 2W output. P—Power supply. FREQUENCY STANDARD TYPE 2111C Size, with cover 10" x 17" x 9" H. Panel model 10" x 19" x 8¾" H. Weight, 25 lbs. Frequencies: 50 to 1000 cycles Accuracy: (±.002% at 15° to 35°C) Output: 115V, 75W. Input: 115V, 50 to 75 cycles. This organization makes frequency standards within a range of 30 to 30,000 cycles. They are used extensively by aviation, industry, government departments, armed forces—where maximum accuracy and durability are required. WHEN REQUESTING INFORMATION PLEASE SPECIFY TYPE NUMBER American Time Products, Inc. Watch Master Timing Systems Telephone: Plaza 7-1430 580 Fifth Ave., New York 36, N. Y. IF YOU INSTALL LAMBDA POWER SUPPLIES NOW, YOU KNOW THEY’LL STILL PERFORM TO GUARANTEE IN 1965 BECAUSE ONLY LAMBDA GIVES YOU A 5-YEAR GUARANTEE! Each Lambda Power Supply carries a written guarantee that warrants full performance to specified ratings for five full years. Lambda, alone, offers this unprecedented guarantee because Lambda specializes in the design and manufacture of just one product — power supplies. Each unit is built from the ground up to rigid quality standards in a modern, completely integrated plant. Nothing is overlooked to provide you with the finest performance. When you buy a Lambda Power Supply, you are assured of a unit that is conservatively rated—a unit designed to provide continuous-duty service at all specified loads and ratings. Lambda power supplies are available in a wide range of rack, portable and bench models for laboratory and production service. Of particular interest to electronic designers are: L-T Transistor-Regulated Series ........... 0-1 and 0-2 AMP, 0-32 VDC Com-Pak Tube Regulated Series ............ 200-400-800-1500 MA 0-200, 125-325, 325-525 VDC Write for free 32-page catalog for complete specifications, dimensions, performance ratings and prices on Lambda’s full line of tube-regulated and transistorized power supplies. LAMBDA ELECTRONICS CORP. 11-11 131 STREET • COLLEGE POINT 56, N. Y. • INDEPENDENCE 1-8500 Watch for Lambda’s exciting new solid-state power supply developments. IRE SHOW—Booths 2318-2320 RESEARCH. Electronics—as our special report in this issue suggests—is an industry which by its very nature is oriented toward research and engineering. All over the world, the industry is supported by laboratory experiment, by mathematical formulation—and, more and more, by pure thought. Expensive as these "overhead" costs are to business, they are the means by which technology grows. As such, research effort is a direct measure of the health and vitality of an industry. And it eventually pays off, as our lead news story this week amply demonstrates. The story discusses molecular electronics, one of the many names being applied to the new art of growing whole circuit complexes en bloc* in solid crystalline form. There are a number of laboratories in this country and abroad working in this promising field. Their labors will ultimately lead to a higher level of circuit reliability under all sorts of environmental conditions, and to more value per dollar spent on electronic systems. Even more important, molecular engineering—if we are to call it that—should ease many of today's production and maintenance problems. When the techniques are perfected, we may see nearly automatic production of electronic systems and plug-in-throw-away maintenance. Coming In Our February 19 Issue . . . SONAR. Antisubmarine warfare requires the ability to penetrate the modern submarine's underwater cloak of invisibility. Sometimes referred to as the fourth dimension at sea, the ocean depths are ideal hiding places for missile-carrying nuclear submarines, for they cannot be penetrated to any extent by visible light, radar or similar forms of radiation. As a result, new and improved sonar systems are being emphasized as an effective method of determining range, bearing and depth of submerged objects. In our next issue, we will publish the first of two articles by G. Rand of Sperry Gyroscope Co. on sonar equipment. The article deals with the range capabilities of both passive and active systems. You'll find design charts for estimating the range of sonar gear and for determining the effect of the various parameters on the equipment range. Rand has been concerned with underwater propagation studies, development of underwater sensing devices and sonar detection systems since he joined Sperry in 1957 as an engineering section head. Prior to that he spent 10 years as a chief project engineer on the design, development and production of sonar transducers and underwater detection devices. You will want to read his informative report on this vital topic. SPOT WELDING. A welding control system that insures consistency and reliability of spot welds over long periods is described by G. R. Archer of The Budd Co. The new system, called voltage restraint, uses a feedback to compare the voltage across a spot weld with a previously determined command voltage indicative of the proper temperature for welding. System is the result of a program to design and assemble a complete control system that would allow uniformly good welds over a wide range of in-process variation without the need for destructive testing. Archer has been associated with the program in the capacity of chief engineer for the Electronic Controls Section. BLUE JACKET VITREOUS ENAMEL-PROTECTED, POWER WIREWOUND RESISTORS. Send for Bulletins: 7410-A (Axial Lead), 7400-A (Tab Type) KOOLOHM® CERAMIC INSULATED-SHELL, POWER WIREWOUND RESISTORS. Send for Engineering Bulletin: 7300-A PERMASEAL® CAST EPOXY HOUSING, PRECISION WIREWOUND RESISTORS. Send for Engineering Bulletin: 7500 SPRAGUE RESISTORS FILMISTOR® PRECISION CARBON FILM RESISTORS. Send for Bulletins: 7000 (Molded shell), 7010-B (Ceramic shell) MEG-O-MAX® GLASS-JACKETED HIGH VOLTAGE, HIGH POWER RESISTORS. Send for Engineering Bulletin: 7200-A SPIRAMEG® HIGH-RESISTANCE SPIRAL ELEMENT RESISTORS. Send for Engineering Bulletin: 7100 SPRAGUE ELECTRIC COMPANY 35 Marshall Street North Adams, Mass. SPRAGUE COMPONENTS: RESISTORS • CAPACITORS • MAGNETIC COMPONENTS • TRANSISTORS INTERFERENCE FILTERS • PULSE NETWORKS • HIGH TEMPERATURE MAGNET WIRE • PRINTED CIRCUITS ELECTRONICS • FEBRUARY 12, 1960 CIRCLE 5 ON READER SERVICE CARD Solid State Reliability IN A 10 mc Counter The CMC 700 Series is the only major breakthrough in counting, timing and frequency measuring equipment in the past 10 years. Here is the first successful application of transistors to high frequency counting and timing. Transistors perform all the functions in CMC's 700 series that required 63 tubes in old style counting equipment. These are the most reliable counters ever made. TRUE DIGITAL LOGIC CIRCUITRY By answering an obvious need for a completely new, up-to-date approach to counting and timing instrumentation, CMC has produced solid state instruments with greatly simplified circuitry, using logic "and" and "or" gates. LIGHT AND SMALL, LOWER POWER DRAIN Each 700 series instrument weighs only 27 pounds, measures 7 inches high, 17 inches wide, and 14 inches deep. Power consumption is a meager 46 watts, 1/10 the amount for vacuum tube models. DO ALL THESE JOBS Measure frequency from dc to 10 mc, time interval from 0.1 µsec, ratio 1 cps to 1 mc and unlimited multiple period selection. Frequency converters available for higher frequencies. The counter also generates time interval marker pulses from 1 µsec to 1 second. Data can be presented on standard decades or inline Nixie tubes. The 700 series will operate digital recording equipment, punches, inline readouts, and other data handling gear. These Features, Too—Decade count-down time base—frequency divider circuits never need adjustment. Accuracy, ±1 count ± oscillator stability. Sensitivity, 0.25 v rms; input impedance, 25 k ohms/volt. And The Price—Higher than vacuum tube models. But you can save the difference on down time in the first year. Model 727A Universal Counter-Timer, $3,500; Model 707A Frequency-Period Meter, $2,700; Model 757A Time Interval Meter, $2,500. Rack mount optional at no extra cost. All prices f.o.b. Sylmar, California. More Information Available — Your nearby CMC engineering representative will be happy to arrange a demonstration and provide you with complete technical information. Or you may write Department 18. Computer Measurements Co. A Division of Pacific Industries 12970 Bradley Avenue, Sylmar, California Phone: EMpire 7-2161 OUT-PERFORMS WIRE WOUND... YET SMALLER IN SIZE, LOWER IN COST New Electra Precision Metal Film Resistor Here is an entirely new achievement in electronic components; one of the biggest steps forward in years. This precision metal film resistor offers you precision and stability that formerly was available only in a wire wound resistor, yet it is much smaller in size, much lower in cost, also has far superior high frequency characteristics. Available in five sizes from 1/8 to 2 watts, the new Electra Precision Metal Film Resistor meets or exceeds Mil-R-10509C, Characteristic C, and can be supplied in any of eight standard temperature coefficient tolerances. Why not let us supply you full details by return mail. Write today! CHECK THESE OUTSTANDING TEST RESULTS \| TEMPERATURE CYCLE \| MOISTURE \| LOAD LIFE 125°C \| SHORT TIME OVERLOAD \| \|-------------------\|----------\|-----------------\|---------------------\| \| Initial \| Final \| % Change \| Initial \| Final \| % Change \| \| 236.9 \| 236.9 \| 0 \| 236.9 \| 236.9 \| 0 \| \| 237.1 \| 237.1 \| 0 \| 237.1 \| 237.1 \| 0 \| \| 236.6 \| 236.6 \| 0 \| 236.6 \| 236.6 \| 0 \| \| 236.8 \| 236.8 \| 0 \| 236.8 \| 236.8 \| 0 \| \| 237.2 \| 237.2 \| 0 \| 237.2 \| 237.2 \| 0 \| \| 237.4 \| 237.4 \| 0 \| 237.4 \| 237.4 \| 0 \| \| 237.6 \| 237.6 \| 0 \| 237.6 \| 237.6 \| 0 \| \| 237.8 \| 237.8 \| 0 \| 237.8 \| 237.8 \| 0 \| \| 238.0 \| 238.0 \| 0 \| 238.0 \| 238.0 \| 0 \| Electra Manufacturing Company 4051 Broadway Kansas City, Mo. ELECTRONICS • FEBRUARY 12, 1960 CIRCLE 7 ON READER SERVICE CARD CANNON PLUGS FOR ANY APPLICATION OR ENVIRONMENT Shock ■ Vibration ■ Acceleration ■ Temperature ■ Altitude ■ Moisture ■ Miniaturization ■ For more than four decades — since the first Cannon Plug inaugurated a new method of electrical connection — Cannon has been solving critical plug problems for every conceivable industrial and military application. Cannon’s research and development, manufacturing know-how, and world-wide service is capable of meeting the problems of modern technology. Whenever a new design or modification is needed, Cannon provides the answer. Whatever your requirement in plugs...for any environment or application...you have assured reliability from the first in the field. Remember, consult Cannon for all of your plug requirements. If you don’t already have a copy, write for CPG-4, “Cannon Plug Guide.” For more information write Department 120 CANNON ELECTRIC COMPANY, 3208 Humboldt St., Los Angeles 31, Calif. ELECTRONICS • FEBRUARY 12, 1960 CIRCLE 9 ON READER SERVICE CARD Regulated HI-VOLTAGE power supplies NJE offers a series of regulated High Voltage Power Supplies designed and built to provide the highest degree of reliability and performance. NJE accomplishes this with a simple, effective circuit and the use of high grade components. \| Model \| Output DC Voltage \| Output Current \| Ripple RMS \| Regulation Line \| Regulation Load \| Panel Height \| Unit Price \| \|-------\|------------------\|---------------\|------------\|-----------------\|-----------------\|--------------\|------------\| \| S-324 \| 800-2000 \| 0-10 MA \| 25 MV \| ± 0.005% \| 60 MV \| 5¾" \| $345. \| \| S-325 \| 500-2500 \| 0-10 MA \| 5 MV \| ± 0.005% \| 60 MV \| 5¾" \| $395. \| \| S-326 \| 500-2500 \| 0-50 MA \| 5 MV \| ± 0.005% \| 0.005% \| 8¾" \| $485. \| \| S-327 \| 500-5000 \| 0-10 MA \| 5 MV \| ± 0.005% \| 0.005% \| 8¾" \| $575. \| Stock subject to prior sale. Custom design quotations submitted on request. Send for technical literature. NJE CORPORATION 20 Boright Avenue · Kenilworth, New Jersey BR. 2-6000 · TWX Cranford, NJ 51 · FAX-FFP Tunnel Diode Factory Production Announced by U. S. and Japanese Firms Tunnel diode production developments made news this week in Syracuse, N. Y., and in Tokyo, Japan. H. E. Fancher, general manager of GE's semiconductor products department, disclosed that his company has started factory production of tunnel diodes in the face of increasing demand and has reduced prices on the devices. He announced that units priced at $60 and $75 have been reduced to $10 and $12.50. Fancher reported that "several thousand" tunnel diodes have already been shipped by GE since samples were made available last September. He said advanced R&D was continuing but "production on a pilot manufacturing line is feasible." In Tokyo, meanwhile, ELECTRONICS learned that Sony plans to make tunnel diode samples available next month. By then, the company expects to be producing the devices at a rate of "several thousand" a day. Kazuo Iwama, director of Sony's semiconductor division, expects the units to be priced at about $10, adds there are no plans set up to export any to the U.S. New Telemetry Transmitter for ICBM's Can Operate During Missile Re-entry Telemetry transmitter has been developed for the Air Force by ITT Laboratories to pierce the highly-conductive ionized plasma around a space vehicle re-entering the earth's atmosphere. The company said the transmitter would be used on all ICBM's and in the Mercury man-in-space program. Extremely high frequencies—somewhere between 30,000 and 300,000 mc—are the key to the transmitter's operation during re-entry. (Absorption of r-f energy by the plasma sheath is inversely proportional to frequency.) Work was done in cooperation with Avco Corporation's R&D division. Radar Signal Takes 17 Minutes To Bounce Off Sun's Outer Corona Scientists at Stanford University's Radioscience Laboratory have just disclosed that they bounced a radar beam off the sun last April using a standard transmitter of 40,000 watts. It took 17 minutes for the 25.6 mc signal to return to earth, and almost 10 months of magnetic tape analysis to verify that the echo received had actually come from the sun's outer corona. The scientists used a four-unit rhombic array for both sending and receiving. It consists of wires strung out on 22 power poles over 14 acres. The transmitter was turned on and off in 30-second pulses for 15 minutes, then kept silent while awaiting the echo. As a result of the 93-million-mile radar bounce, scientists expect to be able to explore the whole solar system with radar. Dutch Market First Electronic Computer; Machine Uses Transistors and Ferrite Cores First Netherlands commercial electronic computer, which uses transistors and ferrite cores supplied by Philips of Eindhoven, has been built by N. V. Electrologica of Amsterdam. Estimated cost of the machine, one of which has been installed at First Netherlands Insurance Co. of The Hague, is $237,000. Nine others have been sold. Computer has a 32,000-word memory, adds and subtracts at a rate of 15,000 characters per second and multiplies and divides at a 2,000-per-second rate. Punch-card input is 14,000 an hour, output 56,000 an hour. ELECTRONICS NEWSLETTER Rosy predictions for industry-wide semiconductor sales this year (ELECTRONICS, p 53, Jan. 1 and p 24, Jan. 29) are being bolstered by announcements of production expansion. Latest comes from Motorola's semiconductor products division, Phoenix, Ariz., which is preparing for an anticipated fivefold increase in sales of mesa transistors and other semiconductor products. The company will spend more than $8 million over 18 months in its second major expansion in two years, will add at least 100,000 sq ft. Japanese electronics industry, which launched a five-year development plan in July 1958, will now revise it. New emphasis will include: cathode ray tubes for color tv, video tape recorders, aviation and medical electronics, microwave tubes. New solid-state business data-processor recently announced by IBM and dubbed the 7080 has been ordered by Southern Railway to handle interline freight accounting, traffic statistics, passenger-use analysis and stockholder records. The 7080 is said to be 10 times faster than its predecessor, the 705. New computer has a high-speed front-end buffer memory for possible direct communications inputs. System is compatible with 705 programming. Sophisticated electronic gear aboard merchant and other ships in requiring better voltage regulation of power-generating equipment and paving the way for more solid-state power supplies. At the winter general meeting of the American Institute of Electrical Engineers, experts on marine a-c systems last week discussed the stricter requirements of modern shipboard electronics. Some engineers recommended neutral grounding of electrical systems, claiming that floating ground sometimes results in overvoltages, accidental grounding to the hull, shock hazard and complicated troubleshooting of faults. New "Beam-X" switch outperforms all solid state, vacuum & magnetic devices IN MULTIPOSITION SWITCHING, COUNTING AND DISTRIBUTING A technological breakthrough in the design of Beam Switching Tubes eliminating external magnets and shields has resulted in a low cost revolutionary device. BEAM-X* outperforms all existing solid state, magnetic and vacuum components for electronic switching applications. In aircraft, missile, commercial instrumentation, control systems and other industrial applications, BEAM-X* offers far superior design flexibility and reliability than existing conventional components. BEAM-X* type BX-1000 is the first of a new family of multiposition electronic switches. THE BEAM-X* IS AN ALL-ELECTRONIC MULTIPOSITION SWITCH The BEAM-X* eliminates Multicomponent size Multicomponent weight Multicomponent power Multicomponent cost and Multicomponent unreliability WRITE TODAY FOR TECHNICAL BROCHURE DESCRIBING THE OPERATION AND COMPLETE MECHANICAL AND ELECTRICAL APPLICATION DATA OF THIS NEW BREAKTHROUGH BEAM-X* SWITCH. ANOTHER ELECTRONIC CONTRIBUTION BY Burroughs Corporation ELECTRONIC TUBE DIVISION Plainfield, New Jersey BEAM-X* APPLICATIONS: COUNTING • CODING • DISTRIBUTING • CONVERTING • GATING • MULTIPLEXING • SWITCHING TIMING • SAMPLING • MEMORY • MATRIXING • PRESETTING • DECODING • DIVIDING TRADEMARK OF BURROUGHS CORPORATION WASHINGTON OUTLOOK CONGRESS moves in for a look at the increasing defense electronics buying programs. The General Accounting Office, congressional agency for review of administration spending, is making a study of present programs for defense electronics procurement. GAO will focus most on spending for communications equipment. Missile systems and other weapons programs are less affected. The organization will be looking for ways to save the government money in equipment purchases. GAO is particularly interested in finding out whether duplicate buying can be avoided by consolidating procurement of certain types of equipment in one military agency for distribution to the other services. The initial review is expected to be completed by March or April. Results will be sent to the individual military services for comment before being presented to Congress. GAO is taking an especially close look at procurement by the Army Signal Corps; Rome Air Materiel Area, Griffiss AFB, N.Y., which handles ground-based USAF electronic equipment; and the Navy's recently-set-up Bureau of Weapons. - GAO chief Joseph Campbell* has sketched some of the areas that will be covered in testimony before the congressional joint economic committee which is looking into general defense buying. Campbell indicated that the GAO is seeking to determine whether government orders for electronic equipment are being met with respect to quantity and quality, and delivery times and places. Furthermore, investigators want to know the extent, if any, of duplication and overlapping of procurement, inventory, production, and distribution. - On another front, electronics equipment manufacturers may benefit indirectly from the Pentagon's current review of procurement policies. The Defense Department aims to put more of its budgeted funds into weapon and spare parts buying while cutting down on purchases of nonessential goods. To this end, the Pentagon is studying a score of proposals for procurement savings. The underlying reason for the special attention to procurement policy is President Eisenhower's rigidly imposed ceiling of $41 billion on defense spending. Present indications point to maintenance of this spending level through June, 1962. The Defense Department is thus faced with a period in which the level of spending will remain constant while weapons costs are rising. This has already resulted in cutbacks and stretchouts in some weapons programs. Now Pentagon officials feel that one way to get more money for combat weapons and spare parts is to cut back on nonessential goods and tighten up on general buying practices. - The joint economic committee hearings have produced attacks on the volume of defense orders going to the West Coast, and on the use of negotiation instead of competitive bidding in awarding defense contracts. Sen. Jacob Javits (R-N.Y.) contended that "to many of us in the east, the so-called missile gap has been translated into the defense order gap" because of the loss of contracts in New York and other eastern states. Javits recommended that the Defense Department be required to place more set-asides for small businesses in labor surplus areas of the East. Cleco's A-1A Reversible Screwdriver—Nut-Runner is the BIG VALUE in small air tools... The plus value tool for assembling small precision equipment; equipment with tight working areas, for continuous production operations. This is a tool with no torque reaction; it is absorbed by a balanced impact mechanism. This is an economical tool. Throttle valve opens automatically when screw is engaged; air consuming "free running" is eliminated. High speed motor gives fast run down, improving production. This is the tool for operator comfort. Pencil-type design, (weight: 11 oz., and length: 6¼ ins.), give the operator complete freedom of movement, especially important on precision production jobs. Tool is muffled for quietness. Exhaust of the A-1A is directed away from the work. Power control is adjustable. Your Cleco® representative will be happy to give you a no-obligation demonstration of this versatile, powerful small tool. In development test... This directly-recorded Visicorder chart shows a canceller test of a number of letters through a new mail-handling machine developed by Emerson Research Laboratories for the U.S. Post Office Department. The Visicorder test took only 3 hours to solve a 3-week problem: Why letters changed speed as they went through the machine. Constant speed is necessary to register cancellation on the stamp every time. Motor speed variations, belt slippage, and letter slippage in the drive rollers were responsible. A synchronous drive motor, a timing-belt drive, and a better grade of rubber in the drive rollers were added to solve the problem at a vast saving in engineering time. The Emerson machine is designed to cancel 30,000 non-uniform letters per hour. It is under evaluation tests in the Post Office Department Laboratory, Washington, D.C. these are records of leadership In production... This comparison test of a production gyroscope was directly-recorded on a Model 906A Visicorder oscillograph by the test department of Whittaker Gyro, Van Nuys, Calif. Whittaker is a division of Telecomputing Corporation. The record shows how the Visicorder compares controlled angular velocities as a reference base to simultaneously-recorded variables, and how a dual static reference trace galvanometer simultaneously establishes a base line and a calibration line on the chart. In these and in hundreds of other scientific and industrial applications, Visicorders are bringing about new advances in product design, computing, control, rocketry, nucleonics and production. For information on how to apply the unlimited usefulness of the Visicorder to your specific problems, phone your nearest Honeywell Industrial Sales Office. The Honeywell Visicorder provides instantly-readable, high-sensitivity data at frequencies from DC to 5000 CPS. There are models with 8, 14, or 36-channel capacities. Reference Data: Write for Visicorder Bulletins 906A and 1012. Minneapolis-Honeywell Regulator Co., Industrial Products Group, Heiland Division, 5200 E. Evans Ave., Denver 22, Colorado AC Seeks and Solves the Significant—Since GM has pledged its resources to this nation's defense, AC plans to forge to the forefront in the international race for technological superiority. The resolution of scientific problems even more complex than Achiever inertial guidance—that's what AC now has on its agenda / This is AC QUESTMANSHIP. It's an exciting creative quest for new ideas, methods, components and systems . . . to promote AC's many projects in guidance, navigation, control and detection / Questmanship is readily apparent in AC Manufacturing, headed by Mr. Roy McCullough, AC Works Manager. His group "offers an outstanding challenge to engineers capable of understanding the most advanced scientific concepts . . . and developing the techniques and tools to implement those concepts on a production basis" / There may be a position for you on our specially selected staff . . . if you have a B.S., M.S. or Ph.D. in the electronics, scientific, electrical or mechanical fields, plus related experience. If you are a "seeker and solver," you should write AC's Director of Scientific and Professional Employment, Mr. Robert Allen, Oak Creek Plant, Box 746, South Milwaukee, Wisconsin. GUIDANCE / NAVIGATION / CONTROL / DETECTION / AC SPARK PLUG The Electronics Division of General Motors Look what Raleigh ships by Air Express in one day! Raleigh makes transcriptions—recordings of advertising commercials and pre-recorded programs to be broadcast from many radio stations. There's no margin for error—one slip-up and the sponsor doesn't get what he paid for. Only AIR EXPRESS gives Raleigh Records receipted, on-time delivery—overnight... coast-to-coast. The big difference is AIR EXPRESS dependability. It's the nation's most complete air-ground shipping service. One phone call arranges everything—and AIR EXPRESS rates are low. Use AIR EXPRESS—jet-age wings of modern marketing—and you're FIRST TO MARKET... FIRST TO SELL. CALL AIR EXPRESS DIVISION OF RAILWAY EXPRESS AGENCY GETS THERE FIRST VIA U.S. SCHEDULED AIRLINES The International Rectifier "Thyrode" Silicon Controlled Rectifier is a three-junction, hermetically sealed semi-conductor device that will block positive anode to cathode voltage as does a thyratron. When a signal is applied to its third (gate) lead, the device rapidly switches to a conducting state and provides the low forward voltage drop of a typical medium power silicon rectifier. Current flow may then be halted, by reversal or removal of the anode voltage. This simplicity of control makes the "Thyrode" applicable to a wide range of control and switching uses. \| Int'l Type Number \| Max. Rep. PIV, Volts \| RMS Input (Sin.) Volts \| Average Forward Current, Amps \| Surge Current (1 Cycle), Amps \| Min. Forward Breakover Voltage, Volts \| Max. Forward & Reverse Leakage, Ma \| Gate Power, Watts \| Gate Current, Ma, Peak \| Forward Gate Voltage, Volts \| Max. Forward Voltage, 1 Cycle, Volts \| \|-------------------\|---------------------\|------------------------\|-------------------------------\|-------------------------------\|--------------------------------------\|---------------------------------\|---------------------\|------------------------\|-----------------------------\|----------------------------------\| \| X10RC2 \| 20 \| 14 \| 10 \| 125 \| 20 \| 45 \| 22 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC3 \| 30 \| 21 \| 10 \| 125 \| 30 \| 40 \| 20 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC5 \| 50 \| 35 \| 10 \| 125 \| 50 \| 35 \| 18 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC7 \| 70 \| 50 \| 10 \| 125 \| 70 \| 30 \| 15 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC10 \| 100 \| 70 \| 10 \| 125 \| 100 \| 25 \| 12.5 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC15 \| 150 \| 105 \| 10 \| 125 \| 150 \| 13 \| 6 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| X10RC20 \| 200 \| 140 \| 10 \| 125 \| 200 \| 12 \| 6 \| 5 \| 0.5 \| 2000 \| 85 \| 10 \| 5 \| 1.25 \| \| Int'l Type Number \| Max. Rep. PIV, Volts \| RMS Input (Sin.) Volts \| Average Forward Current, Amps \| Surge Current (1 Cycle), Amps \| Min. Forward Breakover Voltage, Volts \| Max. Forward & Reverse Leakage, Ma \| Gate Power, Watts \| Gate Current, Ma, Peak \| Forward Gate Voltage, Volts \| Max. Forward Voltage, 1 Cycle, Volts \| \|-------------------\|---------------------\|------------------------\|-------------------------------\|-------------------------------\|--------------------------------------\|---------------------------------\|---------------------\|------------------------\|-----------------------------\|----------------------------------\| \| X16RC2 \| 20 \| 14 \| 16 \| 125 \| 20 \| 45 \| 6.5 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC3 \| 30 \| 21 \| 16 \| 125 \| 30 \| 40 \| 6.5 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC5 \| 50 \| 35 \| 16 \| 125 \| 50 \| 35 \| 6.5 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC7 \| 70 \| 50 \| 16 \| 125 \| 70 \| 30 \| 6.5 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC10 \| 100 \| 70 \| 16 \| 125 \| 100 \| 25 \| 6.5 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC15 \| 150 \| 105 \| 16 \| 125 \| 150 \| 13 \| 6 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| \| X16RC20 \| 200 \| 140 \| 16 \| 125 \| 200 \| 12 \| 6 \| 5 \| 0.5 \| 2000 \| 50 \| 10 \| 3 \| .90 \| FOR DETAILED TECHNICAL DATA, CIRCLE READER-SERVICE CARD NO. 19. SPECIFY THE LEADING LINE... All standard Controlled Rectifier types are available 'off the shelf' from strategically located International Rectifier Authorized Industrial Distributors and through our Industrial Representatives throughout the world. INTERNATIONAL RECTIFIER SYMBOL OF QUALITY IN SEMICONDUCTORS INTERNATIONAL RECTIFIER CORPORATION: EL SEGUNDO, CALIFORNIA • PHONE OREGON 8-6281 • CABLE RECTUSA BRANCH OFFICES: NEW YORK CITY: 1580 LEMOINE, FORT LEE, N. J., WINDSOR 7-3311 • SYRACUSE, NEW YORK: 2366 JAMES STREET, HEMPSTEAD 7-8495 • CHICAGO, ILLINOIS: 205 W. WACKER DRIVE, FRANKLIN 2-3888 • CAMBRIDGE, MASS., 17 QUINSTER ST., UNIVERSITY 4-6520 • ARDMORE, PENNSYLVANIA, SUBURBAN SQUARE BLDG., MIDWAY 9-1428 • BERKLEY, MICHIGAN: 1799 COOLIDGE, HIGHWAY, LINCOLN 8-1144 • CANADA: 1581 BANK ST., OTTAWA, ONTARIO, REGENT 3-6880 The New Ramo-Wooldridge Laboratories in Canoga Park ...an environment dedicated to technological research and development The new Ramo-Wooldridge Laboratories in Canoga Park, California, will provide an excellent environment for scientists and engineers engaged in technological research and development. Because of the high degree of scientific and engineering effort involved in Ramo-Wooldridge programs, technically trained people are assigned a more dominant role in the management of the organization than is customary. The ninety-acre landscaped site, with modern buildings grouped around a central mall, contributes to the academic environment necessary for creative work. The new Laboratories will be the West Coast headquarters of Thompson Ramo Wooldridge Inc. as well as house the Ramo-Wooldridge division of TRW. The Ramo-Wooldridge Laboratories are engaged in the broad fields of electronic systems technology, computers, and data processing. Outstanding opportunities exist for scientists and engineers. For specific information on current openings write to Mr. D. L. Pyke. THE RAMO-WOOLDRIDGE LABORATORIES 8433 FALLBROOK AVENUE, CANOGA PARK, CALIFORNIA FEBRUARY 12, 1960 • ELECTRONICS More Air Firms Go Electronic Reports this week indicate increased activity by aviation companies in electronics firms. - Chance Vought Aircraft, Inc., Dallas, Tex., announces expansion and holdings in automation through a purchase agreement signed with Information Systems, Inc., Skokie, Ill. The aircraft firm will acquire 80 percent of the Illinois company which manufactures readout systems. Information Systems, in an additional negotiation, is acquiring the assets of Panellit and of the Genesys Corporation, the latter being a Chance Vought subsidiary. The surviving company will be Information Systems which will operate as a subsidiary of Chance Vought. - Beech Aircraft, Dallas, is reportedly studying the possibility of entering the electronics field, either through its own R&D efforts or through the acquisition of companies now in electronics. The firm is also studying the possibility of acquiring companies which will complement its present area of operation. - California Eastern Aviation held a special directors meeting recently to discuss a 1960 program for operations in the fields of electronics, electromechanics, space research and other related fields. On the basis of contracts negotiated and in process, the company expects this year to exceed the $30 million gross sales reached in 1959. A five percent stock dividend was voted during January. - National Aeronautical Corp., Fort Washington, Pa., reports that recently completed expansions in manufacturing space and "vastly increased" engineering laboratories allow anticipation of a record year for 1960. For the fiscal year ended 1959, net sales were $6,100,000, an increase of 43 percent over fiscal 1958. Net earnings went from $370,000 in 1958 to $719,000 last year, a gain of 94 percent. Earnings were equal to 87 cents a share for 1959 compared with 45 cents a share in 1958. - Raytheon Co., Waltham, Mass. reports total profits after taxes in 1959 of $13,481,000, equal to $3.89 per share of common stock including a special gain equal to 88 cents a share. In making the announcement, the board of directors declared a five percent stock dividend payable March 18 to holders of record at close of business Feb. 24 this year. The special gain mentioned represents a reduction in the company's reserve for estimated liability for federal income taxes and for renegotiation with respect to government business in the past. - Certificates for Ampex stock will be mailed to shareholders next week following approval of a three-for-one split of the Redwood City, Calif. company's stock. ### 25 MOST ACTIVE STOCKS \| WEEK ENDING JANUARY 29 \| SHARES (IN 100'S) \| HIGH \| LOW \| CLOSE \| \|------------------------\|------------------\|------\|-----\|-------\| \| Gen Electric \| 825 \| 90¼ \| 84¼ \| 85½ \| \| Sperry Rand \| 742 \| 39½ \| 36¼ \| 39 \| \| Beckman Inst \| 714 \| 72 \| 66¼ \| 67½ \| \| A. B. DuMont \| 698 \| 9¾ \| 8 \| 8½ \| \| Elco & Mus Ind \| 606 \| 7½ \| 7 \| 7 \| \| RCA \| 669 \| 63¾ \| 59¼ \| 59½ \| \| Raytheon \| 548 \| 50½ \| 46¼ \| 47½ \| \| Varian Assoc \| 520 \| 45¼ \| 40¼ \| 41¼ \| \| Philco Corp \| 522 \| 36 \| 30 \| 31 \| \| Gen Tel & Elec \| 512 \| 80¾ \| 75¾ \| 75¼ \| \| Avco Corp \| 499 \| 14½ \| 13 \| 13 \| \| IT&T \| 476 \| 37¼ \| 34¼ \| 34¼ \| \| Reeves Snedrft \| 436 \| 11½ \| 10 \| 10½ \| \| El-Tronics \| 432 \| 1¾ \| 1¼ \| 1¾ \| \| Siegert Corp \| 389 \| 34½ \| 30¾ \| 33¾ \| \| Collins Radio \| 382 \| 62 \| 55¼ \| 56½ \| \| Ampex \| 350 \| 102¾ \| 97¾ \| 99 \| \| Victorele Inst \| 342 \| 13¼ \| 12 \| 12 \| \| Dynamics Corp Amer \| 326 \| 11½ \| 10¾ \| 10¾ \| \| Gen Dynamics \| 281 \| 49¾ \| 46¼ \| 46¾ \| \| Univ. Controls \| 274 \| 17¾ \| 16¼ \| 16¼ \| \| Westinghouse \| 264 \| 100¼ \| 98¼ \| 98¼ \| \| Lear Inc \| 258 \| 18¾ \| 17½ \| 17¾ \| \| Zenith \| 257 \| 105½ \| 97¼ \| 98½ \| \| Int'l Resistance \| 248 \| 23¾ \| 20¾ \| 20½ \| The share figures represent the volume of transactions stocks on the New York and American Stock Exchanges. Listings are prepared exclusively for ELECTRONICS by Ira Haupt & Co., investment bankers. ### DIVIDEND ANNOUNCEMENTS \| Company \| Amount per Share \| Date Payable \| \|-----------------------\|------------------\|--------------\| \| Amer Res & Dev Corp \| $.93 \| Feb. 23 \| \| AMP, Pacmor \| .20 \| Mar. 1 \| \| Atomic Fund \| .05 \| Feb. 23 \| \| Fedcor Co \| .25 \| Mar. 1 \| \| IBM \| .75 \| Mar. 10 \| \| Magnavox \| .25 \| Mar. 15 \| \| National Co \| .90 \| Apr. 1 \| \| Nucleon Chem & Elec \| .50 \| Feb. 29 \| FUTURA! CHART-PAK'S NEW DRAFTING FILM Saves Time — Saves Money Now Chart-Pak brings you "Tomorrow's Drafting Film Today"... an amazing new drafting medium on "Cronar."* REDUCED SMUDGING HEAT RESISTANT RESISTS CRACKING HIGH TEAR STRENGTH SUPERIOR TRANSPARENCY MOISTURE PROOF ERASES EASILY VERY FLEXIBLE WON'T YELLOW WITH AGE NO STRETCH OR SHRINK LIES FLAT DOESN'T "DOG-EAR" COSTS NO MORE THAN HIGH-GRADE CLOTH OR VELLUM Write Today for Full Information and Free Sample CHART-PAK, INC. ORIGINATOR OF THE TAPE METHOD OF DRAFTING 242 River Road, Leeds, Mass. DuPont Registered Trade Mark NEED PRODUCT INFO? LOOK IN THE NEW electronics BUYERS’ GUIDE You’ll find detailed facts about the products of almost 700 different advertisers — that’s 42% more than you’ll find in any other electronics directory. There’s also 64 pages of reference data about markets, materials and design — vital information for all working in electronics. Also local sales offices of manufacturers . . . the names, addresses and phone numbers of representatives . . . complete lists of manufacturers . . . registered trade names . . . and also, of course, the most complete listing of all electronic and related products. Tells what you want to know . . . when you’re ready to buy. A McGraw-Hill Publication • 330 West 42nd Street, New York 36, New York FEBRUARY 12, 1960 • ELECTRONICS Your move to reliability... CBS INDUSTRIAL TUBES The CBS tubes described offer you many moves to greater reliability and improved circuit performance. They represent just a few types from the rapidly growing CBS industrial tube line. For v-r... reliable miniature and larger allied receiving... krytron... secondary-emission... ultrahigh-resolution cathod-ray... frame-grid... and special-purpose industrial tubes, call your local sales engineering representative or your Manufacturers Warehousing Distributor. Ask for data. Many new CBS industrial tubes are being developed to help solve your design problems. Your CBS sales engineer will be glad to keep you posted. CBS ELECTRONICS A Division of Columbia Broadcasting System, Inc. Sales Offices: Danvers, Mass., 100 Endicott St., Spring 4-2360 Newark, N.J., 500 Washington Ave., Tel. 2-2789 Miami Park, Fla., 1990 N. Mainhenn Rd., Esteroak 9-2101 Los Angeles, Calif., 2120 S. Garfield Ave., Raymond 3-9081 Atlanta, Ga., Gary Chapman & Co., 672 Whitehall St., Jackson 4-7388 Minneapolis, Minn., The Heimann Co., 1711 Hawthorne Ave., Federal 2-5457 Available now—ceramic “extras” in more than 40 tube types SMALLER SIZE IMPACT SURVIVAL VIBRATION SURVIVAL EXACT DIMENSIONAL UNIFORMITY EXTREME HEAT SURVIVAL LOWER DIELECTRIC LOSS Superior performing Eimac ceramic negative-grid tubes and klystrons are available now for modern equipments. EITEL-MCCULLOUGH, INC. SAN CARLOS, CALIFORNIA Eimac First with ceramic tubes that can take it Cable address EIMAC San Carlos 24 CIRCLE 24 ON READER SERVICE CARD ABSOLUTE MAXIMUM RATINGS AT 25°C \| Parameter \| Value \| \|---------------------------------\|---------\| \| Forward Current \| If \| 50 mA \| \| Minimum Breakover Voltage \| Vbo \| { TSW-30 30V } { TSW-60 60V } \| \| Reverse Breakdown Voltage \| Vr \| { TSW-30 30V } { TSW-60 60V } \| \| Storage Temperature \| \| -65°C to 150°C \| \| Ambient Temperature Range \| \| -55°C to +125°C \| SPECIFICATIONS AND TYPICAL CHARACTERISTICS (At 25°C Unless Otherwise Stated) \| Parameter \| Typical \| Max. \| Test Conditions \| \|---------------------------------\|---------\|------\|-----------------\| \| Saturation Voltage \| Vs \| 1.0 \| 1.5 Volts \| \| Forward Leakage Current \| If \| 0.1 \| 10 μA \| \| Reverse Leakage Current \| Ir \| 0.1 \| 10 μA \| \| Forward Leakage Current \| If \| 20. \| 50. μA \| \| Reverse Leakage Current \| Ir \| 20. \| 50. μA \| \| Gate Voltage to Switch "ON" \| Vg On \| 0.7 \| 1.0 Volts \| \| Gate Current to Switch "ON" \| Ig On \| 0.1 \| 1.0 mA \| \| Gate Voltage to Switch "OFF" \| Vg Off \| 1.2 \| 4.0 Volts \| \| Gate Current to Switch "OFF" \| Ig Off \| 7.0 \| 10. mA \| \| Holding Current \| Ih \| 2.0 \| 5.0 mA \| SPECIALY DESIGNED FOR: - Miniaturized Memory Circuits - Ring Counters - Shift Registers - Controlled Rectifier Driver - Flip-Flop Equivalent - Simplified Information Storage - 0.3 m second Switching Transitron announces a NEW computer element for: Greater Reliability • Circuit Simplicity THE TRANSWITCH The TRANSWITCH is a new bistable silicon device that can be TURNED OFF with gate current. This PNPN latching device “remembers” its last gate signal. High current gain, both turn-on and turn-off, leads to greater circuit simplicity and inherent reliability. Excellent linearity of electrical parameters over a wide current range fulfills both low logic level and medium power needs. Here is a unique device that replaces TWO transistors plus resistors in most bistable circuits and permits increased component density. Furthermore, the transwitch is FAST... requiring only 0.3 microseconds to turn ON or OFF! The TRANSWITCH is now available from Transitron in the popular JEDEC TO-5 package, ready to solve your switch-on-switch-off requirements. For further information, write for Bulletin TE-1357A Transitron electronic corporation • wakefield, massachusetts “Leadership in Semiconductor Technology” F-M Radio Sales Look Up F-M radio set sales in 1960 should show a gain of at least 50 percent over last year, predicts Edward Shafer, market research manager of Blonder-Tongue Labs. He predicts factory sales of domestically produced f-m/a-m table sets will jump from 500,000 sets in 1959 to 750,000 sets this year. However, he also sees possibility that set sales will hit the million mark if foreign imports are included. They are currently running about 100,000 sets per year. EIA recently reported 480,894 f-m/a-m sets were produced through November 1959, almost 60 percent more than the first eleven months of 1958. Popularity of hi-fi gear and increasing interest in quality music are major factors supporting the rising sales curve. - Magnetic recording industry association claims 750,000 entertainment tape recorders with a retail value of $170 million will be sold in 1960. It calculates that 650,000 recorders valued at $140 million were produced in 1959. Four track tape, used for stereophonic recording, is the major influence behind the resurgence of magnetic tape recording business, says Herbert L. Brown, association president. - Raytheon's president, Charles F. Adams adds his voice to the growing list of industry leaders who are calling attention to the expanding market for non-military electronic products. Sales of consumer and entertainment equipment, which he currently estimates at $4 billion, will reach $8 billion in 1970, he says. At that time electronics industry factory sales will total $20 billion, he says. - Business and Defense Services Administration plans to issue in a few months its report on the proportion of shipments of electronic equipment and components originating in each of the 50 states and 20 large metropolitan areas. Study will show separate geographical breakdowns of shipments for total electronic industry products as well as for military, non-military equipment and major components. BDSA findings are based on a government survey of where shipments from over 600 large electronics firms originate and a variety of other government data. - This year, retail sales of TV sets may well reach 6.4 million, says Herbert Riegelman, general manager of General Electric's television receiver department. This estimate compares to sales of 5.8 million sets last year and 5.14 million sets in 1958. Riegelman's forecast leans on expectations of prosperous national economic conditions and recent product refinements which manufacturers count on to spur consumer demand. - Independent investigations by industry market analysts confirm our estimate that silicon controlled rectifier sales, which were $2 million to $3 million in 1959, will move up to the $4 million to $6 million level in 1960. (See Electronics, Jan. 1, 1960, p 22.) However, we are reminded that we slipped a digit in making our SCR unit estimate for 1959. It should have read 30,000 to 50,000 units instead of 300 to 500,000 units. available from inventory for immediate delivery—silicon transistors with ONE WATT POWER DISSIPATION @ 25° C in free air...five watts mounted in an infinite heat sink These Hughes PNP fused junction silicon transistors are especially recommended for small signal current gain, DC amplifier and other applications. They utilize a rugged coaxial package which is unequalled for mechanical strength...thereby providing you with dependable performance under highly adverse environmental conditions. Furthermore, they offer you the following electrical advantages over competitive devices: BV<sub>CEO</sub>, BV<sub>EBO</sub> and BV<sub>CBO</sub> are symmetrical • Lower leakage current • Controlled gain bandwidth • Lower saturation resistance • Higher operating frequency You receive still another benefit from these Hughes silicon transistors: proven reliability. Its reliability life tests at maximum operating temperature show a reliability factor of 10<sup>-4</sup> or only one failure in 10,000 transistor hours of operating life. Order today! These Hughes Transistors are available from inventory at both the factory and at all Hughes distributors. Just call or write your nearest Hughes Semiconductor sales office or distributor...or write Hughes, Semiconductor Division, Marketing Department, 500 Superior Avenue, Newport Beach, California. Specifications* \| Type \| BV<sub>CEO</sub> \| Min. \| Max. \| Max. V<sub>ce</sub> @ I<sub>c</sub> = 10mA \| Maximum I<sub>ceo</sub>* and I<sub>ebo</sub>* \| Typ.f<sub>ab</sub> (MC) \| \|--------\|-----------------\|------\|------\|------------------------------------------\|---------------------------------------------\|------------------------\| \| 2N1239 \| — 15V \| 14 \| 65 \| —0.2 \| —0.1µA \| 1.2 \| \| 2N1240 \| — 35V \| 28 \| 65 \| —0.2 \| —0.1µA \| 1.2 * \| \| 2N1241 \| — 35V \| 14 \| 32 \| —0.2 \| —0.1µA \| 1.2 \| \| 2N1242 \| — 60V \| 28 \| 65 \| —0.2 \| —0.1µA \| 1.0 \| \| 2N1243 \| — 60V \| 14 \| 32 \| —0.2 \| —0.1µA \| 1.0 \| \| 2N1244 \| —110V \| 14 \| 32 \| —0.2 \| —0.1µA \| 0.8 \| Transistors also available with same electrical characteristics in a JEDEC 30 (TO-5) package with a power dissipation of 400 mW Measured at 90% of breakdown voltage Creating a new world with ELECTRONICS HUGHES ©1960 HUGHES AIRCRAFT COMPANY SEMICONDUCTOR DIVISION ELECTRONICS • FEBRUARY 12, 1960 CIRCLE 27 ON READER SERVICE CARD 27 NEW ELECTRO INSTRUMENTS HIGH-SPEED, ALL-ELECTRONIC, ANALOG-TO-DIGITAL CONVERTERS 1000 measurements per second! Transistorized circuitry, one millisecond conversion rate, one megohm input impedance, automatic polarity, one digit sensitivity and resolution These new 7000 Series High-Speed, Analog-To-Digital Converters and Digital Voltmeters accept both positive and negative input voltages and produce binary coded decimal descriptions of their magnitude and polarity. This determination is arrived at by the successive approximation method. Bits are sampled as a function of an internal clock and are successively tried and accepted or rejected. Encoding time is always a fixed millisecond. Output can be applied directly to indicators for visual readout and also to auxiliary devices for controlling entry into recorders and computing systems. The constant encoding time and programmed ranging features make the 7000 Series ideal for systems applications. Ask your EI Sales Engineer for complete specifications. Electro Instruments, Inc. 3540 AERO COURT SAN DIEGO 11, CALIF. Digital instruments for measuring AC/DC voltages, AC/DC ratios, resistances, capacitance, and frequency • X-Y Recorders • DC Amplifiers Superb Stampings from the simplest, tiny washer to precision stamped Rocker Arms —and everything in between! We deliver superb stampings... anything from hub caps for baby carriages to vane locks for aircraft turbines—with tolerances well within the accepted standards of those industries. Point is, we deliver exactly what you want. That’s why we say our Stampings are “superb”. You’ll say it, too. How do we do it? By applying the best of our own three Methods: the “ONE PLUS” METHOD avoids tooling charges... the SHORT RUN METHOD employs our own temporary tooling... the PRODUCTION RUN METHOD uses special dies at nominal cost. Either way, you get exactly the Stampings you need—faster, better, and at surprisingly low cost. Our routine WATCH DOG SERVICE sees to that. QUANTITY? Anything from one to a million! Stampings DIVISION THE LAMINATED SHIM COMPANY, INC. 4602 UNION STREET, GLENBROOK, CONNECTICUT Interesting STAMPINGS FOLDER no. 3 contains helpful information for ordering Stampings most economically. Send for your copy. No obligation. from a single source a balanced line of electronic sealants and molding compounds POLYURETHANE... THIKOL... SILICONE... COAST PRO-SEAL & MFG. CO., long a recognized leader in research and manufacture of superior sealants for the aircraft industry, again assumes the lead in introducing a balanced line of electronic sealants and molding compounds...Thikols... polyurethanes...silicones. No longer must a company using a variety of potting and molding compounds seek a number of sources of supply. COAST PRO-SEAL & MFG. CO. has developed a complete line of products for specific needs...ENCAPSULATING, IMPREGNATING, CABLE SPLICING and MOLDING. These compounds, as with all Coast Pro-Seal products, have been laboratory tested and industry proven. Each has "built in" characteristics engineered for SPECIFIC APPLICATIONS—HIGH TEMPERATURE RESISTANCE... FUEL RESISTANCE... COLD FLOW RESISTANCE... and EXCELLENT ELECTRICAL PROPERTIES. The registered trademark of the Thikol Chemical Corporation Write for complete data. Pro-Seal representatives are in all major cities. Coast Pro-Seal & Mfg. Co. 2235 Beverly Boulevard, Dept. E Los Angeles 57, California Gentlemen: Please send us information regarding your balanced line of electronic industry sealants and molding compounds. Company............................................................... Attention.............................................................. Address.............................................................. City................................................................. Zone State THE ELECTRONICS MAN IDENTIFICATION R D P M 6XK-5XKW SQ WHERE TO FIND HIM The electronics man may be found in any or all of the areas of research, design, production, management. Your problem: sell him (wherever he is) and keep him sold all year long. Here's the simplified key to this job! Use electronics to arouse his interest and create acceptance for your products in the magazine's weekly issues. Use the electronics BUYERS' GUIDE and Reference Issue to be there all year long whenever he is ready to buy. This is the best selling combination in the electronics industry...and the one that carries the most weight! THE ELECTRONICS MAN "BUYS" WHAT HE READS IN... electronics A McGRAW-HILL PUBLICATION 330 West 42nd Street, New York 36, New York FEBRUARY 12, 1960 • ELECTRONICS fast, direct-reading, easy-to-use PANORAMIC LP-1a SONIC SPECTRUM ANALYZER 20 cps-22.5 kc. "Hoot-down" analysis of Gyro Motor by Model LP-1a Spectrum Analyzer. Area A shows decreasing fundamental frequency, resonant rise and decay, and vibration components over 60 successive scans in one minute. The Model LP-1a "quick-look" helps locate and evaluate discrete or random signals faster and easier by scanning the entire spectrum logarithmically from 40 cps to 20 kc. Once every second it automatically separates, measures and plots the frequency and voltage of waveform components on the calibrated X and Y axes, respectively, of a long persistence 5" CRT. For very detailed analysis, linear segments 40 to 5000 cps wide, centerable between 0 and 20 kc, may be magnified on the screen. Amplitude ratios of up to 40 db can be simultaneously measured. High sampling rate and panoramic displays assure 1. Minimum risk of missing weak signals or spectrum holes. 2. Fast measurements by eliminating slow point by point plots. 3. Simultaneous measurement of signals with widely divergent amplitudes and/or frequencies. 4. Continuous analysis of rapid changes in spectral content or design parameters. Proved in hundreds of research, design and production installations, the LP-1a is a valuable tool for Noise and Vibration analysis, Harmonic and IM measurements, General waveform studies, Spectral Power Density analysis, Response Curve Tracing. SUMMARY OF SPECIFICATIONS: Frequency Range: 20 cps—22.5 Kc. (1) Preset linear frequency scans: any segment width of 200, 1000, 5000 cps centerable from 0-20 Kc; Variable from 40 cps to 5000 cps with Auxiliary Function Unit C. (2) Preset Log Scan—40 cps to 20 Kc Frequency Scales: Linear and Log Center Frequency Control: Calibrated 0-20 Kc (used on lin scan) Dynamic Range: 60 db Amplitude Scales: Linear and 2 decade log (Expandable to 60 db) Sensitivity: 500 µv to 500 v for full scale linear deflection Voltage Accuracy: Lin Sweep (40 cps-22.5 Kc): ±5% or ±0.5 db Log Sweep (40 cps-20 KC): ±10% on lin ampl. scale, ±1.5 db on log ampl. scale. Scan Rate: 1/sec., internally generated; adjustable with accessory equipments Resolution: For log scan, automatically optimized. For lin scan, preset 30, 75 and 170 cps at 200, 1000 and 5000 cps sweep-width, respectively. Variable from 10 cps to 1 kc with Auxiliary Function Unit C. Write, wire or phone NOW for detailed information and specifications on the Model LP-1a. Ask for Vols. 1, 2 and 5 of the "Panoramic Analyzer" which contains applications and techniques of Spectrum Analysis, and get on our regular mailing list for this helpful publication. NEW CATALOG DIGEST AVAILABLE 530 So. Fulton Ave., Mount Vernon, New York Phone: OWens 9-4600 Cables: Panoramic, Mount Vernon, N. Y. State DESIGNING A NEW INSTRUMENT? YOU CAN USE A TUBE LIKE THIS! SPECIFICATIONS FOR 3" CATHODE-RAY TUBE FOR MODERN INSTRUMENTS JI-22 CONTACT POST ACCELERATOR Deflection factors: D1D2: 12 to 16 volts DC/in. D3D4: 9 to 12 volts DC/in. Focusing voltage: 25 to 125 V. Post accelerator voltage: 1500 V. Du Mont, the pioneer in space-saving and transistor-powered cathode-ray tubes for radar, now makes possible the same inherent features in a fine instrument read-out tube. The Du Mont electrostatically deflected K1951 provides full scan with deflection voltages of 9-15 volts DC/in. If your cathode-ray tube applications call for even greater compactness and power savings—consult the CRT Engineering Specialists at Du Mont. Daily advances in the state-of-the-art are being recorded for your benefit. A tube to fit your exacting requirements can be designed, developed and produced at Du Mont. Whatever your CRT requirements, check with Du Mont first. WRITE for complete details precision electronics is our business ELECTRONIC TUBES/INDUSTRIAL TV/MILITARY ELECTRONICS/MOBILE COMMUNICATIONS/SCIENTIFIC INSTRUMENTS/AUTOMOTIVE TEST EQUIPMENT ALLEN B. DU MONT LABORATORIES, INC., CLIFTON, N. J., U. S. A. INTERNATIONAL DIVISION • 515 MADISON AVENUE, NEW YORK 22, N. Y. • CABLES: ALBEDU, NEW YORK TRIMPOT® MODEL 220 As many as 17 of these compact units can be mounted in a space of just one cubic inch. Designed for printed circuits and modular assemblies, Trimpot Model 220 measures less than 3/16" x 5/16" x 1". Power rating is 1 watt and maximum operating temperature is 175°C. This Potentiometer meets or exceeds Mil-Specs for humidity, salt spray, fungus, sand and dust, as well as acceleration, vibration and shock. Self-locking 15-turn shaft insures sharp, stable settings...exclusive Silverweld® fused-bond termination and ceramic mandrel provide extreme temperature stability. The Model 220 is available in a wide variety of resistance ranges and a choice of two terminal types—gold-plated Copperweld wire or insulated stranded leads. Stocked by leading electronic distributors across the nation, these units are ready for immediate delivery. Write for complete technical data and list of stocking distributors. AVAILABLE AS PANEL MOUNT UNIT (illustrated at right) with same specifications. Exclusive manufacturers of Trimpot®, Trimit®, Pioneers in potentiometer transducers for position, pressure and acceleration. SIZE 11 SYNCHRONOUS MOTOR Featuring pull out torque efficiency of 50% nominal with 3.4 watts input and 3 watts pull out power, this synchronous motor represents a major achievement in terms of performance for a unit of this extremely small size. Additional advantages made possible by Kearfott's unique design include resistance to environmental extremes, lightweight construction and low unit cost. This motor and its variations are available in production quantities. TYPICAL CHARACTERISTICS R172 \| Excitation: Phase 1 \| Phase 2 \| \|---------------------\|---------\| \| Voltage \| 40V \| 40V \| \| Frequency \| 400 CPS \| 400 CPS \| \| Power \| 2.3 Watts \| 2.3 Watts \| \| Current \| 0.157 Amps \| 0.157 Amps \| Performance: - Synchronous Speed: 8900 RPM - Stall Torque: 0.2 In. Oz. - Pull Out Torque: 0.35 In. Oz. - Pull In Torque: 0.15 In. Oz. Write for complete data. FERRITES Kearfott's Solid State Physics Laboratory formulates, fires and machines permanent magnet ferrite materials of various compositions. Typical high-efficiency array utilizes Kearfott PM-3 ferrite material with specially designed pole pieces to produce a design both smaller and lighter than other arrays of equivalent magnetic field strength. Because magnets may be custom engineered to specific requirements, user is not restricted to stock magnet types, thereby providing greater latitude in parameters for focusing arrays. Pole pieces may also be provided according to specification, with the added assurance that, because of special Kearfott design techniques, B axial magnetic fields approximately 10% higher than those generally obtained in standard types may be produced. TYPICAL CHARACTERISTICS \| Peak Magnetic Field Strength \| 1200 gauss \| \|------------------------------\|------------\| \| Period \| 0.560 in. \| \| Length \| 5.64 in. \| \| Inside Diameter of Pole Pieces \| 0.400 in. \| \| Outside Diameter \| 2.0 in. \| \| Weight \| 3.2 pounds \| Write for complete data. ROTARY SWITCH Kearfott's rotary switching devices for missile and aircraft systems are used to sequence or switch circuitry as a function of time or shaft position. Used in conjunction with sensitive relays or solid state switching techniques, high current loads can be handled. These switches consist primarily of shaft assembly and bearing mounted cylinder divided into conducting and non-conducting segments with continuous track for common input. Multiple conductor "broom" type brushes ride on each cylinder track while number of tracks and segmentation of each is function of the number of circuits and type of "on-off" sequencing required. TYPICAL CHARACTERISTICS P1280-11A Number of switching tracks: 2 Angular Segmentation (both referenced to 0° start): - Track 1 — Non-conducting about 0° + 50° - Track 2 — Conducting 0° -180° Non-conducting 180° -0° Mechanical Accuracy of Segmentation: ±1" (better as required) Starting and Running Torque: 0.1 oz.-in. Current Capacity: 50 ma @ 28V/Brush (suitable for any sensitive relay or solid state switching circuits) Write for complete data. Engineers: Kearfott offers challenging opportunities in advanced component and system development. Two machines can do twice the work of one. Sometimes. With most analog recorders you can only record 24 minutes of 100 kc data on a 14-inch reel of 1-mil tape. The new Ampex FR-600 will record 48 minutes. Same data. Same reel. The reason: greater bandwidth at a given speed. 125 kc at 30 ips for instance. The benefits of this are worth considering. Most data runs these days average out at 30 minutes or more. With conventional equipment this means you need a stand-by recorder to pick up where the first leaves off. Or a dual transport set-up. Or a special machine with 19-inch reels. All expensive. Sometimes just for a few extra minutes of recording time. The FR-600 eliminates all this by doubling the recording time for any given bandwidth. It literally does the work of two conventional machines. You get the extra-wide bandwidth of 250 kc at 60 ips to boot. Nice for special applications. And every FR-600 tape is machine-to-machine compatible without lifting a tool. Good reasons why the FR-600 is the most versatile recorder you can use. A word from you will bring the full story. AMPEX DATA PRODUCTS CO., 934 Charter St., Redwood City, Calif. This instrumentation recorder can do the work of two. Microelectronics Moving Fast Engineering on the molecular and single-crystal level promises to increase reliability, cut production costs MICROELECTRONIC ENGINEERING is moving rapidly out of the research phase and into the realm of applications, paced by the space program's urgent need for smaller electronic systems. The new field of engineering, which deals with electronic phenomena at the molecular and single-crystal level, was the hot topic of discussion at the International Solid-State Circuits Conference, held in Philadelphia this week. Applications of microelectronics could radically change both design and production techniques, and may have a profound effect on the business future of producers of small components. Late last month Westinghouse Electric, one of many organizations working in the new field, revealed its new microelectronic techniques and disclosed some dramatic laboratory results achieved under USAF contracts. Westinghouse scientists recently discovered how to grow multizoned semiconductor crystals in ribbon form (dendrites) and are rapidly following up on applications with Air Force support. Scientists believe the ability to grow multizoned crystal and to perform operations on the crystals as they grow in the furnace can be combined to achieve near-automatic production of electronic systems. The company says it eventually may be able to grow equipment as complex as present radio receivers and amplifiers. Production Is Contract Aim Following the company's disclosure of its work with multizoned semiconductor crystals, the Air Materiel Command's Aeronautical Systems Center at Wright-Patterson AFB, Ohio, announced the award of a contract to Westinghouse to develop production processes for dendritically grown silicon and gallium arsenide crystals. A number of solid-state phenomena are employed in the molecular approach, including Seebeck generation, Peltier cooling, Hall-effect multiplication and the growing of $p-n$ junctions. Interfaces and junctions for the various effects can be grown into the dendrite as it is drawn from the furnace. The final result is a single function-block composed entirely of semiconductor material and doing the work of a circuit or subsystem. System Design Analysis of the system requirements establishes the functions to be performed by the block. After this analysis, a topologist determines the arrangement of the domains and interfaces that will most efficiently control the energy flow in the block. Production of a block starts with a basic semiconductor wafer. Necessary domains and interfaces are produced by using techniques employed in the production of conventional semiconductor devices: diffusion, plating, electron-beam machining, etching, irradiation, alloying, cutting and photographic processing. The finished block is encapsulated for protection against shock, vibration, temperature changes and ionizing radiation. One of the simpler function blocks described by the firm uses domains and interfaces to obtain a d-c power supply from line-power input (see cut). It employs the Seebeck effect for the thermoelectric generation of electricity, first converting 110 volts a-c into heat in a resistive domain. The thermoelectric domain is electrically but not thermally isolated from the resistive domain by an insulating layer. Heat produced in the resistive domain is converted into 9 volts d-c in the thermoelectric layer. A conventional power supply requires five components—a transformer and diode, plus the inductor and two capacitors of a pi-section filter—to do the same job. Voltage output from the molecular rectifier has no ripple. Function Blocks* Westinghouse received a $2-million contract from Air Research and Development Command last spring to develop en bloc circuits. Eight classes of function blocks resulted from the contract: a 5-w audio amplifier, a two-stage video amplifier, a tuned amplifier, various multivibrator circuits, multiposition switches, a variable potentiometer, a two-stage cooler employing the Peltier effect, and an analog-to-digital converter. In a status report to the Department of Defense in late January, the firm demonstrated over twenty working subsystems. Important among these was a tuned amplifier which uses a semiconductor notch filter and requires no inductance coils. The amplifier is highly selective to frequency. Changing the potential applied to the amplifier permits tuning it over a range far wider than the broadcast band. May Solve Problems Colonel W. S. Heavner, chief of Wright Air Development Center's electronic technology laboratory at Wright-Patterson AFB, remarks in the report: "It appears that the majority of present-day military electronic equipment requirements can be satisfied with molecular electronics." Other defense agencies are also known to be keenly interested in microelectronic engineering as a potential solution to problems of reliability, space limitation, heating, and production cost. Many companies and research establishments are actively engaged in pursuing solid-state approaches to microminiaturization. During the 1960 International Solid-State Circuits Conference this week, scientists from Bell Telephone Laboratories, Westinghouse, Fairchild Semiconductor, General Electric, Motorola, Texas Instruments, Stanford Research Institute and Massachusetts Institute of Technology have been discussing their work in this field. It is expected that at least one other firm will demonstrate its progress in microelectronics within the next few months. Observers say that the success of any approach to microminiaturization will depend on the cost, reliability and repeatability of the technique as compared to conventional processes. Westinghouse figures its dendrite crystal-growing offers a jump-off to lower production costs and higher reliability, and claims an acceptably high level of repeatability for the process. Other manufacturers, without giving details, indicate that their approaches to microelectronics will give the same results. One factor common to all methods of reducing size and weight is the elimination of conventional component parts. Low-power conventional resistors, capacitors, diodes and transistors of today appear to have no place in these new circuit design techniques. Military experts feel that conventional components will not be supplanted completely, but that need for components as we now know them will diminish as more manufacturers begin to produce circuits directly from raw materials. Although hailed as revolutionary by some, solid-state approaches to microelectronics are actually evolutionary. Some of the new techniques are adaptations of techniques used in thin-film work, and others are adapted from conventional transistor technology. Increased knowledge of materials and better production methods are also contributing to microelectronic engineering. Miniaturization means only active components G-E subminiatures use to give miniaturization HOFFMAN TACAN: MEETS 780%-HIGHER AGREE* SPEC! Advisory Group on Reliability of Electronic Equipment, which has set a new standard of 150 hours mean-time-to-failure for TACAN. General Electric subminiature tubes with heat-resistant glass have played a key role in advancing the reliability of Hoffman Electronics Corporation’s new ARN-21C to nine times that of older TACAN equipment. Compactness is a feature...transmitter, receiver, and electronic computer functions all are grouped in one “black box” that measures only 8 by 11 by 17 inches. Heat build-up necessarily is substantial. In General Electric subminiature tubes, Hoffman found the answers to their pressing need for tubes that would stand up to heat with no sacrifice in reliability. 28 G-E subminiatures are used in the ARN-21C. heat. Tubes are the that can withstand heat. new heat-resistant glass with reliability. WHAT TESTS SHOW Life tests of G-E subminiatures with new heat-resistant glass prove that high-temperature operation has no adverse effect on reliability. Check the total absence of failures with G-E type 6021 after 1500 hours at 250 C, against the high failure rate of ordinary 6021's under the same conditions! SPECIFY FOR RELIABILITY Small...smaller...smallest! The trend in electronic equipment is down in size, up in reliability requirement. Heat is the gremlin ready to play hob with your circuitry, unless you take steps to specify devices that are completely dependable when temperatures rise sharply. General Electric subminiature tubes are proved performers at high temperatures. And their small dimensions give you the extra compactness you need to meet tight equipment size limitations. Telephone your nearest General Electric Receiving Tube Department office below! NEW YORK Wisconsin 7-4065,6,7,8 CHICAGO SPring 7-1600 LOS ANGELES GGranite 9-7765 Progress Is Our Most Important Product GENERAL ELECTRIC Type 6021 is a key tube in Hoffman's TACAN circuit. Glass electrolysis—cause of 90% of tube failures at high temperatures—has virtually been eliminated by General Electric in the 6021 and other subminiature receiving tubes. Air Agency Plans for '60 YEAR-OLD Federal Aviation Agency, despite a lot of time spent putting out brushfire emergencies, has managed to build a firm framework under the nation's airways control system, and has a raft of big and little plans for 1960. FAA's fiscal 1960 program was based on an appropriation of $118.2-million for establishment of facilities and $48.7 million for R&D. Of this amount, approximately 75 percent of purchases and undoubtedly more of the R&D money is going for electronics. For the coming fiscal year, the administration has requested $65 million for research and development projects and $195 million for acquisition of new facilities or improvement of old ones. Of the $260 million total, approximately the same percentages will go for electronics. The Agency's program concentrates heavily on air navigation facilities, traffic-control systems, and special-purpose buildings, with special attention to radar. Radar and Controls The program for the current fiscal year calls for eight long-range radars at an average cost of over $2.4 million. Three Air Defense Command radars will also be made available for joint use by FAA controllers; in these joint-use situations, FAA pays only for remoting the ADC radar data to FAA control facilities. Sixteen long-range radars will be fitted with scan-conversion equipment, costing $375,900 each; and 24 will be equipped with radar beacons at $106,400 each. Twenty airports will get new traffic-control towers. New instrument-landing systems costing $195,900 will go into 15 airports, four of which will use them for training purposes. Sequenced flashing approach lights will be installed at 54 airports at a cost of $36,200 per system. Terminal-type vhf omnirange (TVOR) equipment will be installed at 18 locations; this special type of VOR is used as an approach aid, where standard VOR is used for enroute navigation. TVOR systems cost $100,600 on an average. Another specialized type of vhf omnirange—Doppler VOR—which minimizes interference from terrain obstacles and buildings, will be installed at 20 locations for $54,900 apiece. A total of 32 direct air-ground communications channels will be established at 19 locations; average cost per channel is $108,700. The Agency will spend $5.6 million to improve teletypewriter service (used for weather and traffic-control data) at 500 locations. An additional $3-million-plus will go toward improving communications in the Caribbean, the Pacific and Alaska. New instrumentation for use in flight-checking the accuracy of nav-aids and traffic controls will be installed in 14 flight-inspection aircraft. Average cost will be $345,475 per aircraft. In actions during the calendar year 1959, besides taking over from Civil Aeronautics Administration, FAA commissioned 365 new navigation aids or traffic-management facilities, an average of one a day. The Agency installed medium-scale computers with large-volume memories at six of the busier traffic-control centers and hooked up four of these to communicate with each other. To keep pace with jet operations, FAA set up a total of 25,455 miles of high-altitude jet routes. With the help of 38 air defense radars which supplemented FAA's own long-range radars, almost the entire jet route structure was placed under radar advisory. At the present time, FAA can track all jet flights from takeoff to touchdown except for a small area in the Rockies. First major action this year was the announcement that almost all transport-category aircraft used by commercial airlines to carry passengers must be equipped with airborne weather radar. The rule excludes the Curtiss C-46 and two non-transport craft, the Douglas DC-3 and Lockheed L-18. 1961 Programs FAA's request of $195 million for new facilities and $65 million for R&D represents less than half the total $654 million in new obligation authority requested for the Agency in the President's 1961 budget. The rest of it is eaten up by expenses, and by construction, operation and maintenance costs for the Washington airports. Included under expenses is a request for 4,547 new employees—primarily technicians, and specifically including more traffic controllers, engineers and electronics mainteELECTRONICS learns that FAA will continue to stress electronics in purchases and establishment of facilities during fiscal 1961. New facilities will be mostly—in order of importance—air traffic-control towers, short-range nav-aids like Vortac (VOR with Tacan-compatible distance-measuring equipment), instrument-landing systems, runway-approach lights, including sequenced flashing lights, long-range radars, radar beacons, terminal-area radar equipment like ASDE (airport surface detection equipment) and PAR (precision approach radar). NAFEC Projects* Most of the $65 million in research and development authority will go to or through the National Aviation Facilities Experimental Center for its development and evaluation programs. More than 70 projects are on the center's agenda for the coming year. The first components of a fully automated air-traffic control system are already installed at the Atlantic City, N. J., test facility. General Precision Laboratory's air-traffic computer, the heart of the system, is now undergoing test. The engineers and scientists at the center will also be looking at automatic landing systems, including Bell Aircraft's AN/GSN-5 (beginning this month) and REGAL later this year. Among radar systems undergoing test will be an experimental quad radar, so called because it combines in one system the four functions of surveillance, precision approach (PAR), airport surface detection (ASDE) and height-finding. The year at NAFEC will also be marked by extension of the space-position range, the precision measurement facility for determining the position of aircraft over the range. Two phototheodolites are currently installed on temporary earth mounts; these will be moved to steel towers and additional units will be installed. The center is putting in a MOPTAR system (multiple-object precision tracking and ranging), a radio complex that can simultaneously track five aircraft with extreme precision up to 200 miles out and derive precision data. A special porcelain body is used in the production of Lapp Resistor Cores. It provides a flawless surface of such nature as properly to receive a uniform deposit of carbon or boro-carbon. It also has a temperature coefficient of expansion matched to that of the deposited film...to provide a constant resistance against temperature change. These resistor cores are produced in close tolerances for straightness, roundness and length...they reflect the same quality of workmanship and materials long associated with Lapp. Write for complete information on Lapp Resistor Cores. Lapp Insulator Co., Inc., Radio Specialties Division, 163 Sumner St., LeRoy, N. Y. People believe most completely in the things that work best for them. That's why 52,000 readers of electronics pay more than $300,000 each year to get the information it gives them. Use electronics to create interest and acceptance for your products, materials or services. The electronics BUYERS' GUIDE gets you there when your customers are making buying decisions... actually sells for you. Exclusive! There is clear evidence that the "GUIDE" carries the most weight: it has 42% more advertisers than any other electronics directory. PICK THE ONE THAT CARRIES THE MOST WEIGHT CLEVITE'S NEW SPACESAVER TRANSISTOR Compared with present power transistors of similar ratings, the new Clevite Spacesaver gives you important new advantages. Better Switching — Its low base resistance gives lower input impedance for the same power gain and lower saturation resistance, resulting in lower "switched on" voltage drop. Its lower cut off current means better temperature stability in direct coupled circuits (such as regulated power supplies) and a higher "switched off" impedance. Better Amplifying — Improved frequency response leads to higher audio fidelity, faster switching and improved performance in regulated power supply applications. Better Mounting — The Spacesaver's simple rectangular configuration and low silhouette make it adaptable to a wide variety of mounting requirements where space is at a premium. In aircraft and missile applications, its low mass (half present type) improves shock and vibration resistance of lightweight assemblies. --- THREE AMPERE SWITCHING TYPES \| TEST \| CTP 1728 \| CTP 1735 \| CTP 1729 \| CTP 1730 \| CTP 1731 \| CTP 1736 \| CTP 1737 \| GTP 1733 \| \|-----------------------------\|----------\|----------\|----------\|----------\|----------\|----------\|----------\|----------\| \| Min BVcbo @ 2 ma (volts) \| 40 \| 60 \| 80 \| 100 \| 40 \| 60 \| 80 \| 100 \| \| Min BVceo @ 500 ma (volts) \| 25 \| 40 \| 55 \| 65 \| 25 \| 40 \| 55 \| 65 \| \| Min BVces @ 300 ma (volts) \| 35 \| 50 \| 65 \| 75 \| 35 \| 50 \| 65 \| 75 \| \| Max Icbo @ 90° C @ Max Vcb (ma) \| 10 \| 10 \| 10 \| 10 \| 10 \| 10 \| 10 \| 10 \| \| Max Icbo @ 2 V (µa) \| 50 \| 50 \| 50 \| 50 \| 50 \| 50 \| 50 \| 50 \| \| D.C. Current Gain @ 0.5A \| 30-75 \| 30-75 \| 30-75 \| 30-75 \| 60-150 \| 60-150 \| 60-150 \| 60-150 \| \| Max Veb @ 3.0 A (volts) \| 1.5 \| 1.5 \| 1.5 \| 1.5 \| 1.5 \| 1.5 \| 1.5 \| 1.5 \| \| Max Vce (sat) @ 3.0A,300 ma (volts) \| 1.0 \| 1.0 \| 1.0 \| 1.0 \| 0.8 \| 0.8 \| 0.8 \| 0.8 \| \| Min fae @ 3.0 A (kc) \| 20 \| 20 \| 20 \| 20 \| 15 \| 15 \| 15 \| 15 \| \| Max Thermal Resistance (c/w)\| 2.5 \| 2.5 \| 2.5 \| 2.5 \| 2.5 \| 2.5 \| 2.5 \| 2.5 \| --- Phone for data and prices. CLEVITE TRANSISTOR 254 Crescent Street Waltham 54, Mass. Tel: TWinbrook 4-9330 ELECTRONICS · FEBRUARY 12, 1960 CIRCLE 43 ON READER SERVICE CARD 43 AMP taper technique points the way to greater reliability Magnetic Amplifiers, Inc., of New York carefully manufactures its Static Inverters with a step-by-step quality control and testing program to build in the reliability required for aircraft and missile applications. It found that AMP Taper Technique simplified this procedure. A high speed AMP Automachine pre-terminates circuit leads with crimp-type, pre-insulated solid Taper Pins. Components are then easily tested in the modular stage before final assembly. Crimping eliminates difficult soldering operations and the danger of burning wound components while Taper Technique permits checking and trouble shooting without destroying the main cable. After final assembly, when the Pins are inserted into the Blocks, this Technique provides rugged vibration resistance and operational reliability. AMP solderless Taper Pins are made in formed and solid types, with or without pre-insulation and mate with a wide range of one or two piece stackable Taper Blocks. You'll find that AMP Taper Technique is ideal for your quality control or circuit density problems too. Write for our new Taper Technique brochure. AMP INCORPORATED GENERAL OFFICES: HARRISBURG, PENNSYLVANIA AMP products and engineering assistance are available through subsidiary companies in: Australia • Canada • England • France • Holland • Italy • Japan • West Germany Air Force has just let a contract for a feasibility study of X-rays as a possible means of space communication. It's part of a look at the whole spectrum for transmission possibilities. Taking several giant steps across the electromagnetic spectrum, the Air Force is investigating X-rays as a possible means of space communication. The feasibility study by Tracelab-Keleket of Waltham, Mass., is part of an overall investigation by the Air Research and Development Command of the entire spectrum from a communications standpoint. Rome ADC has let a $75,000 contract for the study, part of its inquiry into the possibility of using radiation in the transvisual frequencies as supplemental media for conveying information. X-rays are only mildly scattered, can be made highly directional, and so potentially offer the advantage of privacy for space communications. The power needed to swamp them would be prohibitive, so they also offer antijamming advantages. At the present time, power requirements for X-ray transmissions are felt to be about the same as for radio transmission—or a bit higher, according to Joris M. Brinkerhoff, head of the physics department, and Charles A. Ziegler, staff scientist, at Tracelab. Low-energy X-rays—probably below 10 kev—will be used, requiring no more shielding than would otherwise be needed in a manned space vehicle. Lower altitude limit for effective use of this type of communication is believed to be about 30,000 feet. As the most practical experimental altitude, researchers will concentrate on communications in space at 100 kilometers and above. Pulsing May Be Key At the frequencies involved, the quantum nature of electromagnetic radiation becomes important. Since counting of quanta is the basic method of detection, there may be an advantage in pulsing the power source. Unlike radio waves, the quanta can be detected instantaneously, and this gives X-ray pulsing an edge over pulsing of radio waves. In fact, there are indications that pulsing may be the key to practical X-ray communication. Conventional nuclear radiation detectors like ionization chambers and scintillation counters will be modified for receivers. New electronic circuits will be required for demodulation in the context of pulses, no sine waves being involved in the transmission technique. In addition to cosmic background, another kind of noise is involved in X-ray transmission. This is an uncertainty, arising from quantum statistics, as to whether or not a bit of information has been received. In fact, researchers anticipate that the predominant noise in X-ray transmission could arise from the uncertainty in the arrival rate of quanta. As in radio transmission, a compromise will be necessary between noise level and rate of message-sending. In radio, recourse to a wide band lets in more noise. Prospective modes of conveying intelligence are analogous to radio transmission. Intensity modulation, by which there is produced a variation in the rate of arrival of quanta at the detector, is comparable to a-m. Energy modulation, accomplished by changing the energy of the X-rays or electrons, may be considered analogous to f-m. In this technique, if an electron beam is used, it will strike one element and that element's characteristic line of radiation will be detected; then another target will emit a different line of radiation. Scrambling Techniques A combination of intensity and energy modulation will be considered; also scrambling techniques, such as sending a special pattern of quanta. The study will also embrace conditions in outer space, particularly with reference to possible interference by natural phenomena. X-rays are subject to interference only by gross objects which absorb them, while electrons are bent by magnetic fields. On the other hand, the certainty of X-ray reception may be adversely affected by large fluctuations in ambient level, such as that caused by solar flares. Efficiency of transmission—bit-per-second-per-watt—will be investigated along with other factors. After the study, Tracelab may select one or two possible systems of instrumentation and build a local short-distance demonstration model to evaluate various modulation and demodulation techniques. "Old timer" or "new hand" at designing transistorized switching-circuits, you'll find useful information in this 24-page Sylvania booklet now available at no charge! "Sylvania Medium and High Speed Switching Transistors" contains practical data that will help you select the best transistor-type for a specific application. Typical switching circuits, with parts values, are included. Absolute maximum ratings and electrical characteristics under minimum, typical and maximum conditions are tabulated for many of the SYLVANIA NPN and PNP SWITCHING TRANSISTORS frequently specified for electronic computers. Given, too, are mechanical specifications for the TO-5 case in which they are supplied. Briefly described are the extraordinary SYLVANIA quality-controls and production-techniques that assure reliable performance of Sylvania Switching Transistors under severe environmental conditions. Send for your copy today. Specify the "Sylvania Switching Transistor Booklet." Write: SYLVANIA SEMICONDUCTOR DIVISION, Dept.22-2-B, WOBURN, MASS. SYLVANIA Subsidiary of GENERAL TELEPHONE & ELECTRONICS CIRCLE 46 ON READER SERVICE CARD FEBRUARY 12, 1960 • ELECTRONICS Now from Weston— a new package concept for AC-DC laboratory standards time-saving, easy-to-operate calibration consoles—accurate within .05% in D.C. 1% in A.C. There's remarkable simplicity and precision in these Weston AC and DC Calibration Consoles — design simplicity that is readily apparent in clean, uncluttered panels... operating simplicity that now enables one operator to perform the work in ½ the time previously required by two skilled operators... with the precision of Weston engineering and craftsmanship that assures ultimate accuracy. Highly stable and readily adjustable, these consoles are truly value-packed. Unique switching networks automatically divide scales into any number of equal parts regardless of range. It's no longer necessary to calibrate each checkpoint or set up extra equipment to obtain required outputs. Operation is reduced to four or five simple steps. These consoles typify Weston leadership in long-term reliability and stability. All are supplied in blue-gray cabinets or ready for mounting in standard 19" rack panels. Also available: an oil bath which accommodates up to 12 Weston Normal Standard Cells... maintains temperatures within .01C. Removable racks also fit baths at U.S. Bureau of Standards. For full information, contact your local Weston representative... or write to Daystrom-Weston Sales Division, Newark 12, N. J. In Canada: Daystrom Ltd., 840 Caledonia Rd., Toronto 19, Ont. Export: Daystrom Int'l, 100 Empire St., Newark 12, N. J. WESTON Instruments WORLD LEADER IN MEASUREMENT AND CONTROL Visit us at the Daystrom Booth IRE Show N. Y. Coliseum March 21-24. Hams Seek Facsimile Military Affiliate Radio System hams and makers of communications equipment see new products coming from need for sending visual data. Facsimile and slow-scan television are now being considered as possible supplements to the regular civilian ham broadcast activities of the Military Affiliate Radio System (MARS). MARS, now 12 years old and counting a record 5,000 radio amateurs as members, is designed to function as an emergency network for military communications. The system has an active training and education program, which is carried out through various Army headquarters. Right now, interest is focusing on equipment to transmit visual information in the amateur frequencies. E. S. Piller, director of the First Army's MARS technical education program, told ELECTRONICS that new developments in facsimile and slow-scan television would be valuable to the system. He said there was a definite need for transmission of visual material. Cost of visual transmission equipment is the stumbling block. Facsimile gear to transmit photos and drawings would cost the amateur operator about $1,800. Receiving equipment can be bought for about half this amount. Some hope that lower equipment costs can be achieved is expressed by manufacturers. John Long, president of Long Laboratories, Emerson, N. J., for example, says that the cost of facsimile equipment could be brought down if stepped-up market demand brings about increased production. Most of today's facsimile equipment is used by government agen- Inspecting Smog Detector Orange County, Calif., officials have outfitted a step-van truck with $15,000 worth of instruments for measuring air pollution. Monitors, developed by Beckman Instruments, include four air-pollution analyzers (picture) which measure low concentration of nitric oxide, nitrogen dioxide, oxidant and sulphur dioxide; and two infrared analyzers to measure carbon monoxide and hydrocarbons. All monitors can be used for independent or simultaneous analysis. Gear cies, common carriers and police departments. There are indications, however, that more manufacturers are requiring rapid means of transmitting visual information such as drawings, schematics and photos. One observer, citing the more than 200,000 hams now licensed in the U.S., expressed the feeling that interest by large numbers of amateurs could be a force in lowering prices. Help in bringing this about may come from the MARS practice of making surplus equipment available to members, he adds. A number of facsimile transmissions made by technical education directors may serve as the spur to purchase of additional units. Work being done now in single-sideband transmission points to the possibility of using one sideband for voice and the other for transmission of the visual information. MARS members are said to be interested in these possibilities. Some amateurs have been in the MARS system since it was first authorized by the Secretaries of the Army and the Air Force in 1948. Originally membership was limited to military personnel on active duty. Today, less than seven percent of the members are in uniform. Open to Amateurs Membership in MARS is open to any licensed radio amateur who is at least 16 years old. Rules require a minimum participation of six hours for each three-month period of the year. In addition, reports on participation must be filed. MARS officials say the highest percentage of participation in system work is now taking place in the Fifth Army region. This area, which comprises the upper section of the Midwest, has about 600 members with 95 percent active participation in transmitting messages and taking part in training and educational programs. Members of the First Army technical net (New York, New Jersey and New England), who number close to 500, attend talks by manufacturers, users of communications gear and military specialists in this field. BALLANTINE'S MODEL 305A VOLTMETER measures peak, or peak to peak PULSES 0.5μs as short as AT PULSE RATES AS LOW AS 5 pps . . . VOLTAGES OF 1 mv TO 1000 v Also measures Complex Waveforms having fundamental of 5 cps to 500 kc with harmonics to 2 mc. Accuracy is 2% to 5% OF INDICATED VOLTAGE, depending upon waveform and frequency. Scale is the usual Ballantine log-voltage and linear db, individually hand-calibrated for optimum precision. Input Impedance is 2 meg, shunted by 10 pf to 25 pf. Price: $395. THIS "A" MODEL is the result of improvements and new features AFTER 11 YEARS OF MANUFACTURING THE VERY SUCCESSFUL MODEL 305 Write for brochure giving many more details — Since 1932 — BALLANTINE LABORATORIES INC. Boonton, New Jersey CHECK WITH BALLANTINE FIRST FOR LABORATORY AC VACUUM TUBE VOLTMETERS, REGARDLESS OF YOUR REQUIREMENTS FOR AMPLITUDE, FREQUENCY, OR WAVEFORM. WE HAVE A LARGE LINE, WITH ADDITIONS EACH YEAR. ALSO AC/DC AND DC/AC INVERTERS, CALIBRATORS, CALIBRATED WIDE BAND RF AMPLIFIER, DIRECT-READING CAPACITANCE METER, OTHER ACCESSORIES. The Aerocom 1046 Transmitter is designed to give superior performance for all point-to-point and ground-to-air communications. It is now in use throughout the world in climates ranging from frigid to tropical (operates efficiently at $-35^\circ$ to $+55^\circ$ Centigrade). As a general purpose High Frequency transmitter, the 1046 supplies 1000 watts of carrier power with high stability (above $-10^\circ$ Centigrade: $\pm .003\%$ for telegraph and telephone. Temperature controlled oven for FSK). Multichannel operation is provided on telegraph A1, telephone A3 and FSK (Radio Teletype). It can be remotely controlled using one pair of telephone lines plus ground return with Aerocom Remote Control Equipment. Front panel switches and microphone are included for local control. Four crystal-controlled frequencies (plus 2 closely-spaced frequencies) in the 2.0 - 24.0 megacycle range can be used one at a time, with channeling time only two seconds. Operates into either balanced or unbalanced loads. The power supply required is nominal 230 volts, 50 - 60 cycles, single phase. The housing is a fully enclosed rack cabinet of welded steel, force-ventilated through electrostatic filter on rear door. Telegraph keying (A1): Up to 100 words per minute. Model 1000 M Modulator (mounts in transmitter cabinet) is used for telephone transmission; a compression circuit permits the use of high average modulation without overmodulation. Model 400 4 Channel exciter is used for FSK. Output connections consist of 4 insulated terminals (for Marconi antenna) and 4 coaxial fittings Type SO-239, which can be used separately or in parallel in any combination. For 600 ohm balanced load, Model TLM matching network is used, one for each transmitter channel. As in all Aerocom products, the quality and workmanship of Model 1046 are of the highest. All components are conservatively rated. Replacement parts are always available for all Aerocom equipment. Complete technical data on Aerocom Model 1046 available on request. A DALOHM Trimmer Potentiometer Retains its INHERENT STABILITY Even the smashing shock of steel on steel has no effect on the inherent stability that is standard in DALOHM trimmer potentiometers. DALOHM T-Pots not only offer rugged construction to withstand high shock and vibration, but also maintain exceptional stability under load, and operate reliably under extremes of temperature — all in sizes to meet the tightest space requirements. For all applications that demand trimmer potentiometers that meet or surpass MIL specs, you can depend on DALOHM. WIRE WOUND • PRECISION • MINIATURE* DALOHM A10-W TRIMMER POTENTIOMETER A miniature T-Pot that surpasses most military requirements. - Rated at 1 watt up to 70° C. ambient temperature, derating to 0 at 150° C. - Resistance range from 10 ohms to 50K ohms - Standard tolerance ± 5%; closer tolerances available - Resolution 0.1% to 1%, depending on resistance - Operating temperature range from -55° C. to 150° C. - Insulation resistance: 1000 megohms minimum at 500 VDC at room temperature - End resistance not greater than 4% - Temperature coefficient of trimmer unit within 100 PPM - Miniature size, .220 x .312 x 1.250 inches - Screw fully adjustable throughout 25 turn range - Shaft torque 7 inch/ounce maximum - Safety clutch arrangement on movable wiper contact prevents breakage due to over-excursion - Self locking adjustment; wiper will not shift under vibration or shock Write for Bulletin R-32, along with handy cross-reference file card. SPECIAL PROBLEMS? You can depend on DALOHM, too, for help in solving any special problem in the realm of development, engineering, design and production. Chances are you can find the answer in our standard line of precision resistors (wire wound, metal film and deposited carbon); trimmer potentiometers; resistor networks; collet-fitting knobs; and hysteresis motors. If not, just outline your specific situation. from DALOHM Better things in smaller packages DALE PRODUCTS, INC. 1300 28th Ave., Columbus, Nebr. Optional Chopper Stabilization ONLY® REGATRON PROGRAMMABLE POWER SUPPLIES HAVE IT! 0.01% or 0.003 V from no load to full load... this is the conservative regulation specification for Regatron Programmable Power Supplies equipped with chopper stabilization. And just as important, chopper stabilization assures a higher order of regulation and stability at every output voltage, even at fractions of one volt. There are other advantages too: Chopper stabilization provides for exceptionally high repeatability of voltage control settings... enhances remote control operation. And chopper stabilization can be specified at any time. The compact plug-in unit can be installed at the factory as an original accessory, or it can be installed in the field. A complete kit is available for field installations. Instructions and all hardware are included. Ask your local E/M representative for more information, or write... © REGISTERED U.S. PATENT OFFICE. PATENTS ISSUED AND PENDING. ELECTRONIC MEASUREMENTS COMPANY, INCORPORATED EATONTOWN • NEW JERSEY Radiation Used in Brain Study ACADEMIC RESEARCH in electronics continues to probe for new ways to expand the horizons of man's knowledge. Here are some examples: • University of California researchers at the university's medical school are conducting experiments to determine the extent to which electromagnetic stimulation can produce mental, emotional and hormonal responses. Studies are also aimed at determining ways of stimulating nerves, muscles and other body tissues electromagnetically. Locating brain areas which control different types of behavior has been largely accomplished by direct application of electrodes when the skull is surgically opened. Use of electromagnetically induced eddy currents could be accomplished without surgery in a conscious patient. • Rutgers University, New Brunswick, N. J., reports encouraging progress in work being done at its year-old microwave research lab. Research programs currently under way include: development of ceramic radomes for Office of Naval Research; studies for the Air Force on the aging of lead calcium titanate and other ferroelectric ceramics, and studies on the crystalline structure and other properties of these compounds; research on harmonic generation at millimeter wavelength for the Air Force; investigation of slow-wave structure for the Signal Corps and studies on the electrical conductivity of titanium oxide. • Georgia Institute of Technology, Atlanta, reports creation of a new research division. Called the Electronics division, the new organization will consist primarily of the research staffs of the former Radar and Communications branches of the Physical Sciences division. The head of the Radar branch, M. W. Long, has been named chief of the new group. Institute officials say the Electronics division was necessitated by the rapid growth of the Physical Sciences division in recent years. It included many research activities in physics, systems analysis and electrical engineering in addition to radar and communications. Communications studies now in process are devoted to advanced techniques of modulating and detecting radio signals, ionospheric effects on radio signals and mutual interference problems. In radar, research is being pursued to develop greater range and angular resolution. • Syracuse University researchers are at work under a $10,200 grant from the Atomic Energy Commission used for purchase of radiation biology equipment. The gear will be used for student training which will focus attention on radioisotopes from the viewpoint of detection and application to biology. • Arizona State University reports a pledge of $150,000 from Motorola for development of a doctoral program in engineering and physical sciences. To date, $30,000 has already been advanced. The contribution is in memory of Paul V. Galvin, late president of the company. • Polytechnic Institute of Brooklyn has been awarded a grant of $60,000 by National Science Foundation to set up a high-speed computer facility. An IBM 650 will be used in the many science programs being worked on as well as an educational program in computer work. • Case Institute of Technology researchers are exploring the value of thin magnetic metal films in computer memory systems. R. W. Hoffman has described films as little as 50 atoms thick made by evaporating iron, nickel and their alloys in vacuum. The resulting magnetic coatings may become spontaneously magnetized below 600 F and changed fast. COUNTERMEASURES and the horny protoparce sexta “RAVEN” SPELLS TROUBLE FOR THIS FELLOW but it once spelled security for our military The protoparce (Tomato worm) faced with the serious problem of existing from day to day, does so by the logical method of vanishing before the eyes of his enemies. To a Raven, he seems only another part of whatever bush or tree he sits upon. This is an application of basic countermeasures. On the other hand “Raven”, the World War II code name for countermeasures, meant security for us and trouble for our enemies. INSTRUMENTS FOR INDUSTRY, a leader in the field of countermeasures, has long been active in successors to project “Raven.” Recently under development is a device which effectively intimidates counter-interception equipment aboard attacking enemy aircraft, resulting in our planes literally disappearing before the eyes of the enemy. You can’t shoot down a shadow…but that’s all that can be found. Another countermeasure by I.F.I. See us at the IRE Show Booth 1424 INSTRUMENTS FOR INDUSTRY, Inc. 101 New South Road, Hicksville, L. I., N.Y. Graduate engineers with two or more years of circuit application in the fields of electronics or physics are invited to meet with Mr. John Hicks in an informal interview or send complete resume to: Dir. Personnel, IFI, 101 New South Road, Hicksville, New York. MEETINGS AHEAD Feb. 10-12: Solid-State Circuits Conf., AIEE, IRE, Univ. of Penn., Hotel Sheraton, Philadelphia. Feb. 10-12: Cleveland Electronics Conf., ISA, IRE, AIEE, CPS, CIT and WRU, Cleveland Eng. & Scientific Center, Cleveland. Feb. 11-13: Electronic Representatives Assoc., Annual Convention, Drake Hotel, Chicago. Feb. 16-18: Nondestructive Testing of Aircraft & Missile Components, Southwest Research Institute, Hilton Hotel, San Antonio, Texas. Feb. 19-23: Component Parts and Electronic Tubes, International Exhibition, Porte de Versailles, Place Balard, Paris. Feb. 25-26: Scintillation Counter Symposium, AIEE, AEC, IRE, NBS, Hotel Shoreham, Wash., D. C. Mar. 17-18: Synchro Design and Testing Symposium, Bureau of Naval Weapons, Dept. of Navy, Dept. of Commerce Auditorium, Wash., D. C. Mar. 21-24: Institute of Radio Engineers, National Convention, Coliseum & Waldorf-Astoria Hotel, N. Y. C. Mar. 24-25: Human Factors in Electronics, PGHF of IRE, Bell Labs Auditorium, N. Y. C. Apr. 3-7: National Assoc. of Broadcasters, Engineering Conf. Committee, NAB, Conrad Hilton Hotel, Chicago. Apr. 3-8: Nuclear Congress, EJC, PGNS of IRE, New York Coliseum, New York City. Apr. 11-13: Protective Relay Engineers, Annual, A. & M. College of Texas, College Station, Tex. Apr. 11-14: Weather Radar Conference, American Meteorological Society and Stanford Research Institute, San Francisco. Aug. 23-26: Western Electronic Show and Convention, WESCON, Ambassador Hotel & Memorial Sports Arena, Los Angeles. There's more news in ON the MARKET, PLANTS and PEOPLE and other departments beginning on p 134. Exclusive from Trio Labs—laboratory precision measurement of complex waves with VTVM versatility . . . New Model 120-1 overcomes the errors of spikes and harmonics in peak responding and average-reading meters (5% harmonics can effect errors as high as 1.7%) and of phase of harmonic shift in both types. Deflection is directly proportional to square of current through dynamometer meter movement — hence true RMS direct-readings. Incorporates laboratory-standard meter with 7-inch custom-calibrated mirror scale. Ranges: 10MV to 500 RMS volts full scale. Input Z: 1 meg. Freq. Response: 50-2000 cps. Accuracy (above 50% electrical deflection): ¾% f. s. at 400 cps; ½% f. s. at all other frequencies. Power: 115 VAC, 50-400 cps. Price: $985.00. Write for FREE "how-to" ENGINEERING GUIDE on the complete line to Dept. E2 Trio Laboratories, Plainview, L. I., N. Y. when ordinary instruments are too big or inadequate... NEW DRIVER TRANSISTORS SWEEPING THE FIELD Extra-versatile Bendix units beat high costs, design limitations over wide front Called the "workhorse of the transistor industry," the new Bendix* Driver Transistor series is winning the nod from more and more engineers daily. These men find it the answer to audio frequency and switching applications requiring extra performance without extra cost. Here is a special device for use where reliability, versatility, and low cost are primary requirements. The Bendix units combine higher voltage rating and high current gain with more linear current gain characteristics for low distortion and more efficient switching. They're now in high production for rapid delivery in JEDEC TO-9 packages. NEW BENDIX SEMICONDUCTOR CATALOG on our complete line of power transistors, power rectifiers, and driver transistors available on request. Write SEMICONDUCTOR PRODUCTS, BENDIX AVIATION CORPORATION, LONG BRANCH, N. J. For information about employment opportunities write personnel manager. TRADEMARK APPLICATION, PERFORMANCE DATA INDICATE BROAD USAGE \| TYPE NUMBERS \| MAXIMUM RATINGS \| TYPICAL OPERATION \| \|--------------\|-----------------\|-------------------\| \| \| Vce Vdc \| Ic mAdc \| Pc mW \| Tj °C \| T storage °C \| hfe \| fcb \| Vce(Sat) \| \| 2N1008 \| -20 \| 300 \| 400 \| 85 \| -65 to +85 \| 90 \| 1.2 mc \| 0.15 Vdc \| \| 2N1008A \| -40 \| 300 \| 400 \| 85 \| -65 to +85 \| 90 \| 1.2 mc \| 0.15 Vdc \| \| 2N1008B \| -60 \| 300 \| 400 \| 85 \| -65 to +85 \| 90 \| 1.2 mc \| 0.15 Vdc \| \| 2N1176 \| -40 \| 300 \| 300 \| 85 \| -65 to +85 \| 65 \| 1.2 mc \| 0.15 Vdc \| \| 2N1176A \| -40 \| 300 \| 300 \| 85 \| -65 to +85 \| 65 \| 1.2 mc \| 0.15 Vdc \| \| 2N1176B \| -60 \| 300 \| 300 \| 85 \| -65 to +85 \| 65 \| 1.2 mc \| 0.15 Vdc \| Ideal for such applications as: TRANSISTOR DRIVER • AUDIO AMPLIFIER (CLASS A OR B) POWER SUPPLY • SERVO CONTROL • AUDIO OSCILLATOR MOTOR CONTROL • RELAY DRIVER • POWER SWITCH Visit us at the New York I. R. E. Show—Booth #2228 SEMICONDUCTOR PRODUCTS Red Bank Division LONG BRANCH, N. J. West Coast Sales Office: 117 E. Providencia Avenue, Burbank, California Midwest Sales Office: 2NS45 York Road, Elmhurst, Illinois New England Sales Office: 4 Lloyd Road, Tewksbury, Massachusetts Export Sales Office: Bendix International Division, 205 E. 42nd Street, New York 17, New York Canadian Affiliate: Computing Devices of Canada, Ltd., P. O. Box 308, Ottawa 4, Ontario, Canada Standby time is no problem with the new, cool (all-transistorized) Mincom C-100. No blowers necessary — and the new Mincom Model C-100 Instrumentation Recorder/Reproducer is particularly new in its simplicity and reliability. Six speeds record frequencies from 50 cps to 100 kc. Only 500 watts input for 14-track system. No mechanical brakes. Only 0.1% flutter and wow. Instant push-button speed control, no belt changes. Interested? Write Mincom for specifications today. MINCOM DIVISION MINNESOTA MINING AND MANUFACTURING COMPANY 2049 SOUTH BARRINGTON AVENUE • LOS ANGELES 25, CALIFORNIA • TELETYPE: WEST LOS ANGELES 6742 ELECTRONICS • FEBRUARY 12, 1960 ECONOTAPE crossbar contacts are most efficient for electrical relays You can now get ideally efficient crossbar contacts for your electrical relays—with as many contacts as you need, where you need them—with positive assurance of full, contact surface. This has been made possible through the development of Makepeace's new ECONOTAPE, a precision-drawn shaped or rectangular contact wire in either solid precious metal or in laminated metal—in your choice of gold, platinum, palladium, silver and their various alloys. Econotape crossbar contacts are supplied complete, attached to Makepeace blades...attached to blades supplied by you...or Econotape for your own attaching. Econotape is cut off and welded to the blade in one operation. It is no longer necessary to handle and attach individual button type contacts. Positioning of Econotape on the blade is done automatically as the tape is cut off and welded. Permanent attachment is assured by a homogenous metallurgical bond that is undisturbed by expansion and contraction caused by temperature changes. If you are designing a new relay or trying to cut the cost of your present mechanism, send for Econotape literature. D. E. MAKEPEACE DIVISION • PINE & DUNHAM STREET ATTELBORO, MASS. CIRCLE 272 ON READER SERVICE CARD fine wire, thin foils, ribbon and tubing in noble metals and their alloys, for all applications. The unique combination of properties of the noble metals continually recommend them for industrial applications. Our modern melting, wire drawing, rolling and heat treating equipment coupled with long experience in the field is at your service for production of standard and special items. WIRES: Bare drawn wire of ductile materials down to .004"—High temperature thermocouple wires—High temperature furnace windings—Potentiometer and Resistance wires—Platinum clad tungsten wire. FOILS: In platinum, palladium and gold down to .0001"—In iridium and rhodium as thin as .001". TUBING: Seamless in platinum, palladium, gold and their alloys. Sizes from .018" with .004" wall up to 1½" with .042" wall. For complete information write for our leaflets, "Fine Wire, Foils, Ribbons" and "Noble Metal Thermocouple Wire". BAKER PLATINUM DIVISION • 113 ASTOR STREET NEWARK, N. J. CIRCLE 273 ON READER SERVICE CARD DOMESTIC DIVISIONS: AMERICAN PLATINUM & SILVER DIVISION • AMERICAN QUARTZ DIVISION • BAKER CONTACT DIVISION • BAKER DENTAL DIVISION • BAKER SETTING DIVISION • BAKER PLATINUM DIVISION • CHEMICAL DIVISION • EAST NEWARK INDUSTRIAL CENTER • HANOVIA LAMP DIVISION • HANOVIA LIQUID GOLD DIVISION • INDUSTRIAL DIAMOND DIVISION • INSTRUMENTS AND SYSTEMS DIVISION • IRVINGTON-BAKER REFINING DIVISION • D. E. MAKEPEACE DIVISION • NATIONAL ELECTRIC INSTRUMENT DIVISION • RESEARCH AND DEVELOPMENT DIVISION • H. A. WILSON DIVISION. PROMPT PRECIOUS METAL SCRAP RECOVERY SERVICE • ENGELHARD PROCEDURES RECOVER a simplified mirror-bright silver plating process for electrical and electronic components Here is the most efficient, simple procedure to protect electrical electronic and lamp components with a mirror-bright silver finish—through a complete range from flash to heavy deposit. The procedure is easy, economical and non-critical—with little or no polishing required. Silva-Brite is a clear, water-white solution, enabling the operator to observe work as it is being plated. Uniformly good results are attained with current densities ranging from 10 to 40 amperes per square foot. Normal room temperature operation minimizes fumes and tendency toward bath decomposition. Send for descriptive data together with detailed plating procedures. AMERICAN PLATINUM & SILVER DIVISION 231 N. J. RAILROAD AVENUE, NEWARK, N. J. for low cost purification and drying of hydrogen and other gases The Deoxo Catalytic Purifier is combined with an extremely efficient automatically operated drying unit to provide oxygen-free hydrogen that is ideally pure and dry. The combined units are identified as the Deoxo Dual Puridryer. It supplies hydrogen with less than one part oxygen per million—dried to a dew point of -100°F. No inert gas purging is needed. The Deoxo Dual Puridryer can also be used with other gases such as: Nitrogen, Argon, Helium and saturated hydrocarbons, with equally fine performance. Write for descriptive literature. CHEMICAL DIVISION • 113 ASTOR STREET NEWARK, N. J. CIRCLE 275 ON READER SERVICE CARD 100% OF ASSAYED PRECIOUS METAL CONTENT • IRVINGTON-BAKER REFINING DIVISION ELECTRONICS • FEBRUARY 12, 1960 KAY Ligna-Sweep SKV AUDIO—VIDEO—VHF . . . IN ONE INSTRUMENT! Sweeping Oscillator - FREQUENCY RANGE—200 CPS TO 230 MC. - SWEEP REPETITION RATES FROM 0.2 TO 60 CPS. - LINEAR AND LOGARITHMIC SWEEPS. - AUDIO SWEEP—200 CPS TO 20,000 CPS. - 3 Highly Stable Video Bands—1 kc to 12 mc., Variable or in Single Sweep. - RF Output of 1 Volt RMS at 70 Ohms ±.5 db Over Widest Sweep Width. - 8 Narrow Customer Selected Fixed Frequency Bands—20 kc to 12 mc. - 9 Fundamental Frequency, Wide VHF Bands—10 mc 220 mc. SPECIFICATIONS VARIABLE FREQUENCY RANGES: .5-12 mc, .1-12 mc, 10 kc-12 mc, 10-220 mc (9 bands) FIXED FREQUENCIES: Up to max. of 8 center frequencies (20 kc to 12 mc—Customer selected) AUDIO RANGE: 200 cps to 20 kc. SWEEP WIDTHS: Selected for maximum stability 1-10 mc on .5-12 mc band; .2-2 mc on .1-12 mc band; 20-200 kc on 10 kc-12 mc band; 6% to 60% of centered frequency 50 mc to 3 mc to 30 mc above 50 mc on 10-220 mc bands; 2-20 kc on fixed frequencies and audio range. OUTPUT LEVEL: Continuously variable from 1 volt rms down to 65 db below 1 volt, ±5% over widest sweep. AGC. Audio range: variable .5-1 volt rms. IMPEDANCE: 70 ohms nominal (50 ohms on request). Audio range: 600 ohms. SWEEP OUTPUT and REPETITION RATES: Sawtooth for horizontal deflection of oscilloscope. Approx. 4 volts peak to peak—Output impedance 1000 ohms nom. Sweep 60 mc line loaded, fixed 30 cps logarithmic (for audio and video application) 3 cont. var. ranges—.2-.1 cps, 1-5 cps, 5-30 cps. MARKERS: Swept signal available for operation of Vari-Marker SKV Generator. PULSE TYPE MARKERS: Limited number of sharp, crystal-controlled pulse-type markers at customer specified frequencies can be provided. Please inquire before ordering. POWER SUPPLY: Input approx. 220 Watts, 117v (±10%), 50-60 cps. B+ electronic regulation. PRICE: $995.00 f.o.b. factory. Fixed freq. bands add $17.00 per band. The wide range of frequency and repetition rate in the Ligna-Sweep Model SKV make it ideally suited for alignment and testing of a wide variety of electronic instruments—audio amplifiers, filters, communication receivers, radar IF channels, TV receivers and transmitters. The unit is stable and carefully shielded and filtered to prevent spurious signals on beat frequency video bands. A wide range of sweep repetition rates makes viewing easy on conventional oscilloscopes. Low repetition rates used with long persistence screens permit study of high Q circuitry, LF limits of band circuits and observation of the "ring" characteristics of tuned circuits. Ligna-Sweep MODEL CP Sweeping Oscillators Cat. No. 932-A: Variable bands between 100 kc and 175 mc. Price: $150.00 f.o.b. factory. 18 pulse-type markers avail. at customer specified freq., $17.00 each. Cat. No. 932-B: Continuous coverage from 100 kc—150 mc. Price: $150.00 f.o.b. factory. 18 pulse-type markers avail. at customer specified freq., $17.00 each. WRITE FOR NEW KAY CATALOG KAY ELECTRIC COMPANY DEPT. E-2, MAPLE AVENUE, PINE BROOK, N. J., CAPitol 6-4000 SEE US AT THE IRE SHOW BOOTHS #3512, 14, 16, 18 60 CIRCLE 60 ON READER SERVICE CARD NOW SHAMBAR KOP-R-KLAD printed circuit laminate COPPER TEFLON COPPER Here's an end to circuit unreliability caused by "leaking" printed circuitry. Unique bonding methods, developed by Shamban, bond insulator to copper without compromising the electrical properties of either. Even in extreme environments, there is no breakdown of strengths. Tradename of W. S. Shamban & Co. A new, complete line of hi-temperature hi-dielectric strength laminates featuring: Maximum electrical properties * Unique, optimum-performance bonding methods * Highest commercial peel strengths * Availability in sheets and continuous lengths * Full range of types of constructions, sizes New Shamban KOP-R-KLAD laminate presents several distinct advantages to the users of printed circuitry. KOP-R-KLAD offers a complete line, the right constructions for every application; optimum electrical properties through proper bonding, best volume, surface and insulation resistivity, highest dielectric strength; highest peel strength, for sharp bends, rugged environments; continuous lengths, for convenience of user, for wider application. KOP-R-KLAD is available in twelve different types, including copper to Teflon, to Teflon-glass, to Kel-F, and to FEP-fluorocarbon. Each type has specific advantages, all types have the advantage of absolute dependability and predictability within the limitations of the materials specified. KOP-R-KLAD is immediately available, dependent upon type, in widths up to 36", in lengths from 2" to continuous rolls. Write or wire factory for complete data. Engineered Plastic Products W. S. Shamban & Co. 11617 W. Jefferson Blvd. Culver City, Calif. Ft. Wayne, Indiana Floral Park, New York SHAMBAR PRODUCTS FOR ELECTRONICS Seag-proof Teflon grommets. Non-abrasive, chemical resistant and very durable, Shamban snap-in and channel type grommets provide secure holding devices. Stand-off and feed-thru insulators. Absolute insulation for critical circuit tiepoints. Resists high frequency and voltage breakdowns. Teflon and Nylon spaghetti tubing, standard and Microthin wall sizes. Available in all sizes, to meet every tubing need. Consistent quality. How CDF Di-Clad® can solve your printed-circuit problems The CDF line of copper-clad laminates in all grades is now known by a new name—Di-Clad. Di-Clad grades meet the varying needs of design, production, and operation of electronic equipment. Grades other than those described are also available. Di-Clad 2350. An economy paper-base phenolic grade having good tensile, flexural, compressive, and impact strength. Adequate for most non-critical printed circuit applications. Can be cold punched and sheared up to 5/64 of an inch in thickness. Di-Clad 112T. A Teflon* glass-fabric laminate offering the best dielectric properties over a wide temperature and frequency range. Send us your requirements and let our engineers help you select the right grade for your application. Du Pont trademark for its tetrafluoroethylene resin. ### TYPICAL Di-Clad PROPERTY VALUES \| Property \| Di-Clad 2350 \| Di-Clad 26 (NEMA XXXP) \| Di-Clad 28 (NEMA XXXP) \| Di-Clad 28E (NEMA G-10) \| Di-Clad 112T Teflon \| \|-----------------------------------------------\|--------------\|------------------------\|------------------------\|-------------------------\|----------------------\| \| Bond Strength—0.0014" foil (lbs. reqd. to separate 1" width of foil from laminate) \| 6 to 10 \| 6 to 10 \| 6 to 10 \| 8 to 12 \| 4 to 8 \| \| Maximum Continuous Operating Temperature (Deg. C.) \| 120 \| 120 \| 120 \| 150 \| 200 \| \| Dielectric Strength (Maximum voltage per mil for 1/16" thickness) \| 800 \| 900 \| 850 \| 650 \| 700 \| \| Insulation Resistance (Megohms) 96 hrs. at 35°C. & 90% RH (ASTM D257, Fig. 3) \| 500 \| 150,000 \| 600,000 \| 100,000 \| 75,000 \| \| Dielectric Constant 10^6 Cycles \| 4.5 \| 4.0 \| 3.6 \| 4.9 \| 2.6 \| \| Dissipation Factor 10^6 Cycles \| 0.040 \| 0.026 \| 0.027 \| 0.019 \| 0.0015 \| \| Arc-Resistance (Seconds) \| 5 \| 10 \| 10 \| 130 \| 180 \| \| Tensile Strength (psi.) \| 18,000 \| 16,000 \| 12,000 \| 48,000 \| 23,000 \| \| Flexural Strength (psi.) \| 27,000 \| 21,000 \| 18,000 \| 70,000 \| 13,000 \| \| Izod Impact Strength edgewise (ft. lbs. per inch of notch) \| 0.80 \| 0.45 \| 0.42 \| 12.0 \| 6.0 \| \| Compressive Strength flatwise (psi.) \| 32,000 \| 28,000 \| 25,000 \| 62,000 \| 20,000 \| \| Base Material of Laminate \| Paper \| Paper \| Paper \| Medium-weave, medium-weight glass cloth \| Fine-weave, medium-weight glass cloth \| \| Color of Unclad Laminate \| Natural \| Natural greenish \| Natural \| Natural \| Natural \| All these standard grades are available with 0.0014" and 0.0028" or thicker electrolytic or rolled copper foil on one or both surfaces. Other metal foils and other resin-and-base combinations can be supplied on special order. Du Pont Trademark NOW $825 A Full Four-Digit Voltmeter at a Pointer Meter Price! FOR THE FIRST TIME you can have the accuracy, speed and reliability of an NLS digital voltmeter with full four-digit resolution... for the price of a quality pointer meter. That's the dramatic story of the new NLS V64! Only NLS high-volume production techniques make it possible. Use the low-cost, versatile V64 for a wide range of measuring jobs. See the V64 in action... contact NLS today! BRIEF SPECIFICATIONS: Accuracy ± (.02% of reading plus 1 digit)... full 4-digit resolution... measures DC voltages from one millivolt to 500 volts in steps of ±9.999/99.99/500.0... one package design (5¼" high, 15¼" deep for 19" rack)... plug-in accessories permit measuring AC or low-level DC voltages... available from stock for immediate delivery. Originator of the Digital Voltmeter non-linear systems, inc. DEL MAR (SAN DIEGO), CALIFORNIA NLS — The Digital Voltmeter That Works... And Works... and Works! ELECTRONICS • FEBRUARY 12, 1960 CIRCLE 63 ON READER SERVICE CARD It takes less than a minute. But it could be the biggest project of your life when you dial CRestview 4-8884. Your future is literally at your fingertips, and it's an opportunity-filled future with the Bedford Laboratory of Raytheon Company in Bedford, Massachusetts. In this key laboratory of Raytheon Company's Missile Systems Division a man can literally "pick his spot". Ability is quickly recognized... achievement rewarded. And, working and living conditions are the finest available... in the heart of New England. If YOU can qualify for one of the following positions, act NOW! Immediate Openings for: Data Handling Engineers... with experience in high speed, analog-to-digital conversion techniques, logic design, converter and buffer design. Should have thorough knowledge of tape recorder techniques and digital, servo, and digital-computer design. Circuit Design Engineers... with experience in design of high-speed switching circuits, pulse techniques, and computer logic. Should be experienced in one or more of the following areas: navigation, guidance, control circuits, CCM, FM, PCM, PDM, and fusing circuitry. Packaging Engineers... with a knowledge of packaging and production techniques in sheet metal and electronic equipment. Will design electronic portions of guided missiles, radars, computers, test equipment. Should have thorough knowledge of circuitry. Electromechanical Designers... will design electromechanical equipment and electronic portions of guided missiles, including coordination of effort through the shop. Will work closely with Design Engineers in developing electronic packaging philosophies. Knowledge of electronics, electronic components, and ability to read schematics required. Should have experience in sheet metal equipment design and knowledge of current "state of the art" in electronic equipment. Call collect CRestview 4-8884 and ask for Mr. Jerry Morris. He will arrange an appointment for you with key personnel at the Bedford Laboratory. If you prefer, send your postcard or letter to Mr. Morris, Raytheon Company, Missile Systems Division, Bedford, Mass. MISSILE SYSTEMS DIVISION ...creates a climate for talent. Philco has designed and built the world’s largest 3-axis tracking antenna The world’s largest 3-axis tracking antenna was recently completed at the Philco Western Development Laboratories in Palo Alto. It will be used at one of the worldwide satellite tracking stations to receive vast amounts of scientific information from outer space. By employing the unique design feature of tri-axial mounting, this extremely accurate and complex instrument, designed and built by Philco, has complete flexibility of movement and can provide continuous coverage of telemetered information and data from satellites and missiles during any phase of flight. At Philco you will find the skills that come from close association with the involved problems of planning, developing and implementing advanced space communications programs... experience that includes the design and construction of antennas of many types. Each fully meets the stringent specifications of the military and various scientific research organizations. Philco stands ready to fill your specific needs. PHILCO CORPORATION/GOVERNMENT AND INDUSTRIAL GROUP Philadelphia 44, Pennsylvania Communications and Weapons Systems Division Computer Division • Sierra Electronic Division Western Development Laboratories This Philco 3-axis antenna stands 80 feet high and weighs over 130 tons. One of its most unique features is the 60-foot reflector—a solid aluminum skin paraboloidal structure manufactured to a tolerance of 65/1000 of an inch over its entire surface to provide maximum reception under the most severe environmental conditions. The antenna maintains its accuracy in winds up to 60 miles per hour and its mechanical efficiency in winds up to 140 miles per hour. SYNCHROS • SERVO MOTORS • MOTOR TACH GENERATORS 400 Cycle: Many for 125°C operation . . . Higher for special applications Many Immediately Available From Stock in Small Quantities SIZE 8 SYNCHROS Highly Stable. Minimum Error Variation from −55°C to +125°C \| OSTER TYPE \| CLASS \| INPUT VOLTAGE \| INPUT CURRENT AMPS \| INPUT WATTS \| OUTPUT VOLTAGE \| PHASE SHIFT (° LEAD) \| ROTOR RESISTANCE (OHM) \| STATOR RESISTANCE (OHMS) \| Z₁₀ OHMS \| Z₂₀ OHMS \| Z₃₅₅ OHMS \| NULL VOLTAGE (MV) \| MAX. ERROR FROM E.Z. (MIN.) \| \|------------\|-------\|---------------\|--------------------\|-------------\|----------------\|----------------------\|------------------------\|--------------------------\|-----------\|-----------\|-----------\|---------------------\|--------------------------\| \| 4253-01 \| LZ-CT \| 11.8 \| 0.87 \| 0.21 \| 23.5 \| 9.0 \| 157.0 \| 24.0 \| 212+j722 \| 28+j119 \| 263+j69 \| 30 \| ±7 \| \| 4269-01* \| Diff \| 11.8 \| 0.87 \| 0.21 \| 11.8 \| 9.0 \| 35.0 \| 24.0 \| 37+j139 \| 28+j124 \| 47+j13 \| 30 \| ±7 \| \| 4273-01** \| XMTR \| 26.0 \| 1.00 \| 0.54 \| 11.8 \| 8.5 \| 34.0 \| 12.0 \| 48+j255 \| 12+j45 \| 82+j31 \| 30 \| ±7 \| \| 4277-01* \| HZ-CT \| 11.8 \| 0.30 \| 0.073 \| 22.5 \| 8.5 \| 316.0 \| 67.0 \| 500+j1937 \| 79+j350 \| 594+j182 \| 30 \| ±7 \| \| 4261-01** \| Resolver \| 26.0 \| 0.43 \| 0.39 \| 11.8 \| 15.0 \| 162.0 \| 22.0 \| 208+j612 \| 34+j159 \| 243+j77 \| 30 \| ±7 \| Stator as Primary Rotor as Primary SIZE 8 SERVO MOTORS \| OSTER TYPE \| RATED VOLTAGES \| Z = R + jX \| IN. OZ. STALL TORQUE \| RPM NO LOAD SPEED \| WATTS PER PHASE \| GM. CM. ROTOR INERTIA \| LENGTH IN. MAX. \| WEIGHT OZ. \| T/I RATIO RAD/SEC² \| \|------------\|----------------\|------------\|----------------------\|-------------------\|-----------------\|-----------------------\|----------------\|-----------\|------------------\| \| 5004-01 \| 26V \| 288 = 226 + j176 \| .15 \| 6200 \| 2.0 \| .47 \| 0.863 \| 1.2 \| 22,500 \| \| \| 26V \| 294 = 238 + j174 \| \| \| \| \| \| \| \| \| 5004-02 \| 26V \| 288 = 226 + j176 \| .15 \| 6200 \| 2.0 \| .47 \| 0.863 \| 1.2 \| 22,500 \| \| \| 36V \| 526 = 409 + j332 \| \| \| \| \| \| \| \| \| 5004-03 \| 26V \| 288 = 226 + j176 \| .15 \| 6200 \| 2.0 \| .47 \| 0.863 \| 1.2 \| 22,500 \| \| \| 40V \| 715 = 582 + j415 \| \| \| \| \| \| \| \| \| 5004-09 \| 26V \| 230 = 190 + j131 \| .20 \| 6200 \| 2.5 \| .47 \| 0.863 \| 1.2 \| 30,000 \| \| \| 40V \| 519 = 399 + j332 \| \| \| \| \| \| \| \| SIZE 8 MOTOR TACH-GENERATORS \| OSTER TYPE \| RATED VOLTAGES \| Z = R + jX \| IN. OZ. STALL TORQUE \| RPM NO LOAD SPEED \| WATTS PER PHASE \| GM. CM. ROTOR INERTIA \| LENGTH IN. MAX. \| WEIGHT OZ. \| T/I RATIO RAD/SEC² \| GENERATOR VOLTAGE \| INPUT WATTS \| OUTPUT VOLTS PER 1000/RPM \| \|------------\|----------------\|------------\|----------------------\|-------------------\|-----------------\|-----------------------\|----------------\|-----------\|------------------\|------------------\|-------------\|-------------------------\| \| 6204-01 \| 26V \| 230 = 190 + j131 \| .20 \| 6000 \| 2.5 \| .65 \| 1.728 \| 2.5 \| 21,800 \| 26 \| 2.5 \| .25 \| \| \| 40V \| 519 = 399 + j332 \| \| \| \| \| \| \| \| \| \| \| \| 6204-03 \| 26V \| 230 = 190 + j131 \| .20 \| 6000 \| 2.5 \| .65 \| 1.728 \| 2.5 \| 21,800 \| 26 \| 2.5 \| .25 \| \| \| 26V \| 230 = 190 + j131 \| \| \| \| \| \| \| \| \| \| \| The Size 8 400 Cycle Servo Motor Tach Generators listed above have 150° max. cont. frame temperature, 110 MA input current, ±5° phase shift and Null Voltage (Total R. M. S.) of 15 millivolts. OTHER PRODUCTS INCLUDE: Resolvers Computers Indicators Servo Mechanisms Servo Torque Units DC Motors John Oster Manufacturing Co. Avionic Division Racine, Wisconsin EASTERN OFFICE 310 Northern Blvd. • Great Neck, Long Island, New York Phone: HUnter 7-9030 • TWX Great Neck N.Y. 2980 WESTERN OFFICE 5333 South Sepulveda Blvd. • Culver City, California Phone: EXmont 1-5742 • UPTon 0-1194 • TWX S. Mon. 7671 Engineers For Advanced Projects: Interesting, varied work on designing transistor circuits and servo mechanisms. Contact Mr. Robert Burns, Personnel Manager, in confidence. FOR CHECK OUT SYSTEMS Giving ultimate reliability and superb construction for Military use, MIL Spec., of course MODEL 3277 PUNCHED TAPE READER SOLID STATE, MILITARIZED DESIGN SPEED Up to 350 characters per second (200 is standard) CAPACITY 550 feet, 2.5 mil, 1" wide Mylar tape CHANNELS Up to 8 information, plus sprocket RUN-STOP INPUT 10 volts, negative pulse BI-DIRECTIONAL External SPDT contact control START TIME 15 milliseconds to next character at 200 char/sec. STOP DISTANCE On stop character REWIND TIME 1 minute OUTPUT -10 volts for logic "O", 0 volts for logic "1"; 40 mo. maximum DIMENSIONS 17" wide x 11" high x 11" deep POWER INPUT 28 V D.C., 3 amps; 115V 400 cps, 1 phase, 1 amp. MODEL 3303 HIGH SPEED PRINTER SOLID STATE, MILITARIZED DESIGN SPEED Up to 10 lines/sec. CAPACITY Up to 17 columns in 20 column format DATA INPUT (Nominal) -10 volts for logic "1" (Nominal) 0 volts for logic "0" Maximum loading 5 mo. DATA LOADING Parallel or serial at 30 kc max. DATA CODING 4 level; all electronics self-contained to store, compare, and print CONTROL FLEXIBILITY Input/Output signals permit closed loop operation with data source PAPER STOCK Pressure sensitive (no ribbon required) with self-contained supply and takeup DIMENSIONS (Mechanism Only) 8" wide x 8" high x 15" deep ASSEMBLY RETMA panel or separate mechanical and remote electronics POWER INPUT 115 volt, 60 cps, 1 or 3 phase 115 volt, 400 cps, single phase POTTER INSTRUMENT CO., INC. Sunnyside Boulevard, Plainview, L.I., N.Y. Overbrook 1-3200 It could happen... with El-Menco Capacitors! NEW Mylar-Paper Dipped CAPACITORS TYPE MPD INSURE FAILURE-PROOF PERFORMANCE! Only 1 Failure in 7,168,000 Unit-Hours for 0.1 MFD Capacitors* Setting a new standard of reliability! Life tests have proved that El-Menco Mylar-Paper Dipped Capacitors — tested at 100°C with rated voltage applied — have yielded a failure rate of only 1 per 716,800 unit-hours for 1 MFD. Since the number of unit-hours of these capacitors is inversely proportional to the capacitance, 0.1 MFD El-Menco Mylar-Paper Dipped Capacitors will yield ONLY 1 FAILURE IN 7,168,000 UNIT-HOURS. SUPERIOR FEATURES! • Five case sizes in working voltages and ranges: \| Voltage (WVDC) \| Range (MFD) \| \|----------------\|-------------------\| \| 200 \| .018 to .5 \| \| 400 \| .0082 to .33 \| \| 600 \| .0018 to .25 \| \| 1000 \| .001 to .1 \| \| 1600 \| .001 to .05 \| Write for Technical Brochure Giving Complete Information on the El-Menco Tubular Dur-Paper Line. THESE CAPACITORS WILL EXCEED ALL THE ELECTRICAL REQUIREMENTS OF E.I.A. SPECIFICATION RS-16A AND MILITARY SPECIFICATIONS MIL-C-91A AND MIL-C-25A. FOR FAILURE-PROOF PERFORMANCE ... COUNT ON EL-MENCO MYLAR-PAPER DIPPED CAPACITORS FROM MISSILE GUIDANCE SYSTEMS TO DATA PROCESSING EQUIPMENT! Registered Trade Mark of DuPont Co. SPECIFICATIONS • TOLERANCES: ±10% and ±20%. Closer tolerances available on request. • INSULATION: Durez phenolic resin impregnated. • LEADS: No. 20 B & S (.032") annealed copper-weld crimped leads for printed circuit application. • DIELECTRIC STRENGTH: 2 or 2½ times rated voltage, depending upon working voltage. • INSULATION RESISTANCE AT 25°C: For .05MFD or less, 100,000 megohms minimum. Greater than .05 MFD, 5000 megohm-microfarads. • INSULATION RESISTANCE AT 100°C: For .05MFD or less, 1400 megohms minimum. Greater than .05MFD, 70 megohm-microfarads. • POWER FACTOR AT 25°C: 1.0% maximum at 1 KC. THE ELECTRO MOTIVE MFG. CO., INC. WILLIMANTIC CONNECTICUT Manufacturers of El-Menco Capacitors • molded mica • dipped mica • mica trimmer • dipped paper • tubular paper • ceramic • silvered mica films • ceramic discs Arco Electronics, Inc., 64 White St., New York 13, N. Y. Exclusive Supplier To Jobbers and Distributors in the U.S. and Canada CIRCLE 68 ON READER SERVICE CARD CIRCLE 69 ON READER SERVICE CARD Optically as far as the first obstruction. For some, the same applies to mental vision. By seeing beyond the apparent obstacles, established theories or accepted principles, Fairchild Semiconductor Corporation has been able to achieve spectacular product innovations in transistors. Because of this faculty, the company has grown from an original nucleus of eight scientists to a complement of more than fourteen hundred in little more than two years. From continuing research and development work through engineering, tooling, manufacturing and testing of products on the line, the success of Fairchild is built on the abilities of its men to see around the obstacles and move beyond. It has resulted in products more advanced than any others of their type and in a solid reputation for quality workmanship. In a rapidly growing company with many challenging programs (e.g. current work on Esaki diodes and micro-logic circuits), there is a constant need for men who can see beyond the first obstacles. If yours is a relevant background and you find our approach attractive, we would like very much to hear from you. 545 WHISMAN ROAD / MOUNTAIN VIEW, CALIFORNIA / YORKSHIRE R 8161 SENSITIVE RESEARCH THE New MODEL "N" A new series of AC and DC portable electrical indicating instruments — that can also be panel mounted. A New HIGH in accuracy, resolution, and stability. A New LOW in cost. In short, the SRIC Model "N" series probably has just the instrument for which you've been looking HIGH and LOW. The Model "N" is a true "secondary standard" because it is engineered and manufactured by "standards people." GENERAL SPECIFICATIONS: Accuracy: .5% of full scale. Scale: Hand-drawn and mirrored. Scale length: 6". Resolution: 100 to 150 division. Ranges: DC from 50 ua. to 50 A. and 20 mv. to 1000 v. full scale. AC from 10 ma. to 10 A. and 7.5 v. to 750 v. full scale. Availability: 1 to 8 ranges completely switch-controlled and self contained. Construction: DC — double pivoted permanent magnet type. AC — moving iron vane for RMS measurements and germanium rectifier type for average reading measurements. Shielding: Magnetic and electrostatic. Pivots and Jewels: High carbon steel pivots and shock mounted sapphire jewels. (Diamond pivots available on special order). Case: Black moulded bakelite with leather carrying handle. Size 7½" x 6¾" x 3½". The New Model "N" is a stable, rugged, versatile field or laboratory instrument that is spectacular because in every way it offers a little bit "more" than any other instrument in its class. It is available in standard range combinations or in quantities with special ranges and/or special scale markings. The New Model "N" conforms to ASA specs. C 39.1-1951 for .5% class instruments. All SRIC instruments are calibrated against primary standards that have National Bureau of Standards Certificates of Accuracy and are unconditionally guaranteed for a period of one year. SENSITIVE RESEARCH INSTRUMENT CORPORATION NEW ROCHELLE, N.Y. ELECTRICAL INSTRUMENTS OF PRECISION SINCE 1927 CIRCLE 70 ON READER SERVICE CARD FEBRUARY 12, 1960 • ELECTRONICS ALLIED'S New CRYSTAL CAN 5 amp Relay GENERAL FEATURES Contact Arrangement: Two pole double throw. Contact Rating: d-c non-inductive—low-level up to 5 amperes at 29 volts. a-c non-inductive—low-level up to 2 amperes at 115 volts. a-c or d-c inductive—1 ampere at 29 volts d-c and 115 volts a-c. Initial Contact Resistance: .05 ohms maximum. Minimum Operate Sensitivity 100 milliwatts with a contact rating of 2 amperes non-inductive. Ambient Temperature: -65°C to +125°C. Dielectric Strength: 1,000 volts rms at sea level. 450 volts rms at 70,000 feet. 350 volts rms at 80,000 feet. Insulating Resistance: 10,000 megohms minimum. Vibration: 5-28 cps at 0.5 inch double amplitude and 28-2000 cps at 20 g. Shock: 50 g operational. 100 g mechanical. Operate Time: 10 milliseconds or less at rated voltage at 25°C. Release Time: 5 milliseconds or less at rated voltage at 25°C. Maximum Over-all Dimensions: height 1.0" length 0.8" width 0.4". Terminals: 0.2 inch grid spaced. Plug-in printed circuit and hooked type solder terminals. Weight: 0.8 ounces maximum. Write for Bulletin JH-6D ALLIED CONTROL COMPANY, INC., 2 EAST END AVENUE, NEW YORK 21, N.Y. CIRCLE 71 ON READER SERVICE CARD Short-cut system for custom orders To help turn out "job shop" work at assembly line speed, the I-T-E Circuit Breaker Company of Philadelphia has devised a simple "Information Sheet" that does away with considerable retracing and revising of engineering prints. More than 70% of I-T-E orders are for custom-designed equipment using standard components. Revising standard drawings to meet customer specs on each order would saddle I-T-E's engineering department with a nearly impossible work load. So the Information Sheet is used instead. It's an $8\frac{1}{2}$" x 11" tracing form—with printed title blocks—quickly reproduced on the company's Ozalid whiteprinters. Here's how it works: An order comes in—for 5KV metal-clad switchgear, for instance. A fast freehand sketch of the switchgear is drawn on the Information Sheet. Drawing numbers of standard components and quantity of prints needed are noted on the Sheet. Then, copies of the Sheet and the required standard drawings are run in the I-T-E repro room. These, with the shop order, go to Manufacturing. When the order is completed, the Information Sheet is returned to the customer file for reference. This simple short cut with Ozalid whiteprinting saves untold hours of engineering time and gives I-T-E customers faster, more efficient service. Colors speak louder than words A simple way to make your security personnel's job a lot easier is to color-code all classified material by using Ozalid sensitized color-copy papers. Colors don't have to be read. Guards can spot restricted or top-secret prints at a glance. Clerks can't make routing mistakes. To help you devise your own color-coding systems, Ozalid offers papers with eleven image-and-stock color combinations. For example, use black image on yellow stock (instead of traditional blue on white) to code prints of preliminary drawings. Potential uses for color-coding in engineering paper work are virtually unlimited: shop orders, bills of material, spec sheets, change notices, cost estimates, etc., etc. Like a copy of our new Color-Coding Booklet? It tells how a truly versatile, full-range color-coding system can be yours with as little effort as it takes to run prints that are black on white. Just write to Ozalid, Johnson City, New York. Booklet L-2-12 New blue-tint Ozacloth cuts glare, saves eyes It's bad enough to have people glare at you. When your drafting materials glare too, one should take steps. Our research people have—by building a delicate blue tint into our new black-line Ozacloth 101 CZB. It provides excellent contrast between background and dye image—cuts glare, reduces eye strain, makes duplicate originals that are easy to read and work with. Other features? Highest printing speed of any cloth intermediate...and a plastic matte surface on both sides which accepts pencil, ink or typewriter...and keeps sheets from sticking together in files. Write Ozalid at Johnson City, New York, for free descriptive literature on blue tint Ozacloth. Turn up the thermostat! Cornell-Dubilier's Teflon* film dielectric capacitors can take the heat...up to operating temperatures of 250°C without a moment's discomfort. For that matter, C-D's Mylar® and polystyrene are almost equally immune to any environmental, life or performance ordeal to which you may want to put them. C-D film dielectrics also offer the widest selection of electrical ratings, case styles, materials and configurations to satisfy space, weight and cost limitations. And they are immediately available in production quantities. Ask now for C-D engineering assistance and bulletins on all the film dielectrics shown here. Write to Cornell-Dubilier Electric Corporation, South Plainfield, New Jersey. Manufacturers of consistently dependable capacitors, filters and networks for electronics, thermonucleonics, broadcasting and utility use for 50 years. DuPont Reg. T.M. ELECTRONICALLY speaking..... CUBIC DIGITAL SYSTEMS speak for themselves: DC VOLTAGE MEASUREMENT: Models V41, V51, with Control Unit C-1 AC VOLTAGE MEASUREMENT: Models AC-1 (manual ranging), AC-2 (Automatic) AC Converters RESISTANCE MEASUREMENT: Models O-41, O-51 4 and 5-digit Ohmmeters, with Control Unit C-2 VOLTAGE RATIO MEASUREMENT: Models R-41, R-51 Ratiometers, Models VR-41, VR-51 Volt-Ratiometers INPUT SAMPLING: Model MS-2 100-point Scanners, Models MS-1, AS-1 Master-Auxiliary 1000-point Scanning Systems DATA RECORDING: Model PC-Series Printer Controls for use with any quality printer on the market Cubic Corporation manufactures the superior* digital instrumentation, including the Voltmeter with these unparalleled electronic features: - .003% repeatability, precise accuracy again and again - .003% attenuator accuracy for wide-range precision - .003% bridge linearity with carefully matched quality components PLUS maximum noise rejection with the finest input filter known...and unmatched stability through superior noise rejection. The Cubic 4 and 5-digit Voltmeters are only two of the superior instruments in a line of digital instrumentation available on an off-the-shelf basis for simple bench use or for easy development of sophisticated systems for measurement and recording. NEW FROM CUBIC... The Talking Meter, instrumentation that really does speak for itself, instrumentation that provides a new dimension in "readout," measurements or other parameters reported to the ear by a clear human voice. Years-ahead engineering, factory production techniques inspired by pride in the end result, careful quality control and reliability testing...all these factors make Cubic's the truly fine instrumentation...Digital Systems that speak for themselves. CUBIC CORPORATION INDUSTRIAL DIVISION 5575 Kearny Villa Road, San Diego 11, California Electronic Engineering With a Dimension for the Future SCIENCE IS UNIVERSAL and knows no borders. This is particularly true of electronics, which heavily influences all other sciences. Any complacency that might have existed in the United States regarding the scientific abilities of other countries was largely dispelled by Sputnik, has been further dissipated by the recent success of other countries in export markets and, indeed, within the United States itself. Many countries in a disadvantageous economic position immediately following the war are now making up for lost time. Research and development in the field of electronics is rapidly advancing in the countries covered within this special report by men recognized as expert. They were approached personally in ELECTRONICS' behalf by McGraw-Hill's unique World News Bureau, and we are particularly indebted to representatives in London, Paris, Milan, Stockholm, Zurich, Tel Aviv, Tokyo and Melbourne for the result. Not all scientifically important countries are covered, nor are all the important research and development activities within these countries. Security requires some omission abroad as it does at home and commercial reticence is not unheard of. We have carried many stories about specific Russian engineering and will again. But to our request for a roundup the Soviet officially said "Nyet." A West German author promised a report right up to deadline and then cabled that company executives had turned thumbs down. A Netherlands author led us to believe he would cover the Dutch waterfront and then delivered a description of one project preoccupying one company which we will publish conventionally later. We found the reports that follow informative, instructive and fascinating to read in detail. We hope this will be your reaction. W W Mac Donald EDITOR REPRINTS AVAILABLE SEE READER SERVICE CARD © ELECTRONICS, a McGraw-Hill Publication Silicon parametric diode may lead to breakthrough, thin-film magnetic storage approaches ability of brain, waveguide point-to-point system attracts attention and c-w radar returns to the news By ERIC EASTWOOD Chief of Research, Marconi's Wireless Telegraph Co., Ltd., Great Baddow, Essex At no previous time in the history of telecommunications has the rate of progress been greater than at present; neither has there been greater opportunity for new ideas and inventions. This is true whether we consider the basic devices, materials and circuit elements of electronics on the one hand, or, on the other, the association of these components to form the systems necessary to order the fundamental processes of sensing or information collection, information transmission and, finally, data utilization and control. SEMICONDUCTORS AND TRANSISTORS—If the period 1948-1958 were termed the "Decade of Promise" for the transistor, then the period 1958-1968 must surely be the "Decade of Fulfilment". No longer does the research engineer merely assess the contribution that transistors might make; he is too busy seeking new devices the development engineer is demanding for his next circuit design. The British transistor industry is growing rapidly to meet the needs of the new era of electronics. It is not allowing its preoccupation with production and testing techniques to obscure the need for devices capable of operating at higher frequencies and at higher power levels. Structures of all kinds are being investigated. For example, interest in transistor geometries has brought a germanium mesa structure of the 500-mc class to the production stage. (Ed. Note: Illustrated in pictorial section of this Report.) This type of work depends greatly upon the availability of primary material of the requisite purity. Silicon devices, with their promise of better temperature performance, are making steady progress. Progress is likely to be speeded with the wider use of the vapor deposition—floating zone purification technique. In this connection, recent work on the silicon parametric diode may lead to a semiconductor break in the kilomegacycle circuit engineering field. THIN-FILM STORAGE—A transistor capable of operating effectively at frequencies in the order of hundreds of megacycles is required urgently, not only to permit complete transistorization of the tv, vhf and uhf bands, but also to accommodate high speed switching characteristics of thin-film magnetic storage elements which are likely to emerge in the near future. Intense research effort is being devoted to the magnetic properties of thin films, and it is clear that turnover times measured in millimicroseconds will be achieved. It is an impressive thought that storage elements of this type, in association with the new transistors that will be developed to match them, can lead to such a high degree of miniaturization that information storage capacity not incomparable with that of the brain itself may be feasible. The problem of access to these active storage elements still remains formidable, for we have as yet nothing comparable to the elegant electrochemical linkages achieved by the brain and the central nervous system. RADAR—Developments in military radar during the early postwar years were still directed towards the wartime operational objective of countrywide defense against airborne attack. The aim was to achieve comprehensive radar cover and, to this end, transmitter powers were substantially increasing. Low noise-figure receivers and highly directive antennas were also developed, with other components. Many of these developments have now become available for civil use, but their adaptation is a highly selective one since the civil need can depart widely from the original military objective. A military radar seeks for good range performance in spite of electronic countermeasures by an enemy; it is thus not possible for transmitter power to be traded for receiver noise figure. In the civil case, no jamming exists and economics demands that receiver noise figure be improved in order to reduce the need for transmitter power. For this reason, civil interest in parametric amplifiers is great. The threat of jamming influences the military choice of radar frequencies, but civil authorities are free to choose the frequency best adapted to their purpose. This freedom has recently been exercised by the Ministry of Transport and Civil Aviation, who has stated its intention to use the 50-cm band as the wavelength of one of the components of its new Airways Surveillance System. This choice is dictated by the desire to reduce rain and ground clutter effects to a minimum, and so to secure continuous tracking of all traffic in the airways. The new approach to the civil radar problem is illustrated by the 500-kw output stage of a transmitter that has recently appeared. This radar is based on a new 50-cm klystron which permits maximum advantage to be taken of a fully coherent system, achieving both good cover performance and reliable MTI. DIRECTIVE ANTENNAS—Large-aperture antennas which provide very narrow beams in the horizontal plane are as attractive to civil authority as to the military, but in both cases it is necessary that sidelobes be reduced to a minimum. A great deal of effort is, therefore, being applied to the sidelobe problem and in this work simulation studies by electronic computers of required phase and amplitude patterns are making a great contribution. No longer can 20 db be accepted as an adequate first sidelobe. The large antenna also poses many difficult mechanical problems such as the accurate figuring of a large double curvature reflector, techniques of stressed skin fabrication and profile distortion in wind. These problems are being studied not only in radar laboratories but also in communication laboratories because of their interest in tropospheric scatter systems. Airframe manufacturers are interested since it has been found that techniques developed for the accurate fabrication of wings can be of great assistance to the antenna designer. The combination of greatly increased transmitter power and very large antennas for both radar and communications has produced a recognition of how little is known of the effects of microwave radiation upon the human body. Study of the physiological effects of intense fluxes of radio-frequency power has been commenced and considerable thought devoted to the development of foolproof alarm devices to indicate when the flux at any selected location exceeds the 10 mw per square centimeter which is at present regarded as a reasonably safe limit for continuous exposure. CIVIL AIR TRAFFIC CONTROL—The growing volume of air traffic is likely to saturate the present system of procedural control in the not too distant future. At present the safety of aircraft is insured by imposing physical separations which are excessively wide, and this leads to rapid saturation of the airways' capacity. The introduction of ground radar will permit the authority to monitor the content of the airways more accurately. The introduction of this system will demand not only a comprehensive radar data handling system, but also the integration and utilization of both radar data and the present procedural information derived from flight plans and aircraft reports. The problem of utilization of radar data is consequently being studied as a matter of urgency in many radar laboratories, industrial as well as governmental, and much effort is being applied to the development of components that will probably be required. One such component is the symbolic or synthetic display illustrated in Fig. 1; such a display permits the controller to present edited information and position about classes of aircraft selected from the storage device and is more adapted to his task than is a raw radar display. The alphanumeric characters shown upon the screen are derived from a transistorized character generator, and are located about the point on the plan display which is the extrapolated position to be taken up by the aircraft at any selected time. COMMUNICATIONS—Progress towards more information-carrying capacity in the Hartley sense goes on continuously, and methods of improving the signal-to-noise ratio on any circuit are still as earnestly sought after. Improvements in antenna performance are as vital to the communication engineer as to the radar engineer. It is now considered essential, for example, to achieve sidelobe levels in the tropospheric case of not worse than 36 db. Improvements of signal-to-noise ratio by reduction employs a two stage amplifier. The input is 10 to 20 mw at the radiated frequency and the output from the high power stage is 1 kw. Research effort is being applied to other modes of communication, particularly point-to-point. The $H_{o1}$ mode of waveguide transmission is generating a surprising amount of interest for a subject which many engineers might have thought to be wholly academic. This is not the case, however, as the rapid growth of television and the need for large channel capacity to accommodate digital transfer of business data suggests that short-haul $H_{o1}$ linkages into busy city centers may not be too far distant. To meet this need in the future, study is being made of modulation techniques, the virtues of fm and pcm are under scrutiny and special thought is being given to the characteristics of the waveguide and the methods for its fabrication and installation. A recent demonstration by S. T. & Co. of such a circuit transmitting a color-tv signal was most impressive. SATELLITE WORK—The initiation of a modest satellite research program by the government has stimulated interest in satellite communications. Telemetry circuits, or data links that might be used in such work, are being actively considered and the design problems associated with packaged transmitters and power supplies in a satellite environment are in the early laboratory stage. Communication circuits depending upon reflection from families of satellites in the role of passive repeaters are also being examined for feasibility and from the point of view of economics. In this connection the recent successful experiments on moon communication conducted by Jodrell Bank with the aid of a Pye 201-mc transmitter are extremely suggestive. CW DOPPLER STUDIES—Although the effectiveness of cw radar as a research tool was demonstrated by Appleton as long ago as 1923, when he commenced his investigations on the structure of the ionosphere, the technique of radiolocation by use of continuous waves received scant attention during the wartime years compared with its pulsed counterpart. It is, therefore, all the more remarkable that there should have been a great upsurge of interest in cw radar methods during the last year or two. In the first place, developments in high-power pulsed radars for military deployment were rapidly approaching the point where major limitation on performance was produced by the earth's curvature. Or, rather, this would have been the case had not the emergence of high-power microwave jamming devices profoundly changed the whole situation. A sure method of improving the performance of a radar in the presence of electronic jamming is to increase the transmitter power, but this is economically expensive. To increase the mean power of a pulsed radar required that many difficult problems associated with the generation and handling of high-power pulses should be overcome, and so it was attractive to consider the cw approach to the problem of increasing the level of the average power radiated. Again, in certain military situations accurate range information on a target is unnecessary, but initial detection and angular coordinates are essential. In such a case the cw radar, with extraction of the doppler signal, is ideal. Civil interest in cw radar techniques has also grown during this period because of a number of requirements that demand the use of a radar capable of operating over very short ranges, probably in the presence of heavy ground clutter and with the need for accurate velocity extraction. Examples of civil applications are traffic control on both road and rail and velocity-measuring radars applied to the study of automobile performance and braking. Growth of interest was fortunately matched by intensive research work in a number of tube laboratories, aimed at producing the necessary generators of microwave power. This work has resulted in spectacular reduction in am and fm tube noise produced in cw klystrons, accompanied by great increase in power output. The klystron work is also being influenced by the needs of tropospheric-scatter communications, and the net result is that klystron generators are now available having a wide range of power outputs well above one kilowatt, at frequencies which span the whole microwave band from 500 mc to 10,000 mc. The power of the cw approach, also the trend in design of small velocity measuring radars, is illustrated by a new electronic velocity analyser. This unit provides a means of obtaining a continuous and highly accurate record of the velocity-time characteristic of a whole class of moving targets such as aircraft at takeoff or touchdown, road vehicles, projectiles from guns, rockets and guided missiles. The equipment operates at X-band and provides an accuracy in velocity measurement of ±1 ft per sec within any velocity bracket of 115 ft per sec up to 5,000 ft per sec. To facilitate velocity bracket selection, the local oscillator is of switched RC type which can be phase locked to suitable harmonics of a crystal-controlled oscillator. The performance of this type of velocity measuring radar is illustrated in Fig. 3, which shows the velocity-time record yielded by a shell leaving the muzzle of a gun. IMPROVED COMPONENTS—The various subjects discussed above represent some of the more exciting prospects in telecommunications, but it would be inappropriate to close without some reference to the less exciting but equally vital work that is directed towards improvement of the basic elements of the whole technology, the electronic components themselves. Reliability in military electronic equipment is essential, and so it is not unnatural that government establishments should have provided leadership in this field. The military electronic environment is also somewhat exotic, but the component designer in meeting the needs of guided missiles and supersonic fighter aircraft is also helping to provide higher reliability in civil electronic equipment. The industrial laboratories are profiting from this work and they may always be relied upon to temper military desirability with commercial expediency. --- FRANCE Nuclear fuel leaks detected by "sweating", tubes cooled by static water, micro electron-probe analyzes alloys, new ferrites promised, masers measure earth's field and help spectroscopy By GEORGES GOUDET General Manager, Laboratoire Central de Telecommunications, Paris and PIERRE GRIVET Professor of Radioelectricity and Electronics, University of Paris FRANCE participates actively in the world's technical evolution, thanks to increasing research and advanced development. This work is shared among the universities, schools of engineering, the Centre National de la Recherche Scientifique, several large state organizations such as the Centre National d'Etudes des Telecommunications and many private industrial laboratories. Scientific development faces an acute shortage of manpower, and various remedies are being put into action. The universities have introduced a short cycle leading to a simplified Ph.D. thesis in two years; many are building new specialized institutions. WIRELESS COMMUNICATION—The network of 4000-mc radio links for telephone and tv is being extended. For example, a network extends through Corsica and Sardinia to Algeria, and has been interconnected with Switzerland, Germany and Belgium. More recently, a connection between Lille and Folkestone, England, has been established by LMT. Among interesting new developments is a cigar or "sausage" radiator, and a new transmission method combining tropospheric scattering and diversity by variation of frequency. Another is that of STAREC, which has developed a slot antenna for jet airplanes covering 225 to 400-mc and multicouplers for vhf and uhf to be used in Paris airports. WIRE COMMUNICATION—The Marseilles-Algiers submarine cable now handles 60 channels, uses 28 flexible bidirectional repeaters developed by CIT. The use of $H_{o1}$ (or $TE_{o1}$) mode in waveguides for long distance links is being studied. Another recent achievement is a 240-line fully electronic telephone switchboard developed by LCT. Conventional electromechanical switches have been replaced by cold-cathode tubes controlled by a central computer using semiconductor devices and ferrite memories. TELEVISION—Several companies have produced lightweight sets for industrial purposes or for submarine prospecting based on vidicons. There is also color tv designed for surgical operations. RADAR AND NAVIGATION—Laboratories Derveaux has developed a 10,000-mc radar and a range finder for aircraft, 3,000 and 10,000-mc navigation radars for ships, radars for harbor surveillance and associated beacons. CSF is producing the radar used in the Mirage III fighter and has also developed a system of diversity reception for long-range protection; it is used by NATO in various areas of Europe. COMPUTERS, INDUSTRIAL CONTROL—The OMEL2 is a versatile and popular analog computer. It uses 12 d-c amplifiers, 30 potentiometers for the insertion of coefficients or initial conditions, and may be used with a rack of nonlinear elements working as function synthesizers or multipliers or with a shortcut system for inverting matrices. One analog computer uses only passive components and operates with hf currents. Companies des Machines Bull has constructed digital computers utilizing delay lines for memories (capacity 12 bits) in models suitable for accounting. For scientific calculations, these can use a magnetic drum which multiplies capacity by a factor of 1,000. In the domain of industrial control, one good example of progress is CIT equipment for control of the first nuclear power plant of Electricite de France, at Chinon. During a scanning cycle of one minute, it registers temperature of 1,250 rods. Another example is the monitoring and control equipment of the EL3 reactor installed by the National Atomic Agency in Saclay. It performs neutron flux measurements, manual or automatic driving of the reactor, control of the safety rods, measurements of radiation and collects physical data such as temperature, pressure. In the domain of components, SEA is producing an efficient type of d-c servomotor with printed flat windings. REACTORS, ACCELERATORS—Saturne has built a three-gev proton synchrotron. Among the main electronic problems involved was automatic control of frequency, which must continuously vary between 650 kc and 8.5 mc. This was solved by using a Ni-Zn ferrite, the "Fernilite" of LTT. This material is used in the master oscillator as well as in the main accelerating cavity; it has made it possible to reduce the necessary hf power to 1 kw, which is one-tenth that of earlier types. The frequency-control equipment itself, made by SEA, comprises an analog computer, an integrator and a power amplifier. It maintains the instantaneous frequency to an accuracy $10^{-3}$ at the beginning of a cycle and $5.10^{-3}$ at the end. A linear electron accelerator has been built for the new Paris University by CSF. This design provides presently 50-ma peak current, with a repetition rate of 300 per second and a pulse length of 2.5 $\mu$sec at a final output of 250 mev; conversion to higher energies (1 gev) seems impeded only by choice of site, in the center of the Orsay campus. Another achievement is the equipment of CFTH for checking the tightness of nuclear fuel cans. The envelope to be tested is evacuated and placed under high helium pressure. If helium has penetrated during the first phase it "sweats" out during the second and is then detected by a mass spectrometer. A leakage of $10^{-10}$ cm$^3$ at atmospheric pressure is detectable. Many firms are engaged in the field of radiation measurement apparatus. A typical case is that of an equipment made by Intertechnique, in collaboration with the Oil Refining Co. Shell-Berre. It measures the sulphur in oil by absorption of soft x-rays by a Geiger-Muller counter tube. A quantitative measurement is carried out in five minutes instead of the 30 minutes or more required by previous methods. TUBES—The tube field has many new developments. An example of particular interest is the series of Vapotrons by CFTH. These are transmitting power tubes (from 10 to 300 kw) cooled by evaporation of water in a purely static arrangement. There is also a triode capable of delivering three megawatts peak power at 200 mc for long pulses up to 200 $\mu$sec. It is used in the 50-mev injector of the 25-gev strong-focusing proton-synchrotron which reached full energy in November at the CERN in Geneva. It will also be used at Brookhaven in the U.S. for the new 35-gev project. There is a klystron which delivers peak power of 30 mw at 3000 mc. The average power is 20 kw, the bandwidth 60 mc. In the big linear accelerator at Orsay 16 CSF tubes of 20 mw each are supplying power at 3000 mc. A recent model traveling-wave tube uses electrostatic focusing, with two helices at different d-c voltages. A peak power of 1 kw is obtained from 1000 to 2000 mc. There are magnetrons for radar, including one giving 2.5 mw at 3000-mc. There is also a family of carcinotrons, with or without magnetic fields. A recent oscillator covers a continuous range from 1000 mc to 37.5 kmc. In a very different domain, there are two new image converters. The first transforms a radar image into a much brighter television image. Observation can be made at a distance in normally lighted rooms. Furthermore, it is provided with an adjustable memory. The second is made for medical purposes. It transforms an x-ray pattern into a 3,000-times brighter tv image. This high intensification allows an important decrease of the x-ray dose. Input field diameter is 6 inches, while the viewing screen has a diameter of 9 inches. The photomultiplier has taken on increasing importance for the detection of nuclear radiation. A variety have been built by LEP and others, with diameters reaching 8 inches and current gains of the order of $10^5$. Special care has been taken to obtain transit time constant for all possible trajectories of the electrons. La Radiotechnique announces a constancy of $10^{-4}$. In the laboratory of the Paris Observatory, low-noise photomultipliers and infrared photocells for threshold applications are being studied and developed for military as well as for astronomical purposes by Dr. Lallemand. At the Toulouse University the group led by Prof. Dupouy is devoted to electron optics, and their work culminated in the recent development of a very flexible magnetic electron microscope which is currently produced by OPL. In Paris, a joint effort of Prof. Guinier's x-ray laboratory and of the ONERA enabled Dr. Castaing to develop fully his original micro electron-probe, which is now industrially produced by CAMECA. This apparatus is excellent in speed and accuracy for the quantitative microanalysis of alloys. There is also a CFTH simplified mass spectrograph for leak detection, and a high-resolution device for the quantitative determination of heavy elements and especially uranium compounds by CSF. SOLID STATE—In the field of semiconductors, French production is still relatively low. However, valuable performances have been obtained. An example is a Silec silicon power rectifier, rated at 650 peak inverse volts. In the Ecole Normale Superieure, Prof. Aigrain is studying semiconductors; the old PEM effect of Kikoin and Naskov was rediscovered and shaped to form an accurate tool of control and research. The study of magnetism is conducted at the University of Grenoble, where Prof. Neel founded the present theory of ferrites and where new garnets were recently developed. And at CNRS laboratory at Bellevue Prof. Guillaud initiated the development of MnBi magnets and of numerous compositions for ferrites; those are now produced by LTT. An early but fundamental contribution to the art of masers was made in the Ecole Normale by Prof. Kastler, who coined the term "pumping between quantum levels" when discovering with Dr. Brossel the technique of optical pumping. A new contribution was recently added by Abragam's group at CEA Saclay, which synthesized an original medium for a three-level maser in an aqueous solution of paramagnetic salt. It is used for convenient and accurate measurement of the earth's magnetic field. Still another low-frequency maser using two levels and nuclear resonance in a stream of water was developed at the University of Paris and may prove useful in nuclear spectroscopy. Even more recently, an effect similar to Overhauser's was noted by Prof. Ubersfeld of Ecole de Physique & Chimie and explained by Abragam; it is hoped that greater understanding of this phenomenon will help solve problems of nuclear orientation and nuclear spectroscopy. Off to a late start, the electronics industry has nevertheless achieved some component miniaturization, improved magnetic industrial controls, designed and modestly produced high-quality instruments. By PAOLO MARSILI Professor of Electronics and Control Equipment, University of Genoa and GIORGIO QUAZZA Control Equipment Study Center, Finmeccanica, Instituto Ricostruzione Industriale, Rome THE ECONOMIC and political situation in Italy following the war conditioned development of its electronics industry. In many other countries the striving for perfection was dictated by exasperating technical requirements of military problems. In Italy, there were severe limitations in the defense budget. COMPONENT PARTS—The qualifications of manufacturers of components required in military radars and servomechanisms has now made available paper, oil, ceramic and tantalum capacitors of minimum size and maximum stability at high temperatures, as well as reactors, rectifiers, impulse transformers, relays, connectors, synchros and two-phase motors for small servomechanisms. The home market does not yet seem to justify production of high-quality resistors, special iron for magnetic amplifiers or memories. The same situation exists with respect to potentiometers for analog computers as well as to high-quality oscilloscopes and other laboratory electronic instruments. On the other hand, the demand for commercial equipment has led two firms to start manufacturing germanium diodes and transistors. Others are achieving noteworthy production of vacuum, gas and cold-cathode electronic tubes, tv picture tubes, magnetrons, klystrons and x-ray tubes. AUTOMATIC CONTROLS—Automatic controls make up a large part of the activity in electronics. This is noted chiefly in collaboration between manufacturers of machines, or of manufacturing plants, and electronic control-equipment designers. In all sectors of industry, electronic firms contribute in varying measure to solve speed, voltage, current, frequency, power, positioning, axes synchronizing, gage coordination, sequential programming, counting, and data processing problems. A firm has been formed with the exclusive aim of electronics research and development. It is engaged in development of controls for large rolling mills and paper mills, computer servomechanisms, power-plant combustion controls and operational amplifiers, data-processing computers and static-power converters, large alternator voltage-regulators and machine-tool positioning, telecommunications for utility networks and many other things. MAGNETICS AND SERVOMECHANISMS—There is great vitality among small manufacturers of magnetic-amplifier motor controls who base their business more on low-cost production of a few types of widely used equipment than on versatility of applications. One firm lives on the construction of special-application, high-quality electronic servomechanisms. and the building of highly specialized metadynes. Iron and steel plants requiring perfected speed, drawing and thickness controls often use mercury-vapor rectifiers. Printing and paper mills, where analogous problems of synchronization, of continuous regulated-speed operations, constant-draw rolling and unrolling, controlled quick stops and starts have for several years favored electronic solutions (initially thyratrons) today mainly use magnetic amplifiers with transistor pilot stages. There are new longitudinal register controls for rotogravure, with extremely accurate electronic synchronization of the sections, operations and programmers for magnetic-drum paper cutting machines that memorize variable programs and control the automatic cutting cycle. These are also useful in operation of speed boxing machines, box counters, plastic-bag cutting controls, and in cable and rubber production where power, drawing and speed-control problems still exist, especially in rolling and unrolling. Frequent use of variable-speed magnetic amplifiers or program controls is in the textile industry, especially in ring spinners and in hosiery looms and warping machines. These are low cost solutions, except for large man-made fiber production plants, where, next to warp or yarn-regularity electronic controls, there are interdependent operation sequence devices. In the chemical, cement, and glass industry automatic controls have been added to electronic speed regulators in the control of electrolytic bath currents, heat-exchanger power, electric furnace temperatures, glass-mixture composition, plate thicknesses. In electric-energy production and distribution, next to alternator-voltage control, the electronics industry has helped solve auxiliary problems such as stabilized-frequency generators, telecommunications and telemetry. On ships there are many magnetic-amplifier voltage controls. In port there is the magnetic control of lifting equipment. INSTRUMENTS—The production of electronic instruments not strictly connected with automatic control has many voids, but does offer high-quality instruments. For example, still non-existent are professional cathode-ray oscilloscopes, variable-frequency oscillators, electronic voltmeters, while of noteworthy quality, are nucleonic measuring instruments such as radioactivity meters, impulse analyzers, meters for very short time intervals. There are also limited examples of pioneer production of microwave measuring instruments. There is good local production of pH meters, conductivity meters, temperature recorders, electric-field meters, strain-gage elastic stress recorders, electro-acoustic meters, stress meters, zero amplifiers for dimension recorders of grinder parts, photoelectric devices for various uses. Italian companies also make medical x-ray equipment and heavy industrial gear such as resistance welders and induction heating equipment. COMMUNICATIONS—That part of the electronics industry that designs and builds telephone exchanges, carrier-frequency equipment, radio links, telemeters and remote controls is fairly mature and self-sufficient. Technically, it appears to be about at the same level as other advanced countries, but there is less basic research. There is conspicuous national production of specific telephone-equipment components such as relays and selectors. Very active is the production of radio and TV, including transmitting equipment manufactured by at least three large firms for the Italian Radio and Television State Network, for the Navy and other military uses, and for industrial applications. The production of picture tubes has been very active. There is also the beginning of production of sound-reproduction equipment including high fidelity. COMPUTERS—A few years ago, a university designed and built an electronic computer with an extremely large memory capacity. Now the digital technique has just left the laboratories and gone into industrial production. The presentation of a numerical electronic computer for use in banking, mathematical and logical operations by a large manufacturer of office machinery is of this year. Input is on punched tape, card or magnetic tape; output is on punched tape, card, magnetic tape or teletype. The computer has a ferrite-core memory in units from 1 to 8 capable of 120,000 bits each, with $10/\mu\text{sec}$ access time, $10/\mu\text{sec}$ time for elementary logical operation. The computer also has up to 20 tape memories with total capacity of 180,000,000 characters, reading and recording speed of 35,000 characters/sec and magnetic drums each with 120,000 characters and $10/\mu\text{sec}$ access time. MILITARY—Fire-control-room analog computers, which today at least five firms manufacture in various forms, are technically up to date. Also quite modern is firing radar, missile telemetry equipment, countermeasures, two-way aircraft transmitters, ultrasonic equipment. TEACHING AND RESEARCH—Credit must be given to state institutions for their considerable contribution to the foundation of research centers, especially in nuclear electronics. There are today electronics and servomechanism sections in at least four nuclear research centers as well as in many universities, in electrotechnical and telecommunication institutes and microwave research centers. Moreover, the polytechnics and universities, despite relatively unelastic academic programs, are giving courses in electronics, computers, automatic controls, and microwaves. Different but equally important is the operation of the electronic centers of a few large private companies striving to improve their production with the introduction of electronic control equipment and of numerical computers for bookkeeping operations, and studying at times independently their most complex technical problems with the aid of analog computers. In general, the Italian electronics industry today appears to be technically very much alive. Free magnetic rings in plasma may solve thermonuclear-energy problem, computers announce subway-train destinations, transistorized telephones talk loud and a traveling-wave maser is under construction By BERTIL AGDUR Professor, Royal Institute of Technology, Stockholm INVESTIGATIONS of gas discharge plasmas, with the aim of generating thermonuclear energy, are being carried out at the Royal Institute of Technology and at the University in Uppsala. One experiment is intended to produce free plasma rings. A 200 μF, 5-8 kv capacitor bank is discharged in 0.03-3 dry hydrogen between coaxial electrodes, forming a plasma gun. A ring of plasma is formed, and accelerated by electrodynamic force towards the muzzle of the gun. There it encounters a static magnetic field directed radially between the electrodes. A circulating current is induced in the plasma and it leaves the gun as a free magnetized ring. The ring moves in a drift tube of glass, where it is studied by means of Kerr-cell photography and by magnetic measurements. The circuit current represents a rapidly damped oscillation, passing through zero 13 μs after ignition of the discharge. The magnetic flux connected with the ring increases to a maximum, which often appears when the circuit current passes zero, and then decays approximately exponentially. When circuit current passes zero the flux curve is smooth and the direction of the flux does not change. This indicates that a current continues to circulate in the plasma independent of the circuit current for about 40 μs. Recent developments in the field of high-voltage dc transmission at ASEA, representing the electric power industry, include design and construction of mercury-arc valves for 160 mw cross-channel transmission between England and France by cable. Conductors are at ±100 kv with respect to ground. Inauguration of the service is scheduled for 1961. Six tubes handle 800 amp at 100 kv on the dc side. Compared with those in the high-voltage dc transmission system commissioned in 1954 between the Swedish mainland and the island of Gotland, these new tubes provide an eightfold increase in power-handling capacity. ASEA is also studying ac arc discharges at currents of the order of 200,000 amp. Such arc dischargers are used in high-voltage switches. Investigations are made of conditions for rapid deionization of the arc and of fast build-up of dielectric strength in the discharge region after the arc is extinguished. COMPUTERS AND DATA-PROCESSING—Typical companies working in the computer and data-processing field include AB Atvidaberg, SAAB Aircraft, Bo Nyman, Philips Teleindustri, Standard Radio and Telephone and LKB-Produkter. There are also many smaller companies working on data-processing systems for civil and military applications. One interesting example of development work in this area is a type of random-access memory system, the so-called "Carousel". Data are recorded on short pieces of magnetic tape which are wound around 64 separate spools. These spools are arranged in two concentric rows on a large wheel rotating on a horizontal axis. Each spool can also rotate around its own axis. Both the spools and the carousel can easily be removed, which means that data can quickly be exchanged by changing the information-carrying medium as in a conventional magnetic tape unit. Selection is performed in three steps. First a particular carousel is selected; each carousel unit has a storage capacity of about three million alpha numeric characters. Then a particular spool is selected by turning the carousel until the desired spool is directly below the center of the carousel, which is then stopped. The carousel can rotate left or right in order to achieve the shortest possible access time. The free end of each tape is equipped with a small weight, which guides the tape to a magnetic head and a drive capstan located below the carousel. A roller presses the tape against the capstan and the tape unreels downwards and is accumulated in a tank. The third step in the selection process is performed during the read-write operation, while the tape is passing the magnetic head, and at this time specified blocks of information can be selected. Average access time is 1.9 seconds. Rewind starts automatically after each read or write operation. Each spool houses about 8.5 meters of tape. The speed of the tape is five meters per second. There are eight channels, one of which is used for checking. The magnetic head uses ferrite cores and there is a constant air gap between the head and the surface of the tape. "Drop outs," one of the most serious problems encountered in connection with digital magnetic-tape recording, have been virtually eliminated. Bo Nyman is manufacturing a data machine, Wegematic 1000, which is a development of the earlier Alvac III E. This is a binary-series machine with magnetic-drum memory. Commercial developments are under way at SAAB on digital computers, analog computers and numerically controlled machine tools. One example is the BT9 toss-bomb sight. The American M2 bombing system is derived from the BT9 system. LKB has designed an automatic system to record the positions and movements of railway cars. On an axlebox of each car there is a soft iron plate with a shape that corresponds to the number of the car in a binary system. Along the railway line are placed coil systems which can read the number plates when the train passes by, and feed the information to a memory. In one of the systems the movement of 7,000 cars can be recorded. A similar system is being developed for automatic destination-announcements in the subway in Stockholm. Standard Radio and Telephone has developed a system for the Swedish Air Force. Data from radar stations are distributed to remote ppi equipment. Presentation is performed with symbols, symbol numbers and vector lines introduced by the inter-scan method. Data handling and storing is performed in digital equipment using transistors and diode logic, with ferrite-core storage. The output from the digital equipment is in binary-decimal form for in-line direct-view presentation and binary-analog for ppi presentation. Ppi circuits for video, sync, waveform generation and antenna rotation modulation are centralized and distributed to the number of ppi's required. TELECOMMUNICATION AND SPEECH RESEARCH—Like most other companies in the telecommunications field, Telefonaktiebolaget L. M. Ericsson is conducting development work aimed at utilizing electronics in telephone exchangers, telephone instruments and transmission circuits as well as for military purposes. One line of development is an all-electronic telephone system working on the time-multiplex principle. This work is being pursued in collaboration with the Swedish Board of Telecommunications. Electronics is gaining ground in the field of telephone instruments as well. A transistorized telephone, with amplified reception, has recently been marketed. Fundamental studies of the acoustic nature of speech and of speech production and speech perception, and new methods of speech transmission, are being carried out at the Speech Transmission Laboratory. This Laboratory is attached to the Division of Telegraphy-Telephony of RIT. The electronic instrumentation used for the research includes unique designs of spectrographs for continuous short-time frequency analysis. New methods of oscillographic data-recording and instrumentation for production of synthetic speech have been developed. Analysis and synthesis techniques have applications for reducing the bandwidth in speech-transmission links and in speech storage and retrieval systems. Techniques for machine recognition of spoken units and of synthetic speech production have applications in man-machine and machine-man communication systems. SEMICONDUCTOR ELECTRONICS—Fundamental research connected with the semiconductor field is carried out at several university laboratories. At the Institute of Semiconductor Research problems concerning crystallization and treatment of semiconductor materials such as silicon, germanium, III-V-alloys and silicon carbide are being studied. They also develop devices for industry, such as rectifying systems made by diffusion processes, controlled rectifiers and $pn\mu m$-switches. Ericsson is developing transistors to satisfy demand for low failure rate and long life, 15 to 50 years. They have, with some success, used intensive temperature cycling before testing. Work at RIT is partly devoted to studies of temperature, stability and noise properties of semiconductor circuits. There has been some progress in development of methods for analyzing and synthesizing $pn$-transitions. COMPONENTS AND INSTRUMENTS—To meet the demand for extended life on electron tubes, ENGLAND - High-power distributed amplifier, showing anode lines (Marconi) - Electronic velocity analyzer used to measure shell velocities - 500-mc germanium mesa transistor (Associated Transistors) - Klystron output stage of 50-cm radar (English Electric Valve) FRANCE - Mode filter for TE₀₁ mode (CNET) - Cigar or "sausage" radiators for 180 mc (CSF) - Operating desk of the EL3 atomic-reactor's monitoring and control equipment (CFTH) from abroad SWEDEN "Carousel" random-access memory (AB Atvidaberg) SWITZERLAND Frequency-shift remote control for power lines Plug-in unit containing 48 switching circuits Transmitter-receiver for 4,000-mc relay station (TRT) Electronic analog computer (SEA) Transistorized vidicon television camera (LEP) Three-megawatt, 200-mc triode with resonant cavity (CFTH) IDEAS and EQUIPMENT from abroad CONTINUED AUSTRALIA Radio heliogram of sun (right) compared with conventional spectroheliogram Transistorized DME receiver (foreground) compared with tube type 64 paraboloids each 19 feet in diameter are used in crossed interferometer having 1500-foot arms JAPAN NEAC 2201 provides series-parallel decimal presentation (Nippon Electric) 64-character printer handles 300 lines of 120 characters per minute (Oki) HIPAC 101 has parallel-binary presentation (Hitachi) ISRAEL Newly developed coneless woofer and tweeter loudspeakers Ericsson uses passive nickel cathodes. At a silicon content under 0.01-0.02 percent it seems that the interface problem can be almost completely solved even for high-transconductance tubes. Low-noise application of electron tubes might be mastered by means of a refined-grid technique including use of 0.005-mm wire. In experimental tubes grid wires of 0.003 mm diameter have been used. At Chalmers University of Technology low-noise amplifiers, especially for radio astronomy applications, are being developed. A three-level solid-state maser is in operation and a traveling-wave maser is under construction. At the Research Institute of the National Defence in Sweden O-type carcinotrons with bifilar helices giving wide electronic tuning ranges, and parametric traveling-wave amplifiers are being developed. They are also working on ferrite components and have developed one-way attenuators covering the whole bandwidth of a conventional waveguide. Sivers Laboratories have precision swr meters with a residual standing-wave ratio of less than 1.005; they may be driven by a motor for automatic recording. The meters allow direct reading of frequency on a counter and have an accuracy better than 0.1 percent. WAVE PROPAGATION—Studies in the field of wave propagation are being carried out at a number of places. Chalmers University of Technology's wave propagation observatory is equipped with five radio telescopes, average diameter 25 ft, a meteor, solar and satellite-tracking interferometer. Fig. 4 shows a Doppler-shifted 21-cm emission from three spiral arms of our galaxy. At RIND intensive studies are being made of the mechanism of tropospheric scattering. Another program includes continuous measurements of ionospheric conditions. The Institute has developed an aurora all-sky camera. Also being studied is the effect that electrical properties and topography of the ground have on wave propagation in the vhf and uhf regions. Studies of microwave propagation in electron beams are carried out both at CUT and at RIT. The latter is considering the possibility of amplifying microwaves by sending electron beams through a plasma, and some experimental amplifiers have been built on that principle. MEDICAL ELECTRONICS—Work done at Karolinska Institute is one interesting example of developments for medical applications. Several types of "radio pills" for telemetering physiological data from body cavities in animals and humans have been developed. These miniature radiosondes are used to telemeter gastrointestinal pressure waves, pH, temperature, enzyme activity and oxygen-tension values. The lifetime of batteries powering the transmitters varies between a few days and five months. The smallest sonde for telemetering pressure has a volume of 0.9 cm³ and a lifetime of four days. The pills are usually unnoticeable to the patient, and thus permit investigations under almost normal conditions. professors as consultants, and some of their members teach at the universities. Battelle Memorial Institute started work in Geneva in 1952. Initially, the program was limited to investigation of basic properties of semiconductors. The scope was then expanded to include semiconductor devices and circuits. The present program includes logic and memory circuits with transistors and magnetic cores, digital control circuits and fundamental studies of electric motor design and control. Laboratories RCA, Zurich, was established in 1955. Here, an international group of scientists is working in the solid-state field, studying synthesis and properties of insulators and semiconductors. In addition, an Industry Service Laboratory is maintained to provide engineering assistance to European manufacturers, particularly in the field of tv and radio receivers. IBM Research, Zurich, is focussed on exploration of speed obtainable through solid-state materials. The objective is to increase the capabilities of data-processing machines. Thin magnetic films are also the subject of investigation; a physics group is studying the basic physical properties underlying ferromagnetism. Very high speed measuring techniques have been developed, such as a sampling oscilloscope with submillimicrosecond rise time, and an apparatus to measure magnetic-film switching in this range. The same techniques are being used for investigating short-time effects in stimulable phosphors. EDUCATIONAL INSTITUTIONS—Swiss Federal Institute of Technology has six institutes whose research program includes electronics. A. P. Speiser of IBM teaches here. The Institute of Physics, under G. Busch, is carrying out fundamental research on semiconductors, and the Institute of Technical Physics, under E. Baumann, is doing significant work on semiconductor materials, electronic circuitry and television. Eidophor large-screen tv is the result of fundamental research carried out by the former head of the institute, Professor Fischer. The Institute of General Electrical Engineering has focussed its research on automatic-control systems. Eduard Gerecke, head of the institute, is treasurer of the International Federation for Automatic Control. Under M. J. O. Strutt of the Institute for Advanced Electrical Engineering, work is being carried out on instrumentation. Hall-effect wattmeters and flux-density meters are examples. There is also study of such things as stabilization and operating life of transistors. A particularly interesting application is the Hall-effect mixer for broadcast receivers shown in Fig. 5. The Institute of High-Frequency Engineering, under Franz Tank, a former vice-president of the Institute of Radio Engineers, is concentrating on microwaves. The major part of this program is devoted to instrumentation and radar applications. The Institute of Telecommunications has a varied research program, in keeping with the broad interests of its head, Heinrich Weber. Several projects have instrumentation applications; there is analyzing equipment for investigating speech defects, equipment for analyzing teleprinter signals and a transistorized quartz clock with time recording and calibration facilities for field use in goodesy. At the University of Basle, E. Baldinger's Institute of Applied Physics is carrying out work on fast-pulse transistor circuitry and applications to nuclear instrumentation. At the University of Neuchatel and the Swiss Research Laboratory for Time-keeping Devices much significant work is being carried out under the supervision of J. Rossel on electronic instrumentation and masers for time-keeping purposes. These laboratories developed the first maser to operate over a long period of time. At the Institute of Technology of the University of Lausanne work is being carried out on electronic circuitry under R. Dessoulavy. COMPONENTS—Two large firms are engaged in the manufacture of electron tubes, Brown Boveri and Hasler. Manufacturing and research programs are concentrated on transmitting tubes and special devices for microwave and control purposes. Much development work is going on on capacitors, particularly of mica and polystyrol type, by Leclanche, Condensateurs Fribourg, Standard Telephone and Radio. Fundamental research work on new insulating materials is being carried out by CIBA, the chemical industry group in Basle, and by Micafil and Oerlikon and associated wire and cable-insulation companies such as Isola-Werke Breitenbach. Several plants and pilot plants have been set up for manufacturing semiconductor devices, including Ebauches in Neuchatel and Philips in Zurich. As germanium and silicon are already well known, present research is concentrated on semiconducting intermetallic compounds. From theoretical investigations it seems that the most interesting of these materials will be Ag Se and Cu$_2$Se. RADIO AND TELEVISION—Work is being done on uhf variometers by Sondyna AG of Zurich and Autophon AG of Solothurn. SWITCHING TELEPRINTERS—A major application of transistor circuitry is in switching systems. Fig. 6 shows transistorized logical circuits applied to a fully transistorized ARQ system for teleprinter service on radio links. In addition to these circuits using direct-coupled transistor and diode units, the parametron is being studied. New electronic switching circuits enable size of equipment to be reduced one half, power consumption one fourth. One system uses the Van Duuren error-detection principle with a 3-out-of-7 code, error correction being achieved by a 4 or 8-character repetition cycle. Error detection is performed by the code converter itself. Called the Electronic TOR, this system provides four channels by means of time multiplexing. It contains about 1,300 transistors, 3,000 diodes, 11,000 resistors and 160 capacitors and is designed for extreme operating conditions. Reliable operation is maintained even when supply voltages vary by ±50 percent, transistor alphas decrease to 50 percent of nominal value, and cut-off collector-currents increase by a factor of three. Further work is being carried out to eliminate all mechanical elements in the whole teleprinter system. TELEPHONE—Advanced work is in progress in the domain of electronic switching techniques for the development of an all-electronic automatic telephone exchange. In the meantime, semi-electronic systems are being investigated. These are mechanical switching systems combined with electronic central control circuits. The central control circuit consists of two main parts, a memory and a decision-making unit. In the memory, information describing the characteristics of the calling and the called subscriber are stored. In addition, it contains all programs needed to control the decision-making unit. Changes or additional specifications can be easily effected by writing a new program into the memory. Several million bits have to be stored and access time has to be a fraction of a microsecond. Memories using ferrite or permalloy cores are well suited for this purpose. An obvious result of the continuously increasing demand for high-quality long-distance telephone channels at reasonable cost is the use of higher and higher frequencies. This applies not only to radio links but also to cable circuits. The next step in coaxial-cable carrier technique will be the extension of the present 6-mc bandwidth to 12 mc. This future system will provide 2,700 telephone channels, or 1 tv channel plus 1,200 telephone channels, as compared to the present 1,260 telephone channels or one 625-line tv channel. Semiconductor elements are being increasingly employed in channel, group and line equipment. In radio links, work has been started on microwave systems in the 4 to 8,000-mc range. Pulse-code-modulated transmission systems are being particularly studied. It seems that such systems could be competitive with ordinary frequency-sharing systems for short distances. Several firms have recently developed paging systems; Hasler, Autophon, Albiswerk. One system uses pulse-code modulation for selecting the required receiver and the spacing between the pulses is varied, another uses a two-frequency code. OTHER APPLICATIONS—Research work is being carried out on transistorized control and remote-control systems for power plants, railways, pipelines. A laboratory model of one such system uses frequency shift. Frequency shift and bandwidth of the transmitting channel are adjusted to the control speed. A report on electronic research would not be complete without mentioning the development of electronic wristwatches and clocks. Significant work is being carried out by several laboratories. Earlier work enabled the replacement of the watch spring by a miniature battery. The problem now is to design an electronic device to replace the watch escapement. This has not been solved for the wrist watch but much promising work is being done on quartz-controlled clocks, which might lead to a small model for office and home use. Advantage: the clock would run accurately to the minute for over a year. RESEARCH may be broken into two categories, basic and applied. However, it is difficult to say if some of what we call basic electronics research is not in reality research in solid-state physics. SEMICONDUCTORS—In the Department of Experimental Physics of Hebrew University, basic research is carried out on the electrical properties of semiconductor surfaces under the direction of A. Many. The study is concerned mainly with the investigation of the parameters describing the fast states present in semiconductor surfaces and is of fundamental importance in transistor problems. This research has been extended to cover the range of low temperatures. In addition, measurements have been made on the recusation of carrier mobility due to surface scattering. The properties of cadmium sulfide are also being investigated at high temperatures. FERROMAGNETISM—A team has been working on ferromagnetism at Weizmann under the direction of E. H. Frei. Materials are known to be divided into homogeneously magnetized domains. However, in 1945 W. F. Brown, Jr., of the University of Minnesota proved that once a perfect ferromagnetic ellipsoid is magnetized to saturation the energy barrier to change this state of saturation is so high that no domains should be observed under normal laboratory conditions. One of the main results of this discrepancy between theory and experiment, known as Brown's paradox, is that the experimental coercive force is usually two or three orders of magnitude lower than the theoretical one. In 1958 DeBlois measured a coercive force approaching the theoretical value in quite perfect iron whiskers under idealized conditions. This suggests that the paradox can be resolved if materials are made as perfect as required by this theory or if the theory can be modified to include practical materials with their crystal imperfections and non-ellipsoidal shapes. The group in Rehovot is working to resolve the paradox. The experimental part of this work consists of growing single crystals of magnetic materials. In the theoretical work some local lowering of magnetocrystalline anisotropy has been observed in simple models. A team has been working on a new type of ferrite obtained by anion substitution in the known barium ferrite. Theoretically, such ferrites should have a better $BH_{\text{max}}$ product. The properties of Neodymium-Yttrium garnets are also being investigated. NUCLEAR ELECTRONICS—Within the next few years most instruments used in nuclear physics will be transistorized. This will include scalers, linear pulse amplifiers and pulse-height analyzers. Some transistor circuits, such as linear gates and coincidence circuits, have already been constructed and used and have proved very satisfactory. COMPUTERS—Off-line printers, and magnetic tape-to-tape amplifiers, have been built using transistors. Transistors are used in input-output equipment. MEDICAL ELECTRONICS—A team is working on subminiature transistorized electroencephalograph preamplifier and transmitter which could be mounted on the head of a living animal and would transmit normal and epileptic brain waves to a receiver placed at some distance from the animal. The fact that wires are not necessary is important in cases of convulsive epileptic attack and a small transistorized instrument would permit simultaneous transmission from different areas of the brain. ELECTROACOUSTICS—Transistorized amplifiers will soon be constructed to work with a new electroacoustic transducer. This transducer is electrodynamic but has many of the qualities of an electrostatic type and may constitute a turning point in the development of electroacoustical devices. The transducer was the direct result of research in ferrites. The high coercive force of this magnetic material, and its low permeability, gave rise to the possibility of imprinting a magnetic pattern on a ferrite plate to obtain various forms of relatively strong fields outside this plate. This has been done to get multipole flat magnets. It has also resulted in construction of a coneless electrodynamic loudspeaker. This speaker uses a very thin, supple, flat membrane as a radiator. A ribbon of thin aluminum which has a zig-zag form and covers nearly the entire area of the membrane is placed on the membrane by the printed circuit method. At the beginning of research this membrane was stretched above a perforate ferrite plate, where a magnetic pattern was printed according to the geometrical dimensions of the zig-zag in such a way that the vector product of the field by the current had the same direction perpendicular to the membrane over the entire area of this membrane. It was found afterwards to be more efficient to use thin strips of oriented ferrites on a perforate plate, thus constituting a magnetode. The high coercive force of ferrites permitted use of two magnetodes in opposition to each other, and provided a much stronger and more homogenous field. Transducers produced in the laboratory have many interesting qualities and the distortion in the working range of either woofer or tweeter is lower than in the normal cone speaker. The transducers have the qualities of both electrodynamic and electrostatic speakers and may be a step towards higher fidelity. There may be applications of the principle in microphones and stethoscopes and in high-power sound projectors. FUTURE DEVELOPMENT—Industry is concentrated heavily on production of home radio sets. Component manufacture is growing steadily and all except tubes, transistors and resistors are made here. There is some export. The trends for the next few years will include development of printed circuits, high-fidelity chains, stereophony and, with a green light from the government, television; several firms are already training their staffs. Several firms in partnership with foreign companies are beginning to expand in the production of electronic devices. Emphasis is on semiconductors, including tunnel diodes, barrier and "hole" capacitors, but parametrons are coming as computer switching elements and there is considerable interest in electronic language translation. By HIROSHI WADA Chief of Electronics Division, Electrotechnical Laboratory, Ministry of International Trade and Industry, Tokyo A TECHNICAL COMMITTEE was organized to extract germanium from copper sludges or exhausted gasses from coal combustion. Neither process came into industrial use because of high costs, but refining facilities for germanium were a result. Silicon is produced by several chemical plants. Specific resistivity is some hundreds of ohms-per-centimeter. The debut of the transistor radio from Sony stimulated several enterprises to establish transistor manufacturing facilities. At present, progress is directed not only to get cheaper and greater production, but also to develop high-frequency, high-power and switching elements. High-frequency transistors include alloy, drift, grown-diffusion, and diffused-melt-back types, each having 100-150 mc alpha cutoff. In 1960, mesa-types are expected to be incorporated in new TV receivers. In power-handling applications, further development of silicon transistors and rectifiers is the interest. A power rectifier by Toshiba of Tokyo is fabricated by the solid-diffusion technique, and has the rating of 750-v inverse voltage, 400-amp forward current. Solar batteries are under field trial. SEMICONDUCTOR SWITCHING—Recent developments for switching elements are considered very important. There are several activities on silicon $pnpn$ switches and controlled rectifiers. L. Esaki has delineated tunnel-diode operation. He noted the anomalous effect in highly-doped $pn$ junctions, based on the quantum-mechanical penetration of degenerated electrons. In addition to the usual diffusion current by minority carriers, this tunnelling current manifests negative resistance in current-voltage characteristics of the $pn$ junction. The tunnel diode is believed to have no frequency limit, since the mechanism of operation takes practically no time whereas the usual diffusion mechanism takes appreciable time for rearrangement of minority carriers. NEW COMPONENTS—Development of transistor radios and computers has pushed miniaturization of electronic components. Research in capacitors and magnetic ferrites has been great. With the advent of the barium-titanate capacitor in quantity production, Murata of Kyoto has produced i-f filters by using steel balls as mechanical resonators. Figure 7 shows a cross-section of the assembly and bandwidth characteristics. Center frequency is determined by diameter of the balls and bandwidth by the length of the coupling nickel rods. The price is low. Semiconductive properties of barium titanate are also under investigation. The approach visualizes controlling the material with a cerium or niobium-group additive. A barrier capacitor is expected to be developed in the near future. The "hole" capacitor, a name given to the foil-type --- Table I — Characteristics of Parametrons \| \| High Speed \| Standard \| Low Power \| \|--------------------------\|------------\|----------\|-----------\| \| Exciting freq., 2f \| 6 mc \| 2 mc \| 200 kc \| \| Maximum clock freq. \| 140 kc \| 25 kc \| 2 kc \| \| Power for continuous excitation \| 120 mw \| 30 mw \| 5 mw \| \| Dc bias \| 0.6 amp \| 0.6 amp \| 0.6 amp \| \| Maximum number of inputs \| 3 or 5 \| 3 or 5 \| 3 or 5 \| \| Maximum number of output branches \| 12 \| 15 \| 15 \| Table II — Characteristics of Ne-ferrite \| Initial Permeability $\mu_0$ \| Core Loss at 100 kc tan/$\mu_0 \times 10^6$ \| Jordan Coefficients of Core Losses \| Induction at 10 oer B (gauss) \| Curie Point Tc (C) \| \|------------------------------\|---------------------------------------------\|-----------------------------------\|-------------------------------\|------------------\| \| 2000 ± 200 \| 2 \| h/$\mu^2 \times 10^6$ Fn/$\mu \times 10^6$ t/$\mu \times 10^3$ \| 4800 \| 230 \| solid-electrolytic tantalum capacitor developed by Nippon has nearly the same specific capacity as the electrolytic tantalum capacitor. Its schematic cross section is shown in Fig. 8. A thin layer of MgF₂ or CaF₂ is evaporated upon the oxide film of tantalum. Then layers of germanium and of aluminum are evaporated successively. The origin of the capacitor came from Y. Sasaki's idea that the anodic oxide film had a structure similar to a pn junction which one finds in a semiconductor. As the capacitor contains no electrolyte, it can be used over a wide temperature range from −200 to 200 C. The value of leakage current at room temperature is about 0.2 μA/μF/V. Since the invention of the parametron, Tokyo Denki Kagaku has been supplying components. The parametron requires a non-linear reactance. The performance of three hundred kinds of ferrites has been tested to find one that fits best this requirement. Several commercial units, which contain 25 such elements, are shown in Table 1. These can be used as basic elements for computers, automatic-control devices and telephone dialing systems. Products using thin magnetic film are expected to be used in the near future. Ne-ferrite, a Mn-Zn ferrite, has characteristics shown in Table 2. It is mainly used for the filters and coils of telecommunications. COMPUTERS—Research on digital computers was started with a relay computer at the Electrotechnical Laboratory. Then electronic computers using tubes were developed by two groups, and research for transistorized computers followed. E. Goto of Tokyo University later suggested the parametron. If, in the circuit shown in Fig. 9, a direct current and a high-frequency current of 2f are supplied to the primary winding, then an oscillation of frequency f, determined by L and C, will be imposed in the secondary circuit by parametric excitation. In the oscillation thus generated, the phase of the oscillation has either one of the two stable phases which differ by π radians from each other. This difference can represent and memorize one binary digit 0 or 1. The phase of oscillation can be controlled by the phase of the initial current in the secondary circuit. The parametron can also be used to transmit information by a three-phased clock, and to perform various logical operations based on a majority-decision principle by the sum of voltages of odd numbers. The parametron computer PD-1516 and transistor-computer ETL-3 were prototypes. Following these machines, computers ETL-4, M-1 and PC-1 were completed. They are equipped with magnetic drums operating at 18,000 rpm or ferrite-core matrices. Work on them was done by the staffs of university and governmental laboratories. Some manufacturers for telecommunication devices designed their products in accordance with them. Products of Hitachi and Nippon using magnetic drums, together with a high-speed line-printer of Oki, have been exhibited. The first has a word length of 10 decimal digits, 63 instructions, handles 1,040 words. The second has word length of 48 bits, 48 instructions, handles 1,024 words. INDUSTRIAL APPLICATIONS—Mechanization in offices lags in Japan. No products for punched-card systems are available. Under these circumstances, magnetic-tape devices and tape-sorting techniques are under intensive development. Quite recently E. Goto proposed basic circuitry for the realization of an extremely fast computer. His idea is to utilize tunnel diodes as switching elements and to apply the majority-decision principle for logical operations as in the case of parametron circuits. Equipment for automatic control of process industries are actively adopting electronic systems. A system recently revealed by Yokogawa Electric has a unified specification for signal translation. The system, including compensation and stabilization circuits, is well designed. Solid-state devices are used. The transducer for displacement, for example, uses a Hall-effect generator. Low-speed analog computers have been developed by several manufacturers. Research on process control by digital technique is under development. Hokushin Electric has a digital computer intended for use as an on-line machine to calculate operational guides for chemical processes. ETL has developed a high-speed digital computer for on-line control of rotating machinery. It can also be used to investigate computer control by connecting it to an analog computer through analog-digital and digital-analog converters. Digital control of machine tools is under development at Fuji Telecommunication and in other places. OTHER R&D—Other research activities include S. Takahashi's English-Japanese machine translator. Research aims first at the rearrangement of word order in a sentence. A letter-reading machine under study by the author can read both capital and small letters of the alphabet together with other symbols. A voice typewriter for speech and a device designed for character recognition of phonetic letters are under development. Our electronics industry is still too young to have borne much fruit from our own basic research. However, we are optimistic about new research programs. --- AUSTRALIA Radio astronomy is high on the list of developments, embracing electronic telescopes and heliographs. Interest in aviation aids and weather forecasting runs a close second. By E. G. BOWEN Chief, Division of Radiophysics, Commonwealth Scientific & Industrial Research Organization, Sydney The great majority of radio and electronics firms in Australia are associated with or are subsidiaries of British and American firms. They are closely in touch with new technical developments as these occur overseas. One company has, to its considerable advantage, maintained a vigorous local research and development facility for many years. Other manufacturers are starting to open research laboratories to adapt overseas techniques to our environment. The lines along which developments are likely to take place will be broadly similar to those elsewhere, in television services, scatter propagation, solid-state devices, electronic control both for the older and for the newly developing industries, computers for business and scientific purposes. Banks are already keenly aware of special problems they face which are different from those in the Northern hemisphere; in country regions their transactions are spread over a comparatively large number of small branch offices extending over huge areas. AVIATION ELECTRONICS—Aviation has seen rapid growth and aircraft now carry a substantial proportion of the passenger traffic within the country and on its overseas links. The services are operated to very high standards and they have achieved excellent regularity and safety records. A share of the credit must go to excellent communications and navigation systems, some of purely Australian origin. Perhaps the best example is Distance Measuring Equipment, a radar-beacon system giving the pilot direct measurement of distance in miles to his terminal point or to any beacon enroute. See Fig. 10. Maximum range is about 200 miles at 20,000 feet, with an accuracy of one or two miles. Within 10 miles of the landing field the accuracy is ±½ mile. All air routes are covered by the system and the equipment is mandatory on all passenger-carrying aircraft. A transistorized version of DME exists in the laboratory and will shortly be introduced into service. It is half the size, quarter the weight and requires one-tenth the input power; these characteristics might well mean that its use will be extended to smaller private aircraft. In a related field, the application of ground radar to assist the flow of traffic, and of airborne radar for weather avoidance, are well understood and the scene is ripe for developments along these lines. WEATHER AIDS—The Weather Service has suffered from a dearth of information imposed by the vast size of the continent and the comparatively small population. Much of our weather has its origin in the surrounding oceans, and there is almost a complete lack of precise information on the movement of weather systems from these regions. The situation is wide open for the application of radar and other aids. These must be supplemented by observations of events in the high atmosphere which undoubtedly contribute to the vagaries of weather experienced on the surface. Information along these lines is already obtained from Doppler observations of the drift of meteor trails at Adelaide University and from rocket observations at the Woomera Range. SATELLITE OBSERVATIONS—In common with Observatory. They show that the radio disturbances originate in the sun's atmosphere rather than at the surface and give data from which the density and temperature structure of the sun's atmosphere can be obtained out to distances which were inaccessible to the astronomer prior to the advent of radio measurements. Results in the galactic field are no less exciting. In collaboration with Leiden observers, a complete map of the spiral arms of our own galaxy has now been obtained. This is only the beginning and the stage is set for a further attack on the detailed structure of the system. At still greater distances, it appears that signals come mostly from abnormal external galaxies emitting vast amounts of energy in the radio spectrum. Some are known to be on the very edge of the observable universe and many are beyond the reach of the most powerful optical telescopes. Researches began with simple but elegant adaptations of existing radio techniques. Then came the invention of various ingenious devices, of which perhaps the Mills Cross is the most famous. This was followed by the crossed-grating interferometer of Christiansen, the so-called "Chris-Cross". Both instruments are now being duplicated in other parts of the world and in the future still larger and more refined versions will be constructed. Among the plans in hand is one to extend solar observations using a rapid-scan radio heliograph on meter wavelengths. This should provide pictures of the movement of disturbances on the sun and will give us further information on the physical processes involved. For galactic and extra-galactic researches, a new development is the construction of a giant radio telescope, shown in Fig. 11, due for completion in 1961. This instrument is of the alt-azimuth type and incorporates a steerable parabolic reflector 210 feet in diameter. An accuracy of $ \pm \frac{1}{3} $ inch will be maintained over the whole of this surface and the telescope should have first-class performance at the 21-centimeter design wavelength. It is to be controlled by a unique master equatorial system and there is every expectation that a tracking accuracy of one minute of arc will be achieved. A contract for construction was placed in July 1959 and fabrication has started. Research observations should commence towards the middle of 1961. The telescope will be erected on a quiet valley site at Parkes, approximately 200 miles west of Sydney. The electrical noise level in this region is exceptionally low and likely to remain so for the next twenty to thirty years. When complete, the instrument will be unique in the Southern Hemisphere and among the most refined of its kind in the world. It will be used principally for researches into the structure of our own galaxy and of the galactic systems in the universe beyond. It will also provide one of the most important links with planetary and space probes others are now planning and, in association with similar instruments in the Northern Hemisphere, a complete cover of the celestial sphere. Vitramon® SOLID STATE PORCELAIN CAPACITORS FINE SILVER ELECTRODES HIGH GRADE PORCELAIN DIELECTRIC MOLECULAR BONDING ASSURES ABSOLUTE IMMUNITY TO HUMIDITY MONOLITHIC CONSTRUCTION ELIMINATES NEED FOR CASE OR HERMETIC SEAL CUTAWAY OF AXIAL CAPACITOR RADIAL SERIES 0.5 mmf to 1200 mmf 50, 100, 300 and 500 vdc SOLID-STATE RELIABILITY MAKES THE DIFFERENCE The exclusive "Vitramon" Process fuses fine silver electrodes within a solid block of dense dielectric porcelain to form an integral monolithic unit with outstanding electrical and physical properties. The molecular bond of dielectric, electrodes and terminals assures absolute immunity to humidity, low loss, low noise and great stability over a temperature range of -55°C to 200°C. True reliability is inherent in the solid-state construction of all "Vitramon" porcelain capacitors. They are used extensively where quality, precision and dependability are imperatives. VITRAMON® INCORPORATED P. O. Box 544 • Bridgeport 1, Connecticut Get much more information in the ... When designs call for products in the field of electronics, the Guide solves problems in advance. There's a 64-page reference section with up-to-the-minute data on markets, materials, components and applications to make product selection easier. You'll find detailed facts about the products of nearly 700 advertisers — that's 42% more than you'll find in any other similar guide. Also local sales offices of manufacturers...the names, addresses, and phone numbers of representatives...complete lists of manufacturers...registered trade names...and, of course, the most complete list of electronic and related products. Tells you what you need to know when you are ready to buy. Greatest versatility! Highest accuracy! Widest range! NARDA transistorized POWER METER MODEL 440 ... $250 What's most important to you in a power meter? Accuracy? Portability? Independence from line voltage deviations? Wide range? Stability? Rapid warm-up? Not that you have to make a choice...or a compromise...any longer. The Narda Model 440 Power Meter gives you all these features! Completely transistorized and powered by a nickel-cadmium battery, rechargeable during operation or overnight, it offers two low-power scales in addition to the five standard scales (see below), a built-in charger with state-of-charge indicator and protection against overcharging, and freedom from internal heating caused by vacuum tubes. Moreover, the 440 provides up to 18 ma bias current, enabling you to use the widest selection of bolometers and thermistors. In short, the 440 is the most versatile unit available to provide accurate direct-reading measurements of cw or pulsed-power automatically, over any frequency range for which there are bolometer or thermistor mounts. For complete data, contact your nearest Narda representative, or write us directly. Address: Dept. E-11. SPECIFICATIONS POWER RANGES: 7 SCALES 0.01 mw full scale ............... -30 to -20dbm 0.03 mw full scale ............... -25 to -15dbm 0.1 mw full scale .................. -20 to -10dbm 0.3 mw full scale .................. -15 to -5dbm 1.0 mw full scale .................. -10 to 0dbm 3.0 mw full scale .................. -5 to +5dbm 10 mw full scale .................... 0 to +10dbm 4.5 ma bolometers give best results on these scales. Range Switch: 0.01 to 10 mw (full scale) Accuracy: 3% of full scale reading Bolometers & Thermistors: All 100 and 200 ohm, requiring up to 18 ma bias. Battery Charger: Built-in; continuous or overnight. (Battery operable 16 hrs. before recharge required.) SYLVANIA—BASIC SUPPLIER TO THE ELECTRONICS INDUSTRY FIRST WORD IN METAL PARTS —when tube quality is your aim The word? Sylvania. First, Sylvania applies unique quality control techniques in the manufacture of metal parts to assure you mirror-image uniformity lot after lot. In addition, Sylvania gives you speedy delivery from facilities that can custom-stamp millions of metal parts each day. Batteries of multi-slide machines, vertical power presses, multiple plunger presses and four-slide tooling machines stand ready to fabricate from strip, coil or wire—of a wide variety of metals—to meet rigid specifications. Sylvania craftsmen and specialized engineers work with one thought: to produce parts that build the quality of your tube. Because of this philosophy—and the Sylvania technical service to go with it—Sylvania has become the byword, and buyword, for basic supplies throughout the electronics industry. Sylvania, Parts Division, Warren, Pa. SYLVANIA Subsidiary of GENERAL TELEPHONE & ELECTRONICS CIRCLE 102 ON READER SERVICE CARD Three-Element Semiconductor Materials List of possibilities is nearly endless. But the search isn’t a blind one. Certain rules make it possible to predict new compounds. By J. H. WERNICK and R. WOLFE, Bell Telephone Laboratories, Inc., Murray Hill, New Jersey Photomicrograph of polished and etched surface of AgSbTe₂ (light areas) containing small amount of silver telluride (dark areas) Exciting new electronic devices, based on the varied properties of semiconductors, are invented each month in research laboratories around the world. Some of these—tunnel diodes, thermoelectric power generators and refrigerators, solar batteries, Hall-effect gyrators and circulators, parametric amplifier diodes, photoelectromagnetic infrared detectors—may soon be as important to the electronics industry as transistors and rectifier diodes are today. All of these devices are the fruits of the ever-increasing research effort on fundamental semiconductor properties (ELECTRONICS, July 19, 1959). One basic attribute of semiconductors makes this great variety of devices possible—the electrical properties of a semiconductor can be controlled over wide limits. This control is achieved by variation of the density of free electrons or holes in the material by adding to the pure material minute quantities of foreign atoms or molecules, either uniformly throughout a specimen, or locally within a single crystal. The behavior of the electrons or holes can then be controlled by electric, magnetic or electromagnetic fields; by heat; or by mechanical force. For example, in the photoelectromagnetic infrared detector, when a uniform bar of semiconductor is illuminated with infrared light in the presence of a magnetic field, an electric field is produced. For each device, a particular material is chosen which has the best combination of properties. For transistors, high electron mobility and large energy Table I—Ternary Compounds \| Compound \| Melting Point °C \| Energy Gap (eV) \| Room Temp Mobilities (cm²/volt-sec) \| \|-------------------\|------------------\|-----------------\|-------------------------------------\| \| CuFeS₂ \| 875 \| 0.53 \| > 30 \| \| CuFeSe₂ \| 574 \| 0.16 \| < 20 \| \| CuFeTe₂ \| 740 \| 0.16 \| < 50 \| \| AgFeS₂ \| 736 \| 0.23 \| > 250 \| \| AgFeSe₂ \| 680 \| \| > 2,000 \| \| CuAlS₂ \| \| \| \| \| CuAlSe₂ \| \| \| \| \| CuAlTe₂ \| \| \| \| \| AgAlS₂ \| \| \| \| \| AgAlSe₂ \| \| \| \| \| AgAlTe₂ \| \| \| \| \| CuGaS₂ \| 1,040 \| 1.63 \| \| \| CuGaSe₂ \| 870 \| 1.0 \| 60 \| \| AgGaS₂ \| 850 \| 1.66 \| \| \| AgGaTe₂ \| 720 \| 1.1 \| \| \| CuInS₂ \| \| 1.2a \| \| \| CuInSe₂ \| 990 \| 0.92a \| 1,150 \| \| CuInTe₂ \| 780 \| 0.95a \| 100 \| \| AgInS₂ \| \| 1.9a \| \| \| AgInSe₂ \| 773 \| 1.18a \| \| \| ZnSiAs₂ \| \| 2.1a \| \| \| ZnGeP₂ \| \| 2.2a \| \| \| CdGeP₂ \| \| 1.8a \| \| \| ZnGeAs₂ \| \| > 0.6, < 1.1a \| \| \| ZnSnAs₂ \| \| \| \| \| CdSnAs₂ \| \| > 5,600 \| \| Optical energy gaps * Lattice thermal conductivity 0.006 watts/cm—°C gap are desirable. For thermoelectric materials, low thermal conductivity is of primary importance, but high mobility, effective mass and energy gap are also useful. However, for each device, a compromise must be reached, because no known semiconductor is superior in all of the required parameters. Better devices of every type could be made if a material could be found which combined the best properties of the various known semiconductors. For example, if a new semiconductor were discovered which had the electron mobility of indium antimonide, the electron effective mass of bismuth telluride, the thermal conductivity of silver antimony telluride, and the energy gap of silicon carbide, thermoelectric generators could be made with an efficiency of 50 percent as compared with conventional power units like diesel engines which have efficiencies up to 40 percent. It is the search for the best combination of properties for the various devices which prompts the continuing research on new compound semiconductors. This research may also lead to the discovery of new phenomena which may suggest still further additions to the list of devices and which will certainly contribute to the basic understanding of the nature of semiconductors. Many physicists are still engaged in fundamental studies of elemental semiconductors, particularly germanium and silicon. It might be expected that nothing remains to be learned about these materials, but deeper understanding and new devices continue to come from this research. The tunnel diode is a notable example of a recent device based on an effect observed in germanium. New semiconductors continue to be discovered among the many possible binary compounds, such as InSb or Bi₂Te₃. The unique properties of some of these compounds have been utilized in such applications as infrared detectors and thermoelectric devices. Recently the search has been extended to include ternary (three different parts) compounds and more complex materials. The list of possibilities is nearly endless. However, the search is not a blind one; there are rules which enable new semiconducting compounds to be predicted. PREDICTING NEW SEMICONDUCTORS—The nature of the bonding in solids, and therefore the properties of the solids, are determined mainly by the interactions of the valence electrons of the constituent atoms. In the simplest semiconductors, Si and Ge, the bonding is purely covalent. Each atom has 4 valence electrons (Group 4b of the periodic table). In the solid, each Si or Ge atom is surrounded by four other atoms in a tetrahedral configuration, each electron being shared by two atoms to form a covalent bond. There are therefore four covalent bonds per atom, and the electron to atom ratio is 4:1. We can make use of these two facts, valence electron to atom ratio of 4:1 and tetrahedral coordination, for the prediction of semiconducting compounds. For example, if instead of 2 Ge atoms (8 valence electrons) we take a gallium atom (Ga—Group 3) and... an arsenic atom (As—Group 5), we form the semiconducting compound GaAs. All of the 3-5 binary compounds can be formed in this manner. Similarly, by combining Group 2 and Group 6 elements, one can form the semiconducting compounds ZnS, CdSe, and CdTe. This procedure can be extended in a manner shown in Fig. 1. All of these compositions have electron to atom ratios of 4:1. The composition GaAs and ZnGeAs$_2$ are known to be single-phase semiconductors with a tetrahedral atomic configuration. By proceeding in this manner, one can fabricate hundreds of compositions which may yield single-phase semiconducting compounds. The crystal structures of known compounds can be examined and those with tetrahedral atomic configurations can be chosen as possible semiconductors. For example, the minerals Enargite (Cu$_4$AsS$_4$), Tetrahedrite (Cu$_4$SbS$_4$), and Tennantite (Cu$_8$AsS$_8$) have tetrahedral coordination and are semiconductors. As discussed above, elements Si and Ge are covalently bonded. Bonding in the Groups 5b and 6b elements is also essentially covalent. This fact, together with the results of the above discussion, suggests another approach in predicting new semiconductors; that is, compounds containing elements from Groups 4b, 5b, and 6b may have a large covalent component to the bonding, an apparent necessary feature for intrinsic semiconductivity, and such compounds should therefore be semiconductors. In addition, by the substitution of chemically similar elements, for example Se and Te for S, one may make new semiconductors. An example of this approach is the substitution of the heavier Se atom for S in the mineral Matildite, AgBiS$_2$. One obtains the new semiconductor AgBiSe$_2$ having the identical crystal structure. The properties of the new semiconductors will be different from those of the compound on which they are based. In general, as the molecular weight of the compound increases, the melting point, thermal conductivity and energy gap decrease while the carrier mobility increases. There are exceptions to this generalization, but it is widely used, particularly in the search for new thermoelectric materials. Solid solutions of two or more semiconductors will exhibit properties which will be different from those of the end members. Regardless of the procedure one follows in predicting that a given combination of elements should yield a single-phase semiconductor, there is no assurance that this combination will lead to a new compound. It is necessary to do much research to establish that this material after preparation is single phase and an intrinsic semiconductor. Because of the volatility of one or more of the constituents, the compounds are usually prepared in sealed systems made of quartz. Occasionally, the partial pressure of the volatile constituent must be kept constant by maintaining one portion of the sealed system at a constant lower temperature. The single-phase nature of the material can be established by metallographic, x-ray and thermal-analysis techniques. A single-phase semiconductor is one in which the chemical composition and crystal structure FIG. 2—Testing semiconductors for thermoelectric power (A), point-contact rectification (B), resistivity and Hall effect (C) and thermal conductivity (D) are the same at every point, unlike a mixture of two different materials like that shown in the photomicrograph. Each crystalline phase has a characteristic x-ray pattern and within the precision of the method, x-ray analysis will allow the establishment of the single-phase nature of the material. Metallographic observations on polished and etched sections will allow for the detection of amounts of second phase which could not be detected by the powder x-ray technique. Single crystal x-ray analysis will also establish the crystal structure of the material. The thermal-analysis technique determines, in addition to the melting point, if the alloy is single phase (absence of eutectic or precipitate) and if it melts congruently (does not dissociate before complete melting) or incongruently. This latter information is of importance for zone refining and crystal growing procedures. Occasionally a knowledge of the complete phase diagram is required. The phase diagram shows the temperature-composition relationship existing in a given alloy system. After the existence of a new compound is established, it must be zone refined and single crystals grown so that the intrinsic properties can be determined. Following this, the materials are doped with known quantities of impurities so that the extrinsic properties can be determined. A number of semiconducting ternary compounds reported in the literature are listed in Table I on page 104. The information is far from complete, and much important research remains to be done. Only when some of the gaps have been filled will it be possible to add to this table a list of potential applications for each material. EVALUATION OF NEW SEMICONDUCTORS—After a new compound has been predicted and prepared, and when good single-phase specimens are available, various physical measurements are made. These measurements determine whether the compound is a semiconductor. Further experiments indicate how the particular semiconductor may be used in devices. For instance, the mobilities of electrons and holes are measured (if possible) because high mobilities are required for many semiconductor devices. THERMOELECTRIC POWER—One of the first parameters which is quickly measured on each ingot is the thermoelectric power (Fig. 2A). Two metal probes are pressed onto the specimen—one warm and the other cool—and the voltage produced between these probes is measured. The sign of this voltage tells immediately whether the material is n-type or p-type. (If the warm probe is positive, the specimen is n-type; if the warm probe is negative, the specimen is p-type.) If this procedure is refined so that the two temperatures and the thermoelectric voltage can be determined accurately, then the thermoelectric power (Seebeck coefficient) can be measured. Values from a hundred microvolts per degree to several millivolts per degree are common for semiconductors. In metals, the thermoelectric power is usually small (less than ±40 μV/degree C). Insulators cannot be measured with simple equipment because of the high impedance between the probes. RECTIFICATION—Point contact rectification is another characteristic of semiconductors (Fig. 2B). If a potential is applied to a specimen between a large-area metal contact and a sharp metal point, rectification may be observed. If forward current flows when the point contact is negative, the material is p-type and if it flows when the point contact is positive, the material is n-type. Ohmic behavior is typical of metals and impure semiconductors. CONDUCTIVITY AND HALL EFFECT—Electrical conductivity and its temperature variation are basic to the definition of a semiconductor. The measurement of conductivity is therefore essential in any semiconductor investigation. Conductivity depends on the density of free electrons or holes and on the velocity with which they move in an applied field (mobility). This dependence is expressed in the equation $ \sigma = Ne\mu $ where $ \sigma $ is the conductivity, $ N $ the density of electrons or holes, $ e $ the electronic charge and $ \mu $ the mobility. To determine $ N $ and $ \mu $ separately, it is necessary to make further measurements. A most useful property to measure is the Hall coefficient (Fig. 2C). When current flows in a semiconductor and a magnetic field is applied perpendicular to this current, the moving electrons or holes tend to be deflected and a potential difference is established perpendicular to both the current and the field. This is the Hall effect. The Hall voltage is proportional to the current and to the field and the factor of proportionality is the Hall coefficient. The sign of the Hall coefficient is, by convention, positive for $p$-type material, negative for $n$-type material. Its magnitude depends on the density of carriers but not on their mobility. It can be shown that $R = \pm r/NE$ where $R$ is the Hall coefficient, and $r$ is some constant which depends on the subtler properties of the carriers but which is usually close to unity. The measurement of $R$ therefore gives the density of carriers (and their sign). The product $R\sigma$ is the Hall mobility. There are many different methods for measuring the conductivity and Hall coefficient in semiconductors and certain precautions must always be observed. It is important to use separate contacts for the current leads and the voltage probes and to avoid drawing current through the voltage probes. This eliminates the errors which may arise from contact resistance and rectification. In Hall measurements, a poor choice of specimen shape may cause current distortion and may lead to serious errors. Many other pitfalls, such as inhomogeneity, anistropy, surface contamination and nonlinearity may be encountered in these measurements. These are the subjects of continuing research. A wealth of information is contained in the variation of conductivity and Hall coefficient with temperature (Fig. 3). The variation of the density of carriers with temperature depends on the energy gap of the material (the energy required to produce a hole-electron pair) and the activation energy of the impurities. Changes of sign in the Hall coefficient often occur as the temperature is changed. These give information about the ratio of the mobilities of electrons and holes. The variation of mobility with temperature depends on the mechanism by which the carriers are scattered as they move through the material. OPTICAL MEASUREMENTS—Many applications of semiconductors depend on their optical properties—their transparency in the infrared region of the spectrum, the photovoltaic and photoelectromagnetic effects, electroluminescence and photoconductivity. Measurements of these properties on new semiconductors are important not only as indications of their potential usefulness in optical devices but also as means of determining the fundamental properties of the materials (Fig. 4). The absorption coefficient of a thin semiconductor specimen as a function of wavelength gives information about the energy gap, impurity ionization energy, effective mass of the free carriers and the dielectric constant. Photoconductivity measurements yield information on lifetime and diffusion of the electrons and holes. The photoelectromagnetic effect depends on the mobilities of the carriers and on the surface recombination velocity. Solid-state spectroscopy is an important field of research which is indispensable to the study of luminescent materials (which overlap the fields of semiconductors and insulators). Many of the classical experiments of atomic spectroscopy are being adapted to solid-state systems, where they are helping to shed light on many of the properties of electrons bound to impurities. THERMAL CONDUCTIVITY—Thermal conductivity ($K$) of semiconductors is of obvious device importance. For transistors and similar devices, it is desirable to use a material with a high thermal conductivity to minimize local heating. On the other hand, low thermal conductivity is one of the most important requirements for any thermoelectric material. It enables large temperature differences to be maintained in thermoelectric refrigerators and minimizes the conducted heat loss in generators. The thermal conductivity of semiconductors is also of theoretical interest. Heat is carried by the free electrons (as in metals), by the vibrating atoms (as in insulators) and by other mechanisms such as the diffusion of electron-hole pairs down a temperature gradient. It is important to separate the contributions from the various mechanisms because some are intrinsic properties of the material and others depend on the impurity concentrations. The measurement of thermal conductivity is very simple in principle (Fig. 2D). It is merely necessary to pass a measured amount of heat through a specimen of known dimensions and to measure the temperature difference across the specimen. The heat ![Graph showing variation of infrared absorption with wavelength](image-url) FIG. 4—Variation of infrared absorption with wavelength 1. Energy gap (from absorption edge) 2. Electron or hole density and effective mass (from free carrier absorption) 3. Dielectric constant (from thin film fringes) 4. Activation energy of impurities (from peaks which appear at low temperatures) flow per unit area per unit temperature gradient can then be calculated. However, in practice it is difficult to make good thermal contacts between the specimen and the heat source and sink and to assure that all of the heat flows through the specimen. Careful experimental procedure is required, including thermal shielding and operating in an evacuated chamber to eliminate convection and conduction in the air between the source and sink. THE CASE OF AgSbTe₃—By way of summary and illustration, the prediction and investigation of the new ternary compound silver antimony telluride will be described. The starting point for the prediction of this compound was the mineral Miargyrite (AgSbS₂). Nature has supplied other substitute compounds—the minerals Matildite (AgBiS₂) in which bismuth is substituted for the antimony, and Wolfsbergite (CuSbS₂) in which copper is substituted for the silver. It was predicted that if the substitution of tellurium for the chemically similar sulfur would produce a single-phase material, a new semiconducting compound would be found. When pure silver, antimony and tellurium were melted together in the exact proportions and then cooled, the result was a clean solid ingot composed of large crystals—the first promising indication that a new compound had been made. The ingot was zone refined, and again the outward appearance was encouraging. X-ray analysis showed that the crystal structure was the same as that of Miargyrite with one important difference. Both have the structure of rock salt which is a simple cubic lattice with sodium and chlorine atoms alternating in a regular arrangement. In both, the sulfur or tellurium atoms fill the chlorine sites. In AgSbS₂ at room temperature, the silver and antimony atoms alternate regularly in the sodium positions, whereas in AgSbTe₃ they are arranged at random on these sites. The thermal analysis again pointed up this difference. When AgSbS₂ is cooled from a high temperature, there is a heating effect (a bump in the cooling curve) when the compound solidifies from the melt, and another effect as the structure of the crystal changes from disordered to ordered. The second effect is absent in the AgSbTe₃ cooling curve. This compound remains disordered down to room temperature. Metallographic examination revealed that the material was often single phase, but under certain conditions another phase, which was darkly stained by the acid etch, appeared. A specimen which was rich in this second phase was examined by x-ray crystallographic methods and an extra pattern of lines was found besides those of AgSbTe₃. These lines were characteristic of another semiconductor, silver telluride. Measurements of thermoelectric power were the first indication that AgSbTe₃ is a semiconductor. On each ingot the thermoelectric power was positive (p-type material) and its magnitude was close to 200 μV/degree C. No point contact rectification was observed on any of these ingots. The resistivity was fairly low, ranging from 0.004 to 0.01 ohm-centimeters, and it varied only slightly in the temperature range −200 C to +200 C. The Hall effect showed an interesting and, at first, bewildering anomaly. It was sometimes positive, as expected in a specimen with p-type thermoelectric power; but in some specimens, with similar conductivity behavior and still with p-type thermoelectric power, the Hall effect was negative. The variation of this n-type Hall effect as the temperature was lowered added to the mystery: the Hall coefficient decreased to zero at about −200 C and then changed sign to become positive. No simple explanation for this behavior could be found within the usual theory of semiconductors, until it was observed that the n-type Hall coefficient and the dark stained regions on metallographically polished surfaces were found in the same specimens. It then became obvious that the n-type Hall coefficient was contributed by the small fraction of silver telluride in these specimens, even though the other properties were still dominated by the AgSbTe₃. This unusual phenomenon is still being investigated. The most interesting property of AgSbTe₃ is its thermal conductivity. It was suspected that the heavy atoms and the disordered structure would result in low thermal conductivity in this material. Experiments have shown that this is indeed the case. The measured lattice component of the thermal conductivity (the part associated with the atomic vibrations and not the electrons, $ K_L = 0.006 \text{ watts/cm degree C} $) is about one hundred times smaller than that of germanium, and three times lower than the best known thermoelectric compounds such as bismuth telluride. This material is therefore potentially useful for thermoelectric refrigeration and for power generation at moderate temperatures (it melts at 575 C). However, the electrical properties of the compound as it is now made are not good enough to make AgSbTe₃ superior to the best thermoelectric alloys. Research is continuing, and there is reason to believe that these properties can be improved. The compound AgSbTe₃ has been added to the growing list of new semiconductors, but much remains to be learned before it can win a place on the short list of most useful semiconductors. This same statement applies to all of the other ternary compounds in the table. As research on these and other new semiconductors continues, our fundamental knowledge of solid-state science will be enhanced. Novel and improved devices for electronics may be the result. BIBLIOGRAPHY O. G. Folberth, Zeit. Für Naturforschung, 14, p 94, 1959. S. Geller and J. H. Wernick, Acta Cryst., 12, p 48, 1959. C. H. L. Goodman, Journ Phys Chem Solids, 6, p 305, 1958. E. T. Schomleits, H. A. Gorlunova, Journ Tech Phys, 25, p 884, 1955. E. Mooser and W. B. Pearson, Journ Chem Phys, 26, p 893, 1957. J. H. Wernick and K. E. Benson, Journ Phys Chem Solids, 3, p 157, 1957. J. H. Wernick, S. Geller and K. E. Benson, Journ Phys Chem Solids, 1, p 154, 1958. V. P. Zhule, V. M. Sergeeva, and E. L. Shtrum, Soviet Physics, Journ Tech Physics AIP, 3, p 208, 1958. Wire Gage Provides Continuous Measurement Simple circuit for measuring wire thickness uses nondestructive method to provide continuous indication. Instrument can measure copper wire as small as AWG size 46. By KLAUS H. JAENSCH, Senior Electronic Engineer, Stromberg-Carlson Co., Rochester, New York AN ELECTRONIC wire gage using a principle normally employed in proximity detectors offers two essential advantages over conventional methods for measuring tolerance of diameter (or resistance per length) of electrical wire. First, the test is non-destructive in that insulation of the wire need not be removed as for making contact in measuring resistance per length. Second, any length of more than an inch can be measured. This allows continuous measurement of wire as it is discharged from a spool. An instrument with these properties enables manufacturers of precision wire to check their product continuously during the drawing operation. On coil winding machines, the test instrument can be inserted between spool and winding spindle. Meter indication shows the thickness of the wire which is actually used, not only of a sample. Moreover, excessive stress during the winding operation can be detected and output of test instrument may be used to adjust the tension automatically. Principle of Operation A piece of conductive material within the magnetic field of a coil acts like a shorted turn on a transformer, lowering the coil's Q. If the coil is employed as part of the resonant circuit of an oscillator, amplitude of oscillation is decreased by inserting a conductive probe into the coil. This principle has been in use for several years for detecting the presence of conductive parts. In proximity detectors, oscillation ceases if a metal piece comes in the vicinity of the detecting coil. The investigation reported here proved the method to be apt for accurate quantitative measurements as well. Amplitude of oscillation decreases with diameter and conductivity of a wire fed through the coil in an axial direction. With the prototype instrument, copper wire as small as 1.5 thousandth of an inch in diameter can be measured. Oscillator The equipment (Fig. 1) consists of an oscillator including the detecting coil, and a vtvm for measuring the oscillator's amplitude. A tube circuit originally described as a Q multiplier, is used for the oscillator. Employing one pentode only, this cathode follower type oscillator has excellent stability. For measuring smallest wire sizes, a frequency around 1 mc proved best suited. The test coil acts as the inductance of the resonant circuit. Its specifications are: length $ \frac{3}{8} $ in., inside diameter, $ \frac{3}{16} $ in., 110 turns of magnet wire, AWG (American Wire Gage) No. 39. Shielding of the coil requires spe- Table I—Ratio of AWG Numbers \| AWG Number \| D \| D² \| D⁴ \| \|------------\|-----\|-----\|-----\| \| N + 6 \| 1 \| 1 \| 1 \| \| N + 5 \| 1.12\| 1.26\| 1.59\| \| N + 4 \| 1.26\| 1.59\| 2.52\| \| N + 3 \| 1.41\| 2 \| 4 \| \| N + 2 \| 1.59\| 2.52\| 6.35\| \| N + 1 \| 1.78\| 3.16\| 10.01\| \| N \| 2 \| 4 \| 16 \| cial attention. Distance between outer shield and coil should be not less than one inch (Fig. 2). The shield is slit to avoid its acting as a shortened turn. Of the two coil terminations, the one from the inner layer is used as ground. In spite of this precaution, the capacitive effect of the wire probe may be too great, especially when measuring sizes thinner than AWG 40. This is indicated by a variation between readings with the probe grounded and not grounded. As a remedy, a capacitive shield may be installed inside the coil when the instrument is intended to measure extremely thin wire. This inner shield must also be slit. To keep damping action small, this inside shield should be as thin as practicable, not more than 0.0001-in. copper, 0.00015-in. brass, or 0.0002-in. Constantan or similar resistive material. Start of the winding and connection lead of about the same size wire are soldered to the shield. Care should be exercised to produce a small joint with little solder, since any coherent mass of conductive material in this sensitive area reduces the sensitivity by biasing the coil's Q. For the same reason, condensing moisture must be kept from forming a continuous film on the coil's surface. Condensation is prevented by impregnating the coil in wax. Finally, the coil is fixed in place by filling the space between coil and outer shield with wax. Relation of Wire Sizes Table I shows the numerical relation of AWG numbers to wire diameter, D. The ratios shown apply to any sequence of numbers of the AWG system. For example, taking AWG 34 as N, $(N + 6)$ means AWG 40, which has half the diameter, and a quarter of the cross section of AWG 34. The fourth power of diameter, displayed in column $D^4$ of the table, characterizes the attenuating action of a wire probe in the test method described. This fact is considered in measuring stranded wire. Sixteen parallel, insulated strands cause the same attenuation of the oscillator amplitude as a single wire of twice the bare diameter. This is in contrast to usual resistance-per-length measurements, in which four strands are equivalent to a single wire of twice the diameter. Due to this peculiarity, stranded wire with strands not insulated from each other can not be measured properly with this instrument. Readings would vary within the limits of second power and fourth power of the equivalent diameter, depending on incalculable contact between strands. Scale Figure 3 shows the typical attenuation of oscillator amplitude caused by wire probes of different size. Attenuated amplitude, $E_A$, is displayed in percent of amplitude without probe, $E_1$. With different wire sizes and certain corresponding values of $E_1$, proportions are the same over the entire usable region of oscillator amplitude. Therefore, symbol $N$ is used for wire sizes, representing any AWG number. Assuming a linear indication of voltage, the meter scale can be directly derived from this characteristic, as presented on the right-hand side of the diagram. Without probe, meter reading is full scale $(E_A = E_1 = 100\%)$. A probe of AWG number $N$, for example, AWG 40, brings the reading down to point $N$ of the scale. In this case, a probe of AWG 41 will read $N + 1$; AWG 42, $N + 2$; and so on. For size $N$, the scale is subdivided in percent deviation from nominal diameter. With this information it is possible to build a multirange instrument for measuring several AWG numbers, using a common meter scale for all ranges. Individual values of oscillator amplitude without probe, $E_1$, required to produce the scale proportion of Fig. 3 for different wire sizes, are plotted in Fig. 4. Meter sensitivity has to be adjusted in each case to produce full scale reading without probe. Output amplitude of the oscillator, $E$, is adjusted by varying the feedback resistor $R_F$ (Fig. 1). The highest value of feedback resistance with which oscillation occurs, $R_o$, is in the order of 4,000 ohms. By decreasing $R_F$ below this value, amplitude of oscillation increases. The solid line of Fig. 5 shows the empirical relationship of amplitude, $E$, versus ratio $(R_o - R_F)/R_F$. Knowing from Fig. 4 the values of oscillator amplitude required for the different ranges, the corresponding feedback resistor for producing these amplitudes can be found from Fig. 5. Related by amplitude $E = E_1$ in both diagrams, the scale of AWG numbers is entered in Fig. 5. Stability and Sensitivity Dashed lines of Fig. 5 referring to the right hand scale, $\Delta E$ percent, allow calculating the stability of the instrument in different ranges. Considering, for example, AWG 46, a variation of feedback resistance $\Delta R_F/R_F = 0.01$ percent would cause a deviation in amplitude, $\Delta E = 7$ percent. This is equivalent to an error of approximately 4 percent in reading wire diameter. The same line, $\Delta R_F/R_F = 0.01$ percent, indicates an error $\Delta E = 1$ percent for range AWG 43, equivalent to 0.5 percent in wire diameter. Both these calculations assume a resistance variation of 0.01 percent. With the temperature coefficient of 20 ppm per degree centigrade of good quality precision wire resistors, this value would be reached by a temperature change of 5 C. Close tolerances like this are achieved by housing the fixed portion of $R_F$ in a temperature-controlled crystal oven. Oscillator output amplitude may be measured by means of any conventional vtvm circuit. Interaction between the indicating circuit and the oscillator is prevented by attaching a buffer amplifier to the oscillator. The range selector switch is conveniently incorporated in this buffer stage, as shown in Fig. 1. Taps of the cathode resistor are calibrated by using the information shown in Fig. 4. Oscillator amplitude is adjusted by using the continuously variable part of the feedback resistance, $R_F$, to produce full scale reading with no probe. The potentiometer should be of low temperature coefficient wire, and preferably a multiturn type for sufficient resolution in a multirange instrument. Skin Effect For measuring copper wire heavier than AWG 38, influence of skin effect must be considered. Figure 6 shows the behavior of copper wire at three different frequencies. Dashed lines present apparent resistance per length. Full lines show apparent results with the electronic wire gage. In designing a common-scale, multirange instrument, skin effect sets the useful limit toward heavier sizes. Renouncing the possibility of measuring extremely thin wire, heavier wires can be tested by employing a lower frequency. Lower frequency designs of the instrument are less critical as to environmental capacitive influence on the test coil. REFERENCE (1) H. E. Harris, Simplified Q Multiplier, Electronics, p 130, May 1951. Circuit Modifications For Electronic methods are used to accelerate conventional electromechanical counters by a factor of 3 or more without attendant contact and flyback problems. Counter life is not shortened. By RONALD L. IVES, Palo Alto, California The average operating rate of simple electromechanical counters has been increased from about 10 cps in 1940 to about 25 cps at this date. Current instrumentation often requires faster operation, preferably at low cost and at no impairment of operating life. Electromechanical counters can be improved by modifying either the electrical or the mechanical design and construction, both of which set limits on the amount of acceleration which is feasible. Mechanical modifications to increase operating rate are few and relatively simple, but sometimes quite effective. Most common speed-increasing adjustment is to minimize the armature-core spacing, tighten the return spring and oil the pivots. With poorly designed counters, with which the improvement labor cost may exceed the purchase price, such adjustments are quite effective, producing an increase in the operating rate of up to 50 percent. With better grade counters, where attention has already been paid to mechanical design, little improvement in this manner is possible. Counters having flap armatures can be accelerated in many instances by reshaping the armature so that while the magnetic circuit is substantially unchanged, the mass of the armature is reduced. Electrical Modifications Electrical circuit modifications to increase operating speed are numerous, fairly simple, very productive and lead to rate increases of as much as 300 percent. Most effective method—vacuum tube drive—not only increases speed by a factor of up to three, but also virtually eliminates contact troubles and flyback problems. Properly applied, these electrical acceleration means do not shorten the life of the counter and make no inordinate demands on the power supply. All electrical methods of accelerating electromechanical counters depend for their operation on a shortening of the time constant of the electrical circuit of the counter. This is substantially the drive magnet coil, and its time constant can be stated as $L/R$, in which $L$ is the inductance in henries, and $R$ the resistance in ohms. Inductance $L$ here is not a constant as the properties of the magnetic circuit change during the operating cycle. The simplified formula for the buildup of current in an inductance is $T = (L/R) \ln (E/E - Ri)$ where $T$ is the pull-in time of the counter armature, $L$ is the inductance of the coil in henries, $R$ is the resistance of the coil in ohms, $E$ is the voltage across the coil and $i$ is the instantaneous coil currents in amperes. As $L$ in this formula is built into the counter and is not subject to easy alteration, $T$ can be reduced to speed up counter operation only by increasing $E$ or $R$. Both methods are useful and a combination of them using vacuum-tube drive seems most effective. The simplest method of increasing $E$ is by direct overvolting the counter. This speeds up armature pull-in time but tends to slow up armature release due to overmagnetization of the core. Serious flyback troubles are also introduced with attendant shortening of contact life. When adequate spark absorption means are added to the circuit, damping retards armature release so that little net gain results from overvolting and the equilibrium temperature of the counter coil is likely to be abnormally high. To offset the disadvantages of direct overvolting while attempting to retain its advantages, two modifications—pulsed overvolting and sliding overvolting have been developed. Pulsed Overvolting Pulsed overvolting is the application of an excess voltage to a Boosting Counter Speed counter coil for a limited time. The reasoning involved here is that if a counter operates satisfactorily with 100 v applied for 0.1 seconds, it should also operate satisfactorily, and much faster, if 500 v are applied for 0.02 seconds. Experiments show that change of armature pull-in time with change of voltage is approximated by $P_2/P_1 = (E_1/E_2)^{-3}$ where $P_1$ and $P_2$ are the armature pull-in times and $E_1$ and $E_2$ are the corresponding applied voltages. Voltage is applied in each case for the full armature pull-in time. This method of counter acceleration is somewhat superior to direct overvolting, as the core is not overmagnetized. This leads to a reduction in flyback problems, contact burning and coil heating. Insulation and mechanical factors limit the useful acceleration attainable by this method to approximately a factor of three. Sliding overvolting is the application of an excess voltage to the counter coil when the circuit is first completed. This voltage is then reduced, so that it falls to or below the normal operating voltage of the counter during the pull-in time. Methods of applying sliding overvolting to a counter coil are outlined in Fig. 1. It may be noted that these are the familiar relay slammer circuits used during World War II. In the first two circuits (A and B), which require a double-throw contactor, the capacitor value is about 1.2 times the minimum for dependable operation of the counter at the voltage applied. In both instances, the applied voltage is considerably higher than the rated operating voltage of the counter. In Fig. 1C, constants for dependable operation of a counter at about twice the rated speed are shown. Capacitor value here is about 0.6 of that needed to operate the counter if charged at the supply voltage, and then discharged through the counter coil. This circuit operates with no serious trouble from contact failure, flybacks or coil heating. With this circuit, counter life at double rated speed is substantially rated life and this appears to be the optimum non-thermionic counter acceleration circuit using only elevated voltage. Another means of accelerating electromechanical counters to increase the circuit resistance. This also calls for an increase in voltage, so that the equilibrium voltage across the coil is substantially the rated operating voltage. The operating circuit, and relation of coil current, supply voltage and voltage applied across coil, are shown in Fig. 2. Here, curve A shows the current buildup in the counter coil when rated voltage is applied. Counter armature pulls in when the current reaches the pull-in point. Curve B shows current buildup at a greatly elevated supply voltage. Here, the current reaches the operating point in a relatively short time. Curve C shows the current buildup in the counter coil when the switch in the circuit insert is closed. Current here reaches the operating value much faster than when only the rated voltage is applied to the coil, but somewhat slower than when an excessive voltage is applied directly. Change of voltage across the counter coil, during the current buildup of curve C, is shown in curve D. This is a very simple application of the sliding overvolting previously discussed. It is widely used in industrial counting, and gives a counter acceleration by a factor of more than... two without apparent shortening of counter life. Thermionic Counters When use of contacts is undesirable, as in photoelectric drive circuits, thermionic counter drives are commonly employed. These permit use of both pulsing and sliding overvolting and give a maximum of counter acceleration with a minimum of overheating and other damage to the counter coil and mechanism. Circuit of a thermionic counter accelerator, having moderate sensitivity, is shown in Fig. 3A. This accelerator is designed for industrial use, employs off-the-shelf components, and does not need readjustment when the tube is changed. By its use, operating speed of most electromechanical counters can be doubled. Counters used in this circuit commonly last somewhat longer than the rated service life. The power supply can be simplified, and the installation made more compact, by securing the hold-off bias from a Zener diode, as in Fig. 3B. As the sustaining current for the Zener voltage drop comes largely from the screen circuit of the pentode, this circuit simplification does not impair counter operation or life. Sensitivity of this circuit can be increased, by a factor of approximately two, by hand picking the tube used, and then adjusting the plate and screen voltages for optimum operation. This expedient is undesirable in most industrial installations because a counter circuit so adjusted will not necessarily work well with any good tube in the socket. A more sensitive circuit, designed for operation from photocell anemometers at I. G. Y. installations in Antarctica and at other difficult locations, is shown in Fig. 4. This consists essentially of a starved 6AK5 amplifier neon-coupled to a 6AQ5 counter driver. As commercially constructed, these counters were arranged in batteries of six, with a single power supply, energizing not only the counters, but also the anemometer lamps and photocells. Consistently attainable operating speeds of slightly more than 60 cps. were noted with counters rated at a maximum operating speed of 25 cps. Ultimate life of the counters in this service is not known, those used in Antarctica having passed 50 million counts without maintenance. Some power supply difficulty was experienced due to condensation in the selenium rectifiers. This is reducible by use of hermetically sealed selenium rectifiers and seems to be no problem when silicon rectifiers are used. Because accelerated counters respond to 60 cycle signals, a number of precautions were necessary to prevent spurious counts from line pickup; and additional precautions were found desirable to prevent the leakage fields of the counters, particularly on release, from triggering other circuits. Tests with a variety of counters of European and American manufacture show that operation can be accelerated by a factor of 2 to 2.5 by use of the circuit of Fig. 4, or rather obvious modifications thereof, provided the counter is of good mechanical design without shortening its rated service life. Counters of this type are built like a good-grade clock, using steel, brass, and plastic. Poorly-made counters can also be accelerated but their service life at high counting rates is very short as they start ejecting gear-teeth, ratchet pawls and small springs after only a few thousand operations. These counters are constructed of pot metal, soft aluminum and inferior plastics. Improved Circuit Satisfactory though this counter-accelerating circuit has proven in use, advances of the art have made possible some improvements by which a counter can be accelerated safely by a factor of more than three, and the power supply can be simplified at the same time. This improved circuit is shown in Fig. 5. In this circuit, higher voltage operation permitting higher operating speeds, is made possible by use of a 6AU6 in place of the 6AK5 formerly used. The coupling between the 6AU6 and the 6AQ5 is changed to capacitative plus neon so that failure of the 6AU6 will not put the 6AQ5 into continuous heavy conduction leading to tube and counter burnout, with possible damage to the power supply. Replacement of the 6AQ5 grid resistor by a silicon diode corrects the tendency of the tube to bias itself off at high count rates. Lastly, the bias supply is eliminated by use of a pair of Zener diodes in the cathode circuit of the 6AQ5. Dependable count rates with this circuit using a counter rated at 25 cps exceed 75 cps. A small additional rate increase can be brought about by lightening the counter armature, as previously outlined, or by reducing the coupling capacitors until the pull-in pulse is about 0.9 of the pull-in time. Under these conditions, the armature coasts for the last 10 percent of its travel, and the armature bounce (after contact with the magnet core) accelerates the return stroke. This means possible a dependable count rate of about 85 cps. Further armature acceleration is possible by this method, but dependable counts at rates exceeding 85 cps are not now attainable because of coasting of the first number wheel, which produces overcounts of up to four in any operation. A slight redesign of the counter mechanism would probably make sustained counts more dependable at rates slightly exceeding 100 cps. Electronic Tonometer For Glaucoma Diagnosis Glaucoma can be detected by a probe that measures pressure within the eyeball. Early diagnosis makes ultimate cure possible. By R. STUART MACKAY and ELWIN MARG, U. of California Medical Center, San Francisco, Cal. Of all people over forty, it is estimated that two percent are going blind from what is known as simple glaucoma. In this malady an increase in pressure, if not detected and corrected in time, causes irreparable damage to the optic nerve. Its onset is diagnosed earliest by measuring the pressure within the eye using devices called tonometers, after which the pressure can be relieved by drugs or surgery. A new electronic tonometer has been conceived that is so fast and gentle that it does not require anesthetics and yet it is more accurate because it does not respond to extraneous factors that lead to uncertain readings in the classic devices. Conventional tonometers measure pressure in one of two ways. Either a plunger is placed upon the front surface of the eye in a vertical position and the indentation due to loading with a known weight measured, or else the area flattened by pressing a transparent plate against the eye with a given force is measured optically. Bending of the cornea introduces an uncertainty into the reading because of its stiffness or rigidity. The difficulty is increased in an astigmatic eye where the curvature may be different in different directions. Even the surface tension of tears introduces an uncertain component of force. Operational Principles Figure 1 shows the arrangement of the new tonometer. The eye is momentarily flattened beyond the pressure sensitive region. Since the bending takes place at the periphery of the probe, the central plunger is not acted upon by bending forces. Any tension in the tissues is a centrifugal force that does not act on the pressure sensitive area. The probe is a small handheld device that is momentarily touched to the eye. As long as the front surface of the probe remains approximately flat, essentially the only variable that will be recorded is the intraocular pressure. In the present device flatness, as well as freedom from drift caused by changes in amplifier gain or changes in resiliency of the mechanical components, is assured by a feedback mechanism. Any tendency for the plunger to be deflected inwards is sensed by a sensitive motion transducer and is counteracted by the resulting change in current in a direct coupled magnetic actuator. A measure of the pressure is then obtained by recording the current to the feedback magnet. The scale is linear and free from involved calibration because the plunger never moves appreciably and thus changes in transducer sensitivity with deflection, or variations in restoring force field, are not introduced. In using the device the probe is momentarily pressed against the eye. As the force of contact increases, the recorded reading will increase until the plunger, which is about two millimeters in diameter, is covered. The further increase in force will not change the reading until the pressure within the eye begins to be raised by the external force. Thus the reading, which can be recorded on a pen-writer, displays a plateau whose elevation is a measure of intraocular pressure (Fig. 2). Circuit Operation The requirements of the motion-transducer are twofold. It must (1) be extremely sensitive to small displacements and (2) very little force should either be required by the transducer, or reflected back into the system by it. The transducer employed made use of the motion of a ferrite core which altered the inductance of an adjacent coil. The change in inductance is measured by a circuit resembling some types of frequency-modulation detectors. The circuit is shown in Fig. 3. Changes in position of the moving ferrite core cause a signal to be developed across capacitor $C_1$ with signal polarity dependent upon the displacement direction. The signal (voltage) is fed to the d-c amplifier via the shielded probe-cable and the output from the d-c amplifier feeds the transistor-pair, $Q_1$ and $Q_2$. The amplified signal drives the restoring coil which forces the probe into the plane of its surrounding annular plate. The actuator consists of the voice coil and magnet from a small loudspeaker. The signal from the displacement sensor is about 0.1 volt per micron of movement of the plunger. The feedback system applies a force of about 0.8 gram to the eye for a normal intraocular pressure of 16 millimeters of mercury. The deflection of the system is 0.6 microns for a pressure of 40 millimeters of mercury. A change in oscillator frequency from 5 megacycles down to 100 kilocycles decreases the sensitivity to uselessness. In Fig. 3, $T_1$ and $T_2$ are Miller transformers, type 1467, the primary of $T_1$ being $L_n$. Coupled to $T_1$ and $T_2$ are $L_z$ and $L_m$ which each have 16 turns in two layers of No. 22 wire. The primary and secondary of $T_2$ are labeled $L_1$ and $L_2$ respectively. The probe is shielded and the shield has an axial slit so that it does not act as a shorted turn. The probe shield is grounded to the shield of the connecting cable. This connecting cable carries both radio frequency to the probe and the detected signal from it. The plunger diameter is approximately 2 millimeters and the diameter of that part of the eye flattened by the surrounding plate is approximately 3 millimeters. The mass of the moving parts in the device is kept to a minimum so that accelerometer or seismograph effects are minimized in the recording as the probe is moved. Problems of friction were minimized in the design of the probe but their remaining interference was removed by including in the feedback loop a small 200 cycle voltage signal (fed in by $T_2$). The resulting small amplitude motion introduces no noticeable signal but eliminates the effects of static friction. The help of Mr. Raymond Oechsli in perfecting the circuits and the probe is gratefully acknowledged. A compact and economical monoscope tube apparatus which generates well-formed alphanumeric characters for display at high speeds is described in this article. In this apparatus, the character generation function is completely separated from the character display, thus providing a degree of flexibility not available in some other methods. With repetitive digital information, for instance, an ordinary laboratory oscilloscope may be used to display the characters. This monoscope character generator may also be used to drive certain types of hard copy printers, either by optical transfer of visually presented information or directly for special types of electrostatic printing tubes. Input to this monoscope is derived from conventional six-wire parallel binary information. A standardizer at the input supplies uniform and consistent electrical pulses to the generator, permitting the device to be driven by a wide variety of sources. Sources of Information Excellent results have been obtained from two sources of input information. First, a 36-position ring counter employing beam switching tubes has synthesized the necessary digital information on six wires. The beam switching tube outputs were sampled off through diodes to six wires. The device has also worked from information stored in digital form on a six-track magnetic tape. A 360-character message was encoded on the tape and read off at a tape speed of 60 inches per second to give a character rate of approximately 10,000 per second, which is in the speed range of computer and memory devices. (With minimal changes, this rate could be increased to 20,000.) Equally satisfactory performance can be expected from an information source such as a magnetic drum, a paper tape or an electric typewriter. Information can be displayed on commercially available equipment. If the input information is repeated faster than 20 times per second, the characters may be displayed, without flickering, on any conventional cathode ray tube device. For non-repeating or low-repetition rate information, a direct view storage tube is suitable. Portions of a message can be selected and displayed as long as desired on a tube face. Otherwise, the message can run in its entirety for the observer and can be stored if desired. Character quality is limited only by the raster frequencies (and therefore, video bandwidth) and the finite spot size of the monoscope tube. Present techniques utilize about 400 picture elements in the small raster to cover a character and any guard space around it. For a 10 kc character rate (one character each 100 microseconds), about four megacycles of video bandwidth is required. To operate at faster character rates, the switching transistors should be faster and a higher frequency raster should be employed, resulting in wider video bandwidth requirements. Character quality is superior to character display systems utilizing a simple dot matrix for character writing. Monoscope Construction Video waveforms necessary to write characters are generated in a monoscope tube—a fundamental cathode ray device. The tube comprises a conventional electron gun and an aluminum target enclosed in a vacuum envelope. All of the desired characters and symbols are on the metal target in printer's ink. A six-wire system makes up to 64 different characters and symbols available, including one for spaces. These are arranged in eight rows of eight characters each on the target. Preparation of the monoscope target is by common photoengraving and printing techniques. This enables complete freedom in the selection of the style of type, symbols or simple pictures. In the preparation of the target the characters can be located in the matrix in order to conform to six-unit binary codes which vary with machines. Type font or character-selection code can be changed by replacing the monoscope tube (a $100 to $200 item). Monoscope Operation As a constant-intensity electron beam is scanned across a character, the secondary emission current from the target is modulated; aluminum and carbon, the chief ingredient of the ink, exhibit different secondary emission coefficients. The resulting video signal is amplified and used to intensity-modulate the display tube. The monoscope beam is scanned in a television-like raster which covers only one character on the matrix. The display device is swept in synchronism with the monoscope and modulated with the video signal; thus, the character appears on the face of the tube. The character is positioned to its proper sequence in a word by deflecting the display device. A character is selected by positioning the small raster to a specified location on the 8 by 8 character matrix. This requires an accurate horizontal and vertical deflection of the beam. This is accomplished in the course of deriving horizontal and vertical deflection voltages from the digital input information. The vertical component of the raster is a 200 kc sinusoidal or sawtooth voltage while the horizontal component is a once-per-character (10-kc) sawtooth. To select a character from an 8 by 8 matrix, eight distinct voltage levels are needed, both in the horizontal and vertical deflection circuitry. Input information in digital form must be decoded in order to be read as alphanumeric characters. The six-track digital input can be thought of as two three-track sources. From one three-track binary arrangement, eight distinct codes can be derived. Input information channels 1, 2, and 3 therefore may be decoded to determine one of eight horizontal positions merely by adding binary pulses whose amplitudes are in the ratio 1:2:4. Channels 4, 5, and 6 may be decoded similarly to determine one of eight vertical positions. The instability and drift nor- --- Small loop tape recorder (left) drives system. From left, units shown are: power supply; monoscope chassis and decoding and deflection circuitry; companion unit, providing sweep circuits for both monoscope and display scope, and pulse standardization equipment; and small monitoring display scope The 7 and 9 pin Miniature Series, T, TR and NW Series are variously covered by Military Specifications: - MIL-S-9372B (USAF) - MIL-STD-242B (Ships) - MIL-S-19786A (Navy) - SCL-6307/2 (Signal Corps) CINCH HEAT DISSIPATING TUBE SHIELDS for Miniature, Subminiature, Octal and Power Tubes In the Octal and Power Series, the shields are covered by Redstone Arsenal and Signal Corps approval. Where shields are required blackened, Henderlube is supplied as the Standard finish. OCTAL and POWER SHIELDS Under license arrangement with International Electronic Research Corporation. SUBMINIATURE AND MINIATURE SHIELDS Centrally located plants at Chicago, Illinois, Shelbyville, Indiana, City of Industry, California and St. Louis, Missouri. Cinch ELECTRONIC COMPONENTS® Cinch MANUFACTURING COMPANY 1026 South Homan Ave., Chicago 24, Illinois Division of United-Carr Fastener Corporation, Boston, Mass. mally associated with cathode ray tube devices is overcome by a simple but highly effective combination decoding and deflection circuit (see Fig. 1 and 2). The voltages applied to the four monoscope deflection plates are developed across voltage-reference diodes which are driven from constant current sources. This arrangement provides stable d-c levels without amplification. A series string of three voltage-reference diodes is connected to each of the four monoscope deflection plates (see Fig. 1). The three diodes have breakdown voltages in the ratio 1:2:4. A transistor is connected across each diode and is biased so the constant current may be shunted around the diode or through it, depending on whether or not the transistor is saturated. The transistors associated with one deflection plate are all normally biased OFF while the transistors associated with the other plate are biased to saturation. In Fig. 1, the right deflection plate will be B+ potential (less the saturation voltage of the three transistors). Because the transistors are biased OFF and the current is through the diodes, the left plate is 70 volts below B+ potential. The transformers which couple incoming pulses to the transistors are polarized to turn off the saturated transistor and saturate the open transistor. The voltage across either diode string can vary from zero to 70 volts in 10-volt increments. As voltage excursions on each deflection plate are equal in amplitude and opposite in polarity, true push-pull deflection results. Thus eight levels of horizontal deflection are established by eight input pulse combinations. Vertical deflection circuitry is identical to the horizontal except for slightly different Zener voltages of the voltage reference diodes. A higher voltage is required in the vertical direction because of a slightly lower deflection sensitivity. With the eight voltage levels on each deflection axis, 64 characters are defined by the binary information on the six-wire input. Results Excellent stability of deflection has been obtained and is due to three prime factors: the voltage-reference diode is a constant-voltage device with a low incremental impedance; the voltage-reference diodes are driven from constant current sources; and, series-string voltage-reference diodes are used with breakdown voltages near the zero-temperature coefficient point of operation, reducing temperature sensitivity. The system was developed by W. E. Evans, L. J. Kabell, the author and other members of the Video Systems Laboratory of Stanford Research Institute for the A. B. Dick Company of Chicago as a component of that firm's Videograph equipment. Helping to guarantee a vital "something" for a rainy day The effectiveness of America's defense "umbrella"—today and tomorrow—depends on instant availability of superior electronics weapons. For over seven years, the Hallicrafters company has been answering this urgent need with • QRC—Quick Reaction Capability. For your electronic requirements . . . from single circuit to complete system . . . for application on land, sea, air or space . . . Hallicrafters QRC can provide you with this unique design and production service in electronics. ENGINEERS: Join our rapidly expanding QRC team now. For complete information address your inquiry to: William F. Frankart, Director of Engineering. hallicrafters company MILITARY ELECTRONICS DIVISION CHICAGO 24, ILLINOIS URGENT PROBLEMS RELIABLY SOLVED ELECTRONICS - FEBRUARY 12, 1960 Solid-State Electronic Tuning By T. W. BUTLER, JR., and G. A. ROBERTS, University of Michigan Research Institute, Ann Arbor, Mich. LONG-TIME goal of circuit designers is to use solid-state tuning devices (back-biased diodes, ferroelectric capacitors and increductors) in stable, high-accuracy receivers and transmitters. However, stability and set-on accuracy of these devices are limited. A generalized frequency synthesizer is under development that accomplishes this goal. Its frequency stability is determined by a reference crystal oscillator or atomic standard. Typical stability is $10^{-7}$ for crystal control. Frequency precision is determined by the reference oscillator and is typically $10^{-6}$ for crystal control. Choice of components and their state of development determine maximum electronic tuning range. For example, tuning ranges available with existing components are 10 to 1 up to about 10 mc and 2 to 1 up to about 100 mc using increductors, 3 to 1 up to about 100 mc using ferroelectric capacitors, and 2 or 3 to 1 up to about 100 mc using back-biased diodes (varactors). Increments of tuning may be made as small as desired, and the synthesizer costs less than an all-crystal synthesizer. It is easily adaptable to microminiaturization, and remote control is possible over low data rate channels. Tuning to desired frequency is rapid. Operating Principle High stability and set-on accuracies at any desired frequency can be obtained by suitably mixing outputs of crystal oscillators but the cost, weight and size of crystal banks with associated switching mechanisms are prohibitive. Electronic tuning may be obtained with such components as varactors and ferroelectric capacitors in a voltage-tuned oscillator. However, these relatively unprecise components do not generally allow open-loop set-on accuracies greater than about 5 percent. It is possible to achieve the characteristics of a crystal-controlled oscillator with the tuning versatility of a voltage-tuned oscillator by combining them through a phase-lock loop system. For example, an oscillator could tune over a 1-mc range in increments of 1 kc by having 3 decade sets of harmonics and subharmonics of the single crystal reference readily available. The voltage-tuned oscillator can be brought near the desired crystal harmonic where the phase-lock loop takes over and pulls the voltage-tuned oscillator into precise frequency alignment with the selected harmonic. Stability and set-on accuracy are then functions of the reference crystal oscillator. Figure 1 (A and B) is a block diagram of the discrete frequency reference and its frequency spectrum, while Fig. 1C is a block diagram of the phase-lock oscillator. A complete discrete frequency generator (DFG) consists of both. The DFG will have output frequencies that are spaced in frequency an amount equal to the harmonic generator fundamental frequency. The number of different frequencies that can be selected is a function of voltage-tuned oscillator accuracy. About 20 is a reasonable figure. By using several DFG's, each with different incremental steps, combining and taking the mixed output, it is possible to cover as large an overall range with increments as small as desired. The lower limit will be set by noise and reference oscillator stability. Frequency Synthesizer A block diagram of the frequency synthesizer is shown in Fig. 2. The DFG's are decade devices and constructed in module form so that system range can be extended. Synthesizer output may be translated to any desired frequency by heterodyning with a desired fixed frequency. For ssb work, the ssb generator should have fine frequency control so that infinite resolution is possible in the 1-mc range. This provision permits setting on frequencies that might fall in between the 1-kc increments. If additional stability were needed, DFG's with 100-cycle and smaller increments could be added to the system. References (1) R. L. Craiglow and E. L. Martin, HEWLETT PACKARD specifies Tung-Sol tubes for high stability calibration generator The Hewlett-Packard Voltmeter Calibration Generator calibrates high impedance voltmeters and oscilloscopes with extreme accuracy. An exceptionally stable source for a wide range of precision voltages, the premium instrument speeds up production and maintenance testing. To assure high stability and low distortion performance, which are listed among the unit's principal advantages Hewlett-Packard selected Tung-Sol 6550's for the 400 cycle power amplifier. As Hewlett-Packard reports: "Tung-Sol's 6550 shows unusual insensitivity to load changes." What this means, of course, is that under varying loads the 6550 drive, with its tight characteristics, holds to a minimum any change in the unit's already minimal distortion (less than 0.2%). In addition the 6550 helps to provide long-term stability. Like all Tung-Sol components, the 6550's optimum performance and dependability stems from Tung-Sol's deep-rooted component know-how. Every step in the manufacturing process is carefully disciplined. Stringent quality control guarantees uniformly high performance in any one lot or from lot to lot. And exhaustive life tests under severe overload assures adequate safety margins. Maybe you're up against some exacting component requirements. If so, you'll be steering a wise course by getting in touch with Tung-Sol applications engineers. They're component experts who will gladly study your design and recommend the units that will do the job... precisely. Tung-Sol Electric Inc., Newark 4, New Jersey. TWX:NK193. For prompt and competent technical consultation on Tung-Sol components call the Tung-Sol Commercial Engineering office near you. SALES OFFICES: Atlanta, Ga.; Columbus, Ohio; Culver City, Calif.; Dallas, Texas; Denver, Colo.; Detroit, Mich.; Irvington, N. J.; Melrose Park, Ill.; Newark, N. J.; Philadelphia, Pa.; Seattle, Wash. Canada: Montreal, P. Q. Every part that goes into a modern-day missile system must pass a rigid battery of tests and a thorough statistical screening to insure highest possible reliability in action. That's why we're pleased to announce that Bristol Syncroverter choppers play an important role in guidance of the U.S. Army HAWK missile, produced by Raytheon Company, Waltham, Mass., prime contractor for the complete HAWK weapons system. Billions of operations. Bristol Syncroverter® choppers are ideal for applications requiring the utmost in statistical reliability. The Bristol life-test lab has now had miniature Syncroverter choppers running for years without failure—both with and without contact load. Just one sample: five choppers with 400-cycle drive and 12v, 1ma, resistive contact load have completed 26,000 hours (2.96 years) continuous operation—over 37-billion operations! An extremely wide variety of standard models is available—including external coil low-noise types. For complete data, write: Aeronautical Components Division, The Bristol Company, 152 Bristol Road, Waterbury 20, Conn. T.M. REG. U.S. PAT. OFF. Computer Searches for Patents PATENT searching using computer techniques is being tested at the U. S. Patent Office. Officials say it may accomplish in about half an hour a job that formerly took a skilled researcher up to a full day. Patent Commissioner R. C. Watson said the new system is being tested on a mass of patent data covering chemical compounds that form the basis for many commonly used plastics. This area, covering the polymer chemical group, is one of the broadest and most difficult in patent research. If successful, the system could be adapted for most problems in patent searching in the chemical field. The system relies on rapid calculation by a general-purpose computer that can process data cards turned out by a punched card machine. The Patent R & D Office and du Pont adapted the agency's punched-card file, which had been processed only on specially built equipment, to commercially available data-processing equipment. A Bendix G-15 digital computer successfully searched for patent information by a serial scanning technique. Advantages Technology is advancing so rapidly that time-consuming patent searches can deter invention and capital investment in new products and processes. Besides helping patent searchers, the computer records on magnetic tape the areas of invention already mechanized and therefore available for subsequent searches. While searching for specific patents, the computer identifies areas of invention or discovery that may have been overlooked. For instance, in searching for patented chemical compounds capable of performing a certain job, the machine may also be instructed to indicate those compounds that have similar characterFast-Response Overload Protection By FRED W. KEAR, Integrated Dynamics Div., Globe Industries, Inc., Albuquerque, N. M. FIG. 1—Current greater than 3 amp drops base voltage of $Q_2$, rapidly cutting off output THOUSANDS of dollars worth of transistors and other components have been saved with a transistorized overload circuit. Fuses were found to be unsatisfactory because of their time delay. The circuit switches power off much faster than conventional current protection devices. The overload circuit in Fig. 1 is useful for production and maintenance testing with low d-c voltages. Current greater than 3 amp flowing through the 0.47-ohm resistor in the emitter circuit of current-switching transistor $Q_1$ drops voltage on the base of voltage-sensing transistor $Q_2$. This drop causes $Q_2$ to saturate, dropping bias voltage and causing $Q_2$ and $Q_1$ to saturate. Transistor $Q_1$ opens the circuit immediately and keeps it open for the duration of the overload. For complete short circuits, $Q_1$ latches cut-off relay $K_1$, providing positive protection. Even if $Q_1$ were to fail, $Q_1$ would open the circuit, providing more reliable protection. Transistor $Q_1$ also provides protection against fast-rising surges that would tend to damage $Q_1$ before relay $K_1$ could operate. The transistors are provided with heat sinks to protect them from heavy loads or extended periods of undetected overloads. Resistors can be selected to protect circuits using greater load currents or supply or load voltages. --- Properties* \| Property \| US600 \| US500 \| US100 \| \|---------------------------\|---------\|---------\|---------\| \| Dielectric Constant \| 1350 \| 1200 \| 500 \| \| Curie Temperature \| 310°C \| 380°C \| 150°C \| \| Rad. Coupling Coefficient \| 0.46 \| 0.50 \| 0.31 \| \| d Constant ($d_{33}$) \| $-120 \times 10^{-12}$ \| $-170 \times 10^{-12}$ \| $-62 \times 10^{-12}$ \| \| g Constant ($g_{33}$) \| $25.3 \times 10^{-3}$ \| $38 \times 10^{-3}$ \| $31 \times 10^{-3}$ \| --- Transducer elements are intended for use as drivers, resonators, and sensors. Applications include: missile systems, underwater sounding, thickness detectors, depth and liquid level sensing gages, IF filters, ladder networks, microphone elements, and power drivers. For further information write or call: U. S. SONICS CORPORATION 625 McGrath Highway • Somerville 45 • Massachusetts MOnument 6-5100 CIRCLE 125 ON READER SERVICE CARD 125 Small Beam Switch Extends Applications By eliminating heavy external magnets and magnetic shields, and by improving tube performance, and cost and packaging, a new family of beam switching tubes are bidding to claim an extended market. Beam switching tubes are already employed to perform functions of counting, distributing, programming, sampling, frequency dividing, gating, timing, coding, decoding and multiplexing. Smaller, less expensive versions of the beam switching tubes may assure wider use in commercial, military and industrial applications. The BEAM-X switch, developed by Burroughs Corp., Plainfield, N. J., is functionally similar to its predecessor, but is ten times lighter, five times smaller and costs half the price. How Used A few applications are given to show typical ways in which the new decade switch can be used. Each application cited is one in which beam switching tubes are already employed. In Fig. 1, pulses which represent photocell outputs, mechanical switch closures or some frequency, are delivered to the switching grids of the first BEAM-X Counter. The tenth output of this tube is used to advance the succeeding switch one position. Thus, with two tubes, 99 counts can be accumulated and with three, 999 and so on indefinitely. In addition to the cascade or carry output, current is available at each end of the ten positions to activate visual readouts, show the progress of the count, activate a printer for recording information upon command, or to perform useful work at the end of a preset number of events. Multiposition Functions The individual outputs of the new decade tube can be utilized to perform multiposition functions sequentially. In Fig. 2, the tenth output of each switch transfers the beam from switch to switch. Driving pulses are delivered to all tubes in parallel so that synchronous operation is assured. The nine constant-current outputs per switch can be used for multichannel communications in gating, timing or sampling applications such as airborne telemetry. Data Conversion In data processing there is a need to convert information rapidly from one form to another. The small tube handles this job by accepting binary coded decimal instructions and producing either decimal or analog equivalent as a useful output. This technique is used in air traffic control systems to display the decimal equivalent of a plane's binary identification number. The first of this new series of beam switching tubes, the BX-1000, weighs $1\frac{1}{2}$ oz and has a volume of 3 cu in. This device is designed for over 1 mc operation and tested to operate at 55 v supply. The output electrode can be... OUTSTANDING IN PERFORMANCE type D RESIN-COATED SILVERED MICA CAPACITORS Sangamo Type D mica capacitors combine the excellent electrical performance characteristics of silvered mica with a multi-layer, protective case of high moisture-resistant thermo-setting resins. The Type D is designed to operate over the temperature range of $-55^\circ C$ to $+125^\circ C$ at rated working voltage without derating. Available in capacitance tolerance values of $\pm 20\%$, $\pm 10\%$, $\pm 5\%$, $\pm 2\%$, $\pm 1\%$ (or $\pm 1$ mmfd, whichever is greater). The insulation resistance of these capacitors will exceed 3,000 megohms at 125°C. Insulation resistance shall be greater than 1000 megohms as measured in accordance with paragraph 2.6.2 of EIA specification RS-186-A, Method 2. Paragraphs 2.4 and 2.6.1 do not apply. The test shall continue for 10 cycles, as described in paragraph 2.5. Insulation resistance shall be greater than 3000 megohms after being subjected to temperature cycling between $-55^\circ C$ and $+125^\circ C$, as outlined in Method 102-A, Test Condition D, and followed by Method 104-A, Test Condition A, of MIL-STD 202A. Write for Bulletin SC59-10 \| TYPE \| DC WORKING VOLTAGE - VOLTS \| CAPACITANCE RANGE - MMF. \| \|------\|---------------------------\|--------------------------\| \| D-15 \| 500 \| 5-400 \| \| \| 300 \| 5-800 \| \| D-20 \| 500 \| 100-2000 \| \| \| 300 \| 100-4000 \| \| D-30 \| 500 \| 1000-10000 \| \| \| 300 \| 1000-20000 \| SANGAMO ELECTRIC COMPANY SPRINGFIELD, ILLINOIS SC-59-10 TWO NEW "800" OHM ALLOYS for Potentiometers and Precision Wire Wound Resistors HOSKINS Chromel-R A modified 80-20 type nickel-chromium alloy possessing optimum uniformity of all physical properties required for close tolerance electronic control applications. Possesses electrical resistivity of 800 ohms/cm² at 20°C, and a low temperature coefficient controlled within 0 ± 10 ppm/°C. Performance characteristics include remarkably low noise level plus exceptional linearity and stability from −65° to +150° C. HOSKINS Alloy 815-R A lower density, higher resistivity iron-chromium-aluminum composition that gives you 14% more ohms per pound than nickel-chromium resistor alloys. It possesses high strength, good ductility, excellent resistance to wear and corrosion. Specific resistance is 815 ohms/cm² at 20°C, and temperature coefficient is inherently controlled within 0 ± 10 ppm/°C. over the range from −65° to +150° C. If you make potentiometers or precision wire wound resistors, these alloys are right for you—right for your customers, too. Complete technical data—the most comprehensive ever offered—are available upon request, as are sample spools of both alloys taken from current production material. Send for them today! HOSKINS MANUFACTURING COMPANY 4451 Lawton Avenue • Detroit 8, Michigan Custom-Quality resistance, resistor and thermo-electric alloys since 1908 Ferroelectric Is Also Ferrimagnetic The National Bureau of Standards has discovered a series of materials that show simultaneously both ferroelectric and ferrimagnetic* properties. The generalized composition is a barium niobate containing any one of several rare earths plus iron oxide, and has a single-phase crystalline structure. As the two properties seem to be mutually dependent in these materials, the composition should find application in new electronic components where a coupling between dielectric and magnetic effects is desirable or where a magnetic material having a high dielectric constant would be useful. Both Properties Ceramics with magnetic properties have been known for a long time, but until now none had been known to show both ferroelectric and magnetic properties at the same time. Whether or not such a material could exist is not stated by theory. P. H. Fang and R. S. Roth of the Bureau's mineral products laboratory postulated the existence of the material and then examined nearly 90 dielectric compositions before finding one with both types of properties. Once this composition was discovered, additional similar structures were postulated by substituting other atoms at appropriate locations in the crystal. Samples of the material are made following the usual ceramics laboratory procedures: the constituents are mixed, pressed into pellets, and fired until the materials sinter. Its formula is (Ba_{0.8} R_{0.2}) (Nb_{0.8-x} Fe_{1.3x})O₃, where R is a rare earth and x varies from 0 to 1. The structure is described in terms of the tungsten-bronze structure. * Ferrimagnetism as distinguished from ferromagnetism, is the property of a material with only partial resultant magnetization because of the presence of antiparallel magnetic spins. Some of these compositions were made with the rare earths neodymium, samarium, europium, or gladdolinium in one part of the structure, and with varying amounts of iron in another part of the structure. All show both ferroelectric and magnetic properties in a single-phase crystalline form. The presence of both the rare earth and the iron seems to be necessary for the material to show both properties simultaneously; the kind of rare earth has a substantial effect on the Curie points of the composition. Verification Ordinarily, ferroelectrics have high dielectric permittivity and very small magnetic permeability; ferrimagnetics have high magnetic permeability and very small dielectric permittivity. The material discovered by the Bureau has both of these properties to an appreciable extent. Ferroelectric properties have been confirmed by the presence of the dielectric hysteresis loop and the piezoelectric resonance. The piezoelectric effect was measured on a polarized ceramic disk. Ferrimagnetic properties are verified by the presence of the remanent magnetization and the effect of the replacement of different rare earth ions on the ferrimagnetic Curie temperature. Rare Earth Ions There seems to be some correlation between the ferroelectric and magnetic properties. When the rare earth ion is Nd$^{3+}$, both effects exist only below room temperature. On the other hand, with the rare earth ions Sm$^{3+}$, Eu$^{2+}$, and Gd$^{3+}$, the materials show both properties above room temperature. However, when the rare earth ion is La$^{3+}$, neither ferroelectric nor ferrimagnetic effects were observed down to the temperature of liquid nitrogen. The relaxation dispersions of the dielectric permittivity and the magnetic permeability occur in the same frequency region, around 20 megacycles. The ferroelectric coercive field shows a strong temperature dependence and large temperature hysteresis. The magnetic coercive field is nearly 1,000 oersteds. --- New HIGH POWER 60 AND 400 CPS TRANSI-MAGS® STANDARD LINE OF FAST RESPONSE TRANSISTOR MAGNETIC AMPLIFIER FOR SERVO MOTOR CONTROL For driving AC servo motors in industrial and military servo systems which require hi-power, fast response and (static) high reliability. A COMPLETE SERVO AMPLIFIER DRIVE SYSTEM IN A SINGLE PACKAGE - TEMPERATURE STABILIZED - FAST RESPONSE - HIGH GAIN - INSTANTANEOUS START - SMALL STANDBY POWER - RUGGED - HIGH RELIABILITY - QUADRATURE REJECTION - SERVO MOTOR COMPATIBILITY --- PERFORMANCE SPECIFICATIONS • AC INPUT • TRANSI-MAG 400 CPS TRANSISTOR MAGNETIC SERVO AMPLIFIERS \| MODEL \| TMA 4501B \| TMA 4571B \| TMA 4601B \| TMA 4701B \| TMA 4801B \| TMA 4901B \| \|-------------\|-----------\|-----------\|-----------\|-----------\|-----------\|-----------\| \| MAXIMUM POWER OUTPUT \| 85 Watts \| 130 Watts \| 200 Watts \| 850 Watts \| 1500 Watts \| 2500 Watts \| \| TYPICAL SERVO MOTOR LOAD \| Diehl FPF 49-19-1 \| Diehl FPF 66-26-1 \| Diehl FPF 85-16-1 \| Diehl ZP 105-2217-1 \| Diehl ZP 143-2256-1 \| Diehl ZP 162-2209-1 \| \| POWER SUPPLY \| 115V 400 CPS 1 Phase \| 115VAC \| 220VAC \| \| MAXIMUM OUTPUT VOLTAGE \| \| \| \| \| INPUT IMPEDANCE \| \| 10,000 OHMS \| \| \| MAXIMUM POWER GAIN \| $1 \times 10^7$ \| $1.5 \times 10^7$ \| $2.2 \times 10^7$ \| $1 \times 10^7$ \| $1.7 \times 10^7$ \| $2.7 \times 10^6$ \| \| SENSITIVITY \| \| \| 0.3VAC INTO 10,000 OHMS FOR FULL POWER OUTPUT \| \| RESPONSE TIME \| \| \| .01 SECONDS \| \| AMBIENT TEMPERATURE \| \| \| $-55^\circ C$ to $+71^\circ C$ \| --- MAGNETIC AMPLIFIERS INC. 632 TINTON AVENUE • NEW YORK 55, N.Y. • CYPRESS 2-6610 West Coast Division 136 WASHINGTON ST. • EL SEGUNDO, CAL. • OREGON 8-2665 CIRCLE 129 ON READER SERVICE CARD 129 Wheel in Oven Makes 150 C Tests WHEEL-SHAPED component carrier is employed in a test set recently developed and put into use by Sperry Semiconductor Division, Sperry Rand Corp., South Norwalk, Conn. The equipment classifies silicon diodes according to reverse current characteristics at high temperature. The wheel is upright and enclosed in an oven. This design promotes close temperature control in the oven, mechanical simplicity and compactness. With manual loading and automatic testing and unloading, the machine has a top rate of 1,200 diodes an hour. The rate depends on test specifications. Normally, the diodes are held under high potential for 20 minutes at 150 C before classification, to eliminate drift. At this cycle, the testing rate is 720 an hour. The machine can be adapted to test other parameters. A second model under construction will have automatic loading. Additional machines with the same concept are planned. The wheel has 200 loading positions, each a pair of lead-holding clips. The circumference is divided into 40 segments, each carrying 5 diodes and equipped with a resistor to protect the other 4 diodes should the fifth diode malfunction or be improperly loaded. During the warmup and stabilization period, the diodes are under load from a common power supply. This supply simultaneously loads 32 segments, or 160 diodes, as they approach the test positions. During test, each diode is individually loaded from a second power supply. Loading and Operation The operator places each diode's leads across 2 bar contacts to check polarity. The diode is then placed in a slide bar which is pushed into the oven, placing the diode leads adjacent to the wheel. As the wheel steps, a pair of contact clips lift the diode off the slide bar. As the wheel continues to step, a cam presses against the leads, seating them in the contacts. The entry portion of the oven is overheated to compensate for the diodes entering at room temperature. Temperature is then stabilized during the half-revolution leading to test positions. The oven is electrically heated. Thermostatic controls are placed at 3 points, with an additional safety thermal overload. Diodes are stepped through 4 test An entirely NEW concept in ceramic capacitors CEROL* HIGH CAPACITY HIGH RELIABILITY ROLLED CERAMIC CAPACITORS* TYPICAL CEROL CAPACITOR TEMPERATURE CHARACTERISTIC TYPE CL 90 \| Port Number \| Cap. Mfd. \| D Max. in. \| L Max. in. \| \|-------------\|-----------\|------------\|------------\| \| CL90V104AM \| .1 \| .210 \| .690 \| \| CL90V254AM \| .25 \| .260 \| .690 \| \| CL90V504AM \| .5 \| .350 \| .690 \| \| CL90V105AM \| 1.0 \| .480 \| .690 \| \| CL90V205AM \| 2.0 \| .400 \| 1.44 \| Cap. Tol. = ±20% P.F. = 2% Max. T.C. (0 Voltage) = +15% -25% over temperature range of -55°C to 125°C. T.C. (100 V. applied) = +15% -35% Working Voltage = 100 VDC at 85°C, derate to 50 VDC at 125°C Test Voltage = 300 VDC Insulation Resistance = 100 Meg.—Mfd. minimum Series Resistance < .25 ohms at 8 to 10 mc. Other requirements per MIL-C-11015B Leads axial #22 gauge 1½" ±¼" long New from Hi-Q... a major breakthrough in the design and construction of ceramic capacitors that provides extremely high capacity ceramic units in ranges previously unattainable. CEROL capacitors are rolled ceramic capacitors in the high capacitance range of paper and plastic film dielectrics but in much smaller physical sizes and with superior electrical characteristics. Designed for general applications in bypass-coupling, filtering and blocking circuits, CEROL capacitors offer excellent electrical characteristics for critical applications in decoupling and pulse circuits where low series resistance at high frequencies together with extremely miniature sizes are required. The extremely low series resistance of CEROL capacitors makes them ideal for computer applications. CEROL capacitors are currently available in capacitance ratings of .1, .25, .5, 1.0 and 2.0 mfd for operation at 100 VDC at temperatures between -55°C to 85° and at 50 VDC up to 125°C. Capable of withstanding severe environmental conditions CEROL capacitors will meet or surpass all the applicable requirements of MIL-C-11015B. Write today for detailed information on these remarkable new capacitors to... AEROVOX CORPORATION OLEAN, NEW YORK ELECTRONICS • FEBRUARY 12, 1960 For research and more accurate production of hyperpure materials for semi-conductors—the new Lindberg Floating Zone Scanner This newly developed Floating Zone Scanner is a product of Lindberg's technical staff, widely recognized for many significant developments in the application of heat to industry. Expertly designed, it provides more accurate and more precise production of semi-conductors as well as serving as ideal research equipment. Already, a number of important companies in the semi-conductor field are using this equipment for research and production. Lindberg Induction Heating Units have been specifically designed for use as research and production equipment for crystal growing and zone refining of semi-conductors and other materials. Write for our Bulletin No. 1600. High Frequency Division LINDBERG ENGINEERING COMPANY 2457 WEST HUBBARD STREET, CHICAGO 12, ILLINOIS Los Angeles Plant: 11937 South Regentview Avenue, at Downey, California In Canada: Birlefco—Lindberg, Ltd., Toronto Instrument cabinet contains meters and relays. Power supplies are mounted on other side positions, each representing a reverse current level. Each time a diode enters a test position, the indexing mechanism locks the wheel's position. The contacts are isolated from the common power supply and grasped by air-actuated test contacts leading from the test power supply. If the diode meets the level set at the first position, it is pulled from the contact by hook-shaped fingers and is dropped into a bin. Otherwise, it is retested at the next positions. Instrumentation consists of 3 Sensitrol microammeter relays, associated relays to operate the selectors, the power supplies, calibrating fixtures, trouble and indicator lamps and buzzers, and failsafe relays. Power supplies are provided with voltage range selectors and are kept in calibration with voltmeters. The meter relays are monitored by precision resistors and periodically calibrated with instruments. Test ranges are variable. A typical group of classifications is: up to 4.5 μA, 4.5 to 13 μA, 13 to 27 μA and 27 to 50 μA. Poly-Glycol Improves Acid Flux Performance SOLDER FLUX of the zinc chloride type, as well as resin flux (Electronics, p 96, July 31, 1959), is improved by using polyethylene glycol as solvent, according to the Tin Research Institute, Greenford, Middlesex, England. With polyethylene glycol, the flux spatters very little, spreads the solder over a larger area and flux residues can be easily washed off. with water. Water solvent, by comparison, spatters severely, evaporates and retards solder spread, and develops a hard to wash film, contributing to subsequent corrosion problems. Polyethylene glycol is also reported to be a good vehicle for hydrazine hydrochloride and bromides, organic amine hydrochlorides and acidified resinous fluxes as well. Adjustable Punch Set Uses Peg-Hole Plates Punching setup installed in press Setup can be put together and adjusted with template outside of press PUNCH and die kits suitable for multiple-hole punching of short-run sheet metal parts are announced by O'Neil-Irwin Mfg. Co., Lake City, Minn. The kits contain dies and punches in a variety of shapes, die holders and die sets with holes on $ \frac{3}{16} $-inch centers. Holes can be punched anywhere within the area of the die sets by changing the location of the holders or rotating the dies in the holders. Setups can be made either in the press or outside it. The punches and dies are chrome steel and will punch material as thick as 16 gage, in mild steel, or $ \frac{1}{4} $ inch, in aluminum, to 0.005-inch tolerances. For basic research, pilot plant studies and the more efficient production of semi-conductors—new Lindberg Diffusion Furnaces Here is a new Lindberg Furnace designed specifically for basic research, pilot plant work, or production of solid state devices. It is offered in a variety of sizes and capacities to enable industries in the semi-conductor field to have higher powered equipment adequately insulated and designed for its specific use. With this type of furnace available, it is not necessary for industry to attempt to adapt ordinary furnaces to the highly specialized requirements of the semi-conductor field. For complete information on Lindberg's standard line of furnaces specifically designed for gaseous and solid diffusion uses write for our Bulletin No. T-1081. Pilot Plant Equipment Division LINDBERG ENGINEERING COMPANY 2457 WEST HUBBARD STREET, CHICAGO 12, ILLINOIS Los Angeles Plant: 11937 South Regentview Avenue, at Downey, California In Canada: Birlefo—Lindberg, Ltd., Toronto LINDBERG heat for industry CIRCLE 133 ON READER SERVICE CARD 133 Inverter single phase TEMCO AIRCRAFT CORP., P. O. Box 6191, Dallas 22, Texas. New single phase inverter is designed specifically to supply accurate 400 cycle power to rate gyro packages. It is suitable for any application requiring small quantities of 26 ± 1.0 percent volt a-c power. Occupying only 20 cu in., the unit is capable of delivering 20 w at 400 cycles ± 0.1 percent with an input voltage of 28 ± 4 v. Distortion is less than 4 percent. Efficiency exceeds 60 percent at full load. The single phase inverter meets or exceeds all applicable portions of MIL-E-5272. CIRCLE 301 ON READER SERVICE CARD Terminals feed-through type CAMBRIDGE THERMIONIC CORP., 445 Concord Ave., Cambridge 38, Mass. Designed for use in plug-in components, the Cambion 1030 and 1031 feed-through terminals fit into 7-or 9-pin miniature sockets. These have pin diameters of 0.040 in. and are available in six different mounting shank lengths. Terminal 1032, with a pin diameter of 0.032 in., is available in five different mounting shank lengths. All are of quality brass, finished with 0.0003 in. silver plate and coated with water dip lacquer. Terminals are quality-controlled and guaranteed. CIRCLE 302 ON READER SERVICE CARD D-C Amplifier differential type ALLEGANY INSTRUMENT CO., INC., 1091 Wills Mountain, Cumberland, Md. Chopper stabilized, the model 516 true differential d-c amplifier features low noise of 14 μv rms over the entire bandwidth of d-c to 25 kc. High output of ±100 ma at 10 v, with continuous variable gain to 1,000 X, makes it a flexible general-purpose instrument. It is available in individual case or eight to a 19 in. rack. CIRCLE 303 ON READER SERVICE CARD Switch Attenuator high power KEARFOTT CO., INC., 14844 Oxnard St., Van Nuys, Calif., announces a ferrite switch attenuator that may be used in high power transmitter circuitry. Model W662-3A-2 is ideal for range adjustment, static testing and slow modulation up to 250 kw. Frequency range is 8.5 to 9.6 kmc; maximum attenuation, 35 db min.; minimum attenuation, 0.5 db max.; vswr max., 1.3 bilateral; peak power, to 250 kw; average power, at 250 w; switching time, 20 msec; switching rate at 2 cps; driving power, 100 w max.; weight, less than 4 lb. CIRCLE 304 ON READER SERVICE CARD Spectrum Analyzer log-linear scan PROBESCOPE CO., INC., 8 Sagamore Hill Dr., Port Washington, N. Y. Telemetering analyzer with automatic optimum logarithmic display of subcarrier channels and simultaneous linear display of individual channels is available in the model TA-100L-120L. All f-m/f-m sub-carrier channels will be linearly displayed along the horizontal axis of the log scan crt and at the same time individual portions of the spectrum can be analyzed on a second crt. Frequency ranges are 350 cycles to 85 kc or 120 kc logarithmic display and 13 cycles to 85 kc or 120 kc linear display. Sweep width, 150 cycles to 22 kc. Other features: 60 db dynamic range, 500 μv sensitivNEW FROM JFD LUMPED CONSTANT DELAY LINES Meet the newest addition to the growing family of JFD precision electronic components. Designed with compactness, ruggedness and reliability in mind, new JFD lumped constant Delay Lines upgrade your prototype or production project. Compare the advantages of the standard JFD lumped constant delay lines: - High delay-to-rise time ratio with minimum signal attenuation. - Tolerance of ±5% max. on delay and characteristic impedance. - Temperature range of −55° C to +125° C. - Delay time thermal stability of 50 parts per million per degree centigrade. - Up to 25 Mc bandwidth. - Virtually linear phase shift. - Hermetically sealed metal cases for maximum resistance to shock, vibration and humidity. - Meet all applicable MIL specs. Typical Standard Delay Line Characteristics \| Delay Time \| Rise Time \| Size \| Rise Time \| Size \| Rise Time \| Size \| \|------------\|-----------\|----------\|-----------\|----------\|-----------\|----------\| \| 5 µ sec. \| 1.0 \| 1½"x1½"x2½" \| 2.0 \| 1½"x1½"x3" \| 5.0 \| 1¾"x1¾"x2¾" \| \| \| 1.5 \| 1¾"x1¾"x2¾" \| 1.0 \| 1¾"x1¾"x2¾" \| 2.5 \| 1¾"x1¾"x3½" \| \| \| 3 \| 1¾"x1¾"x2¼" \| .6 \| 1¾"x1¾"x3¼" \| 1.5 \| 2½"x2½"x4¾" \| \| \| 1.5 \| 2¼"x2¼"x4¼" \| .3 \| 2¼"x2¼"x4¼" \| .75 \| 2¾"x2¾"x5½" \| Range of characteristic impedance: 50 ohms to 2000 ohms ±5%. Attenuation: Less than 1db per µ sec. up to 3 µ sec. delay; 6db max. up to 50 µ sec. delay. Temperature stability: 50 parts per million per degree C from −55° to +125° C. Whether your application calls for standard or custom-built lumped constant or distributed constant delay lines, our engineering staff will be glad to review your needs and submit recommendations. Closer tolerance delays and impedances are available, in forms, sizes and terminal designs to match your needs. Write for Bulletin No. 213A. Pioneers in electronics since 1929 ELECTRONICS CORPORATION 1462 62nd Street, Brooklyn, New York JFD International, 15 Moore Street, New York, New York JFD Canada Ltd., 51 McCormack Street, Toronto, Ont., Canada Steel Clamp self-aligning TIMBER-TOP, INC., 36 Brooklyn Ave., Freeport 8, L. I., N. Y., has designed a self-aligning stainless steel clamp that permits the secure fastening of components with base flanges or grooves. Synclamps are available in 8 different sizes with the same o-d of 0.390 maximum. Ideally suited for hard-to-reach places, these standard fasteners are self-locking and withstand extreme environmental conditions. Self-alignment is quickly and easily accomplished because of a nylon insert. When the screw is tightened, the nylon insert is stripped which allows the clamp to self-align perfectly. The more the screw is turned the tighter the clamp seats. CIRCLE 306 ON READER SERVICE CARD Tube Socket for planar triode INSTRUMENTS FOR INDUSTRY, INC., 101 New South Road, Hicksville, L. I., N. Y., has designed a tube socket for GE's GL6299 uhf planar triode that enables the designer to quickly realize practical uhf lumped constant circuitry with absolute accuracy and linear and logarithmic amplitude scale. CIRCLE 305 ON READER SERVICE CARD Automatic Miniaturized Silicon Power Rectifiers Small to fit your space requirements \| JEDEC TYPE NO. \| PEAK INV. VOLTAGE (V) \| MAX. AVG. RECTIFIED CURRENT (mA) @ 25° C. \| MINIMUM SATURATION VOLTAGE @ 100° C. (VOLTS) \| MAXIMUM REVERSE CURRENT @ 25° C. (μA) \| MAXIMUM VOLTAGE DROP @ 400 ma DC @ 25° C. VOLTS DC \| \|----------------\|-----------------------\|------------------------------------------\|-------------------------------------------\|--------------------------------------\|--------------------------------------------------\| \| 1N6-15 \| 225 \| 400 \| 150 \| 275 \| 0.2 \| 15 \| 1.0 \| \| 1N646 \| 300 \| 400 \| 150 \| 360 \| 0.2 \| 15 \| 1.0 \| \| 1N647 \| 400 \| 400 \| 150 \| 480 \| 0.2 \| 20 \| 1.0 \| \| 1N648 \| 500 \| 400 \| 150 \| 600 \| 0.2 \| 20 \| 1.0 \| \| 1N649 \| 600 \| 400 \| 150 \| 720 \| 0.2 \| 25 \| 1.0 \| Resistive or inductive load We've shrunk the size, but not the quality. All the outstanding characteristics and reliability you expect of products from General Instrument Corporation are present in these miniaturized units. Data sheets on these and other Automatic silicon rectifiers are available upon request. Semiconductor Division GENERAL INSTRUMENT CORPORATION 65 Gouverneur Street, Newark 4, N. J. Midwest office: 5249 West Diversey Ave., Chicago 39 Western office: 11982 Wilshire Blvd., Los Angeles 25 GENERAL INSTRUMENT CORPORATION INCLUDES F. W. SICKLES DIVISION, AUTOMATIC MANUFACTURING DIVISION, SEMICONDUCTOR DIVISION, RADIO RECEPTOR COMPANY, INC. THE HARRIS TRANSDUCER CORPORATION, MICAMOLD ELECTRONICS MANUFACTURING CORPORATION AND GENERAL INSTRUMENT — F. W. SICKLES OF CANADA LTD. (SUBSIDIARIES) ELECTRONICS • FEBRUARY 12, 1960 Draftsmen can straighten up and do better, faster work! Bruning's all-new Neoglide drafter literally puts draftsmen in a position to do better, faster work. It provides complete maneuverability on any board at any angle without adjustment! Reinforced U-Beam construction offers amazing rigidity, strength, and accuracy. Yet resistance-free movement of the counterweight and vertical beam provide free-and-easy "floating" action. Touch-control protractor head gives automatic, pin-point angle selection. Mail coupon now to put yourself in a position to save time and money with all-new Neoglide! Study of 300 draftsmen showed 35% savings on drawing time—1/5 the backaches—on vertical or near vertical boards. Charles Bruning Co., Inc. Dept. 2-Z 1800 Central Rd., Mt. Prospect, Ill. Offices in principal U.S. Cities In Canada: 103 Church St., Toronto 1, Ont. Please send me more information about your all-new Neoglide drafters. Name________________________Title_____________________ Company______________________________________________ Address_______________________________________________ City____________County________State______________ Leak Detector ultrasensitive Crosby-Teletronics Corp., Westbury, N.Y., has developed an ultra-sensitive mass spectrometer leak detector with important military and civilian applications. The machine accurately detects, locates and measures the size of leaks in vacuum, pressure or hermetically sealed systems, devices or components. Leak rate sensitivity is $10^{-14}$ standard cubic cm per sec. The equipment is sensitive to helium only. Presence of other gases in the equipment cannot give false indications. Audio Tube with rattle control General Electric Co., 212 N. Vignes St., Los Angeles 54, Calif. The 7581 is a 30-w beam-power pentode for high quality audio power output. It has a low-loss, mica-filled base which also proTHE ELECTRONICS MAN... HIS BUSINESS IS DIFFERENT A man in one industry usually sells to another. A steel man sells to the automotive industry. A packaging man sells to a soap company. The electronics man is different. The electronics man sells a large portion of his output to another electronics company. The design man may be the key to selling a President. The President of one company may sell to the design man of another. The project engineer is responsible for product design, but may also determine market potentials for new products. Look at the badge above. It reads Research-Design-Production-Management. The interests of the electronics man are in any or all of the four areas. No matter where you find the electronics man his engineering background enables him to influence the purchase of electronic components and equipment. Your advertising must reach him to sell electronic goods. And...in selling the electronics man, your basic buy is the one book edited weekly for all electronics men—whatever their title or job functions. The basic book is electronics—52,000 electronics men pay to read it every week. THE ELECTRONICS MAN "BUYS" WHAT HE READS IN... electronics A McGraw-Hill Publication 330 West 42nd Street • New York 36, N.Y. and in the electronics BUYERS' GUIDE Where do you go for fast electronic parts delivery? To the telephone! And call your local RAYTHEON distributor. and at factory prices RAYTHEON COMPANY DISTRIBUTOR PRODUCTS DIVISION • WESTWOOD, MASS. RECEIVING AND INDUSTRIAL TUBES • SEMICONDUCTOR PRODUCTS MECHANICAL COMPONENTS • RAYTHEON/MACHLETT POWER TUBES VOLTAGE REGULATORS • CAPTIVE HARDWARE Serving Key Markets Include Missouri Kansas City Burstein-Applebee Company Baltimore 1-1155 New Mexico Alamogordo Radio Specialties Company, Inc. HEmlock 7-0307 Albuquerque Radio Specialties Company, Inc. AM 8-3901 New York Mineola, Long Island Arrow Electronics, Inc. Pioneer 5-8686 New York City H. L. Dalis, Inc. EMpire 1-1100 Milo Electronics Corporation BEekman 3-2980 Ohio Cincinnati United Radio Inc. CHerry 1-6530 Cleveland Mar Line Cleveland, Inc. Express 1-1800 Pioneer Electronic Supply Co. SUPERior 1-9411 Columbus Buckeye Electronic Distributors, Inc. CA 8-3265 Dayton Sreco, Inc. BALdwin 4-3871 Oklahoma Tulsa S & S Radio Supply CHerry 2-7174 Oregon Portland Lou Johnson Company CAPitol 2-9551 Pennsylvania Braddock Mayne Parts Company Electric 1-1314 Philadelphia Almo Radio Company Walnut 2-5918 Rath Electric Service Co. WAlnut 5-5840 Reading The George D. Barbey Co., Inc. FR 6-7451 Tennessee Knoxville Bondurant Brothers Company Texas Dallas Graybar Electric Company Riverside 2-6451 Houston Harrison Equipment Company CAPitol 4-9131 Utah Salt Lake City Standard Supply Company EL 5-2971 Virginia Norfolk Priest Electronics MA 7-4534 Richmond Meridian Electronics, Inc. RIchmond 5-2834 Wisconsin Appleton Electronic Expeditors, Inc. REgina 3-1755 Green Bay Electronic Expeditors, Inc. HEmlock 2-4165 Menasha Twin City Electronics PArkyway 2-5735 Milwaukee EX-EL Distributors, Inc. Mitchell 5-7900 Electronic Expeditors, Inc. FLagstone 2-2070 Milwaukee Electronic Expeditors, Inc. WOODruff 4-8820 Oshkosh Electronic Expeditors of Oshkosh, Inc. BEverly 5-8930 CIRCLE 203 ON READER SERVICE CARD National, known for its communications and electronics equipment, is equally recognized as a manufacturer of quality components. Sockets, plate caps, grid grips, terminal components, and assemblies are a few of the items available from catalog stock. Most are made to meet JAN-SPECS. A representative catalog listing: Sensing Element* solid state BERKELEY/DYNAMICS, 2831 — 7th St., Berkeley, Calif. Model P-CE solid state photorelay sensing element features direct operation of relays or counters without amplification. The photosensitive unit will operate standard relays or solenoid actuated devices wired in series with model P-CE. Coil resistance of solenoid or relay should be approximately 2,200-2,800 ohms, maximum current 50 ma at 115 v a-c. The sensing element is a compact, rugged, moisture resistant cadmium sulfide cell. Model P-CE has threaded mounting hub for $ \frac{3}{8} $ in. conduit. List price, $10. CIRCLE 310 ON READER SERVICE CARD Digital Test Units low-speed DIGITAL EQUIPMENT CORP., Maynard, Mass., announces a low-cost line of coordinated low-speed digital test equipment building blocks. The 3000 series operate at speeds up to 500 kc, as compared with speeds up CIRCLE 141 ON READER SERVICE CARD MARCONI FM SIGNAL GENERATOR Covers 10 to 470 mc on fundamentals Model 1066A offers a unique combination of features essential to the exacting tasks required of a precision fm generator. Its wide range is covered with the complete absence of spurious sub-harmonics. Directly calibrated stepped and continuous incremental tuning, supported by exceptional frequency stability, bring new ease and accuracy to bandwidth measurement. Deviation up to ± 100 kc is produced at either of two modulation frequencies by a ferrite modulator. Other major features are the Marconi-patented contactless range turret, and a piston attenuator giving a high-quality 50-ohm output. MARCONI FM SIGNAL GENERATOR MODEL 1066A Abridged Specifications FREQUENCY RANGE: 10 to 470 mc in five bands—all on fundamentals. FREQUENCY STABILITY: Better than 0.002% over 24 hours; better after warm-up. INCREMENTAL FREQUENCY CONTROLS: Variable 0 to ± 20 and 0 to ± 100 kc stepped and 10 and 15 kc modulation. MODULATION: 0 to 20 and 0 to 100 kc deviation monitored and continuously variable amplitude control with a gain decay up to 40% is also obtainable. MODULATION FREQUENCIES: 1 and 5 kc output; 0.1 µs to 100 µs at a 50Ω termination. OUTPUT ACCURACY: Incremental, 0.2 db within 2.5% overall; LEAKAGE: Negligible; allows full use of 0.1 µs pulses. TUBES: 5Z4G, 6AK6, 6CD6G, 6AK5, 5861, 6C4, 6L6G, 12AT7, 6X4, 6AU6. Marconi FM Deviation Meters 791D and 934/2 are companion instruments. Send for leaflet B159 for full details. Panel Meters military style Helipot Division of Beckman Instruments, Inc., 2500 Fullerton Road, Fullerton, Calif., has announced a line of 3½ in. panel meters. Built in accordance with MIL-M-10304A, the new line features voltmeters, ammeters, microammeters, and milliammeters. There are 61 standard models now available. CIRCUIT BREAKER low impedance General Electric Co., Circuit Protective Devices Dept., Plainville, Conn. ALB-1C low impedance, high shock circuit breaker is available in ratings 5-75 amperes, and features inverse time delay action. A quick-make, quick-break device, it has a thermal bimetal and independent... NEW CURVE TRACER with Tube Adapter Transistors Tetrode Transistors Diodes Vacuum Tubes 30-Amps Continuous 50-Amps Intermittent 450 Watts Dissipation For complete information write for bulletin #TT108. Baird Atomic Baird-Atomic, Inc. 33 UNIVERSITY RD. CAMBRIDGE 38, MASS. CIRCLE 204 ON READER SERVICE CARD STEREO LEVEL INDICATOR PLASTIC EDGewise METER PLASTIC PANEL METER KEW Around the world it's KEW KYORITSU ELECTRICAL INST. WORKS, LTD. NO. 120, Nakane-cho, Meguro-ku, Tokyo, Japan. Cable Address "KYORITSUKEIKI TOKYO" CIRCLE 205 ON READER SERVICE CARD TRANSISTOR CIRCUITRY ENGINEERING "KNOW HOW" AND PRODUCTION • How to get the optimum performance and reliability from an electronic component is often directly related to research and engineering "know-how" of transistor circuitry. The Acme Electric research and engineering staff have a wealth of experience to develop assemblies in this specialized field of manufacturing. A letter outlining your problem will have our prompt attention. ACME ELECTRIC CORPORATION 312 WATER ST. CUBA, N. Y. West Coast: 12822 Yukon Avenue • Hawthorne, Calif. CIRCLE 143 ON READER SERVICE CARD Custom transformers for printed circuits are now available from ADC in five standard case sizes with terminals and inserts on 0.1" grid multiples. Audio, power, and ultrasonic transformers and inductors with maximum electrical performance for each size are being custom designed for transistor and vacuum tube circuitry. Raised mountings prevent moisture from being trapped. Available in Mumetal cases. They meet MIL-T-27-A Grade 5 Class R or S Life X, and can be designed to meet 500 and 2,000 cps vibration. TYPICAL RATINGS \| Fig. \| Description \| Primary \| Secondary \| Maximum Level (CPS) \| \|------\|-------------\|---------\|-----------\|---------------------\| \| 1 \| Output \| P P collectors 600 ohms CT \| 150 ohms \| +33 dbm (2w) ±2db 250-10,000 cps \| \| 2 \| Output \| 5000 ohms 5ma DC \| 50 250 600 ohms \| +10 dbm (10mw) ±1db 100-10,000 cps \| \| 3 \| Output \| P P collectors 1000 ohms CT \| 4 8 16 ohms \| +25 dbm (300mw) ±1db 250-10,000 cps \| \| 3 \| Interstage \| Collector, 5000 ohms 1ma DC \| P P bases 3000 ohms CT \| +5 dbm ±1db 250-5,000 cps \| \| 4 \| Input \| 50 250 600 ohms \| 50,000 ohms \| +2 dbm ±1db 250-10,000 cps \| \| 5 \| Output \| P P collectors 500 ohms CT \| 4 8 16 ohms \| +20 dbm (100mw) ±1db 250-10,000 cps \| \| 5 \| Interstage \| Collector 7500 ohms 1ma DC \| P P bases 5000 ohms CT \| 0 dbm ±1db 250-10,000 cps \| INDUCTORS \| Fig. \| Description \| Rating \| \|------\|-------------\|--------\| \| 3 \| Audio \| 200 hys 1v 1000 cps 0 DC \| \| 5 \| Power \| 500 mhy 1v 400 cps 10ma DC \| WAVE FILTERS \| Fig. \| Description \| Rating \| \|------\|-------------\|--------\| \| 3 \| Low pass \| 600 ohms input 600 ohms output \| +10dbm f cutoff 50kc Attenuation 18db per octave \| \| 3 \| High pass \| 10,000 ohms input 10,000 ohms output \| +10dbm f cutoff 2kc Attenuation 18 db per octave \| POWER \| Fig. \| Description \| Primary \| Secondary \| VA \| Regulation \| \|------\|-------------\|---------\|-----------\|----\|------------\| \| 4 \| Filament \| 7.5v 380-420 cps \| 6.3v .6a \| 4.0 \| 10% \| \| 5 \| Dual filament \| 26v 380-420 cps \| (1) 6v 5ma (2) 6v 5ma \| .2 \| 2% \| Note: Other combinations are available with 400 cps max, volt ampere ratings up to 15 for Fig. 1, 10 for Fig. 2, 6 for Fig. 3, 4 for Fig. 4, and 1 for Fig. 5. WRITE TODAY FOR COMPLETE INFORMATION MINIATURE TRANSFORMERS FOR Transistor AND PRINTED CIRCUIT APPLICATIONS Automatic Test Set multiconductor cable PESCHEL ELECTRONICS, INC., Towners, Patterson, N. Y. Automatic unit features dielectric testing up to 5 kv rms; any number of test positions to order; voltage continuously adjustable up to 5 kv rms; test duration or dwell time adjustable from 1 to 120 sec; maximum short circuit current limited to 5 ma. Other features include built-in test cage with terminal board for components, easily accessible through top lid of cabinet, with electrical interlock to prevent application of high voltage when lid is open. Built-in test cage obviates need for an auxiliary cabinet with interconnecting cables at high voltage. Coaxial Switch manually-operated JERROLD ELECTRONICS CORP., 15th and Lehigh Ave., Philadelphia 32, Pa. Designed for use wherever r-f energy is to be switched from one circuit to another, the CS-250 transfers r-f energy from 0 to 1,000 mc with a maximum vswr of 1.1 up to 500 mc and 1.2 up to 1,000 mc. The unit is a four terminal, two position magnetic trip element and is available with or without an auxiliary switch. Interrupting capacity: 5 ampere breaker—1,500 amperes, 125 v a-c and d-c; 800 amperes, 300 v a-c; and 10-75 ampere breakers—1,500 amperes, 125 v a-c and d-c; 2,500 amperes, 300 v a-c. Unit meets MIL-C-17588. CIRCLE 313 ON READER SERVICE CARD CIRCLE 314 ON READER SERVICE CARD CIRCLE 144 ON READER SERVICE CARD STACKPOLE Coldite 70+ fixed composition RESISTORS Today's slickest looking resistors... and every bit as good as they look! Designed to MIL-R-11 specifications, they're unmatched for load life and moisture resistance. And now, for the first time, you can get such resistors in a full line of RC-42 (2-watt); RC-32 (1-watt) and RC-20 (½-watt) types IMMEDIATELY from distributors' stocks. Now! PICK 'EM OFF DISTRIBUTORS' SHELVES! ...for military prototypes, small runs, production emergencies or "hurry-up" projects ...in any standard value or tolerance Complete stocks—and we mean complete—in the hands of the 28 selected Stackpole distributors listed below help you handle every job with highest quality resistors, fully proved and accepted for critical applications. Baltimore, MD. Kann-Ellert Electronics, Inc. Battle Creek, Mich. Electronic Supply Corp. Birmingham, Ala. MG Electrical Supply Co. Boston, Mass. Sager Electrical Supply Brooklyn, N.Y. Electrical Equipment Corp. Cleveland, Ohio Pioneer Electronic Supply Co. Dallas, Texas Wholesale Electronics Supply Co. Dayton, Ohio Srepea, Inc. Denver, Colo. Denver Electronics Supply Co. Glendale, Calif. R.V. Weatherford Company Indianapolis, Indiana Radio Distg. Co. Kansas City, Mo. Burstein-Applebee Co. Melbourne, Florida Electronic Supply Miami, Florida Electronic Supply New York, N.Y. Harvey Radio Co. Philadelphia, Pa. Alma Radio Co. San Diego, Calif. Radio Parts Co. Scranton, Pa. Fred P. Purcell Seattle, Wash. C & G Radio Supply Co. St. Louis, Mo. Interstate Supply Co. Syracuse, N.Y. Morris Electronics of Syracuse Tacoma, Wash. C & G Radio Supply Co. Washington, D.C. Electronic Wholesalers, Inc. Waterbury, Conn. Band Radio Supply Co. Inc. West Palm Beach, Fla. Gaddard Distributors, Inc. Wichita, Kansas Interstate Electronic Sup. Corp. Wilbraham, Mass. Industrial Components Corp. Winston-Salem, N.C. Dalton-Hege Radio Supply ...and G C/STACKPOLE, TOO! Attractively packaged by G-C Electronics for service replacement uses, Coldite 70+ Resistors are also available through over 800 G-C distributors. He's discovered the cable that'll do the trick! It's Hickory Brand Microphone Cable! - EXTRA LIMP - LONG FLEX LIFE - LOW CAPACITANCE - HIGH TENSILE STRENGTH These plastic-insulated cables with non-marking jackets are lightweight, weatherproof and highly resistant to abrasion. Use Hickory Brand Microphone Cables for all stage and studio work. Excellent for audience-participation programs. Use also for outdoor extensions. All Hickory Brand Electronic Wires and Cables are quality-engineered and precision manufactured to meet the most exacting requirements. Write for complete information on the full line of HICKORY BRAND Electronic Wires and Cables Manufactured by SUPERIOR CABLE CORPORATION, Hickory, North Carolina transfer switch, coaxially mounted in a rugged die cast metal frame. Its dull wiping, self aligning switch contacts are silver, mounted in a Kel-F dielectric. Insertion loss is less than 1/10 db to 1,000 mc. Isolation between the open and closed circuit is 58 db at 100 mc and 40 db at 1,000 mc. CS-250 is available with BNC connectors at 50 ohm impedance. CIRCLE 315 ON READER SERVICE CARD Trimming Pot stackable HANDLEY, INC., 2030 Colorado Ave., Santa Monica, Calif., announces model 1W-STK trimming pot. Two to ten may be stacked in a row, firmly held together by a steel bolt and nut. Because of the unusual configuration, placement and length of the leads, lead screw position (on top), this stacking feature offers outstanding accessibility and compactness. Trimmer withstands 100 g acceleration, exceeding MIL-R-19; withstands 50 g shock, exceeding NAS 710, Proc. III; and temperature range is from -55 C to 140 C with 1.3 w at 40 C. Small worm gear adjustment, free of back lash, delivers high friction loading. CIRCLE 316 ON READER SERVICE CARD H-V Rectifier for radar use PEK LABS, INC., 4024 Transport St., Palo Alto, Calif. Rated at 75 kv piv and 800 w average plate dissipation, the PEK 5973 h-v rectifier or surge limiting diode is ideally suited to applications where low tube drop is important such as in NOW TEST TRANSISTOR BETA IN THE CIRCUIT WITHOUT POWER ON! New Sierra 219A Transistor Tester reads Beta directly in the circuit; also measures $I_{CO}$ and Beta out of circuit. Simple operation, completely portable. Battery powered; easily used anywhere. Consider the real advantages of testing transistors in the circuit. Downtime, and damage to transistors, is greatly reduced. Complete assemblies are quickly checked out. Quality control "ounce of prevention" is simplified during manufacture. Beta is read simply and directly in or out of the circuit; a basic function of the Sierra 219A is to electrically isolate the transistor under test permitting accurate in-circuit tests. $I_{CO}$ is measured on a straightforward dc basis; collector potentials of 3, 6, or 12 vdc may be selected. For complete information and demonstration, telephone your Sierra representative now. SPECIFICATIONS Test Ranges: Beta, 10 to 100, $I_{CO}$, 0 to 50 $\mu$amp Accuracy: In-circuit ± 20% for external loadings above 500 ohms Out-of-circuit ± 10% Power: Mercury or zinc-carbon battery; 1,000 hrs. average service life; meter indicates battery output Temperature Range: 32° to 122°F for specified accuracy Dimensions: 9" high x 7$\frac{3}{8}$" wide x 6$\frac{1}{2}$" deep; weight 16 pounds Accessories: Test leads supplied Price: $250.00 f.o.b. factory Delivery from stock Data subject to change without notice SIERRA ELECTRONIC CORPORATION A Division of Philco Corporation 6251A BOHANNON DRIVE • DAVENPORT 6-2060 • MENLO PARK, CALIFORNIA, U.S.A. Sales representatives in all principal areas Canada: Atlas Instrument Corporation, Ltd., Montreal, Ottawa, Toronto, Vancouver Export: Frazier & Hansen, Ltd., San Francisco, Los Angeles IT'S WHAT'S IN HERE THAT COUNTS Do you know, for instance...which electronic stocks are hottest? Who's in the news and why? About "Three Approaches to Microminiaturization"? About the newest product ideas hitting the market? What's up in production? Opportunities overseas? What's going on in Washington? It pays to know more than the next man! The questions above are just 6 reasons why you should subscribe to electronics. IF YOU'RE ON THE TAG END OF A ROUTING SLIP, get your own subscription. Knowing what's going on is the first step to going up. Fill in the coupon below right now...it will pay big dividends. FIND WHAT YOU NEED IN...electronics ____Renew my subscription for 3 more years. ____Enter my new subscription. U.S. Subscription Rates: ____3 years $12. ____1 year $6 Canadian rates $10 for 1 year. Foreign rates $20 for 1 year. Name__________________________________________________________ Street__________________________________________________________ City_____________________________Zone_____State_______________ Company_______________________________________________________ Street__________________________________________________________ City_____________________________Zone_____State_______________ Your Title________________________Department_____________________ Product Manufactured or Service Performed_________________________ Mail reply to: electronics, 330 West 42nd Street, New York 36, N.Y. Power Package brushless ELECTRIC MACHINERY MFG. CO., Minneapolis 13, Minn. A 400 cycle motor-generator set complete with controls serves as the power supply for a new computer. The m-g set is rated 7.5 kw at 1,714 rpm with a three phase, 208 v output. An a-c exciter and silicon diodes replace the usual commutator and brushes. The free standing control provides complete starting and protective arrangements with metering and voltage regulator. CIRCLE 318 ON READER SERVICE CARD Solenoid a-c or d-c GUARDIAN ELECTRIC MFG. CO., 1621 W. Walnut St., Chicago 12, Ill. No. 28 midget solenoid is available a-c HERE'S WHY CENTRICORES ARE PROBABLY THE MOST CONSISTENTLY UNIFORM CORES YOU CAN BUY: The exceptional uniformity you get in tape-wound Centricores is not easy to come by. It's the result of painstaking precision at every stage of the manufacturing process—and, in fact, before manufacturing. Three principal factors help produce Centricore uniformity: Careful classification of materials—Raw alloys are first "pedigreed"—meticulously selected, then tested for some 14 parameters, and classified by magnetic properties. We're the largest buyer of nickel alloy magnetic materials in the world...which permits us to choose material for Centricores from an unusually wide distribution of magnetic properties. Special winding machines—We build our own machines, to die-making tolerances, for winding magnetic alloy tape into cores. We also build our own machines for applying insulating coating to the tape. These machines give us far greater uniformity in dimensions, insulation and ultimate performance of Centricores. Closely-controlled annealing—Annealing—perhaps the most critical phase of the core-making process—is done under precisely regulated atmospheric and temperature stabilized conditions to hold Centricore magnetic performance to uniformly high levels. Exceptional uniformity from core to core and lot to lot is further assured with Super Squaremu "79", a new high-performance alloy we've developed. It has outstanding magnetic qualities and is remarkably uniform in squareness, thermal stability and gain. Super Squaremu "79" offers an effective solution to problems of variation in magnetic performance. WRITE FOR BULLETIN C-3 \| SIZE \| MATERIAL \| THICKNESS \| \|------\|--------------------------------\|-----------\| \| 1 \| HIGH NICKEL \| .001* \| \| \| Hymu 80 \| \| \| \| Squaremu 79 \| \| \| \| Super Squaremu 79 \| \| \| THRU \| LOW NICKEL \| \| \| \| Squaremu 49 \| \| \| \| Carpenter 49 \| \| \| 225 \| GRAIN-ORIENTED SILICON \| .004* \| \| \| Crystaligned \| \| \| \| Microsil \| \| Special sizes, shapes and thicknesses quoted on request. NEW! LOW-COST D-C POWER SUPPLIES Standard output voltages from 3 to 1000 VDC at 30% intervals... Maximum powers of 50, 100, 200, 400, 750, 1500, 3000** watts These new Sorensen MD supplies form one of the most comprehensive and economical power supply series on the market. More than 130 catalog models to choose from, providing 20 output voltages in the range from 3 to 1000 vdc, inclusive. Sorensen engineers welcome the opportunity to develop non-catalog models for unusual requirements. Simple, rugged design features magnetic voltage regulator to obtain ±1% regulation against input line variations plus low-impedance silicon rectifier for good load regulation. (Typical regulation from 50% load to full load is 2% to 10%, depending on load current rating. Additional data can be supplied upon request.) Ripple: 1% rms max. (Some units can be supplied with 0.5% max. ripple.) Dependable, tubeless construction. All parts are conservatively rated for continuous duty. Units will withstand output short circuits without damage to components. 19-inch rack-panel mounting for all units simplifies application in lab or custom-built equipment. Rugged MD series supplies are just one example of the outstanding power-supply models offered by Sorensen. Sorensen controlled power equipment, with the widest line, enables you to make the wisest selection. Included are: regulated d-c supplies, regulated a-c supplies, variable frequency power sources (frequency changers; for example, 60 to 400 cps), high voltage supplies (to 600 kv, ac or dc), and miniature converters and inverters. Available in an extremely wide variety of input-output combinations. Write for complete specs. Sorensen & Company, Richards Ave., South Norwalk, Conn. Voltage range, these sizes: 6.3 to 1000 vdc Voltage range, this size: 12 to 1000 vdc Analyzer 256-channel Nuclear Data, Inc., 145 No. Washington, Wheaton, Ill., announces a completely solid state 256-channel analyzer with ferrite core memory. Analyzer weighs 37 lb. is 1 cu ft in size. Circuits float in foam rubber in stainless steel cabinet. Power required for portable operation on batteries is 9½ w. Performance is said to exceed vacuum tube models. Live time is printed out. Logarithmic and linear analog outputs. Pressure Switch subminiature The Bristol Co., Waterbury 20, Conn. Adjustable pressure switch exceeds requirements of MIL-E-005272B for performance up to 200 psi under vibration, shock, and acceleration test conditions. Volume of unit is less than 1 cu in. and weight slightly over 1 oz. A Ni-Span C pressure capsule and a This NEW resistance soldering tool practically eliminates tip replacement...provides void-free solder joints. ...MEETS CRITICAL DEFENSE STANDARDS FOR SOLDERJOINT RELIABILITY, QUALITY For aircraft and missile applications—where solder-joint reliability is a precious must—General Electric's new Resistance Soldering Tool provides the void-free solder joints necessary for reliable, high-quality connections. This new tool is particularly useful for soldering multiple-prong plugs. It heats work evenly throughout, allowing complete solder melting in only one operation. Call your nearby G-E Apparatus Sales Office or write for Bulletin GEA-6588, General Electric Company, Schenectady 5, N. Y. ZIRCONIUM-TUNGSTEN TIP does not wear down, does not require periodic redressing or replacement; heats work evenly throughout. INTERNAL SWITCH is hand-operated, requiring no foot pedals or buttons; provides positive heat control; eliminates tip arcing. Model HFM Multicouplers facilitate the operation of from four to 16 receivers from one antenna. They consist of a broadband, low noise, high gain amplifier followed by a passive multicoupler distribution system. The multicoupler can be used in any type of communications system where it is necessary to operate two or more receivers from one antenna while maintaining maximum isolation between receivers. TYPICAL PERFORMANCE CHARACTERISTICS \| Model \| HFM-6(AC)-326 \| HFM-12(A)-50110 \| \|----------------\|---------------\|-----------------\| \| Frequency Range\| 30—265 MC \| 500—1100 MC \| \| Gain \| > 10 DB \| > 10 DB \| \| Noise Figure \| 6 to 8 DB \| 7 to 11 DB \| \| Number of Outlets \| From 4 to 16 \| From 4 to 16 \| \| Isolation* \| > 30 DB \| > 30 DB \| \| Peak to Valley Ratio \| ±1.5 DB over band \| ±2 DB \| Input and Output Impedance: 50 ohms VSWR Input: 1.5 VSWR Output: 1.75 Power Requirements: 105-125 VAC 60 CPS To customer requirements. Effective component protection is hard to supply under conditions of violent acceleration, high ambient temperature, and vicious vibration. But in military electronic gear, transistors must get unfailing protection against these threats to reliable operation. They get it, most fully, with atlee mounting clips.* atlee clips are provably better in three ways: HOLDING POWER. Under severe shock and vibration, these clips actually mold themselves tighter to the transistors. There's no visible shifting or twisting, no lead-breaking resonance, and the dislodging force actually increases. COOLING EFFICIENCY. With atlee clips, this approaches to within 10% of “infinity” — the ideal derating curve for a transistor with an infinite heat sink which keeps the case temperature from rising above the ambient level. ELECTRICAL INSULATION. When required, these clips can be coated with Dalcoat B — an exclusive high-dielectric enamel that has twice the dielectric strength of Teflon but conducts heat as well as mica. There are still more reasons why engineers who seek perfection choose atlee transistor clips. They know that Atlas E-E is the pioneering company in the development of component holders of all types, with unequalled years of specialized experience, and a complete line of clips for all case sizes and mounting requirements. They have learned it costs no more to get the best . . . and that Atlas E-E makes these “little things” as though they were the biggest things in the circuit. DESIGN FOR RELIABILITY WITH atlee — a complete line of superior heat-dissipating holders and shields, plus the experience and skill to help you solve unusual problems of holding and cooling electronic components. atlee corporation (Fomerly Atlas E-E Corporation) 47 PROSPECT STREET, WOBURN, MASSACHUSETTS snap-action switch are contained in an all stainless steel housing. The pressure-setting adjustment is simple, and the setting is positive, even under severe shock and vibration conditions. CIRCLE 321 ON READER SERVICE CARD Stepping Motor high precision G. H. LELAND, INC., 123 Webster St., Dayton 2, Ohio. The Syncramental motor is used to rotate potentiometers, counters, rotary switches, tape advances and various control mechanisms. A special clutch mechanism replaces the use of ratchets, permitting the translation of pulses to incremental shaft positions with 99.9999 percent accuracy and dependability. Motor is available in unidirectional and bidirectional models, ranging in weight from 4 to 13 oz, with angular increments of 36 deg per pulse, up to 15 steps per sec, and load capacity up to 2 lb-in. starting torque. CIRCLE 322 ON READER SERVICE CARD Radiators for semiconductors WAKEFIELD ENGINEERING, INC., 11 Broadway, Wakefield, Mass. Range of the R-5000 series is 5 w to 100 w dissipation ratings, with forced convection. Illustrated are: R-5010 with rectifier, R-5020 with Zener diode, and R-5030 with transistor. With forced convection applied to the R-5030, for example, a transistor nominally rated by its manuDesigned for Application INSTRUMENTATION OSCILLOSCOPE One Inch Miniaturized basic packaged panel mounting Cathode Ray Oscilloscope for instrumentation use replacing "Painter Type" meters. Panel bezel matches 2" square meter. No. 90901 uses ICPI tube. No. 90911 uses IEP1 tube. Power supply No. 90202 available where application requires. JAMES MILLEN MFG. CO., INC. MALDEN MASSACHUSETTS CIRCLE 206 ON READER SERVICE CARD HOLD TIGHT! Even under severe conditions of shock and vibration, CAMBION® Vibration-Proof Battery Holders hold tight. Unique "locking strap" keeps "D" size mercury battery securely in place. Suitable for use in all types of circuits, the holders are available in two models. No. 2570 is insulated at one end only. No. 2870 is insulated at both ends to permit "floating" of the voltage supply, and is ideal for transistorized circuits and other low-voltage applications. Both holders are designed so that leads can be easily brought up to terminals even though unit is flush-mounted. Write Cambridge Thermionic Corporation, 437 Concord Avenue, Cambridge 38, Mass., for full details on these and other products in the wide line of CAMBION® The guaranteed electronic components CIRCLE 207 ON READER SERVICE CARD I wish I knew what Power Sources was up to. POWER SOURCES, INC. Burlington, Massachusetts CIRCLE 153 ON READER SERVICE CARD NEW SILICON "PILL" VARACTOR GREATLY REDUCES THE PACKAGE AS A FACTOR IN CIRCUIT DESIGN specifically developed for: - amplifiers at the higher microwave frequencies (1000 mc and above) - travelling wave parametric amplifiers - microwave computers as sub-harmonic generators - amplifiers in which stray susceptance effects must be minimized - applications of varactors to stripline circuits - modulators for frequency synthesis \| Experimental quantities are available with these nominal specifications \| TYPE NUMBER \| CAPACITANCE TOLERANCE (Zero Bias) \| TYPICAL Q AT -6 VOLTS \| \|------------------------------------------------------------------------\|-------------\|-----------------------------------\|-----------------------\| \| MA-4255X \| 0.5-1.4 μf \| 60-80 \| \| MA-4256X \| 1.2-2.5 μf \| 50 \| \| MA-4257X \| 2.5-4.0 μf \| 30 \| Package shunt capacitance ~ 0.2 μf. Series lead inductance <10⁻³ henries. Write or call: MICROWAVE ASSOCIATES INC. BURLINGTON, MASSACHUSETTS BRowning 2-3000 • TWX Burlington, Mass. 942. CIRCLE 323 ON READER SERVICE CARD Shift Register transistorized NAVIGATION COMPUTER CORP., 1621 Snyder Ave., Philadelphia 45, Pa. Model 308 is a 5-stage transistorized shift register capable of operating at 300 kc. Set and reset inputs, and ONE and ZERO outputs are available for each flip-flop stage. Units may be serially cascaded to assemble a shift register of any length. A common buss is provided to reset all stages simultaneously. Model 308 is fabricated on a 5 in. by 6 in. glass-epoxy p-c card ½ in. thick, and is used with an 18 pin p-c receptacle. Only one voltage, −12 v. is required. Standard output levels are −6.8 v for ONE and −0.2 v for ZERO. CIRCLE 324 ON READER SERVICE CARD Rotary Trimmer super compact SUBMINIATURE INSTRUMENTS CORP., 3705 Sunnyside Drive, Riverside, Calif. "Trimquate" trimmers are easily mounted in two planes without additional brackets. They have a shaft torque of 3 oz in., friction clutch to assure precise setting, are adjustable from either side. Self phasing segmented units can be easily ganged. Three types of housing are available—aluminum, aluminum and molded, and stainless steel. A newly developed method of applying installation material to metal housing provides an inexpensive easily controlled procedure of application which yields a very high dielectric strength insulation. Process is used on both aluminum and stainless steel. CIRCLE 325 ON READER SERVICE CARD Servo Amplifier subminiaturized M. TEN BOSCH, INC., Pleasantville, N. Y. Model 1800-0900 is a potted, hermetically sealed, plug-in transistor servo amplifier. It is primarily intended to receive signals from a synchro control transformer and to operate a 400 cycle, 3.1 w per phase servo motor or equivalent. The amplifier is designed to meet the environmental requirements of MIL-E-5400. CIRCLE 326 ON READER SERVICE CARD Pressure Transducer high g level WHITE AVIONICS CORP., Terminal Road, Plainview, L. I., N. Y., offers to control, telemetry, and propulsion pressure measurement systems designers a potentiometric output type transducer capable of performance under high g level environment (to 50 g's at 3 kc) and yielding infinite resolution. ...and now for the sealing test! If the pots you need must function in a dust or sand environment, you could build 'em yourself to make sure they stay clean! But before you move heaven and earth while testing your creation, exactly what have you planned, to give you a tight seal, yet low torque? And if that isn't enough of a problem, how do you keep foreign matter out of the bearings? But why move heaven and earth, mostly earth, to test your own dirt-free pot, when Ace has the pots with the dust-free features? Special O-rings seal sand, dust and other foreign matter eliminating abrasion damage. Our wound nylon packing delivers excellent sealing with lowest torque. Also, a special silicone-type grease, located in shaft pockets, captures foreign particles before they ever get a chance to do any damage. So if grit's a problem for you, come to Ace for the answer. See your ACErep! This 5" AIA Acepot (shown 1/3-scale), meeting all MIL spec's on sealing, incorporates these exclusive anti-dirt and dirt-trapping features. Mandrels are also fungicide-varnished, to insure long life. ELECTRONICS · FEBRUARY 12, 1960 The triple-threat pot line Pick the pot to suit your system... be it esoteric or plebeian... from the triumvirates of HELIPOT single-turns (1/2" to 3" dia.) or multi-turns (7/8" to 3-5/16" dia.). No need to overspecify... pay only for what you need! Par exemple? The three HELIPOT 17/16" single-turns, each the leader in its own milieu: Series 5400 with plastic case to 80°C all-metal Series 5410 to 125°C all-metal Series 5420 to 150°C Pay only for what you need in single-turn and multi-turn precision pots. You'll find your circuit solution... for a system hot or cold... in Data File A-72. Beckman® Helipot® Helipot Division of Beckman Instruments, Inc. Fullerton, California Engineering representatives in 29 cities potentiometers: dials: delay lines: expanded scale meters: servomotors: breadboard parts 100 is available for pressure ranges 0-15 to 0-300 psi for corrosive and noncorrosive liquids and gases in a configuration 1½ by 1½ by 1½ in., weighing 6 oz or less. CIRCLE 327 ON READER SERVICE CARD Bolometer Bridge for microwave use N. V. PHILIPS' GLOEILAMPENFABRIEKEN, Eindhoven, Nederland. Type PP4460 bolometer bridge is designed for low power measurements of microwave fields. It operates on the principle of a change of resistance being produced in a thermistor, when exposed to a uhf power field. The thermistor is part of the feedback circuit of an oscillator, the output of which is inversely proportional to the microwave power striking the thermistor. Power is measured in three ranges: 0-0.1; 0-1 and 0-10 mw with an accuracy of better than 5 percent at full scale deflection. The meter is as well calibrated in db (0 db = 1 mw). CIRCLE 328 ON READER SERVICE CARD Preset Counters transistorized OXFORD ENGINEERING Co., 47A River St., Wellesley Hills 81, Mass. The 2000 series preset counters are available from 3 to 6 digits, and from 1 to 6 preset banks. Illustrated is the model 2044, 4-digit and 4-bank counter. Some features include: all electronic circuitry; solid state power supply; plug-in output Lepel HIGH FREQUENCY INDUCTION HEATING EQUIPMENT For Hardening • Annealing • Soldering Brazing • Zone Refining • Crystal Growing ELECTRONIC TUBE GENERATORS: 1 kw; 2½ kw; 5 kw; 10 kw; 20 kw; 30 kw; 50 kw; 75 kw; 100 kw. SPARK GAP CONVERTERS: 2 kw; 4 kw; 7½ kw; 15 kw; 30 kw. WRITE FOR THE NEW LEPEL CATALOG 36 Illustrated pages of valuable information. LEPEL HIGH FREQUENCY LABORATORIES, INC. 55th ST. & 37th AVE., WOODSIDE 77, N.Y. CIRCLE 208 ON READER SERVICE CARD ...IT GLOWS when the FUSE BLOWS! NEW INDICATING 3AG FUSE POSTS EXAMINE THESE FEATURES 1. New patented knob design to assure high degree of illumination for instant blown fuse indication. 2. Positive finger grip for knob extraction. 3. Quick service bayonet lock. 4. Constant tension beryllium copper coil & leaf spring for positive contact & lower millivolt drop. 5. Optional—at extra cost—neoprene “O” ring to assure splash-proof feature. 6. New high degree vacuum neon lamp for greater brilliance & visibility. 7. Impact black phenolic material in accordance with MIL-M-14E type CFG. 8. One piece brass hot tin dipped non-turning bottom terminal. 9. Double flats on body to permit mounting versatility. ACTUAL SIZE SPECIFICATIONS: \| PART # \| VOLTAGE RANGE \| \|--------\|---------------\| \| 344016 \| 2 1/4 - 7 volts \| \| 344012 \| 7 - 16 volts \| \| 344024 \| 16 - 32 volts \| \| 344125 \| 90 - 125 volts \| \| 344250 \| 200 - 250 volts \| Maximum current rating 20 amps. PHYSICAL CHARACTERISTICS—Overall length 2 3/8" with fuse inserted • Front of panel length 1 3/16" • Back of panel length 1 1/16" • Panel area front 1 3/16" dia. • Panel area back 1 3/16" dia. • Mounting hole size (D hole) 3/8" dia. flat at one side. TERMINAL—Side—one piece, .025 brass—electro-tin plated • Bottom—one piece, lead free brass, hot tin dipped. KNOB—High temperature styrene (amber with incandescent bulbs—2 1/2 thru 32 volts—and clear with high degree vacuum neon bulbs—90 thru 250 volts) • Extractor Method—Bayonet, spring grip in cap. HARDWARE—Hexagon nut—steel, zinc cronak or zinc iridite finish • Interlock lock washer—steel, cadmium plated • Oil resistant rubber washer. MILITARY SPECIFICATIONS—MIL-M-14E type CFG. Fungus treatment available upon request per Jan-T-152 & Jan-C-173. TORQUE—Unit will withstand 15 inch lbs. mounting torque. LITTLEFUSE DES PLAINES, ILLINOIS CIRCLE 159 ON READER SERVICE CARD TIC Precision Servo Potentiometers have all 13 features Your Assurance of Superior System Performance A few of the many applications of TIC Precision Servo Potentiometers are as input-output transducers in servo systems for airborne navigation and flight control, fire control, fuel control, shipboard gun directors, missile aiming and flight control, analog computing, air traffic control and telemetering. TIC Precision Servo Potentiometers are available in 21 types with diameters from $\frac{1}{2}$" to 3", giving design engineers a wide range from which to select. Included are single and multi-turn types with either wirewound or infinite resolution metallic film resistance elements, as well as types designed for ganging without a shaft. And TIC Precision Servo Potentiometers are engineered to withstand the severe environmental conditions imposed by military equipment operation. Test Set semiautomatic Peschel Electronics, Inc., R.F.D. No. 1, Patterson, N.Y. Semiautomatic test set features integral construction for nondestructive hipot testing a group of components or multiconductor cable. The table top cabinet houses a 0-10 kv a-c/d-c sensitive hipot tester, a central panel for control of testing sequence, and a heavy lucite interlocked door leading to a large compartment for the group of components to be tested. The electronic fault relay is adjustable in six steps from 5 to 5,000 $\mu$A. Dual scale panel meters show voltage and leakage current on each component being tested. Cabinet measures 22 in. wide by 36 in. high by 15 in. deep. Miniature Filters for telemetering Kelvin Electric Co., 5907 Noble Ave., Van Nuys, Calif. Series F-185 telemetering band-pass filters meet MIL-T-27A for long term stability and reliability under adverse en Kodak Recognizes... Proper Color Control Requires PRECISE Timing ...Just one of Countless Applications of SPECIAL TIMERS by Standard What's your timing need? For precise printer time measurements—as used with Kodak IV-C and 5S Color Printers? Accurate test timing (to tolerances of ±.001 seconds)—as with numerous electronics and missile manufacturers? For these and countless other applications, STANDARD stands ready to develop the exact elapsed time indicator to meet the most stringent requirements. Inquire today. Request Catalog No. 198A covering the full line of Standard Precision Timers ...portable or panel mounted. THE STANDARD ELECTRIC TIME COMPANY 89 LOGAN STREET • SPRINGFIELD, MASSACHUSETTS "Splitting the Split Second... Precisely" Environmental conditions. All channels (23) from 400 cps to 70 kc are available. Series measures $\frac{3}{4}$ by $1\frac{1}{2}$ by $2\frac{1}{4}$ in., channels 1 through 6; $\frac{3}{4}$ by $\frac{3}{4}$ by $1\frac{1}{8}$ in., channels 7 through 18 and A through E. Insertion loss is less than 6 db; relative attenuation is less than 3 db in pass-band, greater than 15 db at adjacent channel edges; impedance range 1K to 100K ohms, in and out; operating level, 10 v max. CIRCLE 331 ON READER SERVICE CARD Solenoid Driver transistorized COMPUTER CONTROL CO., INC., 983 Concord St., Framingham, Mass. Model SD-102 contains two independent solenoid driver circuits. Each circuit amplifies the output of a standard M-PAC to drive a solenoid load of up to 50 v at 150 ma. A diode in each circuit suppresses inductive kick. Power requirements: +20 v at 12 ma, -90 v at 4 ma. Input requirements: approximately 0.5 ma into a 30 K input impedance. CIRCLE 332 ON READER SERVICE CARD Motor Generator small, lightweight KEARFOTT CO., INC., 1500 Main Ave., Clifton, N. J. The M863-05 integrating motor generator is only $2\frac{1}{2}$ in. long, is thermistor compensated, and yields an output of 0.5 v per 1,000 rpm. Having a linearity of but 0.06 percent through the 0-3600 rpm range, the component is ideally qualified for missile applications where space and weight considerations are critical. It weighs 3.9 oz, and can operate through a temperature range of -54 C to +105 C. CIRCLE 333 ON READER SERVICE CARD Time Analyzer 32-channel ELDORADO ELECTRONICS, 2821 Tenth St., Berkeley 10, Calif. System 0900 time analyzer accurately counts and stores digital information in 32 consecutively gated channels. Channel scalers have a double pulse resolution of 1 μsec. Channel widths are variable from 1 μsec to 0.08 sec. They can be identical or weighted with last channels up to 8 times the width of the first channel. Each of the 32 channels has two electronic counting units and a 4-digit mechanical register to provide storage of $10^6$ digits. Each channel accurately counts pulses beginning no more than 1 μsec apart at rates to 2,500 counts per sec. CIRCLE 334 ON READER SERVICE CARD Modulation Monitor self-calibrating GATES RADIO CO., Quincy, Ill. Model 5693 broadcast modulation monitor will read the time values of positive and negative peaks regardless of the presence of carrier shift. With the fastest meter allowable, it will ASTRON SOLID TANTALUM CAPACITORS stability COMPARE IN CAPACITANCE STABILITY WITH QUALITY PAPER CAPACITORS The temperature coefficient of capacitance of Astron Tantalum Solid Electrolyte Capacitors is typically less than 500 ppm/°C, and production capacitance tolerances are available as close as ±5%. The estimated shelf life is 20 years. - 125° C operation - Rugged construction - Capacitance stability - Subminiature - Dry, solid construction - Meets MIL specifications FOR COMPLETE INFORMATION WRITE TODD FORD, CIRCUIT E-675A AND FOR ASTRON'S DESIGN ENGINEER PUBLICATION TECHNIQUES, VOL. 59, NO. 2 ASTRON CORPORATION 255 Grant Avenue East Newark, New Jersey SPECIALISTS IN CAPACITOR MINIATURIZATION CIRCLE 163 ON READER SERVICE CARD Now...from Sonotone—4 Big Improvements in the quality stereo cartridge Sonotone 8TA cartridge replaces 8T as industry standard The new Sonotone 8TA cartridge gives greater than ever stereo performance...has 4 big extras: - fuller, smoother frequency response - higher compliance than ever before - lighter tracking pressure - practically eliminates dust pile-up ONLY $14.50 Sonotone 10T unitized stereo at lowest price ever New 10T cartridge sells at record low price of $6.45.* And it covers the complete high fidelity range. 10T's unitized construction makes it easiest to install, easiest to replace. Low price means more sales—more profits. SPECIFICATIONS \| \| 8TA \| 10T \| \|----------------\|----------------------\|----------------------\| \| Frequency Response \| Smooth 20 to 20,000 cycles. Flat to 15,000 with gradual roll-off beyond. \| Flat from 20 to 15,000 cycles ± 2.5 db. \| \| Channel Isolation \| 13 decibels \| 18 decibels \| \| Compliance \| 3.0 x 10⁻⁶ cm/dyne \| 1.5 x 10⁻⁶ cm/dyne \| \| Tracking Pressure \| 3-5 grams in professional arms. 4-6 grams in changers \| 5-7 grams \| \| Output Voltage \| 0.3 volt \| 0.5 volt \| \| Cartridge Weight \| 7.5 grams \| 2.8 grams \| \| Recommended Load \| 1-5 megohms \| 1-5 megohms \| \| Stylus \| Dual jewel tips, sapphire or diamond. \| Dual jewel tips, sapphire or diamond. \| including mounting brackets Sonotone makes only 6 basic ceramic cartridge models...yet has sold over 9 million units...used in over 662 different phonograph models. For finest performance, replace worn needles with genuine Sonotone needles. Silicon Rectifier controlled output SOLID STATE PRODUCTS, INC., One Pingree St., Salem, Mass. Precise power control of loads up to 300 w with extremely low losses can be achieved reliably with a new line of miniature silicon controlled rectifiers. At 100 C these units control up to 1 ampere (continuous) per cell with an input signal level of only 2 ma. Switching efficiency on the order of 98 percent is typical. At 2 amperes the maximum drop is 2.5 v. The junction is ruggedly mounted on high alumina ceramic, positively insulating all active elements from the case. Minimum heat sink requirements with peak recurrent ratings to 30 amperes are afforded with very low internal dissipation, operating over a temperature range of −65 C to +50 C. Power Supplies airborne type SOUTHWESTERN INDUSTRIAL ELECTRONICS Co., 10201 Westheimer Rd., Houston 27, Texas, announces models TPC-18A and 19A airborne transistorized power supplies. They are designed for direct, plug-in reUPGRADE YOUR EQUIPMENT with C-A-C MIL. SPEC. LAMINATED TRANSFORMERS and REACTORS FILTER REACTORS FILAMENT TRANSFORMERS CONVERTER TRANSFORMERS Because we custom design and build to your needs, we welcome the difficult assignments as well as the more common requirements. Quality—reliability are plus factors when you specify C-A-C. COMMUNICATION ACCESSORIES COMPANY LEE'S SUMMIT, MISSOURI The Sphere Co., Inc., 25 Amity St., Little Falls, N. J. New line of terminals has an outer insulator of solid Teflon. Features: extremely high dielectric strength, insulation resistance; non-carbon tracking; high resistance to temperature extremes; tough physical characteristics, abrasion and chemical resistance; extremely low moisture absorption. They range in size from $ \frac{1}{4} $ in. to $ \frac{3}{8} $ in. diameter. Head styles are turret, slotted, double and single lug. Terminals range in height from $ \frac{1}{4} $ in. above the mounting panel to $ \frac{3}{8} $ in. CIRCLE 338 ON READER SERVICE CARD Rotary Pots subminiature Subminiature Instruments Corp., 3705 Sunnyside Dr., Riverside, Calif. "Turnquate" precision pot features include: subminiature size, humidity sealing, dual outputs or two circuits in this tiny size, 10 or 20 output circuits in 2$ \frac{1}{2} $ in. long, 200 C stainless steel high temperature units, 200 K resistance, 40 g vibration, self phasing without placement of D-10A dynamotors as the power supply for aircraft communications and navigation receivers, using a transistor multivibrator circuit to deliver voltage at high efficiency with good regulation and provide protection against overload or short-circuit. CIRCLE 337 ON READER SERVICE CARD KLYSTRON POWER SUPPLY with a MEMORY No more fiddling with reflector voltage adjustments when you switch between cw and square wave...because of just one of the typically advanced features of this low-medium voltage Klystron Power Supply. Ever double-mode a Klystron? Not with our model 809! Again, the thoughtful engineering that goes into every PRD product assures the user of self-protection against errors. Even little things like the built-in beam voltage and current meter prevent guessing and doubt during runs. And when it comes to 'scoping the Klystron...compare the CRT display of a tube powered by the 809 and you'll see for the first time what a really sharp trace looks like! For complete details, send for our data sheet F-10. P.S. In case you don't have our latest catalog, E-8...160 pages, chock-full of useful data...dash off a note on your company letterhead. Polytechnic Research & Development Co., Inc. 202 Tillary St., Brooklyn 1, N.Y. ULster 2-6800 CIRCLE 165 ON READER SERVICE CARD FOR SUPER-FINE CUTTING OF HARD, BRITTLE MATERIALS... We don't recommend slicing up the family's fine Limoge China, but this does illustrate the precisely controlled cutting action of the S. S. White Airbrasive Unit. Note how clean the edge is, and how the delicate ceramic decoration is unharmed. The secret of the Airbrasive is an accurate stream of non-toxic abrasive, gas-propelled through a small, easy-to-use nozzle. The result is a completely cool and shockless cutting or abrading of even the most fragile hard materials. Airbrasive has amazing flexibility of operation in the lab or on an automated production line. Use the same tool to frost a large area or to make a cut as fine as .008"!...printed circuits...shaping and drilling of germanium and other crystals...deburring fine needles...cleaning off oxide coatings...wire-stripping potentiometers...engraving glass, minerals, ceramics. Jobs that were previously thought impossible are now being done. Send us samples and specs on your difficult jobs and let us test them for you. SEND FOR BULLETIN 5705A ...complete information S. S. WHITE INDUSTRIAL DIVISION • Dept. EU • 10 East 40th Street, New York 16, N. Y. Exclusive representatives for Arizona and California • WEIGHTMAN AND ASSOCIATES, Burbank, Calif. loosening screws or clamp bands, superior noise level with new principle in design, improved moment of inertia due to compact design, less torque values with seals. CIRCLE 339 ON READER SERVICE CARD Toroidal inductors center tapped UNITED TRANSFORMER CORP., 150 Varick St., New York 13, N. Y. Type TQA precision inductors provide an ideal solution to stable oscillators for frequencies from 400 cycles to 75 kc. They are center tapped for oscillator circuits and employ an extremely stabilized core for maximum temperature stability. Units are available as stock items in 19 inductance values ranging from 7 mhy to 22 henries, laboratory adjusted to 1 percent accuracy. Maximum Q is approximately 160 at 7.5 kc ranging down to 20 at 400 cycles and to approximately 30 at 75 kc for low inductance values. CIRCLE 340 ON READER SERVICE CARD Wire Tester abrasion-scrape PESCHEL ELECTRONICS, INC., R.F.D. No. 1, Patterson, N. Y. Conforming to MIL-W-19583 (Navy) requirement to determine abrasion resistance of film-insulated high-temperature magnet wire, model TS-2 abrasion-scrape tester is a device that repeatedly scrapes the wire with the cylindrical surface of a No. 11 steel needle. Length of the scrape motion in one direction is 3/8 in. The device is equipped with an electrical circuit providing 12 v at approximately 5 ma between the needle and the wire sample. Circuit is so designed that failure is detected and the device shut off when the film is worn through to the copper for approximately 1/4 of the stroke length. CIRCLE 341 ON READER SERVICE CARD Microwave Amplifier general-purpose ALFRED ELECTRONICS, 897 Commercial St., Palo Alto, Calif. Model 549 offers broad band amplification with gain of 30 db and 10 mw output from 10.5 to 16 mc. Use of a twt, permanently magnet-focused, also provides versatile modulation features. Phase modulation, which is used in Doppler shift simulators and Doppler radars, may be accomplished simply through a front panel connector which is capacitively coupled to the twt helix. Simple, rugged design, plus use of quality components in a conservative circuit, assures virtually troublefree operation. CIRCLE 342 ON READER SERVICE CARD Converter 4 decimal digit EPSCO, INC., 275 Massachusetts Ave., Cambridge, Mass. Model MTD-704 is a 4 decimal digit voltage-to-digital converter. It translates input analog voltages into 4 binary-coded decimal digits, plus sign and overflow digits. Input full-scale voltage range is ± 10 v d-c. ARMY'S MOST ACCURATE SURFACE-TO-SURFACE MISSILE - automatically normalizes current in DC teleprinter signal loops - eliminates metering and manual adjustments - effects savings in maintenance costs - may be used on polar or neutral DC circuits - requires no supplemental power supply TRANSISTORIZED LOOP CURRENT CONTROL Type 238: Model 1, for 60 ma DC loops; Model 2, for 20 ma DC loops Pace-Setters in Quality Communication Equipment NORTHERN RADIO COMPANY, inc. 147 WEST 22nd ST. NEW YORK 11, NEW YORK In Canada: Northern Radio Mfg. Co., Ltd. 1950 Bank St. Billings Bridge, Ottawa, Ontario. Write on your letterhead for free literature to Dept. E-2. ELECTRONICS · FEBRUARY 12, 1960 CIRCLE 167 ON READER SERVICE CARD BH100 Series MILLI-V-METER™ The instrument with the TAPE-SLIDewire for Test Cells Ground Support Flight Deck Telemetry ...to accurately indicate TEMPERATURE R.P.M. PRESSURE FLOW 1/10 THE SIZE, 10 TIMES THE ACCURACY (of standard instruments)! These 3x5" and 3"-dia. instruments are used by many manufacturers and laboratories to measure every parameter...in missile, aircraft or ground support. Ideally small and matched for panel uniformity, they offer reading of the individual phenomenon or of many states through selector switching...with fast response. Every measurement exhibited on the counter is a definitely calibrated value. Component density of the 3"-dia. BH183 is the highest obtainable combined with ultimate instrument accuracy. Manufactured to MIL-E-5272 and MIL-I-6181 specs. Produced by the makers of JETCAL® jet engine Analyzer...in worldwide military and airline use! Full information is available for the asking! B & H INSTRUMENT CO., INC. 3479 West Vickery Blvd., Fort Worth 7, Texas FEATURES: 1) Accuracy 1 part in 1000. 2) Laboratory precision for the military or industry. 3) Compatibility with any transducer—AC or DC. 4) For strain gage, linear differential transformer, thermocouple, thermistor, resistance thermometer, pulse or variable frequency circuits or systems. 5) Available with re-transmitting slidewire. 6) Every scale unit a calibrated value. 7) Operates directly from 60- or 400-cycle power. Integrated Test Unit for lab use RESEARCH COUNCIL INC., 1062 Main St., Waltham 54, Mass. The Thermion integrated test unit (Thermion I.T.U.), model B210 combines in one unit the functions of: (1) supplying variable power to up to several Thermions, and (2) monitoring and selecting the internal test points in any of the Thermions under evaluation. Among the many test conveniences available, the unit incorporates techniques of design and construction to assure accurate performance of temperature measurements. The effects of ambient temperature variations are eliminated and undesirable stray thermocouples avoided. Bandpass Filter high shape factor CONTROL ELECTRONICS Co., INC., 10 Stepar Place, Huntington Station, L. I., N. Y. The BF-121 high shape factor bandpass filter is designed for video i-f strip applications. It has a center frequency of 40 kc, a 1-db bandwidth of 3.4 kc and a 50-db bandwidth of 22 kc. It has a shape factor of 6.5. Input impedance is 51K ohms, and the output is to the grid of the next strip stage. Special high voltage 1,900 v d-c terminals are employed as a safety factor. Case size is 1½ by 2 by 4 in. high. CIRCLE 345 ON READER SERVICE CARD Terminal Boards 3/32 or 1/8 in. thick CAMBRIDGE THERMIONIC CORP., 445 Concord Ave., Cambridge 38, Mass. All-set terminal boards with Cambion No. 1010 castelated terminals are now available in 13¼ in. lengths and widths of ½, 1½, 2, 2½ and 3 in. One-half-inch wide board has single row of 25 No. 1010 terminals; all others, double rows of 25 terminals each. Terminals are mounted on ¾ in. centers. All boards are sectioned into five 2½ in. sections for convenient breaking into suitable lengths. Each section is drilled for 14 terminals with 10 mounted. Hole size is 0.120 No. 31 drill. Terminals are brass per QQ-B-626a, furnished in 6 shank lengths from ¼ in. to ¾ in. CIRCLE 346 ON READER SERVICE CARD Accelerometer remote indicating M. TEN BOSCH, INC., Pleasantville, N. Y. Model 0013-5100 remote indicating accelerometer is a highly accurate instrument designed for automatic aircraft and missile control. CIRCLE 347 ON READER SERVICE CARD FIRST PRACTICAL RADIOISOTOPIC FUELED GENERATOR FROM THE DESIGN LABORATORIES OF Gyra Precise control of voltages Superior voltage regulation at competitive prices incorporating a new method of control, the result of 25 years of design work by the staff of Gyra Electronics Corporation. All designs incorporate the Bucknam circuit for superior long-term regulation. Tubes are accessible from the rear with removing the instrument from a standard relay rack. Cabinet panels are available at extra cost. MODEL V-200. 0 to ±2000 volts, continuously variable, with up to 5 MA output in a single chassis. MODEL V-201. ±300 to ±2000 volts, continuously variable with 5 MA output. MODEL V-202. 0 to ±2000 volts, continuously variable, with up to 5 MA output. MODEL V-206. Dual 0 to ±2000 volts, continuously variable, 5 MA output. Identical to having two Model V-202 units in single 3½ inch chassis. MODEL V-104. 0 to ±500 volts, continuously variable, 1 MA output. Specially designed for fission chamber control. Optional design allows 0 to ±800 volts, 1 MA. MODEL V-106. —300 to +800 volts, and 0 to +100 volts, continuously variable, 1 MA output. Specially designed and dual polarity unit for compensated ion chambers. MODEL V-301. —500 to +3000 volts, continuously variable, with 5 MA output. Gyra Electronics also offers a line of fixed 300 volt power supplies with 175 MA output. Stability: 1 part in 10⁴ without oven; ±5 parts in 10⁴ with oven. Impedance: 1 ohm with meter, 0.001 ohm without meter. Four models available, write for prices and literature. \| OUTPUT CHARACTERISTICS \| V-104 \| V-106 \| V-200 \| V-201 \| V-202 \| V-206 \| V-301 \| \|------------------------\|-------\|-------\|-------\|-------\|-------\|-------\|-------\| \| Impedance, ohms \| 0.1 \| 0.1 \| 0.1 \| 0.1 \| 0.1 \| 0.1 \| 0.1 \| \| Ripple voltage \| 5MV \| 5MV \| 5MV \| 5MV \| 5MV \| 5MV \| 7MV \| \| Output change with 1 volt line change \| 0.01% \| Note A \| 500µV \| 10⁻⁶ \| 100µV \| 500µV \| 10⁻⁶ \| \| 60 cycle sola regulation \| Yes \| Yes \| No \| No \| Yes \| Yes \| No \| \| INPUT REQUIREMENTS \| \| \| \| \| \| \| \| \|------------------------\|-------\|-------\|-------\|-------\|-------\|-------\|-------\| \| Line Voltage \| 100 to 130 \| 105 to 125 \| 100 to 130 \| 105-125 \| \| Power consumed, watts \| 60 \| 60 \| 60 \| 60 \| 60 \| 70 \| \| PRICES, F.O.B. LaGrange, Ill. \| $500. \| $650. \| $325. \| $325. \| $650. \| $525. \| $400. \| Note A: Less than 5% ripple by (dv/dt) less than 1 mv per second. Gyra ELECTRONICS CORPORATION P.O. BOX 184 • LAGRANGE, ILL. • Fleetwood 4-4644 CIRCLE 169 ON READER SERVICE CARD WHEN YOU HAVE extraneous common mode signals AND WANT TO MEASURE 0.1 to 100 millivolts full scale AND THEN AMPLIFY CHOOSE THE NEW HONEYWELL D-C AMPLIFIER AccuData II wide-band differential all-transistor D-C Amplifier for strain gages and thermocouples - Full Scale Input: Unbalanced: ±100 μv to ±100 mv Differential: ±3 mv to ±100 mv Open Loop: Below drift level - Full Scale Output: ±2 v at 50 ma, dc to 10 kc - Frequency Response: to 20 kc - Output Impedance: Less than 0.5 ohm at dc on all ranges - Input Impedance: Unbalanced 3 to 100 mv ranges; greater than 20 megohms in parallel with 350 micromicrofarads. Differential: Greater than ± 2 megohms - Equivalent D-C Input Drift: Less than 2 μv/10°F ambient temp change on 0.1 to 30 mv input ranges - Equivalent Input Noise: 4μv peak-to-peak on 100 μv to 300 μv range (10-10 cps). 8μv rms on 10 to 30 mv ranges (10 to 100 kc) - Common Mode Rejection: 200,000 at 60 cps on 3 to 30mv ranges The new Honeywell AccuData II is a completely transistorized D-C Amplifier designed for use in high accuracy data handling systems as a wide-band pre-amplifier for strain gages and thermocouples. Its output can be fed to electronic or electromechanical analog-to-digital converters and simultaneously recorded on galvanometer oscillographs or magnetic tape. Either differential or single-ended input modes can be selected by an eleven position range switch. This switch changes the gain in three-to-one steps. Intermediate gains with high resolution are provided by a ten-turn potentiometer. Write for AccuData II Bulletin to Minneapolis-Honeywell, Dept. E-7, Boston Division, 40 Life Street, Boston 35, Mass. Transistor Pad precision molded The Delbert Blinn Co., P.O. Box 757, Pomona, Calif. New transistor pad provides an excellent mounting base for standard transistors. The spherical feet act as standoffs and provide the necessary lead length and air-space for thermal insulation when hand or dip soldering. The five holes will accept most combinations of lead wires. The transistor pads are precision molded of mineral-filled diallyl phthalate, which conforms to MIL-P-14D (MDG type) and Mil-P-4389. Nominal tensile strength is 5,500 psi. They resist continuous exposure to 400 F, and offer very high dielectric strength. X-Y Recorder integral input F. L. Moseley Co., 409 N. Fair Oaks Ave., Pasadena, Calif. Model 2D Autograf recorder has a new, integral a-c/d-c input and a built-in X-axis time base. It operates directly from a transducer and eliminates the need for an extra a-c converter. D-C ranges provide accuracy and resolution of better than 0.2 percent. Input range is 7.5 mv to 150 v on the X-axis; 5 mv to 100 v on the Y-axis. The X-axis time base (5 steps) is 7.5 to 750 sec. Input resistance is 200,000 ohms/v. The recorder provides a zero offset and a vacuum paper hold-down. Pen speed is 20 ips for each axis. CIRCLE 349 ON READER SERVICE CARD Digital Voltmeter 4-digit resolution Non-Linear Systems, Inc., Del Mar, Calif. The V64 is designed for a wide range of d-c measuring jobs and, with accessories, a-c and low-level d-c measurements. It features full 4-digit (0.01 percent) resolution, high input impedance and an average measuring time of 0.75 sec per reading. Its range without accessories is 500 v d-c in steps of ±9.999/99.99/500. Price is $825. CIRCLE 350 ON READER SERVICE CARD Crystal Mounts video detector American Electronic Laboratories, Inc., 121 N. Seventh St., Philadelphia 6, Pa. A new group of crystal video detector mounts provide a high tangential sensitivity down to 50 mc. The miniature octave band units utilize a slow wave structure to create a physically short line to match the crystal impedance to the input line. Ranges covered are 50 mc to 120 mc, 120 mc to 300 mc, BUILT BY MARTIN Missile Master FIRST FULLY OPERATIONAL ELECTRONIC AIR-DEFENSE SYSTEM Complete LACING CORD line designed by Electronic Specialists for ELECTRONICS Who knows better about lacing cord than the leader who has been designing and manufacturing electronic wire and custom cables for 38 years? - round cords and flat-braided tapes - nylon, dacron and teflon-fiberglass - finishes: micro-crystalline wax, synthetic rubber-like resin, teflon resin, synthetic rubber - meets industrial specifications and MIL-I-713A - each spool individually packaged in boxes Available for immediate delivery from your local Alpha Wire electronics parts distributor. Write for free Catalog I-59. ALPHA WIRE CORPORATION • 200 Varick Street, New York 14, N. Y. In the 13 Western States, call on our new Pacific Division: 1871 South Orange Drive Los Angeles 19, California CIRCLE 171 ON READER SERVICE CARD news about SPECIAL ISSUES! I.R.E. SPECIAL A preview of exhibits and exhibitors and highlights of what they have to offer in new products. Just one more reason why you should subscribe to electronics (or renew your subscription). Fill in box on Reader Service Card. Easy to use. Postage free. FIND WHAT YOU NEED IN electronics and 300 mc to 600 mc. Tangential sensitivities better than -55 dbm with selected MA408B or selected 1N23B crystals can be achieved. CIRCLE 351 ON READER SERVICE CARD D-C Amplifier solid-state VIDEO INSTRUMENTS Co., Inc., 3002 Pennsylvania Ave., Santa Monica, Calif. Model 93 solid-state d-c amplifier features a combination of differential input and wide bandwidth. Another feature is that the amplifier can drive any of the new h-f galvanometers. Maximum current output is 100 ma at an output impedance of less than 1 ohm. Input impedance is higher than 50 K ohms, isolated by at least 100 megohms from the cabinet. Recovery time from overloads is 100 millisec maximum. CIRCLE 352 ON READER SERVICE CARD Hermetic Terminal nonturning LUNDEY ASSOCIATES, 694 Main St., Waltham 54, Mass., has available model 599 hermetic terminals for electronic components such as transformers and capacitors in the intermediate voltage range (1,500 v operating). The nonturning terminal is a single-unit assembly (no loose parts). It has been proved by independent laboratory tests to meet MIL-T-27A specifications. CIRCLE 353 ON READER SERVICE CARD Jack Panels video and r-f NEMS-CLARKE CO., 919 Jesup-Blair Drive, Silver Spring, Md. Type 921 jack panel (illustrated) is made of aluminum and is 19 in. wide by only $1\frac{1}{2}$ in. high. It is provided with 12 type 925 jacks and is especially useful where space is at a premium. Types 928 and 929 are similar in construction to the 921 but are both $3\frac{1}{2}$ in. high. Type 928 provides for 24 type 925 jacks; and type 929 provides for 48. CIRCLE 354 ON READER SERVICE CARD Overspeed Monitor uses tach-generator ROHDE & SCHWARZ, 111 Lexington Ave., Passaic, N. J. Type FDW overspeed monitor provides a means for accurate and very fast protection of steam turbines from excessive speeds. It employs a tach-generator which is directly coupled to the turbine shaft and delivers an a-c voltage whose frequency is strictly proportional to the rpm. Vector Electronics manufactures a complete line of structures for mounting circuitry easily, compactly and with good accessibility. Vector experience and facilities guarantee delivery, performance and economical prices. Write for catalog to: VECTOR ELECTRONIC COMPANY 1100 FLOWER STREET, GLENDALE 1, CALIFORNIA TELEPHONE: CHapman 5-1076 TARZIAN M-500... a high efficiency silicon rectifier commercially priced 500-ma ferrule rectifier connects easily to standard clips The Sarkes Tarzian M-500 silicon rectifier is rated at 500 milliamperes dc, with a peak inverse voltage rating of 400 volts. This was the first commercially priced silicon rectifier, and more M-500's are now in use than any similarly rated unit. The Tarzian M-500 is a cartridge type rectifier with end ferrules that snap quickly and easily into standard clips. These silicon rectifiers are made by a special Tarzian process that provides optimum forward to reverse ratios and long, useful life. For additional information, practical application assistance, and prices on the M-500, write to Section 4393K, Semiconductor Division, Sarkes Tarzian, Inc., Bloomington, Indiana. M-500 Characteristics \| DC amps (100° C) \| Peak Inv. Voltage \| Tarzian Type \| Max. RMS Volts \| Max. Recurrent Peak Amperes (100° C) \| Max. Surge Amps 4MS \| JEDEC No. \| \|------------------\|-------------------\|--------------\|----------------\|--------------------------------------\|---------------------\|-----------\| \| 0.5 \| 400 \| M-500 \| 280 \| 5 \| 30 \| 1N1084 \| SARKES TARZIAN, INC. SEMICONDUCTOR DIVISION BLOOMINGTON, INDIANA In Canada: 700 Weston Rd., Toronto 9, Ontario Export: Ad Auriema, Inc., New York City CIRCLE 174 ON READER SERVICE CARD Helitron 250-500 mc WATKINS-JOHNSON Co., 3333 Hillview Ave., Stanford Industrial Park, Palo Alto, Calif. The WJ-207 helitron is a lightweight oscillator continuously voltage tuned over the 250-500 mc frequency band, providing a minimum of 3 mw output power. Maximum power output variation over the band is 6 db. It is electrostatically focused and therefore requires no focusing magnet. It is designed to withstand the severities of modern environmental extremes by utilizing rigid construction throughout, for example the metal-ceramic envelope. CIRCLE 356 ON READER SERVICE CARD Pulse Mixer high speed HARVEY-WELLS ELECTRONICS, INC., East Natick Industrial Park, East Natick, Mass. New pulse mixer consists of two gating transistors with a common output pulse transformer. It differs from the pulse gate in that two pulse rates can be mixed and amplified. Electrical specifications are: input/output, negative 4-v, 1/10 μsec pulses; supply voltages and currents, negative 15-v, 30 ma; plus 10-v, 0.3 ma; and clamp voltage negative 4 v at +20 ma. CIRCLE 357 ON READER SERVICE CARD Transistor Amplifier hushed type MILLIVAC INSTRUMENTS, Division of Cohu Electronics, Inc., Box 997, Schenectady, N. Y. The VS-64A hushed transistor amplifier has an rms noise voltage, referred to the shorted-input terminals, of 50 milliμv (10⁻⁶ v), when used with a bandpass of 20 cps—300 cps. A typical audio-bandpass of 20 cps-14 kc generates only 160 milliμv noise voltage. The total available bandpass of the amplifier is 2 cps—180 kc. Noise voltage with this larger bandpass is in the order of 620 milliμv, which is 20 to 30 db less than can be expected of low-noise vacuum tubes operating under comparable conditions. CIRCLE 358 ON READER SERVICE CARD P-C Connector 44 contacts CINCH MFG. Co., 1926 S. Homan Ave., Chicago 24, Ill., announces a ELECTRONICS • FEBRUARY 12, 1960 p-c connector with wire wrap termination. It maintains positive contact with a p-c board over a dimensional range of 0.57-0.072. It contains 44 contacts which are phosphor bronze with a silver plate of 0.0002 and 0.00003 gold plate finish. A plastic polarizing key is furnished with each connector, and this key can be inserted into the contact slot of any contact position. CIRCLE 359 ON READER SERVICE CARD BOESCH* semi-automatic toroidal winders - Wind #20 to #42 AWG wire with constant uniformity at speeds up to 1200 RPM. - Finished coils from 7/32" ID through 5" OD. Low cost, high production winders. Core oscillated manually, clamped manually. Both machines wind standard size cores without additional attachments — use interchangeable shuttle heads. Capabilities identical except that TW-251 has built-in turns counter and variable speed motor. Accessory, electronic, predetermined turns counters available for both machines ... automatically stop winding at a preset number of turns — results in faster winding because operator does not have to watch counter. Both the TW251 and TW201 are bench-type machines with the following standard equipment: motor, core holder, shuttle opening lever, wire tension device, predetermined mechanical linear counter, reversing switch, 3" shuttle head, choice of 3" standard or 3S shuttle with slider. Write for complete data. BOESCH MANUFACTURING COMPANY, INCORPORATED BOESCH DANBURY, CONNECTICUT CIRCLE 176 ON READER SERVICE CARD D-C Bridge high speed INDUSTRIAL INSTRUMENTS AUTOMATION CORP., 89 Commerce Road, Cedar Grove, N. J. Model AB-4-5 high speed d-c bridge can sort resistors into three groups automatically at rates of 5,000 pieces per hr. The equipment is capable of sorting to high accuracy over an extremely high and low range extending up to 100 megohms. A built-in 7-dial resistance decade is provided for setting to the nominal value of resistance being checked. Tolerance limits for the 3 bins is set by means of plug-in units. Resettable electromechanical counters tally the number of pieces in each bin. CIRCLE 360 ON READER SERVICE CARD Gold-Antimony Alloy for transistor mfrs. ENGELHARD INDUSTRIES, INC., 113 Astor St., Newark, N. J., has developed a new gold-antimony alloy (No. 1549) with improved properties for transistor manufacturers. Available in rod or whisker wire in diameters from 0.25 in. down to 0.001 in. and in sheet down to 0.0015 in. thickness, the material consists of high purity gold containing 1 percent antimony and features completely homogeneous dispersion of the antimony-rich phase throughout the gold matrix. It is designed for use in imparting a controlled impurity (antimony) into a semiconductor crystal, usually silicon, by evaporation techniques. In this method of crystal doping, the alloy is heated until the antimony boils off, creating an antimony atmosphere from which the crystal picks up the desired impurity. CIRCLE 361 ON READER SERVICE CARD Disk Capacitors in 86 values CENTRALAB, A Division of Globe-Union, Inc., 900 E. Keefe, Milwaukee 1, Wisc. The ID 500 vdcw disk capacitors are available in 86 values ranging from 3.3 μf to 0.05 μf. Depending on the capacity the size varies from ½ in. to ¾ in. diameter. CIRCLE 362 ON READER SERVICE CARD Tantalum Capacitor 300 v foil-type GENERAL ELECTRIC CO., Schenectady 5, N. Y. New Tantalytic units, available for operation at 85 C (to 300 v) and 125 C (to 250 v), are FOUR OPERATIONAL MISSILES BUILT BY MARTIN what’s in a name? Your name gives you identity. A tool has a “name”. A part has a “name”. A switch has a “name”. Everything in your plant has a name, from works for you to the ELEVATOR. Engrave all these “names” permanently with unskilled labor by Engravo Graph. Send for booklet ZT-3 new hermes ENGRAVING MACHINE CORP. 154 WEST 14TH STREET NEW YORK 11, N.Y. CIRCLE 177 ON READER SERVICE CARD Reduce rejects and operational failures caused by airborne dirt, dust, lint, abrasives with the HONEYWELL ELECTRONIC AIR CLEANER When microscopic airborne dirt, dust, lint and abrasives get into precision electric and electronic devices, the results are unusually high reject rates, poor product quality, operational failures. The Honeywell Electronic Air Cleaner will protect your precision components from the damaging action of these airborne spoilers by trapping particles 100 times as small as those stopped by mechanical filters. Pays for itself! Send coupon today to learn how the Honeywell Electronic Air Cleaner soon pays for itself in: - Electronic Manufacturing - Controlled environment rooms for precision manufacturing, processing and assembly - Switch & relay rooms - Computer rooms - Transmitter rooms - Precision component manufacturing designed primarily for applications in missiles, computers and airborne electronic equipment where minimum size and weight are major considerations. Representative units illustrated are rated at 300 v and 1, 7, and 25 μf. A capacitance change of no more than 10 percent is specified after 2,000 hr operation at rated voltage and temperature. At -55 C the maximum capacitance change is 20 percent. Capacitance tolerances are tightened to ±15 percent. Units are available in seven case sizes and in both polar and nonpolar construction. CIRCLE 363 ON READER SERVICE CARD Count Rate Meter transistorized INTERSTATE ELECTRONICS CORP., 707 E. Vermont Ave., Anaheim, Calif. Model 502 completely transistorized count rate meter has eight scale ranges, extending to 600,000 cpm, displayed on a large three-in., easy-to-read meter scale. Three time constants are available, from a front panel mounted switch, as is a calibration control. The input is sensitive to a 250 mv negative pulse. Provisions are available on the rear of the instrument to drive either a 0-1 ma or 0-10 mv recorder. Price is $295. CIRCLE 364 ON READER SERVICE CARD Oscillators voltage-controlled TELE-DYNAMICS INC., 5000 Parkside Ave., Philadelphia 31, Pa. Transistorized voltage-controlled oscillators are available for use in Electromechanical Components and Systems Capability AIRESEARCH POSITIONING CONTROL SYSTEMS One of the many types of high speed positioning control systems produced by AiResearch, the system above amplifies electric signals from an inertial guidance source and adjusts the control surfaces of the missile or drone to maintain a predetermined course. AiResearch diversification and experience provide full capability in the development and production of electromechanical equipment and avionic controls for aircraft, ground handling, ordnance and missile systems. A.C. and D.C. Motors, Generators and Controls • Inverters • Alternators • Linear and Rotary Actuators • Power Servos • Hoists • Electrical Pyrotechnics • Antenna Positioners • Positioning Controls • Temperature Controls • Sensors • Williamsgrip Connectors • Static Converters. Your inquiries are invited. THE GARRETT CORPORATION AiResearch Manufacturing Division Los Angeles 45, California CIRCLE 219 ON READER SERVICE CARD FEBRUARY 12, 1960 • ELECTRONICS extreme missile environments. Type 1250 are used for 0 to +3, 0 to +5, ±1.5, or ±2.5 v signals. Type 1251, operating with 0.25 v inputs is compatible with the recently available ±v output pressure transducers. Type 1252, requiring only 20 mv for full deviation, are ideal for bridge instrumentation. Environmental characteristics are: -55 C to 125 C, 100 percent relative humidity, 30 g rms random vibration, 150 g acceleration, and 200 g shock. Types 1250 and 1251 occupy only 2.7 cu in.; type 1252, 3.7 cu. in. CIRCLE 365 ON READER SERVICE CARD VHF-UHF Cavities miniature size RESDEL ENGINEERING CORP., 330 S. Fair Oaks Ave. Pasadena, Calif. A series of six vhf-uhf cavities offering 36 combinations of size, power and function, operating in the frequency range 220 through 1,150 mc. have been developed for missile-borne and ground-based r-f equipment. All are designed for heat sink type mounting and the components are derated to provide maximum trouble-free operation. Features are minimum weight, optimum surface conductivity, corrosion protection, stable mechanical tuning, negligible backlash and built-in ruggedness for operation under severe environmental conditions. CIRCLE 366 ON READER SERVICE CARD Readout Display Unit four decimal places ADVANCE INDUSTRIES, INC., 640 Memorial Drive, Cambridge, Mass. New digital converter readout display unit provides rapid, positive, visual numerical readout of the out- This is the new knight-kit® ac vtvm. It marks a major achievement in instrumentation...and a breakthrough in the professional instrument price barrier. Here is the only vtvm with automatic range selection ...featuring a self-seeking mechanism which automatically selects the proper range when probes are touched to the circuit under examination. Simultaneously, a front panel light indicates the range in use. There are 11 ranges from 3 millivolts to 300 volts full scale; frequency response to 2.5 mc. Reads as low as 100μv. This precision instrument is an exclusive knight-kit development, designed for easy assembly. There is nothing like it on the market, in any form or at any price. Available only from Allied Radio ........$99.50 Send for special bulletin covering the remarkable specifications and full details of the Knight-Kit AC VTVM with Automatic Range Selection sold exclusively by ALLIED RADIO 100 N. Western Ave., Chicago 80, Ill. Dept. 66-B the knight-kit DC Lab Oscilloscope with Interchangeable Vertical Amplifiers For the first time—a triggered sweep DC lab scope with plug-in interchangeable vertical amplifiers, in easy-to-build kit form. Highlights: crystal-controlled timing markers; DC amplifiers in both horizontal and vertical channels; electronic regulated power supply. Three interchangeable vertical preamps available: high-gain differential; wide-band (to 10 mc); and dual-trace (also blank plug-in chassis for your own circuitry). The only instrument of its kind in kit form. The performance is truly impressive. The price (less preamps) ........$285 for full details, ask for descriptive bulletin CIRCLE 179 ON READER SERVICE CARD MODEL DS SERIES DEKASTAT®—Precision decade resistors for panel mounting, featuring the exclusive ESI DEKADIAL® concentric dial assembly for convenient straight line readings. Total resistance values available from 1,200 to 120,000 ohms with accuracy of ±0.05%. Power rating, ½ watt per step. 3 or 4 decades of resolution. Standard units available from stock. Prices: $63.00 to $110.00. MODEL DB SERIES DEKABOX®—Precision decade resistors similar to Model DS series DEKASTAT® units, but conveniently mounted on an adjustable base with binding posts. Features ESI DEKADIAL® design for straight line readings. Total resistance values available from 12,000 ohms to 1.2 megohms with accuracy of ±0.05%. 3 to 6 decades of resolution. Power rating, ½ watt per step. Standard units available from stock. Price: $73.00 to $151.00. MODEL RS SERIES DEKASTAT®—Rack-mounted precision decade resistors. Adjusted to very close tolerances for use as laboratory resistance standards. Independently operated dials provide both coarse initial steps for quickly approximating the required value and progressively finer steps for more exact settings. Less than 10 ppm/C° temperature coefficient. Total resistance values to 1.2 megohms. Accuracy, 0.02%. Six decades of resolution. Power rating, ½ watt per step. 30-day delivery; Price: $550.00. SEND FOR DESCRIPTIVE LITERATURE See our display at the MARCH IRE SHOW Booth 3010-3011 Electro Scientific Industries 7524 S.W. MACADAM • PORTLAND 19, OREGON formerly ELECTRO-MEASUREMENTS, INC. ESI has outstanding job opportunities for experienced design and applications engineers. Call or write C. Davis. CIRCLE 180 ON READER SERVICE CARD put of analog-to-digital converters. Model 411 Digitester is designed to readout the outputs of converters using binary-coded decimal codes and having outputs in the form of either open-or-closed contacts or suitable voltage level changes. It presents directly readout of the converter output to four decimal places. CIRCLE 367 ON READER SERVICE CARD Coaxial Switch miniaturized MICRONOT INC., 220 Pasadena Ave., South Pasadena, Calif. Type 90-01 spdt coaxial switch for 50 ohm cable is designed so that it can be directly inserted into a miniaturized circuit without adapters. Positive toggle action assures proper switching and panel mounting provides easy installation. Mating connectors are available for application at above or below 400 mc. Vswr is less than 1.2 to 2 kmc and insertion loss is 0.5 db at 2.0 kmc. Voltage rating is at 1,500 v maximum (60 cps) and crosstalk is better than 60 db to 2.0 kmc. Operative life is at a minimum of 50,000 cycles. Switch weighs only 1 ¼ oz. CIRCLE 368 ON READER SERVICE CARD VHF Preamplifier low noise COMMUNITY ENGINEERING CORP., P. O. Box 824, State College, Pa. Designed for use in the 50 to 200 mc range, the model 1001 vhf preamplifier has a noise figure of better than 3 db at 85 mc and 4.5 db at 200 mc with a nominal gain of 30 db. Unit is fixed tuned to required frequency. Bandwidth is 10 mc. Amplifier subassembly and inIntegral power supply are mounted in light weight, solidly constructed chassis, on a standard size rack panel protected by an easily removable dust cover. Input and output impedance is 50 ohms. Standard type N connectors are used. CIRCLE 369 ON READER SERVICE CARD Silicon Rectifiers highly stable DALLONS SEMICONDUCTORS, 5066 Santa Monica Blvd., Los Angeles 29, Calif., announces 20-35 ampere, 60-600 piv silicon rectifiers which contain solders within their construction which have a melting point in excess of 600 C. The $ \frac{1}{4} $ in. stud construction houses a pure silver, heavy spring lead anode assuring ruggedness and high resistance to shock and vibration. Units have less than 5 ma reverse current, and maximum forward drop voltage at a test temperature of 25 C at 20 amperes, d-c, is 0.65 v. CIRCLE 370 ON READER SERVICE CARD Long-Life Encoder high-resolution DATEX CORP., 1307 So. Myrtle Ave., Monrovia, Calif. Model C-804 shaft position encoder provides an output of 3,600 quanta per revolution in Datex code or 4,096 quanta per revolution in Gray code. It is capable of unlimited readout cycles and the sampling rate is limited only by the readout device. Use of brush contacts gives high current carrying capabilities and long life. A useful life of over 10 million revolutions has been obtained from this encoder. Unit uses an 8$\frac{1}{2}$-in. disk. It is 9 in. in diameter and 3$\frac{1}{2}$ in. high, exclusive of the shaft. CIRCLE 371 ON READER SERVICE CARD AC RATIO MEASUREMENTS? THERE'S A NORTH ATLANTIC INSTRUMENT TO MEET YOUR REQUIREMENTS, TOO... Now—from North Atlantic—you get the complete answer to AC ratio instrumentation problems—in the laboratory, on the production line, in the field. Specialists in ratiometry, North Atlantic offers the only complete line of precision instruments to handle any ratio measurement task. All are designed to meet the most demanding requirements of missile age electronics—provide high accuracy, flexibility, component compatibility and service-proven performance. Some are shown above. If your project demands total solution to ratio measurement problems, write for Date File No.10H It provides complete specifications and application data and shows how North Atlantic's unparalleled experience in ratiometry can help you. 1. RATIO BOXES: Both laboratory standards and general duty models. Ratio accuracy better than 0.1%. Operation from 25 cps to 10 kc. 2. COMPLEX VOLTAGE RATIOIMETERS Integrated, single-unit instruments where phase relations are critical. Accuracy better than 0.1%, unaffected by quadrature. Three frequency options. Direct reading of phase angle in milliradians or degrees. 3. PHASE ANGLE VOLTMETERS: Versatile readout system for all ratiometry applications including direct reading of phase, null, quadrature, in-phase and total voltage. Multiband, single-, or multiple-frequency operation. 4. RATIO TEST SETS Ratio reference and readout in one convenient package for production line and similar applications. Can be supplied with any combination of ratio box and phase angle voltmeter. NORTH ATLANTIC INDUSTRIES, INC. 603 MAIN STREET, WESTBURY, N.Y. • EDGEWOOD 4-1122 CIRCLE 181 ON READER SERVICE CARD HARD-ANODIZED POWER TRANSISTOR WAFERS FOR SUPERIOR DIELECTRIC INSULATION AND HEAT DISSIPATION Featuring thermal conductivity 400% greater than conventional mica heat sinks and dielectric properties equal to the best insulating materials, this thin, hard-anodized aluminum wafer insulates power transistor from chassis and dissipates the substantial heat generated at rated capacities. Unlike mica that is fragile and difficult to handle, the hard-anodized aluminum wafer is extremely durable with high abrasion and corrosion resistance. In diamond, round or square shape to suit any base, the wafer is installed between transistor and chassis, heat sink or other surface on which the transistor is mounted. Write for bulletin. NEW BOOKS Encyclopedic Dictionary of Electronics and Nuclear Engineering By Dr. Robert I. Sarbacher Prentice-Hall, Inc., New York, 1959, 1,417 p, $35. This massive volume, issued as 1959 drew to a close, is a valuable addition to the rapidly expanding library of technical reference works. Electronics engineers — particularly those working in the nucleonics area — would benefit from owning or at least having access to this book; technical writers and editors should consider it a professional "must". Nontechnical people will find the lucid definitions most helpful. Abbreviations (including those authorized by the military), acronyms, vernacular expressions, British words and proprietary terms are accurately defined. Illustrations are used somewhat sparingly, approximating one per page, probably because of length considerations. Cross references are adequate. Many acoustical, chemical, electrical, mechanical, and mathematical terms applicable to electronics are included. Most important contribution of the work may be its comprehensive standardization of terms. Definitions given are approved by every official professional society concerned, including the IRE and AIEE as well as by government agencies such as the AEC and FCC and by industrial groups such as RETMA and SMPTE. The publisher's claim that ALL current terms are listed is a bit presumptuous. A cursory examination revealed that thin films, microminiaturization, vacuum deposition, tunnel diodes, sampling oscilloscopes, biaxial ferrite elements and varactors are not mentioned. Also, medical electronics is not listed nor are any of the terms related to this field. One factor prospective buyers should consider is the relative expense of the book if their interest is solely in electronics. Less expensive, if not more complete, purely electronic dictionaries are available. —W.E.B. Magnetic Materials Verlag Chemie, GMBH, Weinheim/Bergstrasse, West Germany, 1959, 580 p, $85.50. This is a new volume in the comprehensive "Gmelins Handbook of Inorganic Chemistry." It is Supplement 2 of System No. 59, Iron, Part D, and also supplements Systems 58 (Cobalt), 57 (Nickel), 56 (Manganese) and 52 (Chromium). The volume presents an exhaustive review of magnetic and electrical properties, applications, literature and patents, bringing up to date the volumes published in the 1930's. Data is introduced by a chapter on ferromagnetic theory and a new section covers ferromagnetic semiconductors. Text is in German, but there are English marginal notes, table of contents and index. The text is supplemented by 308 graphs and numerous tables. Organization of data follows the Gmelin classification. The high price is apparently necessary because of the tremendous effort which has been expended to assemble information gathered from so thorough a literate search. G.M. Linear Network Analysis By Sundaram Seshu and Norman Balabanian. John Wiley & Sons, Inc., New York, 1959, 571 p, $11.75. This book is intended for graduate students of electrical engineering who have completed course material on the theory of functions of a complex variable and the theory of the Laplace transformation. The text reviews basic network theory and presents, in a mature fashion, a clear discussion of general network analysis including topological theorems such as "block diagrams" and "signal-flow graphs." It has adequate discussions on feedback notions and stability theory. Two-port networks are clearly discussed, as is also general filter theory. The convolution theorem approach to network system problems is developed on the basis of the Laplace transformation and then... do you know what's expected from semi-conductor materials? There were more than a dozen articles on semiconductor materials in electronics in recent months. Each was specially edited to give you all key facts, ideas or trends—and there's more coming! Accurate electronics' reporting tells you what's happening now...what's expected in materials and components. Don't miss dozens of articles on basic subjects edited to keep you informed, help make your research, development, sales and marketing plans pay off. It pays to subscribe to electronics (or renew). Fill in box on Reader Service Card now. Easy to use. Postage free. FIND WHAT YOU NEED IN electronics redeveloped using the superposition principle for linear systems and the impulsive response of the network. The DuHamel form of the integral for the response in terms of the step response of the network is also developed. Methods employed in the text do not disregard mathematical subtleties. In those few cases in which to have maintained an exact mathematical presentation would have meant digressing too far afield from the immediate subject, the authors were careful to indicate source material where the precise mathematical theorems may be found to substantiate their statements. This is the case, for example, in the discussion of the impulse function. I was pleased to note the abundance of excellent examples and subsequent careful discussions. Attention may also be directed to the excellent appendices which provide the relevant results of complex variable theory and Lapace transformation needed for the text. David Jagerman, System Development Corp., Lodi, N. J. THUMBNAIL REVIEWS Analog Methods—Computation and Simulation. By W. J. Karplus and W. W. Soroka, McGraw-Hill Book Co., Inc., New York, 1959, 483 p., $12.50. This second edition of a widely used book shifts emphasis from mechanical to electrical and electronic equipment. Reader learns to utilize linear and nonlinear computer elements in both special- and general-purpose design before he begins to interconnect these elements to solve laboratory problems. The chapter on network analyzer techniques has been completely revised and augmented to reflect recent developments. Discussion of mechanical differential analyzers and other outmoded techniques has been shortened considerably, but chapter dealing with mechanical computing elements has been retained. Crystals and Crystals Growing. By A. Holden and P. Singer, Doubleday & Co., Inc., New York, 1960, 320 p., $1.45. Written primarily for students, this book suggests experiments to develop laboratory skill at growing crystals and for building simple models and apparatus. Theory and practice of art of modern crystallography is explained and problems are given to sharpen visualization and thought. CHECKLIST OF NEW McGRAW-HILL BOOKS SERVOMECHANISM FUNDAMENTALS Provides needed focus in a straight-forward manner in enable you to solve problems in the design of automatic control devices, and achieve efficient, smooth operating automatic control systems. By Henri Lauer, Leslie E. Matson, both of Radio Corp. of America; and Robert N. Lesnick, Belfone Research Lab. 2nd Ed., 190 pp., 273 illus., $10.00 LINEAR CIRCUIT ANALYSIS A helpful working reference on analysis of linear, lumped-constant circuits. Logically treats the theory and explains the formulation of network equations and their solution in the time domain by classical methods, and in the frequency domain by Fourier and Laplace transforms. By R. L. Leve, NYU; Samuel G. Lutz, Hughes Aircraft Company; and Charles F. Rehberg, NYU. 608 pp., 560 illus., $12.50 CATHODIC PROTECTION Gives you basic principles and successful techniques for protecting underground and submerged pipelines, tanks, and other structures. Covers instruments, field surveys and tests, system, design, installation, and tests. By Lindsay M. Applegren, Lt. Col., U. S. Army, Retired. 2nd Ed., 68 illus., $7.00 ELECTRICAL ENGINEERING FUNDAMENTALS A Unified Introduction to Electrical Engineering Presents clear and thorough explanations of basic electrical engineering concepts: electromagnetic characteristics, vector concepts, magnetic fields, electric network analysis, etc. By James P. Neutl, Professor of Electrical Engineering, Univ. of Ill. 100 pp., 269 illus., $8.50 CONTROL-SYSTEMS ENGINEERING Provides full coverage of the mathematical aspects of modern control-systems engineering. Covers matrix methods, numerical analysis, basic statistical theory, various analyses in phase space, elements of game theory, and much more. By William W. Seifert, Mass. Inst. of Tech.; and Carl W. Stengel, Boeing Airplane Systems Laboratory. 900 pp., illus., $15.00 ADVANCED ENGINEERING MATHEMATICS Covers advanced mathematics as applied to engineering and physics. Covers determinants, matrices, differential equations, Fourier series, partial fractions, vector analysis, etc. By C. R. Wylie, Jr., Professor of Mathematics, University of Utah. 2nd Ed., 525 pp., 193 illus., $9.00 SEE THESE BOOKS 10 DAYS FREE McGraw-Hill Book Co., Dept. L-2-12 327 W. 41st St., N.Y.C. 36 Send me (circle check below) one (1) place shipment on attached bill. If I like it, I will send for book(s) I keep plus ten cents for delivery costs, and return unwanted book(s) prepaid. (We pay shipping costs if you return with this coupon—same return privilege.) Lauer, et al.—Servomech. Fund. $10.00 Applegren—Linear Circ. Anal. $12.50 Applegren—Cathodic Protection $7.00 Neutl—Elec. Eng. Fundamentals $8.50 Seifert & Stengel—Control.Sys. Eng. $15.00 Wylie—Adv. Eng. Mathematics $9.00 (PRINT) Name ........................................... Address ........................................... City ............................................ Zone State Company ........................................ Position ........................................... For price and terms outside U. S. write McGraw-Hill Int'l., N.Y.C. CIRCLE 221 ON READER SERVICE CARD Concise Dictionary of Science. By F. Gaynor, Philosophical Library, Inc., New York 1959, 546 p, $10. This up-to-date encyclopedic-like dictionary provides concise definitions of terms and concepts pertaining to all fields of science. Newer sciences of virology, enzymology, cytogenetics, radio-chemistry, high energy and solid-state physics, and the like are covered. In committing an undertaking of such sweep to so few pages, many terms were omitted; however, the electronics man dealing in cross-field applications will find the definitions succinct and entirely comprehensible. Ham Radio Handbook. By Robert Hertzberg, Arco Publishing Co., Inc., New York, 1959, 144 p, $2.50. This book tells how to learn code, select the proper transmitter and receiver and get a ham ticket. A special section treats mobile equipment. Magnetism and Electromagnetism (71 p, $1.80) and Advanced Magnetism and Electromagnetism (96 p, $2.25). Edited by A. Schure, John F. Rider Publisher, Inc., New York, 1959. The first book is an elementary treatment of the major theoretical considerations of magnetism, magnetic circuits and electromagnetism. The second is a college-level companion volume which penetrates deeply into the various underlying details of the forces acting on a charge moving through a magnetic field, induction lines, and magnetic flux. Biot's Law, Faraday's Law, Lenz's Law, the Curie-Weiss Law and the domain theory of magnetism are also discussed as are the cyclotron, the mass spectrograph and terrestrial magnetism. Masers. By G. Troup, John Wiley & Sons, Inc., New York, 1959, 168 p, $2.75. This monograph, written by an Australian, is probably the first book published on masers. A discussion of the stimulated emission process is followed by a brief outline of methods used to obtain the conditions necessary for amplification. A review of the experimental work done is given together with a comprehensive bibliography. Fourier's Series. By W. E. Byerly, Dover Publications, Inc., New York, 1959, 287 p, $1.75. This elementary treatise, originally written in 1893, is recognized as one of the most useful and practical expositions of Fourier's series, and spherical, cylindrical, and ellipsoidal harmonics. Reader is given 190 problems to solve. An appendix provides six tables of surface zonal harmonics, hyperbolic functions, and Bessel's function. Electronics men will find book helpful in all forms of advanced engineering, especially in communications, radar and acoustics. Convair 880 jet airliners are first to use new, advanced design anti-fogging, anti-icing heat control systems developed by Magnetic Controls Company. Windshield icing and fogging problems are considerably more complex for jet airliners, compared with those encountered by piston engine planes. With jet travel, faster speeds, higher altitudes, plus greater extremes in temperatures and atmospheric pressures, many new problems develop. Magnetic Controls proportional heat control systems have helped to solve these problems for Convair 880 jets. The Magnetic Controls systems offer four distinct advantages: - Automatic compensation for windshield aging. - Ultra-reliable, fully static design. - Elimination of thermal shock for longer windshield life. - Superior accuracy — proportional plus integral control. For complete information about this system and how it may help you obtain precise, accurate heat control for many applications, write or phone Magnetic Controls Company today. MAGNETIC CONTROLS COMPANY 6405 CAMBRIDGE STREET • MINNEAPOLIS 26, MINN. • WEST 9-4691 Heat Control Systems • Static Inverters • Voltage Monitoring Systems CIRCLE 185 ON READER SERVICE CARD 185 Low-Frequency Amplifier Systems. Edited by A. Schure, John F. Rider Publishers, Inc., New York, 1959, 70 p, $1.80. This companion to another Rider book (Low-Frequency Amplifiers) treats, with special emphasis, coupling methods suitable for low-frequency range, phase inversion and inverse feedback as well as circuit design using vacuum tubes and transistors. Mathematics has been kept simple, but analyses are sufficiently extensive to permit technicians or students to fully comprehend pertinent theory. The Physics of Television. By D. G. Fink and D. M. Lutwyens, Doubleday & Co., Inc., New York, 1960, 160 p, $0.95. Written primarily for students, this book gives account of how men have learned to control electrons, photons, and electromagnetic waves to produce instantaneous moving pictures at great distances. Principles of physics as applied to television systems are discussed. Proceedings of the Fourth Symposium on Magnetism and Magnetic Materials. American Institute of Physics, McGraw-Hill Book Co., Inc., New York, 1959, 322 p, $10.00. Some 150 papers reporting advanced research and development in the field are included. A third of the papers will interest design engineers, covering new materials or techniques for computer components, microwave amplifiers, waveguide, permanent magnets and magnetic instrumentation. Several papers discuss effects of environments or processing on the properties of magnetic materials. The Measurement of Power Spectra. By R. B. Blackman and J. W. Tukey, Dover Publications, Inc., New York, 1959, 190 p, $1.85. This graduate-level book, reprinted from Bell System Technical Journal, views the topic from the point of view of communications engineering and such related fields as oceanography, aerodynamics, meteorology, seismology, economics, guided missiles, radar and acoustics. Authors explain various ways of getting practically useful answers in the measurement of power spectra using results from both transmission theory and the theory of statistical estimation. An appendix reviewing fundamental Fourier techniques, an index of notation and a glossary of terms are also included. Calculus is used extensively along with some advanced mathematics. Printed Circuits. By Morris Moses, Gernsback Library, Inc., New York, 1959, 224 p, $2.90 (soft cover), $4.60 (hard cover). This is a how-to-do-it book written for radio hams, television and radio technicians, and experimenters. Development of the art and advanced applications are also discussed. The Gentle Art of Mathematics. By D. Pedoe, The MacMillan Co., New York, 143 p, $3.50. Fascinating, thought provoking little volume wholly devoted to bringing modern mathematics into perspective. Those who wish to develop their mental capacities and those whose business is mathematics should read this book. Of particular interest to electronics men will be the chapter on "Automatic Thinking", a non-Carrollian approach to symbolic logic. Moon Base—Technical and Psychological Aspects. By T. C. Helvey, John F. Rider Publisher, Inc., New York, 1960, 72 p, $1.95. Although professedly not science fiction, this booklet does little to dispel the aura of sensationalism and fantasy surrounding moon travel. Avowed aim of the author is to show only problem areas in the construction of a moon base, but this limitation is not used to good advantage—the text is sketchy and the illustrations infantile. The erroneous impression that the discussion is on a high scientific level is conveyed by an early reference to the two-page glossary of technical terms and expressions used in the text. Model Radio Control. By E. Safford, Jr., Gernsback Library, Inc., New York, 1959, 192 p, $2.65. This handbook covers all aspects of radio control from theory to construction of coders, decoders and other complex components as well as complete systems. The Magnetodynamics of Conducting Fluids. Edited by Daniel Bershader, Stanford University Press, Stanford, Calif., 1959, 145 p, $1.50. Third volume in a series growing out of a Lockheed-sponsored symposium on magnetohydrodynamics. Seven authoritative analyses of underlying behavior of conducting fluids in magnetic fields are given. Electronics men involved in handling of hot plasmas in magnetic bottles and ducts will find this book particularly valuable. Microwave Data Tables. By A. E. Booth, Iliffe and Sons, Ltd., London, England, 1959, 61 p, 27s 6d ($3.85). A collection of 26 tables for use by engineers engaged in research and development of waveguides and similar transmission lines. Technical Writing Manual. By C. K. Arnold, Electronic Periodicals, Inc., Cleveland, Ohio, 1959, 87 p, $2.50. This glossary of correct and approved usage of commonly misused words and phrases for engineers and scientists is a valuable contribution to the literature on the art of technical writing. Unfortunately, the high price tag may keep this short monograph from those who most need it. "WHO NEEDS TO KNOW ABOUT GRAPHIC METERS?" You can't afford to "hide" from the facts and figures about your operations, no matter what they involve—research, design, production or what not. Neither can you afford to "hide" from information about the best, the simplest and the most effective way to obtain these important facts and figures. We're referring to "the Esterline-Angus way" with E/A direct writing graphic instruments that will accurately measure most any quantity and record it while you are busy with other tasks. Easiest way to learn how E/A graphic meters can help you is to discuss your own facts-and-figures requirements with the Esterline-Angus Specialist in your locality. He's trained and he's experienced. He will diagnose your problem and if E/A instruments are your answer, he will prescribe the equipment best suited to your needs. He will also follow through to see that it operates to your complete satisfaction. Behind every E/A Specialist is an organization which has been designing and building fine graphic instruments for more than 50 years. Let the E/A Specialist in your area help you. Just drop us a line and say when. No obligation, of course. The Esterline-Angus Company No. 1 in fine Graphic Instruments for more than 50 years. DEPT. E, P.O. BOX 596, INDIANAPOLIS 6, INDIANA Airborne Time Code Generator illustrates high-density packing obtainable with T-Series circuits. Hinged arrangement of mounting panel facilitates accessibility. The finished package weighs only 20 lbs., measures 5" x 8" x 203/4". Unit generates 14-digit Pulsed Mugu code, modulating a 1 kc carrier on two lines. Three sine wave and four pulse outputs are also provided, all with only 96 T-Series circuits and 77 watts of input power. FROM SYSTEM SPECS TO BREADBOARD TO FINISHED PRODUCT IN 75 DAYS! That's the record set by the manufacturer of this complex airborne Time Code Generator — thanks to the compatibility of proven EECO T-Series Circuit Modules and the flexibility of the EECO Breadboard Kit. Designed and developed for testing the fire control of manned supersonic aircraft under actual flight conditions at altitudes up to 80,000 feet, this Time Code Generator employs T-Series circuits throughout. Required accuracy of 1 part in $10^5$ was easily obtained. HIGH DENSITY, LIGHT WEIGHT The total package contains 96 T-Series Circuits, 14 filament-type EECO Minisig Indicators, and power converters (the beginning of our line of compact 12-volt EECO Power supplies for use with T-Series circuits) — all within a volume of ½ cubic foot. In spite of this terrific packing density, the equipment still retains extreme ease of accessibility and weighs only 20 lbs. No cooling is required. T-SERIES VS. VACUUM TUBE CIRCUITS The use of T-Series transistorized Germanium circuits throughout resulted in great savings as against equivalent equipment designed around vacuum tube circuits. Here are some startling comparisons: \| T-SERIES \| VACUUM TUBE \| \|-------------------\|--------------------------------------------------\| \| SIZE \| 800 cu. in. \| \| WEIGHT \| 20 lbs. (including power converters) \| \| POWER \| 77 watts \| \| \| 8,000 cu. in. \| \| \| 160 lbs. (plus fan and power supply) \| \| \| 650 watts (plus power for fan) \| SAVE TIME AND MONEY You, too, can develop the most complex equipment in record time with these proven EECO circuits and systems development aids. They'll save you time and money in four major areas: 1. DESIGN — You can devote full time to system design problems or unusual circuit requirements, knowing that routine circuit detail has been compatibly pre-engineered and packaged for you. 2. BREADBOARD — The unique EECO Breadboard Kit and plastic circuit cards enable you to set up, change, or take down experimental arrangements quickly — without waste of time or materials. Unit contains all necessary permanent wiring to accommodate any regular T-Series circuit. All other circuit inter-connections are made by patch cords or plugs, with prepunched circuit cards to guide you. 3. PRODUCTION — Your production problem is reduced to one of mounting sockets on panels or chassis and providing simple socket-to-socket wiring. Plug in the appropriate circuits and the system is complete. 4. CHECKOUT — The extreme reliability of T-Series circuits eliminates the need for circuit "debugging." Checkout time is reduced to a bare minimum. Why not let proven EECO T-Series circuits and systems development aids help you solve your equipment design problems? If you have not already requested your copy of our new Catalog No. 859, write us today on your company letterhead. ENGINEERED ELECTRONICS COMPANY (a subsidiary of Electronic Engineering Company of California) 506 East First Street • Santa Ana, California ANALOG COMPUTERS. Computer Systems, Inc., 611 Broadway, New York 12, N. Y. A 4-page brochure describes the company's line of general and special purpose analog computers and accessories. CIRCLE 380 ON READER SERVICE CARD REINFORCED PLASTICS. General Electric Co., Missile and Space Vehicle Department, 3198 Chestnut St., Philadelphia, Pa., has available a collection of scientific papers on the behavior of reinforced plastics at very high temperatures. CIRCLE 381 ON READER SERVICE CARD STATIC INVERTER. Magnetic Amplifiers, Inc., 632 Tinton Ave., New York 55, N. Y. Bulletin S-1035 covers the SIS3-40613 static inverter which features operating temperatures at 100 C, automatic voltage regulation, short circuit protection with automatic recovery, and reverse voltage protection. CIRCLE 382 ON READER SERVICE CARD ENVIRONMENTAL TESTING. Stavid Engineering, Inc., Plainfield, N. J., offers a 10-page brochure describing the company's environmental test laboratory. CIRCLE 383 ON READER SERVICE CARD CHOPPERS. Airpax Electronics Inc., Cambridge, Md. Series 310 choppers for operation at high temperatures are described in bulletin C-52. CIRCLE 384 ON READER SERVICE CARD TIME DELAY RELAYS. Master Specialties Co., 956 E. 108th St., Los Angeles 59, Calif. Bulletin 376-100REV covers 18 standard electronic time delay relays including factory preset, internally adjustable, and externally adjustable units. CIRCLE 385 ON READER SERVICE CARD AUDIO RESPONSE PLOTTER. Southwestern Industrial Electronics Co., 10201 Westheimer, Houston 19, Texas. Model ARP-2 audio response plotter, which provides permanent pen-written frequency response curves of any audio-range equipment, is described and WHY your own copies of ELECTRONICS WILL BE WORTH EVEN MORE in 1960 YOU WILL WANT TO KEEP A FILE OF THESE SPECIAL REPORTS FOR 1960 ELECTRONIC MARKETS — Today and the Future . . . U. S. and Export . . . Plant Concentration . . . Distribution Patterns . . . Manpower ELECTRONIC ENGINEERING Around the World . . . What's Important Now . . . What's Ahead I.R.E. SPECIAL ELECTRON TUBES Today and Tomorrow . . . A Technical Look Ahead MODERN MICROWAVES . . . Advanced Theory . . . Practical Systems . . . Components . . . Test Gear SEARCH and PROBE SYSTEMS . . . Radio Astronomy . . . Radar . . . Sonar . . . Ultrasonics . . . Infrared . . . Others . . . Military, Research and Industrial Applications WESCON SPECIAL ELECTROMECHANICAL SYSTEMS and DEVICES . . . Emphasizing Servos and Related Equipment and Components MINIATURIZATION METHODS . . . Concepts . . . Equipment . . . Components Plus, plus, plus — Serials, Foldouts, Tabulars ELECTRONICS now brings you . . . . . . latest technological and economic data and what it means to the Electronics Industry — and to you! Every Friday, your personal subscription to ELECTRONICS delivers to you an ALL-IN-ONE fact-packed manual — the easy, practical way to make sure you're among the best informed electronic engineers and managers. Specifically, each issue will cover Design, Research, Engineering, and Manufacturing . . . plus the Trends, Economic Developments, Markets, and Production Figures which interrelate with the Technical Breakthroughs and Economic Pressures. YOUR PERSONAL SUBSCRIPTION TO "ELECTRONICS" TOTALY INFORMS YOU . . . and HERE'S WHERE TO ORDER ↓ NEW subscription ORDER FORM O.K. — put me down for a personal subscription to "electronics" . . . giving me benefit of your weekly editorial service... Check here □ for 1 year $6 □ for 3 years, $12 (Save $6) Name__________________________Position________________Dept.__________ Company Name_____________________________________________________ Street________________________City_______Zone_______State___________ Products Mfrd.________________________Number of Employees________ Or Service_________________________________________________________ □ Check here if you want Publication sent to your home Street________________________City_______Zone_______State___________ (Foreign Rates (1 year) Canada $10, Other Foreign: $20 Please fill out card completely for best service HOW TO GET YOUR PERSONAL COPY of electronics ISSUED WEEKLY . . . EVERY FRIDAY ENGINEERING TECHNICAL DATA EVERY ISSUE This expanded service of ELECTRONICS provides the most comprehensive and fastest aid for the executive and engineer vitally concerned with every phase of the electronics industry... New Ideas! . . . New Circuits! . . . New Engineering Techniques! Improved Manufacturing Methods! . . . New Design! . . . Advances in Research! Prompt word about New Electronic Equipment, Parts and Components! Latest Word on New Industrial, Scientific and Military Applications! Sales and Production Figures! . . . Economic Trends to help you make sounder Investment Decisions! . . . All the Vital News! All subscribers get copies of the GIANT BUYER'S GUIDE SUBSCRIBE PERSONALLY for top-value in IDEAS and NEWS electronics Established 1930 — Pioneer Publication in the Field FIRST CLASS PERMIT No. 64 New York, N. Y. BUSINESS REPLY MAIL No postage stamp necessary if mailed in the United States McGRAW-HILL PUBLISHING CO., Inc. electronics 330 WEST 42nd STREET NEW YORK 36, N. Y. Use this handy order card... to profit by ELECTRONICS weekly editorial service L 2-12-60 illustrated in a single-sheet bulletin. CIRCLE 386 ON READER SERVICE CARD P-C RESIN FLUXES. London Chemical Co., Inc., 1535 N. 31st Ave., Melrose Park, Ill. A recent bulletin describes and illustrates results of a U.S. Signal Corps mirror test for corrosiveness of printed circuit resin fluxes. CIRCLE 387 ON READER SERVICE CARD DIRECTIONAL COUPLERS. Waveline, Inc., Caldwell, N.J. A six-page technical brochure describes microwave directional coupler design and operation. CIRCLE 388 ON READER SERVICE CARD DIGITAL COMPUTER. Autonetics, a division of North American Aviation, Inc. Industrial Products, 3584 Wilshire Blvd., Los Angeles 5, Calif., has published a 4-page bulletin describing RECOMP II, a general purpose, small size, all-transistor, single address, digital computer which solves problems of extreme complexity with split-second accuracy. CIRCLE 389 ON READER SERVICE CARD CONTROL UNIT. Wintriss, Inc., 20 Vandam St., New York 13, N.Y. A new electronic control unit that automatically stops injection molding machines when molds are closing improperly is described in detail in an illustrated bulletin. CIRCLE 390 ON READER SERVICE CARD INDUCTION HEATING. Lepel High Frequency Laboratories, Inc., Woodside, N.Y. Vol. 1, No. 6 of the Review features a story on the floating zone method for growing germanium or silicon crystals, zone refining and zone leveling. Also included is an article on coil design and construction. CIRCLE 391 ON READER SERVICE CARD MINIATURE CONNECTORS. H.H. Buggie Division, Burndy Corp., Toledo 1, Ohio. A catalog sheet illustrates and lists Bantam connectors now in production for prompt delivery of 18, 20 and 22 shell size configurations. Also discussed are modified Bantam connectors. CIRCLE 392 ON READER SERVICE CARD Look beyond the obvious... ...as you consider your future in the electronics industry. First, what is the obvious? It's obvious that you're in demand. You don't have to worry about getting your material wants satisfied. But, when you look beyond the obvious, you realize that you want something more than simple "want satisfaction" out of your career. You want pride—pride in the importance of your personal, individual contribution. At Melpar, where we are now working on 120 advanced defense and space exploration projects, we have significant opportunities for the professional engineer or scientist who wants to be proud of his contribution to advancing the state of electronic art. Senior-level positions are available in the following areas at this time: Reconnaissance Systems Detection & Identification Systems Airborne Equipment Antenna & Radiation Systems Ground Data Handling Equipment Physical Sciences Laboratory Simulation & Training Systems Production Engineering Communication & Navigation Systems Quality Control Ground Support Equipment Field Service Engineering INTERVIEWS ARRANGED IN YOUR LOCALE For details wire collect or write to: Professional Employment Supervisor MELPAR INC. A SUBSIDIARY OF WESTINGHOUSE AIR BRAKE COMPANY 3323 Arlington Boulevard, Falls Church, Virginia In Historic Fairfax County 10 miles from Washington, D.C. Dorne: 10 o'clock scholar An unique group of 10 o'clock scholars thrice weekly sets aside engineering tools, measuring devices and test equipment to return to the classroom. All on company time and premises. "We believe that all work and no study makes for rusty engineers," explains Arthur Dorne, president of Dorne and Margolin, Inc., Westbury, L. I., electronics firm which specializes in airborne antennas. "Our bringing the campus to the plant is an easy way to keep our staff up to date with space age technology." The company was organized by Dorne and a wartime colleague, Joseph Margolin, 12 years ago as a partnership. It was incorporated in 1953. Today employees are included in the stockholding group. Education—the company pays for both in-plant and on-campus education—and stock ownership are but two of the programs undertaken by the firm in an approach designed to "make this the best place to work." The 140 employees and their families also are covered by a broad, company-paid medical, health, accident and life insurance program. Dorne a graduate of the University of Pennsylvania (1939), first handled antenna design for a Washington engineering firm serving broadcast license applicants. Later he helped another firm pioneer in aircraft instrument landing systems. When the war came, he joined the radio research laboratory operated by Harvard for the government. It was here Dorne and Margolin first worked together. This team shifted after the war to Airborne Instruments Laboratory in Mineola and then pooled resources and talents to form its own firm in 1947. The company expects to gross about $2 million in the fiscal year ending next March 31. The research, development and manufacturing firm is broadening its activities to undertake contracts for electronic systems and components, Dorne says. Dorne, who now lives in Glen Head, L. I., with his wife and child, is chairman of the Antenna and Propagation professional group of the Institute of Radio Engineers. He was an early writer in the field and was among those who authored and edited a two-volume book, "Very High Frequency Techniques" (McGraw-Hill, 1948), which incorporated findings of the wartime research at Harvard. Temco Appoints Exec V-P & G-M IN A MOVE to accelerate its projected program of getting more heavily into missile, electronic and aerospace system work, Temco Aircraft Corp., Dallas, Texas, has named Clyde Skeen, executive vice president and general manager. He comes to Temco from Boeing Airplane Company's Aerospace division, where he was vice president for weapons system program management. Skeen took over Feb. 1 the day-to-day management of the company that is trying to convert from an airframe manufacturer and subcontractor into an aerospace corporation. AIL Names Two Vice Presidents AIRBORNE INSTRUMENTS LABORATORY, Deer Park, N. Y., recently announced the appointment of E. G. Fubini and G. C. Comstock as vice presidents. Both were formerly co-directors of AIL's Research and Engineering division. Fubini was named vice president in charge of Research and Systems Engineering; and Comstock, vice president in charge of the Electronic Systems and Techniques division. Both divisions are newly created within AIL. Ang Takes New Post at Mallory CHOH-YI ANG was recently appointed director of the Materials Laboratories of P. R. Mallory & Co., Inc., Indianapolis, Ind. He will direct research and development of new structural and electronic materials and processes, heat resistant and semiconducting intermetallics, neutron and gamma radiation. A member of the Mallory research and development staff since 1954, Ang has supervised investigations of tungsten alloys, copper alloys, electrical contact materials, New Ketay size 15 integrating motor tachometer— - REQUIRES NO WARM-UP TIME - USES NO HEATERS - HAS 1.0000 VOLT PER 1000 RPM OUTPUT GRADIENT OVER ENTIRE TEMPERATURE RANGE The Ketay integrating motor tachometer, Type 105P2Y has many features that assure high output-to-null voltage ratios and extreme accuracy over the entire temperature range of -55°C. to +80°C. In addition, the Type 105P2Y will give instantaneous response, for no warm-up time is required at any temperature within the operating range. The Ketay design uses no heating elements, mechanical thermostats, amplifiers or external heat sources and, as a result, the unit has increased life, less weight and less power drain on the over-all system. The unit will pass military environmental specifications called out in MIL-S-17306. CHECK THESE GENERATOR SPECIFICATIONS: \| Specification \| Value \| \|----------------------------------------------------\|------------------------\| \| Excitation (volts) \| .115 \| \| Frequency (cps) \| .400 \| \| Input current (nominal) amps \| .0275 \| \| Input power (nominal) watts \| 1.9 \| \| Output gradient per 1,000 RPM @ 25°C \| 1.0000 ± .05% \| \| Phase shift with speed to input at 3600 RPM @ 25°C\| 0° ± 10° \| \| Null voltage at 25°C (Maximum Values) \| \| \| IN-phase Fundamental at zero speed \| 2MV \| \| Quadrature Fundamental at zero speed \| 6MV \| \| Total harmonic \| 10MV (RMS) \| \| IN-phase axis error \| 1MV \| \| Quadrature axis error \| 1MV \| \| Linearity —0 to 4000 RPM (percentage of voltage output at 3600 RPM) \| ± .06% \| \| Variation in output gradient with variation in ambient temperature (-55°C. to +80°C.) \| ± 0.2% \| \| Variation in axis error with variation in ambient temperature (-55°C. to +80°C.) \| ± 7MV ± J8 MV \| \| Variation in phase shift with variation in ambient temperature (-55°C. to +80°C) \| ± 1° \| \| Warm-up time \| None \| \| Total unit weight (motor and generator) \| 15 oz. \| Write for detailed specifications and drawings, or for information on other Ketay integrating and dampening tachometers. for CONTINUOUS RELIABILITY... INSTALL HOYT PANEL METERS Quality meters on the panel indicate quality throughout—and HOYT Panel Meters are quality in appearance and function—the complete line of matching AC and DC Meters for original equipment and replacement applications. Get accuracy, readability, and reliability; plus economy. Specify HOYT Electrical Instruments—compatible components for production, research, and test requirements. Model 1035 Transparent Polystyrene Model 647 Black Bakelite Moving coil, rectifier, and repulsion types available promptly in a wide assortment of sizes, ranges, cases, shapes, and colors; some with parallax-free mirror scales—all with standard mounting dimensions. Or, custom designed to the most exacting specifications. Send for latest fully illustrated brochure with descriptions, engineering data, and moderate prices. Hoyt ELECTRICAL INSTRUMENTS BURTON-ROGERS COMPANY Sales Division—Dept. E 42 Carleton St., Cambridge 42, Mass., U.S.A. Company Ends Facilities Move AIRBORNE INSTRUMENTS LABORATORY, a Division of Cutler-Hammer, Inc., recently completed the move of all of its facilities to Deer Park and Melville, Long Island, N.Y. A 157,000-sq ft building on a 92-acre plot in Melville houses most of AIL’s research and development facilities. The 493,000-sq ft building on a 102-acre plot in Deer Park houses general administration, engineering, and production facilities. Epsco Worcester Names Sterling APPOINTMENT of Howard T. Sterling as chief engineer of the Epsco Worcester division is announced. This division, located in Worcester, Mass., manufactures a complete line of graphic recording instruments for both industry and medicine. Sterling has recently specialized in transistorized circuit development and brings with him an abundance of experience gained through 18 years in electronic and industrial instrumentation. His most recent affiliation was with Fischer and Porter Co., Hatboro, Pa., as chief electronic engineer. Prior to that, he was president and founder of Waveforms, Inc., in New York. MINIATURIZED RF COAXIAL LOBING SWITCH - WEIGHT 0.7 lbs. (SP2T model) - BROADBAND — DC to 10,000 mc - NOISE-FREE OPERATION - "MAKE" before "BREAK" operation - LIFE 1,000 hours - HERMETICALLY SEALED - CHOICE OF "N," "HN" or "SC" RF Connector - LOW COST - EXCELLENT RF CHARACTERISTICS square-wave operation $236 (quantity prices upon request) DON-LAN electronics, inc. 1101 olympic blvd., santa monica, calif. division of Regan Industries, Inc. CIRCLE 223 ON READER SERVICE CARD WHO DROPPED THE BINOCULARS? With everybody watching each other along the DEW line and the Iron Curtain these days, electronics has replaced binoculars. What's happening in the giant markets for missile controls, radar and communications equipment? electronics tells how things are going, keeps you informed of developments as they occur. This is a good time to subscribe or renew your subscription. Just fill in box on Reader Service Card. Easy to use. Postage free. FIND WHAT YOU NEED IN... electronics FEBRUARY 12, 1960 · ELECTRONICS Yardney Electric Promotes Feld Sheldon L. Feld has been named applications engineer at Yardney Electric Corp., New York City. With the firm since 1956, Feld was formerly a project engineer in design and development. He will now handle Yardney Electric sales and applications engineering in the Midwest and Southeast. News of Reps Rep firm of Cooper-Di Blasi, Port Washington, N. Y., which has served the electronics industry since 1930, will now be known as John DiBlasi Associates Inc., Personnel and address remain unchanged. The Waltham Precision Instrument Co., Waltham, Mass., has appointed R. C. Dudek & Co. of Beverly Hills, Calif., to represent it in the states of California and Arizona. Essex Electronics, division of Nytronics, Inc., Berkeley Heights, N. J., appoints George W. Meeker of Seattle, Wash., to handle its standard line of electronic components in Washington and Oregon. Mid-Eastern Electronics, Inc., Springfield, N. J., names Houser Associates of Washington, D. C., to handle its line of power supplies, ultra high resistance measuring instruments and test equipment in Virginia, Maryland, West Virginia and the District of Columbia. Appointment of Production Methods Co. of Chicago as its sales rep for a wide midwest area has been announced by The Narda Ultrasonics Corp., Westbury, N. Y., volume producer of ultrasonic cleaning equipment. Territory to be covered is Illinois, Indiana, Wisconsin and the Michigan peninsula. Central Engineering Sales Co. of Chicago, Ill., has been appointed exclusive midwest rep for Radiation Instrument Development Laboratory, Inc., Chicago. Here are the Cinema Resistors used in the nation's finest instruments. Precision wire-wound MICRO-MINIATURE RESISTORS for applications where space is at a premium. - axial and printed circuit types - temperature coefficient of resistance ±20 PPM. - for −65° to +150° C operation - epoxy encapsulation - accuracies 1% to .01% Write for Bulletin LC1066 CINEMA ENGINEERING 1100 Chestnut • Burbank, Calif. DIVISION AEROVOX CORPORATION CIRCLE 231 ON READER SERVICE CARD There’s going to be a meeting Who’s going to get together and what are they going to talk about? Electronics men are meeting all over the country to talk about everything from ultrasonics to quantum electronics. electronics tells you where and when “Meetings Ahead” gives you the highlights later on. Another reason why it will pay you to subscribe to electronics (or renew your subscription) right now. Fill in the box on Reader Service Card. Easy to use. Postage free. FIND WHAT YOU NEED IN... electronics VALUABLE 64 PAGE REFERENCE SECTION in electronics BUYERS’ GUIDE Prepared especially by the 25-man editorial staff of electronics, this 64-page section is designed to assist the buyer by providing him with market data, electronics applications, market distribution, market reports and books, industry organizations and services. electronics BUYERS’ GUIDE and Reference Issue A McGraw-Hill Publication 330 West 42nd Street New York 36, New York COMMENT Transistor Oscillators ... W. E. Roach's article "Designing High-Power Transistor Oscillators" (p 52, Jan. 8) is well organized and most informative. However, there seems to be a mistake somewhere. Step 5 on p 55 says that Fig. 4D is "drawn as a grounded collector circuit." The drawing shows a transistor with the emitter grounded... ROBERT R. RANSOM LOS ANGELES Reader Ransom picked up an obvious drafting error that author Roach had also called to our attention. There's another, less obvious, that goes with it: connection points E and C, which lead to the bias power supplies, are also switched. Kudos: No. 1 In your Dec. 18 '59 issue of ELECTRONICS, you carried a story on the Stratolab Flight #4 on p 41. This article is the most concise and accurate we have read about the flight... FRED B. FERSON FERSON OPTICAL CO. OCEAN SPRINGS, MISS. Kudos: No. 2 The feature article on microwave fault-finding equipment ("Pulse-Coded Fault Alarm in Microwave Systems," p 83, Jan. 1) was well handled. Thanks for your careful editing... J. B. BULLOCK MOORE ASSOCIATES REDWOOD CITY, CALIF. Author Bullock's article was a pleasure to handle. Kudos: No. 3 I'd like to compliment you on your clearly written article "Modern Communications Methods," p 93, Oct. 23 '59. This should give you a rough idea of how far behind I am in reading ELECTRONICS, but you know how the telephone business is—it's all over the place and you've got to go where it is... I think we need more of this sort of article, as I think we assume too often nowadays that the younger engineers have a much fuller appreciation of the art than they actually do. In fact, you've got to crawl before you can walk, so maybe it wouldn't hurt to show some of the present working systems so that when new ones come out, some of the old wrinkles that have been showing through for so long will be gone. ROBERT G. STONEMAN MILWAUKEE Profilometer In the December 18 '59 issue of ELECTRONICS ("Electronics Assists in Highway Construction," p 69), we note that you improperly used the term Profilometer. Profilometer is a trademark, registration #371,385, belonging to the Micrometrical Manufacturing Co., and is used to identify the instruments that company manufactures for measuring surface roughness. Hereafter, if you desire to use our trademark, please . . . make suitable notation . . . CHARLES H. GOOD MICROMETRICAL DEVELOPMENT CORP. ANN ARBOR, MICH. Minification I'm answering your request for takers (Comment, p 104, Jan. 22) . . . I'm with Frank Smith as regards adopting minify to replace miniaturize. This field of technology is not unique in adopting -ize words, but it's one of the worst offenders. And between -ize and -wise (as in "voltage-wise, the circuit is . . ." etc.), I've begun to believe that the lyricist of "My Fair Lady" was right when he wrote "In America they haven't spoken it (English) for years." With all the richly varied ways of expressing ideas that this language legitimately provides, I can't figure out why we have to put up with neologisms . . . R. L. MULLEN NEW YORK CITY Beattie Oscillotron DIRECT VIEW Oscilloscope Recording Camera Now, get a direct view with both eyes of the cathode ray tube while you're recording. No mirrors — full, clear vision. The new Beattie Direct-View Oscillotron is the most versatile instrument of its kind — actually three cameras in one — and the only system to offer all these important features: - Direct binocular view of CRT. - Non-reversed image. - Camera swings back for easy access to lens and shutter, or lifts off completely. Can be rotated. - Available with large Polaroid® Land back or electrically-pulsed 35mm or 70mm camera for automatic, remote operation. Adapters for film pack or sheet film available. - Continuous motion magazines available for 35mm and 70mm models. - Electric shutter actuator optional with Polaroid® Land model. - Attaches easily to bezel of 5" CRT. Adaptable to other sizes. No special tools. "Polaroid"® by Polaroid Corp. Write today for full details BEATTIE-COLEMAN INC. 1000 N. Olive St., Anaheim, California Branch: 437 Fifth Ave., New York, N.Y. EMPLOYMENT OPPORTUNITIES SYSTEMS ENGINEERS ELECTRONICS ENGINEERS CHALLENGING R & D OPPORTUNITIES Fundamental and applied research in the fields of hydrodynamics, acoustics, electronics, network theory, servomechanisms, mechanics, information theory and noise reduction. Also design of electronic instrumentation for underwater ordnance and application of analogue and digital computers. Opportunities for Graduate Study Faculty Appointments for Qualified Applicants Excellent Working and Living Conditions Send Resume to ARNOLD ADDISON, PERSONNEL DIRECTOR ORDNANCE RESEARCH LABORATORY THE PENNSYLVANIA STATE UNIVERSITY BOX 30, UNIVERSITY PARK, PA. HAMMOND ORGAN COMPANY INVITES YOU To investigate your opportunity to join the industry's recognized leader. Professional Opportunities for ELECTRONIC ENGINEERS to Express Themselves DEVELOPMENT Our company's expanding program offers opportunities for professional engineers with BSEE degrees and 6 to 8 years of experience in development of special inductors and magnetic devices or electronic circuits using transistors. Responsible assignments, skilled assistants and technicians, industrial level salaries, well equipped laboratories at your disposal and desirable working conditions complement an atmosphere conducive to the further development of personal competence and broad experience. RELOCATION ASSISTANCE, ADVANCED COMPANY BENEFITS PROGRAM. Please Forward Detailed Resume To: MR. J. J. PROSSER Manager of Personnel HAMMOND ORGAN COMPANY 4200 W. Diversey Avenue, Chicago 39, Illinois POSITION WANTED Field or Liaison Engineer Dr. of Engineering European Universities. Age 50. Wide experience R & D Power Electronics manufacturing licenses and patents. Foreign engineering connections. Speaks fluently French, Italian. Interested in overseas assignments. Best references. Present salary $14,000. PW-3610. Electronics Class. Adv. Div., P.O. Box 12, N. Y. 36, N. Y. GENERAL SALES MANAGER Seeking position in electronics companies manufacture. Proven man with solid experience selling electronic sub-systems to military and commercial customers. Experienced manager to join a well-established firm with immediate expansion plans. Ideal living conditions in a no-smog, no-dust climate. Please communicate with letter. Write: P-3554, Electronics 520 N. Michigan Ave., Chicago 11, Ill. new openings in Florida with Vitro Vitro Laboratories' Florida operation, Vitro Weapons Services, is rapidly expanding its technical staff to operate the nation's newest missile test range, the Eglin Gulf Test Range. High level supervisory and staff opportunities exist at five Florida locations: 1. Fort Walton Beach; 2. Port St. Joe; 3. Tarpon Springs; 4. Naples and 5. Key West. Electronic Engineers - with degree(s) and several years experience in Automatic Tracking Radar, Electronic Instrumentation, Phase Comparison Space Positioning Systems, Data Recording and Conversion, Telemetry or Missile Range Instrumentation. Radar Technicians - with training and experience in one or more of the following radar systems: MSQ-1, MSQ-1A, FPS-16, Nike, M-33, MPS-9, SCR-584, CPS-6, FPS-3 and FPS-20. Electronic Technicians - with a solid background in electronics and several years experience in Telemetry, Data Converters, Oscillograph Recorders or Range Electronic Instrumentation. For your opportunity to relocate in Florida with an electronics industry leader, address a confidential resume to D. D. Cox, Personnel Director, Vitro Weapons Services, 119 East Main Street, Fort Walton Beach, Florida Dept. AW. Vitro LABORATORIES DIVISION OF VITRO CORPORATION OF AMERICA Other laboratories at West Orange, N. J., and Silver Springs, Md. FEBRUARY 12, 1960 • ELECTRONICS Expanding Computer Memory to Hold One Million Bits of Information Some time ago IBM scientists were handed a knotty problem: how to develop a magnetic drum memory (for a system similar to the 650 computer) that retains present speeds and access time but multiplies storage capacity. Starting with the existing system—operated at 125 kc., 50 bits per inch, and 20 tracks per inch—the engineers aimed at a memory of 500 kc., 200 bits per inch, and 40 tracks per inch. After months of work, they achieved their goal with advanced transistor circuitry and an improved read-write head which affords greater bit and track densities. The result: Drum capacity has been expanded eight times to an unprecedented one million bits. Although this accomplishment satisfies an immediate need, work is in progress to develop magnetic memories of even superior capabilities. Several positions are available on the development team assigned to this project. In addition, interesting opportunities exist on projects involving microwaves, optics, semiconductors, inertial guidance, and human factors engineering. If you have experience in these fields and a degree in engineering, mathematics, or one of the sciences, you may find the career you've been looking for in IBM systems development. For more information, write—outlining your background and interests—to: Manager of Technical Employment, Dept. 554N2 IBM Corporation 590 Madison Avenue, New York 22, N.Y. grow with kollsman! Opportunities in GROUND SUPPORT EQUIPMENT PROJECT ENGINEERS—EE & ME Experience in ground support equipment built to military specifications. ENGINEERS—EE Some experience desired in ground support equipment. SEMICONDUCTOR ACTIVITIES PHYSICISTS or CHEMISTS Significant educational and experience background in solid state physics. For participation in the following areas of research and development. - Photoelectronics - Magnetics (ferrites, magnetic films, etc.) - Dielectrics - Semiconductor Materials (as applied to diodes, transistors & Hall-effect devices) - Microminiaturization Kollsman is seeking a limited group of exceptional men to participate in its continuing growth in the field of automatic navigation and flight instrumentation. These openings offer unusual opportunity with an organization intimate enough to allow individual recognition, yet large enough to assure stability. Please send resumes to T. A. DeLuca Honeywell First in Controls RATES "Employment Opportunities" Displayed—The advertising inch is $34.33 per inch for all advertising appearing on other than a contract basis. Contract rates quoted on request. An advertising inch is measured 7/8" vertically on a column—3 columns—30 inches to a page. Subject to Agency Commission. Undisplayed—$2.40 per line, minimum 3 lines. To figure advance payment count 5 average words as a line. Discount of 10% if full payment is made in advance for 4 consecutive insertions. Not subject to Agency Commission. Send NEW ADS to ELECTRONICS, Class. Adv. Div., P. O. Box 12, New York 36, N. Y. MANUFACTURERS' REPRESENTATIVES IN THE ELECTRONIC INDUSTRY SAMUEL K. MACDONALD, INC. manufacturers representatives over 25 years 1531 SPRUCE STREET, PHILA. 2, PA. Territory: Pennsylvania • New Jersey Delaware • Maryland Virginia • West Virginia District of Columbia Other Offices: Pittsburgh Baltimore Washington, D.C. Announces NEW HEADQUARTERS UNDER CONSTRUCTION IN SUBURBAN BOSTON MITRE'S new headquarters will serve a Technical Staff that has established, and is daily adding to, its reputation in large-scale systems design and development. Located in Bedford, Massachusetts it affords engineers and scientists an opportunity to pursue long-term, challenging work programs with an expanding organization in an environment of professional and cultural excitement. If you possess a baccalaureate or advanced degree in engineering, mathematics or the physical sciences, you are invited to inquire about career opportunities in these areas: - SYSTEM DESIGN - COMPONENT DEVELOPMENT - RADAR SYSTEMS - OPERATIONS ANALYSIS - WEAPONS SYSTEM INTEGRATION - COMMUNICATION SYSTEMS - ELECTRONIC WARFARE - INTEGRATED SYSTEM EVALUATION - COMPUTER PROGRAMMING - PROGRAM LOGISTICS To arrange an immediate confidential interview, direct your inquiries to: Dana N. Burdette, Personnel Director, Dept. 17-WS THE MITRE CORPORATION 244 Wood Street – Lexington 73, Massachusetts SEARCHLIGHT SECTION (Classified Advertising) BUSINESS OPPORTUNITIES EQUIPMENT - USED or RESALE DISPLAYED The advertising is $24.75 per inch for all advertising other than on a contract basis. AN ADVERTISING INCH is measured 3/8" vert. and 1-1/2" horiz., or 30 inches in all. EQUIPMENT WANTED OR FOR SALE ADVERTISEMENTS acceptable only in Displayed Style. Send NEW ADVERTISEMENTS or Inquiries to Classified Adv. Div. of Electronics, P. O. Box 12, N. Y. 36, N. Y. The publisher cannot accept advertising in the Searchlight Section, which lists the names of the manufacturers of resistors, capacitors, rheostats, and potentiometers or other names designed to describe such products. UNDISPLAYED $2.40 a line, minimum 3 lines. To figure advance payment count 5 average words as a line. BOX NUMBERS count as one line additional. DISCOUNT of 10% if full payment is made in advance for four consecutive insertions. --- RELAYS Thousands of Different Types—Most Makes LATEST SUB-MINIATURE HERMETICALLY SEALED RELAYS - SIGMA 22RJ AND SIMILAR - SP & DP SENSITIVE SEALED TYPES Light weight; extremely small; hermetically sealed in solder dipped brass can; 3 oz. max. weight; 6 relay contacts rated at 2 amp. Solder lug connections. SIGMA 22RJ TYPES \| Signal # \| Ohms \| Contracts \| Oper. \| Rel. Dif. \| Min. Stk. \| Univ. Stk. \| Price Ea. \| \|----------\|------\|-----------\|-------\|-----------\|-----------\|------------\|---------\| \| 22RJAS00G \| 500 \| 2C \| 9.0 \| 3.1 \| 3.0 \| R618 \| 7.50 \| \| 22RJC500G \| 500 \| 2C \| 9.0 \| 3.1 \| 3.0 \| R830 \| 7.75 \| \| 22RJC500G \| 500 \| 2C \| 9.0 \| 3.1 \| 3.0 \| R619 \| 8.50 \| \| 22RJC800G \| 800 \| 2C \| 2.3 \| 0.8 \| 0.8 \| R617 \| 8.50 \| SIGNAL ENGINEERING TYPES \| Signal # \| Ohms \| Contracts \| Oper. \| Rel. Dif. \| Min. Stk. \| Univ. Stk. \| Price Ea. \| \|----------\|------\|-----------\|-------\|-----------\|-----------\|------------\|---------\| \| 100x10 \| .054 \| 1A \| 2000 \| .45 \| \| \| \| \| 100x11 \| 1C \| 100 \| 1.0 \| \| \| \| \| \| 100x12 \| 1C \| 100 \| 1.0 \| \| \| \| \| \| 100x13 \| 250m \| 1C \| 1.0 \| 1.4 \| \| \| \| \| 100x14 \| 250m \| 2C \| 2.3 \| 0.8 \| 0.8 \| \| \| MICRO-MINIATURE RELAYS Extremely small relays for printed circuit and other circuit applications. These are designed to meet the requirements of MIL-R-106 and 2757. Temp. range: rated 100° F. to 125° C. Operating shock 100G. Vibration 5G up to 500 cps. Mechanical shock 175G. Contact rating: 2 amps resistive at 28 VDC or 115 VAC. All relays listed are DPDT. WRITE OR PHONE WALKER 5-9257 FOR YOUR RELAY REQUIREMENTS Among the wide variety of RELAYS • SOLENOIDS • CONTACTORS • STEPPERS, ETC. We have in stock, we can usually fill every need. When inquiring, give as much information as possible. ALL MERCHANDISE IS GUARANTEED AND MAY BE RETURNED FOR FULL CREDIT Prices listed with asterisk () are subject to QUANTITY DISCOUNTS 1-9 ...........as quoted 50-99 ............15% 10-19 ..........10% over 100 ............20% Prices subject to change without notice SEND FOR CATALOG EB CAPACITORS OIL—Fm 50 V. to 100 KV MICA—All Types GLASSEAL—Complete Selection CAPACITOR SPECIAL 1 mfd—25 KV @ $59.50 120 mfd—3 KV @ 49.50 9 mfd—10 KV @ 85.00 ALSO "V" & "J" type potentiometers @ $.75 POWER RESISTORS & RHEOSTATS QUOTATIONS RENDERED MONMOUTH RADIO LABS BOX 159 — OAKHURST, N. J. CAPITAL 2-0121 ART HANKINS, Owner CONTACTS FOR THE FIELD OF ELECTRONICS "How to Make Money in Mobile Radio Maintenance" AUTHORITATIVE GUIDEBOOK ABOUT TOOGAN IN THE FIELD OF MOBILE RADIO. GIVES FACTS, FIGURES, PAY RATES, ETC. FREE LAMPKIN LABORATORIES, INC Instruments On BRADENTON, FLA. CIRCLE 461 ON READER SERVICE CARD SEARCHLIGHT Equipment Locating Service NO COST OR OBLIGATION This service is aimed at helping you, the reader of "SEARCHLIGHT", to locate Surplus new and used electronic equipment and components not currently advertised. (This service is for USER-BUYERS only). How to use: Check the dealer ads to see if what you want is not currently advertised. If not, send us the specifications of the equipment wanted on the coupon below, or on your own company letterhead to: Searchlight Equipment Locating Service c/o ELECTRONICS P. O. Box 12, N. Y. 36, N. Y. Your requirements will be brought promptly to the attention of the equipment dealers advertising in this section. You will receive replies directly from them. CIRCLE 460 ON READER SERVICE CARD CIRCLE 462 ON READER SERVICE CARD Your Inquiries to Advertisers Will Have Special Value... ...for you—the advertiser—and the publisher, if you mention this publication. Advertisers are especially this evidence that the publication is read. Satisfied advertisers enable the publishers to secure more advertisers—and—more advertisers mean more information about more products or better service—more value—to YOU. INDEX TO ADVERTISERS - AC Electronics ........................................ 17 - AMP Incorporated .................................... 44 - Ace Electronics Associates, Inc. .................. 157 - Acme Electric Corporation .......................... 143 - Acton Laboratories ................................... 206 - Aeronautical Communications Equipment, Inc. .... 50 - Aerovox Corp. ......................................... 131 - Air Express Division of Railway Express .......... 18 - Allied Control Co., Inc. ............................. 71 - Allied Radio ........................................... 179 - Alpha Wire Corp. ...................................... 171 - American Time Products, Inc. ....................... 2 - Amperite Co., Inc. .................................... 161 - Ampex Data Products Co. ............................ 35 - Applied Research, Inc. ............................... 151 - Astron Corp. .......................................... 163 - Atlee Corp. ............................................ 152 - Audio Development Co. ............................... 144 - Avon Gear and Engineering Co. ...................... 161 - B & H Instruments Co., Inc. ........................ 168 - Baird-Atomic, Inc. .................................... 143 - Ballantine Laboratories, Inc. ....................... 49 - Beattie Coleman Inc. ................................ 197 - Bendix Aviation Corp. Red Bank Transistors ....... 56 - Boesch Mfg. Co., Inc. ................................. 176 - Bourne, Inc. .......................................... 33 - Bristol Company, The ................................. 124 - Bruning Co., Inc., Charles .......................... 138 - Burroughs Corp. ....................................... 12, 13 - CBS Electronics ........................................ 23 - Cambridge Thermionic Corporation .................. 153 - Cannon Electric Co. .................................. 8, 9 - Chart-Pak, Inc. ....................................... 21 - Cinch Mfg. Co. ........................................ 119 - Cinema Engineering .................................... 195 - Cleco Div. of Reed Roller Bit Co. .................. 15 - Clevite Transistor, Div. of Clevite Corp. .......... 43 - Coast Pro-Seal & Mfg. Co. ............................ 30 - Communication Accessories Co. ...................... 165 - Computer-Measurements Corp. ....................... 6 - Continental Diamond Fibre Subsidiary of the Budd Company ........................................ 62 - Cornell-Dubilier Electric Corp. ..................... 73 - Coto-Coil Co., Inc. ................................... 196 - Cubic Corp. ............................................ 74 - Dale Products, Inc. ................................... 51 - Delta Air Lines ........................................ 183 - Don-Lan Electronics, Inc. ............................ 194 - Dumont Laboratories Inc., Allen B. ................. 32 - Dynacore Div., Sprague Electric Co. ................. 187 - Eitel-McCullough, Inc. ................................ 24 - Electra Mfg. Co. ...................................... 7 - Electro Instruments, Inc. ............................ 28 - Electro Motive Mfg. Co., Inc. ....................... 68 - Electro Scientific Industries ......................... 180 - Electronic Engineering Co. of Calif. Engineered Electronics Co. Sub. ......................... 188 - Electronic Instrument Company (EICO) .............. 197 - Electronic Measurements Company, Inc. ............. 52 - Electronics ............................................. 139 - Englehard Industries, Inc. ........................... 58, 59 - Esterline-Angus Company ............................. 187 - Fairchild Semiconductor Corp. ....................... 69 - Garrett Corporation, The ............................. 178 - General Electric Co. Apparatus Dept. ............... 151 Receiving Tubes ........................................ 38, 39 Defense Electronics Div. ............................. 175 - General Instrument Corp. Semiconductor Div. ....... 137 Distributor Div. ...................................... 136 - Grayhill, Inc. ......................................... 161 - Gyra Electronics Corporation ......................... 169 - Hallicrafters Company, The .......................... 121 - Helipat Div. of Beckman Inst., Inc. ................. 158 - Heiland Div. Minneapolis-Honeywell .................. 16 - Hoskins Mfg. Co. ...................................... 128 - Hoyt Electrical Instruments .......................... 194 - Hughes Aircraft Co. ................................... 27 See advertisement in the June, 1959 Mid-Month ELECTRONICS BUYERS GUIDE for complete line of products or services. \| Company Name \| Page \| \|--------------------------------------------------\|------\| \| Instruments for Industry, Inc. \| 54 \| \| International Electronic Research Corp. \| 148 \| \| International Rectifier Corp. \| 19 \| \| International Resistance Co. \| 3rd Cover \| \| J F D Electronics Corp. \| 135 \| \| Kay Electric Co. \| 60 \| \| Kearfott—Division of General Precision Inc. \| 34 \| \| Kip Electronics Corp. \| 53 \| \| Kyoritsu Electrical Instrument Works. Ltd. \| 143 \| \| Lambda Electronics Corp. \| 3 \| \| Laminated Shim Co. \| 29 \| \| Lampkin Laboratories, Inc. \| 6 \| \| Lapp Insulator Co., Inc. \| 41 \| \| Lepel High Frequency Laboratories, Inc. \| 159 \| \| Lindberg Engineering Co. \| 132, 133 \| \| Link Aviation \| 186 \| \| Littlefuse \| 159 \| \| Lomatron Electronics, Inc. \| 26 \| \| MacDonald Inc., Samuel K. \| 200 \| \| Magnecraft Electric Co. \| 204 \| \| Magnetic Amplifiers, Inc. \| 129 \| \| Magnetic Controls Company \| 185 \| \| Magnetic Metals \| 149 \| \| Malec Mfg. Co. \| 136 \| \| Marconi Instruments, Ltd. \| 142 \| \| Martin Co., The. \| 167, 169, 171, 173, 175, 177 \| \| McGraw-Hill Book Company \| 184 \| \| Melpar, Inc. \| 191 \| \| Microwave Associates, Inc. \| 154 \| \| Millen Mfg. Co., James \| 153 \| \| Minneapolis-Honeywell Regulator Co. Boston Division \| 178 \| \| Minnesota Mining & Mfg. Co. \| 57 \| \| Monadnock Mills \| 182 \| \| N J E Corp. \| 10 \| \| Narda Microwave Corporation \| 101 \| \| National Company \| 196 \| \| National Radio Co., Inc. \| 141 \| \| New Hermes Engraving Machine Corp. \| 177 \| \| Non-Linear Systems, Inc. \| 63 \| \| Norden-Div. of United Aircraft Corp. \| 183 \| \| North Atlantic Industries, Inc. \| 181 \| \| Northern Radio Co., Inc. \| 167 \| \| Oster Mfg. Co., John \| 66 \| \| Ozalid—Division of General Aniline & Film Corp.\| 72 \| \| Panoramic Radio Products, Inc. \| 31 \| \| Philco Corp. Government & Industrial \| 65 \| \| Polytechnic Research & Development Co., Inc. \| 165 \| \| Potter Instrument Co. \| 67 \| \| Power Sources, Inc. \| 153 \| \| Radio Corporation of America \| 4th Cover \| \| Semiconductor Products \| 183 \| \| Ramo Wooldridge, Div. of Thompson-Ramo-Wooldridge Inc. \| 20 \| \| Raytheon Company \| 64 \| \| Rye Sound Corp. \| 197 \| \| Sangamo Electric Co. \| 127 \| \| Sarkes Tarzian, Inc. \| 174 \| \| Segal, Edward \| 159 \| \| Sensitive Research Instrument Corp. \| 70 \| \| Shamban & Co., W. S. \| 61 \| \| Sierra Electronic Corp. \| 147 \| \| Somatone Corp. \| 164 \| \| Sorensen & Co. \| 150 \| \| Sprague Electric Co. \| 5 \| \| Stackpole Carbon Co. \| 145 \| \| Standard Electric Time Co. \| 162 \| \| Superior Cable Corp. \| 146 \| \| Sylvania Electric Products, Inc. \| \| \| Electronic Div. (Semiconductors) \| 16 \| \| Parts Div. \| 102 \| \| Syntronic Instruments, Inc. \| 206 \| \| Technology Instrument Corp. \| 160 \| \| Thermal American Fused Quartz Co. \| 48 \| \| Transitron Electronic Corp. \| 25 \| \| Trio Laboratories, Inc. \| 55 \| \| Tung-Sol Electric, Inc. \| 123 \| \| U. S. Sonics Corporation \| 125 \| \| See advertisement in the June, 1959 Mid-Month ELECTRONICS BUYERS GUIDE for complete line of products or services. \| WALLSON 20 AMPERE DYNAMIC RECTIFIER ANALYZER* Model 141A* FOR - INCOMING INSPECTION - ON-LINE INSPECTION - LABORATORY USE This dynamic rectifier test set, with independent forward current and reverse voltage controls, is completely self-contained and measures average forward voltage drop and reverse current of any type of semi-conductor rectifier rated to 20 amperes forward current and 1000 volts PIV, in accordance with proposed JEDEC specifications. OTHER WALLSON PRODUCTS - Automatic High Vacuum Exhaust Equipment - TWT Power Supplies - Dynamic & Life Test Equipment for Semi-conductor Rectifiers with Ratings up to 500 Amperes and 2500 Volts Wallson also produces a 5 Ampere Rectifier Analyzer SURGE TEST UNIT AVAILABLE WRITE TODAY* For Free Descriptive Literature WALLSON ASSOCIATES, INC. 912-914 WESTFIELD AVE. ELIZABETH, NEW JERSEY Flanders 1-0700 CIRCLE 205 ON READER SERVICE CARD 205 TIME-SAVING GUIDE FOR... SPECIFYING DEFLECTION YOKES Helps speed your project. Eliminates confusion in choosing the right yoke. Engineers have saved countless hours, many dollars and numerous headaches by using this simple Guide Sheet For Specifying Deflection Yokes. Offered as a public service to engineers by SYNTRONIC INSTRUMENTS, INC., YOKE SPECIALISTS, the only firm devoted primarily to deflection yoke manufacture; therefore preeminently qualified to help you specify the correct yoke for your application. Complete line for every military and special purpose—in production quantities or custom designed to your specific requirement. Phone our nearest rep. today for your Time-Saving Guide Sheet. New York Area: Jules J. Bressler Co. Phone: N.Y. Oxford 5-0255; N.J. Union 4-9577 Philadelphia Area: Massey Associates Phone: MOhawk 4-4200 Washington-Baltimore Area: Massey Associates Phone: GRahite 4-2071 Indianapolis: Joe Murphy Phone: Victor 6-0359 Los Angeles: Ash M. Wood Co. Phone: CUMberland 3-1201 No obligation. We are glad to help you. syntronic INSTRUMENTS, INC. 100 Industrial Road, Addison, Illinois Phone: Kingswood 3-6444 CLASSIFIED ADVERTISING F. J. Eberle, Business Mgr. EMPLOYMENT OPPORTUNITIES 198-201 EQUIPMENT (Used or Surplus New) For Sale .......................... 202, 203 WANTED Equipment .......................... 203 ADVERTISERS INDEX Barry Electronics, Corp.............. 203 Blan—The Radio Man, Inc............ 203 Engineering Associates ................ 203 Hammond Organ Company ............. 198 Honeywell .......................... 200 International Business Machines Corp.. 199 Kollsman Instrument Corp., sub. of Standard Coil Products, Inc......... 200 Legri S. Company .................. 203 Liberty Electronics, Inc............. 203 Lockheed Electronics Co. Stavid Division .................. 200 Mitre Corporation .................. 201 Monmouth Radio Labs ............... 202 Pennsylvania State University ....... 198 R. W. Electronics .................. 203 Universal Relay Corporation .......... 202 Vitro Weapons Services ............. 198 Western Engineers .................. 203 See advertisement in the June, 1959 Mid-Month ELECTRONICS BUYERS GUIDE for complete line of products or services. This Index and our Reader Service Numbers are published as a service. Every precaution is taken to make them accurate, but ELECTRONICS assumes no responsibilities for errors or omissions. Another FIRST from ALI NEW PACKAGED PRECISION DRIVE UNITS Now, for the first time you can obtain precision drive units complete in one package. ALI Drives save time, trouble and money—eliminate the need of procuring and assembling separate indicators, dials, knobs, gears, etc. And, with ALI Precision Drive Units you know you have a unit that will do the required job without a possibility of error. Use them with rotary components such as resolvers, syncros, potentiometers, inductors, capacitors, tuning coils, etc. REQUIRE NO ADDITIONAL PARTS TYPE PDW-1 PRECISION DRIVE, WORM . . . Accuracy 1 min. of arc A worm gear, shaft positioner for accurate repeatable positioning and indicating. Features: no backlash, compact design, long life, rugged construction. Mechanism utilizes a 180-1 ratio allowing direct dial readings of 1 minute of arc. Has hairline indicator, large engraved dials. ARE COMPLETE UNITS IN THEMSELVES TYPE DSD-2 DUAL SPEED DRIVE . . . Accuracy 6 mins. of arc A precision gear reducer for highly-accurate repeatable positioning and indicating. Inner dial for coarse positioning, outer dial for fine positioning plus in-line input and output shafts and the availability of a wide range of ratios through 72-1. No backlash. 3 TYPES AVAILABLE all meet Mil-E-4970 (USAF) TYPE DSD-3 DUAL SPEED DRIVE . . . Accuracy 12 mins. of arc An economical precision gear reducer for positioning and indicating. Features: no backlash, in-line input and output shafts, compact design. Like Type DSD-2, it has inner dial for coarse positioning, outer dial for fine positioning. Write for full details ACTON LABORATORIES INC. Subsidiary of Technology Instrument Corp. 517 MAIN STREET, ACTON, MASS. COLONIAL 3-7756 CIRCLE 229 ON READER SERVICE CARD FEBRUARY 12, 1960 • ELECTRONICS NO DERATING with IRC Resisteg Coated Power Resistors Exclusive RESISTEG COATING accounts directly for the ability of IRC Power Resistors to operate at full rated power—even at high resistance values. Resisteg Coating is cured at less than 205°F. This is more than 1000° lower than is required for other power resistor coatings. With Resisteg low-temperature curing there is no tendency for wire turns to shift, no necessity for tight windings, no hot spots from arcing-over, no appreciable change in temperature coefficient or resistance. Resisteg Coating permits the use of close spacing, large wire diameter, and maximum number of turns. This increases the transfer of heat from the interior of the IRC resistor to the terminals—providing a safety margin for surges and minimizing any need to derate at high ambient temperatures. For the “inside story” of power resistors, write for Bulletin C-1C. IRC Resisteg Resistor □ Comparative wattages of vitreous resistors in popular 20 watt size. RCA MESA COMPUTER TRANSISTORS 2N1300 • 2N1301 feature HIGH POWER DISSIPATION FAST-SWITCHING TIMES AT LOW COST AVAILABLE NOW... IN QUANTITY \| RCA TYPE \| Collector-to-Base Volts \| Collector-to-Emitter Volts \| Emitter-to-Base Volts \| Collector Mo. \| Transistor Dissipation Milliwatts \| \|----------\|-------------------------\|----------------------------\|-----------------------\|---------------\|----------------------------------\| \| \| \| \| \| \| at 25°C \| at 55°C \| at 71°C \| \| 2N1300 \| -13 \| -12 \| -1 \| -100 \| 150 \| 75 \| 35 \| \| 2N1301 \| -13 \| -12 \| -4 \| -100 \| 150 \| 75 \| 35 \| For collector mo = -10 and collector-to-emitter volts = -3 Characteristics: Common-Emitter Circuit, Base Input—Ambient Temperature = 25°C \| RCA TYPE \| Minimum DC Current Transfer Ratio \| Gain-Bandwidth Product Mc \| \|----------\|----------------------------------\|---------------------------\| \| 2N1300 \| at collector mo = -10 \| 30 \| \| 2N1301 \| at collector mo = -40 \| 40 \| RCA’s Germanium P-N-P Mesa Transistors 2N1300 and 2N1301 combine low-cost and quantity availability with these major benefits for designers of switching circuits: - high power dissipation—150 milliwatts maximum at 25°C, 75 milliwatts maximum at 55°C - fast switching times—made possible by high frequency response and low total stored charge - rugged Mesa structure—with an extremely small base width to insure top performance at high frequencies - high current transfer ratio—permits high fanout ratios (number of paralleled similar circuits per driver-stage output) - high breakdown-voltage and punch-through voltage ratings—result of the diffusion process - high current ratings—improves overall system speed - especially well suited for use at pulse repetition rates up to 10 Mc - rugged overall design—units have unusual capabilities to withstand severe drop tests and electrical overloads - electrical uniformity—a result of the diffused-junction process used by RCA in the manufacture of Mesa Transistors Contact your RCA Field Representative for prices and delivery. For technical data, see your HB-10 Semiconductor Products Handbook, or write RCA Commercial Engineering, Section B-19-NN-2, Somerville, N. J. RCA FIELD OFFICES East: 744 Broad St., Newark, N. J. HUMboldt 5-3900 Northeast: 64 "A" Street, Needham Heights 94, Mass. Hillcrest 4-7200 East Central: 714 New Center Bldg., Detroit 2, Mich. TRinity 8-2000 Central: Suite 1154, Merchandise Mart Plaza, Chicago 54, Ill., WHITEhall 4-2900 West: 6355 E. Washington Blvd., Los Angeles 22, Calif., Raymond 3-8361 Southwest: 7905 Empire Freeway, Dallas 7, Texas Fleetwood 2-8663 Gov'ts: 224 N. Wilkinson Street, Dayton, Ohio BALDwin 6-3366 1625 "K" Street, N.W., Washington, D.C. District 7-1260 Also available through your local RCA Semiconductor Distributor.
N-gram based Language Identification of Individual Words Oluwapelumi Giwa and Marelie H. Davel Multilingual Speech Technologies, North-West University, Vanderbijlpark, South Africa firstname.lastname@example.org, email@example.com Abstract—Various factors influence the accuracy with which the language of individual words can be classified using n-grams. We consider a South African text-based language identification (LID) task and experiment with two different types of n-gram classifiers: a Naïve Bayes classifier and a Support Vector Machine. Specifically, we investigate various factors that influence LID accuracy when identifying generic words (as opposed to running text) in four languages. These include: the importance of n-gram smoothing (Katz smoothing, absolute discounting and Witten-Bell smoothing) when training Naïve Bayes classifiers; the effect of training corpus size on classification accuracy; and the relationship between word length, n-gram length and classification accuracy. For the best variant of each of the two sets of algorithms, we achieve relatively comparable classification accuracies. The accuracy of the Support Vector Machine (88.16%, obtained with a Radial Basis function) is higher than that of the Naïve Bayes classifier (87.62%, obtained using Witten-Bell smoothing), but the latter result is associated with a significantly lower computational cost. Index Terms: text-based language identification, smoothing, character n-grams, Naïve Bayes classifier, support vector machine. I. INTRODUCTION Code switching is the act of mixing words from different languages within a single sentence. In running text, words occurring side-by-side originally may have come from different languages; that is, in a code-switched sentence or phrase, it is typical to find individual words from one language (referred to as the embedded language) embedded within a larger sentence of a different language (referred to as the matrix language) [1]. In regions with multiple languages, it is a common act amongst younger generations to use mixed language within text conversations [2]. A specific scenario is the use of numerical digits, where people often prefer using numbers from a different language to the matrix language. In speech and text processing systems, identifying code-switched words is important for applications such as machine translation, speech synthesis, information extraction and pronunciation prediction. For example, Bhargava and Kondrak [3] showed that text-based language identification (LID) can be used to improve the accuracy of grapheme-to-phoneme conversion, and both Font Llitjos and Black [4] and Church [5] demonstrated that being able to identify the language of origin of a name (from its orthography) is important in being able to predict the possible pronunciations of that name. In related work, Basson and Davel [6] showed that the ability to identify code-switched words from orthography alone can be a useful step in building optimised grapheme-based automatic speech recognition (ASR) systems. We approach the LID task using n-grams. N-gram based methods are widely used ([7], [8], [9]), and highly suitable for LID with small training sets and short test samples [9]. Botha and Barnard [10] applied Naïve Bayes (NB) classification and Support Vector Machines (SVMs) using n-grams to 11 South African languages, using segments of 15 to 300 characters each. While smoothing was not investigated in [10], Vatanen et al. [9] demonstrated the importance of smoothing when using n-gram models to classify short text segments. The current study extends previous contributions by comparing the two classification techniques mentioned above (SVMs and NB classifiers) for the classification of individual words rather than general text segments; and by analysing the effect of smoothing, specifically. In this context, the relationship between word length, n-gram length and classification accuracy is also investigated. The paper is structured as follows: Section II provides an overview of different LID techniques. Section III describes the methods used during experimentation in more detail. Section IV provides an overview of the experimental approach followed and describes the data set used. Section V presents the various experiments and results. A summary of main findings concludes the paper in Section VI. II. BACKGROUND In principle, the language origin of an input text string can be estimated by creating a model per language (from training data) and selecting the best-fitting model to predict the language source of the input string. Statistical text-based LID techniques include: Naïve Bayes classification [10], ranking methods [7], Markov models [8], [11], support vector machines [12], decision trees [13] and k-nearest neighbour classification [14]. Many LID experiments have adopted character n-gram models; such techniques have demonstrated good performances over a variety of applications. To the best of our knowledge, limited previous work has focused on identifying the language of generic words in isolation, with more results available with regard to LID of running text. (Exceptions being [15], [13], [3], discussed below.). When classifying longer text segments, accuracy quickly approaches 100% given enough text; for example, Cavnar et al. [7] used rank difference to predict the distance between the most frequent n-gram in the language model and the test document. They extracted their evaluation set from Usenet newsgroup articles written in 14 different languages. They achieved an accuracy of 99.8% on text of 300 characters or more, while retaining the first 400 most common n-grams up to length 5. In related work, Kruengkrai et al. [12] showed a similar result when classifying 17 languages with average length of 50 bytes, while ignoring character-encoding system during processing (that is, irrespective of the number of characters, 50 bytes of data were used). They achieved an accuracy of 99.7% with an SVM classifier. Classification of a short textual fragment is more complex due to the lack of contextual information. Short text segments include proper names, generic words in isolation and very short sentences (less than approximately 15 characters). Vatanen et al. [9] used the Cavnar ranking method and SVMs to identify short text segments. They experimented with 281 languages using a fairly small training set, and for test samples within the range of 5-21 characters, they obtained an accuracy of less than 90%. Similarly, Bhargava and Kondrak [3] used SVMs to classify proper names while training on a small data set of 900 names and testing on 100 names. They obtained their best identification rate of 84% using a support vector machine with a radial basis function (RBF). Not all methods can be applied to words in isolation, with linguistic models (such as the stop words used by Johnson [16] or the closed grammatical classes used by Lins and Gonçalves [17]) not applicable to this task. One technique that is not n-gram based that is worth mentioning, is the use of a data compression model for LID, as introduced by Hatlegan et al. [15]. They evaluated the performance of the algorithm on individual names and isolated words from 6 European languages, and reported an accuracy of above 80% on two-best results. N-gram based method has been compared directly to other LID approaches in a number of studies. Hakkinnen and Tien [13] compared a decision tree and n-gram methods. They concluded that the n-gram based method perform better on longer text samples while decision trees do better on short words like proper names. They also emphasised that the decision tree method does well with learning lexical structure information. In recent work, Baldwin and Lui [18] used SVMs and the Naïve Bayes algorithm for LID. Their experiment was carried out on 17 European languages from Wikipedia. They observed that SVMs performed better on short text segment and concluded that the shorter the text the more difficult it is to classify. Similarly, Vatanen et al. [9] experimented with two classifiers and smoothing techniques in identifying short text segments. Their reports show that Naïve Bayes classification outperformed a ranking method on sample text length in the range of 5 to 21 characters. To increase identification accuracy they test different smoothing techniques such as Katz smoothing, absolute discounting, and modified Knæser-Ney discounting. They observed the best result with absolute discounting. Apart from the sample text length, the accuracies of these approaches depend on various other factors. Botha and Barnard [10] discussed different factors that could influence text-based LID accuracy. These factors included: size of the training data, input text size, n-gram size, LID techniques employed and language similarities. III. METHODS In this section, we describe the n-gram based classifiers and smoothing techniques employed in further experiments. A. LID using a Naïve Bayes classifier A Naïve Bayes classifier uses the concept of Bayes’ theorem [19]. This classifier assigns the most likely class to an input string, based on the highest a posteriori probability, given the input string. For T-LID, a Naïve Bayes classifier can be constructed using n-grams as features. Let T be a set of training samples and let each sample be represented by $n$ feature vectors, $X = x_1, x_2, ..., x_n$, with their class labels. Let there be m classes: $K_1, K_2, ..., K_m$. To predict, a sample X is selected to belong to class $K_i$, if and only if: $$P(K_i \mid X) > P(K_j \mid X); \text{for } 1 \leq j \leq m; \ j \neq i$$ \hspace{1cm} (1) where $P(K_i \mid X)$ is the probability of a class $K_i$ given a sample. Bayes’ theorem states that: $$P(K_i \mid X) = \frac{P(X \mid K_i)P(K_i)}{P(X)}$$ \hspace{1cm} (2) where $P(X \mid K_i)$ represents the likelihood of a sample X belonging to class $K_i$, and $P(X)$ does not influence model comparison. The class a priori probability, $P(K_i)$, represents the count relative frequency in the sample set. According to the Naïve Bayes assumption, statistical independence of features is assumed, and the class $K_i$ is selected such that $\prod_j P(x_j \mid K_i)P(K_i)$ is optimised, where $P(x_j \mid K_i)$ is then the likelihood of a specific n-gram being observed in a given language, and the word being classified consists of $j$ n-grams. B. Support Vector Machines Support vector machines estimate a linear hyper-plane, which separates two binary classifiers while maximising the distance from the hyper-plane to the class samples. It was first introduced by Vapnik [20]. Data normalisation or scaling is a sensitive part of SVM training and testing. Normalisation is a way of reducing the weight of frequent n-gram counts by preventing larger n-gram counts from dominating smaller n-gram counts. Various benefits are associated to normalising data, which include speeding up training and avoiding computational complexity with numerical values. For further details on which SVM libraries were used, the size of n-gram models and how we normalised our data, see Section V-D. C. Smoothing in the context of LID Rare or unseen n-grams can result in poor probability estimates. ‘Smoothing’ refers to a range of techniques that re-distribute probability density among rare or unseen tokens [21]. As maximum likelihood estimates (MLE) are used by Naïve Bayes classifiers to estimate class probability, smoothing can help to address poor probability estimates, which result in zero probability of missing n-gram sequence models. Different smoothing techniques have been proposed and applied to the T-LID task [21], such as: Laplace smoothing (simply adding one count across a data set), Katz smoothing [22], Witten-Bell smoothing [23], absolute discounting [24], Kneser-Ney discounting [24] and Jelinek-Mercer [25] methods. We exclude Modified Kneser-Ney (regarded as a state-of-the-art smoothing technique) from further experiments due to the small vocabulary size of this task, since Modified Kneser-Ney assumes a larger vocabulary size [21]. Rather, this work focuses on three of the above-mentioned smoothing techniques, as discussed below: 1) Katz backoff with Good-Turing discounting: In the speech recognition domain, Katz smoothing is a widely used smoothing technique [21]. It uses Good-Turing discounting to calculate adjusted counts, $C^$, which determine how much probability density goes to unseen n-grams. Katz smoothing can be represented as: $$P^(x_i \mid x_{i-N+1}^{i-1}) = \frac{C^(x_{i-N+1}^i)}{\sum_{x_i} C^(x_{i-N+1}^i)}$$ \hspace{1cm} (3) where $C(x)$ counts how many times $x$ appears in the training set, $x_i$ represents the position of the $i^{th}$ character in the given context, $N$ is the n-gram parameter, and $C^$ is an adjusted count, with: $$C^(x_{i-N+1}^i) = d_i C(x_{i-N+1}^i) \text{ if } C(x_{i-N+1}^i) > 0$$ $$= \alpha(x_{i-N+1}) P(x_i) \text{ otherwise}$$ \hspace{1cm} (4) where $\alpha$ represents a back-off weight and $d_i$ a discount ratio according to the Good Turing estimate. This distributes leftover probability density to lower-order n-grams. 2) Witten-Bell discounting + interpolation: Witten-Bell discounting defines models recursively in terms of linear interpolation between the $n^{th}$ and $(n-1)^{th}$ order maximum likelihood models. The discounted probability density is evenly distributed in the training set among previously unseen words with the same history. This can be represented as: $$P_{WB}(x_i \| x_{i-N+1}^{i-1}) =$$ $$\lambda_{x_{i-N+1}^{i-1}} P_{MLE}(x_i \mid x_{i-N+1}^{i-1})$$ $$+ 1 - \lambda_{x_{i-N+1}^{i-1}} P_{WB}(x_i \mid x_{i-N+2}^{i-1})$$ \hspace{1cm} (5) where $x_i$ represents position of the $i^{th}$ character in the given context, $\lambda_{x_{i-N+1}^{i-1}}$ is the discounted probability density, and $1 - \lambda_{x_{i-N+1}^{i-1}}$ is the probability mass that needs to be distributed evenly to previously unseen types. 3) Absolute discounting + interpolation: Absolute discounting uses a fixed discounting parameter, $D$, to reduce probability mass of seen types by subtracting a fixed value. Absolute discounting interpolates higher and lower-order n-grams by using information from lower-order n-gram models. For each seen type we subtract any fixed value between 0 and 1 from the higher-order n-gram. The estimated leftover probability mass is assigned to lower-order n-grams. $$P_{abs}(x_i \mid x_{i-N+1}^{i-1}) = \frac{\max(C(x_{i-N+1}^i) - D, 0)}{\sum_{x_i} C(x_{i-N+1}^i)}$$ $$+ (1 - \lambda_{x_{i-N+1}^{i-1}}) P_{abs}(x_i \mid x_{i-N+2}^{i-1})$$ \hspace{1cm} (6) where $x_i$ represents position of the $i^{th}$ character in the given context, $D$ is the discount weight, and $\lambda$ is a normalising constant (that is, probability mass we have discounted). IV. APPROACH In our experiments, we aim to identify the language origin of a single word at a time. Experiments are carried out on generic text corpora from four South African languages, and results are presented in terms of LID accuracy across all languages. Using empirical measurements, we analyse the following aspects: - For NB classifiers: the difference in classification accuracy when using word types or tokens at different training corpus sizes; and the implications of different smoothing techniques for different n-grams at different training corpus sizes. - For SVMs: the effect of kernel choice. - For both classifiers: the interplay between n-gram length, word length and classification accuracy. A. Experimental Data Generic text in three South African languages (Afrikaans, Sesotho, and isiZulu) were obtained from a pre-release of the NCHLT text corpora [26]. These corpora were collected from “gov.za” which is a South African government domain. This domain is devoted to topics relevant to South Africa, which include: news events across South Africa or beyond, speeches, and statements from various cabinet meetings in the Republic of South Africa. All text are encoded in UTF-8 format to accommodate special characters found in Afrikaans and Sesotho. Our South African English data was obtained from a broadcast news corpus [27]. We performed text normalisation to remove punctuation marks, numbers, formulae, dashes and brackets. The data is randomly partitioned into a training set (80%), development set (10%) and test set (10%) based on the number of characters in running text. Table I contains the original data set, while Table II contains the newly partitioned set after removing overlapping words from the test and development set: that is, only unique words are retained. For the purpose of this experiment, we will refer to Tables I and II as the “all” and “unique” data sets, respectively. In Tables I and II, “A” represents Afrikaans, “E” English, “S” Sesotho, and “Z” isiZulu. The rationale behind the experiment is that the text-based LID (on running text) results are typically obtained by retaining multiple copies of the same token; word-based TABLE I Original data that contain all words. The number of unique words, total number of characters and average word length per language are shown. \| \| Total characters \| Unique words \| Average word length \| \|------------------\|-----------------\|--------------\|---------------------\| \| Training Set \| \| \| \| \| A - 1 840 494 \| A - 15 727 \| A - 13.64 \| \| E - 1 840 547 \| E - 15 765 \| E - 13.21 \| \| S - 1 840 697 \| S - 15 680 \| S - 13.80 \| \| Z - 1 840 570 \| Z - 15 799 \| Z - 13.38 \| \| Total \| 7 362 308 \| 62 971 \| \| \| Development Set \| \| \| \| \| A - 221 040 \| A - 2 101 \| A - 13.28 \| \| E - 221 023 \| E - 2 060 \| E - 13.86 \| \| S - 221 087 \| S - 2 073 \| S - 12.78 \| \| Z - 220 933 \| Z - 2 114 \| Z - 12.87 \| \| Total \| 884 083 \| 8 348 \| \| \| Test Set \| \| \| \| \| A - 221 140 \| A - 2 110 \| A - 13.28 \| \| E - 221 125 \| E - 2 080 \| E - 13.86 \| \| S - 221 007 \| S - 2 120 \| S - 13.18 \| \| Z - 221 041 \| Z - 2 111 \| Z - 13.80 \| \| Total \| 884 313 \| 8 421 \| \| TABLE II Repartitioned data set after removing repeated words. The number of unique words, total number of characters and average word length per language are shown. \| \| Total characters \| Unique words \| Average word length \| \|------------------\|-----------------\|--------------\|---------------------\| \| Training Set \| \| \| \| \| A - 204 456 \| A - 15 727 \| A - 13.00 \| \| E - 204 751 \| E - 15 765 \| E - 12.99 \| \| S - 204 764 \| S - 15 680 \| S - 13.06 \| \| Z - 204 925 \| Z - 15 799 \| Z - 12.97 \| \| Total \| 818 896 \| 62 971 \| \| \| Development \| \| \| \| \| A - 9 911 \| A - 740 \| A - 13.39 \| \| E - 9 931 \| E - 717 \| E - 13.85 \| \| S - 9 940 \| S - 738 \| S - 13.49 \| \| Z - 9 933 \| Z - 710 \| Z - 13.99 \| \| Total \| 39 715 \| 2 905 \| \| \| Test Set \| \| \| \| \| A - 9 901 \| A - 744 \| A - 13.30 \| \| E - 9 961 \| E - 723 \| E - 13.78 \| \| S - 9 910 \| S - 745 \| S - 13.30 \| \| Z - 9 935 \| Z - 707 \| Z - 14.05 \| \| Total \| 39 707 \| 2 919 \| \| LID results are typically obtained on dictionaries, where only unique tokens are retained automatically. In order to investigate the relationship between identification accuracy and training-set sizes; we conduct a series of experiments, which involves creating different training subsets of different sizes from the “all” data set, namely: 250KB, 500KB, 1M, and 1.8M. To avoid bias of having the same words that run across both the test set and train set; we construct different data subsets by removing the identical words from our data sets. Finally, our respective unique words extracted for training are 75KB, 113KB, 162KB, and 204KB in size, which are equivalent to our respective subset sizes; that is, 250KB, 500KB, 1M, and 1.8M. Table II shows our largest unique data set (240KB) together with development and test data set. We use the test and development set from Table II across all experiments. B. General discussion of our setup In order to identify the language origin of individual words, we generate text of different n-gram character lengths from our corpus. All generic words include word boundary markers (indicated by #); while we do not show the results here, we observed that using word boundary markers improve classification accuracy in all experiments. For each word all n-grams are extracted, for example, the word #BREAD# would be represented by the tri-grams #BR, BRE, REA, EAD, AD#. During NB training, an n-gram model is trained, estimating the probabilities of the individual n-grams based on their counts in the training corpus. For example, when training a tri-gram model, a vector of tri-gram values is created in a T-dimensional space, where T is the number of possible tri-grams, and each tri-gram in the vector is associated with a specific likelihood estimate. For SVM training, a similar T-dimensional vector is constructed, but now each training sample is represented by the number of times a specific n-gram occurs in that training sample. These T-dimensional training samples are then used to train an SVM. C. Evaluation To evaluate the performance of the models, we evaluate their accuracies based on certain criteria. We measure LID accuracy as the proportion of correctly identified tokens in the test set, compared to the total number of tokens in the test set. Each test is characterised by the following parameters: language, training data used (size, unique/all), adopted n-gram model and technique employed. V. EXPERIMENTS This section focuses on evaluating the accuracy achieved by the different n-gram models on the test set. We examine how each n-gram model performs with different word lengths and LID classifiers. We also investigate the influence of different training data set sizes on accuracy. Our n-gram models range from 1 to 7. A. Naïve Bayes Baseline Back-off Our baseline model is based on a back-off technique that estimates word probability for tokens with n-gram length greater than or equal to the word length. Using standard Naïve Bayes, the probability of an n-gram with length greater than or equal to the word length becomes zero. We need some form of back-off to be able to produce results over all test types regardless of word length or n-gram parameter (prior to investigating more sophisticated smoothing techniques). Our baseline strategy adopts a technique that backs off to an n-gram size of the word length ($W$), minus a fixed value ($x$), where $x$ is a small value. To obtain a suitable estimate for our fixed value, $x$, we performed 5-fold cross validation on our training data. We evaluate the effect of subtracting a fixed value, $x$, by setting $x$ in the range of $W/2$ to $W$, where $W$ represents total character counts per word without word boundaries. For our baseline model, we find $W/2$ as the most suitable parameter for $x$. This back-off technique is used in our baseline model in order for us to be able to compare with more advanced techniques in later experiments. ### Table III Classification accuracy of baseline Naïve Bayes systems trained on unique types at different training sizes and evaluated on test set \| \| 1-gram \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| 6-gram \| 7-gram \| \|-------\|--------\|--------\|--------\|--------\|--------\|--------\|--------\| \| 250K \| 58.715 \| 75.176 \| 78.488 \| 78.488 \| 77.384 \| 76.614 \| 75.878 \| \| 500K \| 58.884 \| 73.771 \| 78.555 \| 78.554 \| 77.852 \| 77.018 \| 76.447 \| \| 1M \| 58.983 \| 75.176 \| 79.793 \| 80.262 \| 79.258 \| 79.258 \| 79.190 \| \| 1.8M \| 58.983 \| 76.179 \| 80.261 \| 80.763 \| 80.027 \| 79.425 \| 79.190 \| #### B. Unique words vs. All words In our first experiment, we examine the influence of using unique words (types) against all words (tokens). A reduction in computational complexity and training time are some of the benefits of using types over tokens. We therefore examine the influence of using only types against tokens in training, using the data set from Table I, partitioned into different sized subsets. For proper comparison, we use only one test data set across all different subsets. Figure 1 shows the classification accuracies obtained between “unique” and “all”, as defined above. It also shows the relative difference in identification accuracy using different n-gram models across various training data sizes. This baseline result is based on the Naïve Bayes baseline back-off described in Section V-A. Most notable, smaller n-gram models benefit most from the use of types while higher-order models perform better under all tokens, with less than 1% percentage gain over its lower-order models. These accuracies can be considered to be corpus-dependent. For further experiments, we use types over tokens (Table II), as this results in reduced training time and computational complexity. ![Graph showing the difference in LID accuracy when comparing the baseline n-gram models trained using only unique tokens with one trained on all tokens. Results are provided at different training set sizes.](image) Fig. 1. Difference in LID accuracy when comparing the baseline n-gram models trained using only unique tokens with one trained on all tokens. Results are provided at different training set sizes. Table III shows the overall identification accuracy of the Naïve Bayes Baseline back-off method across various data sizes. With smaller data sets, higher-order n-gram models show poor performance, with 7-gram model producing the worst performance. With more training data we observe an improvement in accuracy across all models; however, we notice larger improvement on lower-order n-gram models with 4-gram model benefiting most. Note that classification accuracy increased for all models with more training data. This means that we were unable to achieve asymptotic performance with the data sets available, which implies that higher accuracies are possible with increased training data. #### C. Smoothing Analysis Tables IV, V, and VI show classification accuracies of three smoothing techniques using different n-gram models and training data sets. The question is how we handle unseen features in our test data. Smoothing methods help to better deal with data sparseness by borrowing probability mass from higher-order n-grams and redistributing it among lower-order n-grams. We estimate the discount parameter value (D) for absolute discounting by applying 5-fold cross-validation on the training set. To avoid overfitting and reduction in likelihood of the held-out data, we opt for a discount value that keeps the likelihood of the held-out data stable across the validation set. As expected, with more training data, performance increases with n-gram length. Also, we observe a reduction in classification accuracy from the 6-gram model upwards. This is probably due to both the average word length of our data set, and increased data sparsity of higher order models. In conclusion, for the current data set, the 4-gram model seems to be the preferred model across all three modelling techniques employed. The best accuracy (of 87.72%) was obtained using Witten-Bell discounting. This outperformed both absolute discounting and Katz smoothing, even though the difference in accuracies are relatively small. In general (for higher-order models) we see an 8% accuracy increase, when comparing Witten-Bell discounting to the same n-gram model without smoothing. Table IV Classification accuracy using Witten-Bell smoothing at different n-gram lengths, evaluated on test set. \| \| 1-gram \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| 6-gram \| 7-gram \| \|-------\|--------\|--------\|--------\|--------\|--------\|--------\|--------\| \| 250K \| 72.767 \| 82.239 \| 85.212 \| 85.781 \| 85.279 \| 84.811 \| 84.778 \| \| 500K \| 72.968 \| 82.971 \| 84.978 \| 85.781 \| 85.480 \| 84.978 \| 84.711 \| \| 1M \| 72.700 \| 83.673 \| 86.417 \| 86.250 \| 86.785 \| 86.283 \| 86.183 \| \| 1.8M \| 73.202 \| 84.008 \| 86.584 \| 87.722 \| 87.387 \| 87.554 \| 87.287 \| Table V Classification accuracy using Katz smoothing at different n-gram lengths, evaluated on test set. \| \| 1-gram \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| 6-gram \| 7-gram \| \|-------\|--------\|--------\|--------\|--------\|--------\|--------\|--------\| \| 250K \| 72.533 \| 82.937 \| 84.343 \| 85.212 \| 84.778 \| 84.744 \| 84.744 \| \| 500K \| 72.968 \| 83.272 \| 85.012 \| 85.614 \| 85.882 \| 85.547 \| 85.547 \| \| 1M \| 64.604 \| 82.168 \| 85.079 \| 84.778 \| 84.911 \| 85.212 \| 85.212 \| \| 1.8M \| 73.202 \| 84.108 \| 86.751 \| 87.487 \| 86.986 \| 86.818 \| 86.818 \| ### Table VII LID accuracy using a Linear Kernel at different n-gram lengths, evaluated on the test set. \| \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| \|-------\|--------\|--------\|--------\|--------\| \| 250K \| 83.506 \| 85.012 \| 84.744 \| 77.651 \| \| 500K \| 83.473 \| 85.547 \| 84.778 \| 83.473 \| \| 1M \| 84.610 \| 86.718 \| 86.317 \| 84.778 \| \| 1.8M \| 84.744 \| 87.454 \| 86.986 \| 83.674 \| ### Table VIII LID accuracy using an RBF kernel at different n-gram lengths, evaluated on the test set. \| \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| \|-------\|--------\|--------\|--------\|--------\| \| 250K \| 85.012 \| 86.685 \| 85.012 \| 81.632 \| \| 500K \| 85.982 \| 86.283 \| 85.881 \| 82.335 \| \| 1M \| 86.952 \| 87.086 \| 87.621 \| 84.744 \| \| 1.8M \| 87.789 \| 88.123 \| 88.157 \| 84.376 \| ### Table VI Classification accuracy using Absolute Discounting (d = 0.24) at different n-gram lengths. Accuracy evaluated on test set while d calculated by applying 5-fold cross-validation on training set. \| \| 1-gram \| 2-gram \| 3-gram \| 4-gram \| 5-gram \| 6-gram \| 7-gram \| \|-------\|--------\|--------\|--------\|--------\|--------\|--------\|--------\| \| 250K \| 72.800 \| 83.400 \| 85.012 \| 85.315 \| 84.848 \| 84.410 \| 84.007 \| \| 500K \| 72.800 \| 83.208 \| 85.970 \| 85.531 \| 85.346 \| 84.543 \| 84.343 \| \| 1M \| 72.432 \| 83.607 \| 86.517 \| 86.183 \| 86.417 \| 85.982 \| 85.647 \| \| 1.8M \| 72.968 \| 84.008 \| 86.484 \| 87.354 \| 86.685 \| 86.618 \| 85.714 \| ### D. Support Vector Machines Using SVMs for classification involves four decisions vital to proper classification: (1) data preparation, which concerns selection of the training data and deciding how many classes should be included; (2) converting samples to feature vectors, where each sample in a training set should contain a class label, feature index and feature value; (3) data normalisation or scaling, in order to prevent over-fitting (which results in a zero mean and unit variance); and (4) proper choice of SVM kernels for good classification accuracy. Determining the kernel of choice also involves setting various hyper-parameters associated with a specific kernel. Predicting ideal hyper-parameter values can be achieved using a grid search on the development set. This experiment employed two SVM libraries namely: LibSVM [28] and LIBLINEAR [29] for classification. These two libraries use different methods for multi-class classification. LIBSVM uses one-against-one method for classification, where one SVM is constructed for each pair of classes, classification is performed using a voting strategy. LIBLINEAR uses one-against-rest, where a classifier is trained for each class, prior to voting. We carried out our experiment with two widely used kernels namely: a Radial Basis Function (RBF) and a linear kernel. The SVM training set was generated by combining n-grams across languages to create the SVM training set. Each selected model has a feature dimension equal to the number of n-gram combinations. After creating our feature vectors, we scaled each attribute in the range of $[0, 1]$ in our training set. The same scaled value used in building our model was used to scale the test and development set. We performed 5-fold cross validation on our training set to obtain optimal kernel parameters. We limit our n-gram model length to 5-gram due to the longer training time and extensive resource usage associated with higher-order n-gram models. Tables VII and VIII show classification accuracies obtained using the RBF kernel and linear kernel (respectively) and different n-gram models across various training data sets. All models improved with more training data, and the RBF kernel produced higher accuracies than the linear kernel. ### E. Effect of various corpus sizes on LID accuracies The size of the training data set is one of the factors that influences LID accuracy of running text. For the task at hand, what effect does different training data sizes have on accuracy, TABLE IX A comparison of LID accuracy for different classifiers investigated. \| \| NB (n=4) \| Wt (n=4) \| Katz (n=4) \| ABS (n=4) \| Linear (n=3) \| RBF (n=3) \| \|-------\|----------\|----------\|------------\|-----------\|--------------\|-----------\| \| 250K \| 78.488 \| 85.781 \| 85.212 \| 85.915 \| 85.912 \| 86.685 \| \| 500K \| 78.555 \| 85.781 \| 85.614 \| 85.313 \| 85.547 \| 86.283 \| \| 1M \| 80.462 \| 86.250 \| 84.778 \| 86.183 \| 86.718 \| 87.086 \| \| 1.8M \| 80.763 \| 87.722 \| 87.488 \| 87.354 \| 87.454 \| 88.123 \| and how do different classifiers and smoothing techniques perform as training data sizes increase? In Table IX, results are shown for different LID techniques and training set sizes. On the smallest data set, NB with smoothing techniques outperformed the SVM with a linear kernel. For all training set sizes investigated, an SVM with an RBF kernel performed best. NB classification with absolute discounting and Witten-Bell smoothing, as well as SVM classification with a linear kernel, shared similar performance across a range of training data sizes. F. Effect of word length on classification accuracy In this section, we want to analyse another factor that influences classification accuracy. We are interested to see how classifiers and smoothing techniques perform at different word lengths. Using our largest and smallest training data sets, figures 2 and 3 show the classification accuracy achieved when evaluating words with different character lengths. The fluctuation in classification accuracy for words shorter than 5 characters could be due to the fact that these words occur fairly infrequently in the test data set. The SVM RBF classifier obtained the highest performance for short character lengths (5 characters or less), while Naïve Bayes classification achieved the worst performance for the same scenario. With longer words (20 characters or more), the Naïve Bayes classifier achieves a classification accuracy of 100%. This result confirms previous results on using Naïve Bayes for long sentences. Also, with 12 or more characters, the difference in classification using Naïve Bayes with and without smoothing becomes insignificant. VI. CONCLUSION This paper examined two different classification methods which can be applied to LID of generic words in isolation. The analysis showed that high LID accuracy can be achieved on words in isolation using both methods. We investigated the difference between using unique types and all tokens when training a Naïve Bayes classifier, and found a computational win when using unique types (that was not offset by any loss in LID accuracy). As our baseline classifier, we used a Naïve Bayes back-off method for words with character length greater than or equal to the n-gram model parameter. This method produced good results across all n-gram lengths. The Naïve Bayes classifier achieved an identification accuracy of 80.76% using a 4-gram model when tested on unique types. To reduce the total number of unseen tokens as the n-gram model increases, we experimented with Katz smoothing, Absolute Discounting and Witten-Bell smoothing. Amongst the three smoothing techniques, Witten-Bell smoothing achieved the best performance, but only with a small margin (less than 1%). Smoothing improved average classification accuracy for higher order n-grams by a percentage of approximately 8% (from 79.89% to 87.55%). The highest classification accuracy of 88.12% was obtained using an SVM with an RBF kernel and an n-gram length of $ n = 3 $. This classifier clearly outperformed the Naïve Bayes classifier without smoothing, with NB classification accuracy improving considerably when smoothing is used. To conclude, our experiments show that better identification can still be achieved with more training data. Achieving an asymptotic performance depends not only on the available data set; it also depends on the word length, with longer words achieving very high accuracies on the 1.8M data set. For the four language task studied, the accuracy of the Support Vector Machine (88.16%, obtained with a Radial Basis function) was higher than that of the Naïve Bayes classifier (87.62%, obtained using Witten-Bell smoothing), but the latter result was associated with a significantly lower computational cost. As the computational cost increases as the training set size increases, both techniques will be usable in future work, based on the trade-off between accuracy and computational cost required for a specific application. VII. ACKNOWLEDGEMENT We are grateful to Martin Puttkammer of North-West University and Thomas Nielser of Stellenbosch University, who provided us with the text data used in this analysis. REFERENCES [1] P. Li, “Spoken word recognition of code-switched words by Chinese-English bilinguals,” Journal of memory and language, vol. 35, pp. 757–774, 1996. [2] P. Auer, Code-switching in conversation: Language, interaction and identity. Routledge, 2002. [3] A. Bhargava and G. Kondrak, “Language identification of names with SVMs,” in Proc. NAACL-HLT, 2010, pp. 693–696. [4] A. F. Litjós and A. W. Black, “Knowledge of language origin improves pronunciation accuracy of proper names,” in Proc. INTERSPEECH, 2001, pp. 1919–1922. [5] K. Church, “Stress assignment in letter to sound rules for speech synthesis,” in Proc. ACL, 1985, pp. 246–253. [6] W. Basson and M. H. Davel, “Category-based phoneme-to-grapheme transliteration,” in Proc. INTERSPEECH, 2013, pp. 1956–1960. [7] W. B. Cavnar, J. M. Trenkle et al., “N-gram-based text categorization,” Ann Arbor MI, vol. 48113, no. 2, pp. 161–175, 1994. [8] T. Dunning, Statistical identification of language. Computing Research Laboratory, New Mexico State University, 1994. [9] T. Vatanen, J. J. Väyrynen, and S. Virpioja, “Language identification of short text segments with N-gram models,” in Proc. LREC, 2010, pp. 3423–3430. [10] G. R. Botha and E. Barnard, “Factors that affect the accuracy of text-based language identification,” Computer Speech & Language, vol. 26, no. 5, pp. 307–320, 2012. [11] A. Xatopoulos, C. Kotropoulos, G. Almanpidis, and I. Pitas, “Language identification in web documents using discrete HMMs,” Pattern recognition, vol. 37, no. 3, pp. 583–594, 2004. [12] C. Kruengkrai, P. Srichavivattana, V. Sornlertlamvanich, and H. Ishahara, “Language identification based on string kernels,” in Proc. ISCIT, 2005, pp. 926–929. [13] J. Hakkinen and J. Tian, “N-gram and decision tree based language identification for written words,” in Proc. ASRU, 2001, pp. 335–338. [14] P. Sibun and J. C. Reynar, “Language identification: Examining the issues,” in Proc. SDAIR, 1996, pp. 125–135. [15] A. Hategan, B. Barliga, and I. Tabus, “Language identification of individual words in a multilingual automatic speech recognition system,” in Proc. ICASSP, 2009, pp. 4357–4360. [16] S. Johnson, “Solving the problem of language recognition.” Technical report, School of Computer Studies, University of Leeds, Tech. Rep., 1993. [17] R. D. Lins and P. Gonçalves, “Automatic language identification of written texts,” in Proc. ACM symposium on Applied computing, 2004, pp. 1128–1133. [18] T. Baldwin and M. Lui, “Language identification: The long and the short of the matter,” in Proc. ACL, 2010, pp. 229–237. [19] N. Friedman, D. Geiger, and M. Goldszmidt, “Bayesian network classifiers,” Machine learning, vol. 29, no. 2-3, pp. 131–163, 1997. [20] V. N. Vapnik, Statistical learning theory, ser. Adaptive and learning systems for signal processing, communications, and control. Wiley-Interscience, 1998. [21] S. F. Chen and J. Goodman, “An empirical study of smoothing techniques for language modeling,” in Proc. ACL, 1996, pp. 310–318. [22] S. Katz, “Estimation of probabilities from sparse data for the language model component of a speech recognizer,” in Proc. ICASSP, 1987, pp. 400–401. [23] A. M. M. Hasan, S. Islam, and M. A. Rahman, “A comparative study of Witten Bell and Kneser-Ney smoothing methods for statistical machine translation,” Journal of Information Technology, vol. 1, pp. 1–6, 2012. [24] H. Ney, U. Essen, and R. Kneser, “On structuring probabilistic dependences in stochastic language modelling,” Computer Speech & Language, vol. 8, no. 1, pp. 1–38, 1994. [25] F. Jelinek, “Up from trigrams,” in Proc. EUROSPEECH, 1991, pp. 1037–1040. [26] N. J. De Vries, J. Badenhorst, M. H. Davel, E. Barnard, and A. De Waal, “Woetzela-An open-source platform for ASR data collection in the developing world,” in Proc. INTERSPEECH, 2011, pp. 3177–3180. [27] H. Kamper, F. De Wet, T. Hain, and T. Niesler, “Resource development and experiments in automatic SA broadcast news transcription,” in Proc. SLTU, 2012. [28] C.-C. Chang and C.-J. Lin, “Libsvm: a library for support vector machines,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 2, no. 3, p. 27, 2011. [29] R.-E. Fan, K.-W. Chang, C.-J. Hsieh, X.-R. Wang, and C.-J. Lin, “Liblinear: A library for large linear classification,” The Journal of Machine Learning Research, vol. 9, pp. 1871–1874, 2008.
Mobile Footprinting: Linking Individual Distinctiveness in Mobility Patterns to Mood, Sleep, and Brain Functional Connectivity Cedric Huchuan Xia University of Pennsylvania Ian Barnett University of Pennsylvania Tinashe Tapera University of Pennsylvania Zaixu Cui State Key Laboratory of Cognitive Neuroscience and Learning-Beijing Normal University Tyler Moore Perelman School of Medicine https://orcid.org/0000-0002-1384-0151 Azeez Adebimpe Annenberg Public Policy Center, University of Pennsylvania, Philadelphia https://orcid.org/0000-0001-9049-0135 Sage Rush-Goebel University of Pennsylvania Kayla Piwaa University of Pennsylvania Kristin Murtha University of Pennsylvania Sophia Linguiti University of Pennsylvania Ellen Leibenluft National Institute of Mental Health Melissa Brotman National Institute of Mental Health Melissa Martin University of Pennsylvania Monica Calkins University of Pennsylvania David Roalf University of Pennsylvania https://orcid.org/0000-0002-1728-9782 Daniel Wolf University of Pennsylvania Danielle Bassett University of Pennsylvania https://orcid.org/0000-0002-6183-4493 David Lydon-Staley University of Pennsylvania Justin Baker Harvard Medical School https://orcid.org/0000-0002-6629-2195 Lyle Ungar University of Pennsylvania Theodore Satterthwaite (✉️ email@example.com) University of Pennsylvania https://orcid.org/0000-0001-7072-9399 Article Keywords: smartphone, GPS, mobile phenotyping, fMRI, mood, affective instability Posted Date: May 10th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-465623/v1 License: This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Mobile Footprinting: Linking Individual Distinctiveness in Mobility Patterns to Mood, Sleep, and Brain Functional Connectivity Cedric Huchuan Xia\textsuperscript{1,2}, Ian Barnett, Tinashe M. Tapera\textsuperscript{1,2}, Zaixu Cui\textsuperscript{1,2}, Tyler M. Moore\textsuperscript{1,2}, Azeez Adebimpe\textsuperscript{1,2}, Sage Rush-Goebel\textsuperscript{1,2}, Kayla Piiwaa\textsuperscript{1,2}, Kristin Murtha\textsuperscript{1,2}, Sophia Linguiti\textsuperscript{1,2}, Ellen Leibenluft\textsuperscript{1}, Melissa A. Brotman\textsuperscript{1}, Melissa Lynne Martin, Monica E. Calkins\textsuperscript{1,2}, David R. Roalf\textsuperscript{1,2}, Daniel H. Wolf\textsuperscript{5,6,18}, Danielle S. Bassett\textsuperscript{5,6,7,8,9}, David M. Lydon-Staley\textsuperscript{10,20}, Justin T. Baker\textsuperscript{11,12}, Lyle Ungar\textsuperscript{13,14,15,16,17}, Theodore D. Satterthwaite\textsuperscript{1,2,18,19} 1. Penn Lifespan Informatics and Neuroimaging Center, Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 2. Penn/CHOP Lifespan Brain Institute, University of Pennsylvania, Philadelphia, PA 19104, USA 3. Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 4. National Institute of Mental Health, Intramural Research Program, Bethesda, MD 20892 5. Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, PA 19104, USA 6. Department of Physics and Astronomy, University of Pennsylvania, PA 19104, USA 7. Department of Electrical & Systems Engineering, University of Pennsylvania, PA 19104, USA 8. Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 9. Santa Fe Institute, Santa Fe, NM 87501, USA 10. Annenberg School of Communication, University of Pennsylvania, Philadelphia, PA 19104, USA 11. Department of Psychiatry, Harvard Medical School, Boston, MA 02115 12. McLean Institute for Technology in Psychiatry, McLean Hospital, Belmont, MA 02478 13. Department of Computer and Information Science, School of Engineering and Applied Science, University of Pennsylvania, PA 19104, USA 14. Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, PA 19104, USA 15. Department of Genomics and Computational Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 16. Department of Operations, Information and Decisions, Wharton School, Philadelphia, PA 19104, USA 17. Department of Psychology, School of Arts and Sciences, Philadelphia, PA 19104, USA 18. Center for Biomedical Image Computation and Analytics, University of Pennsylvania, Philadelphia, PA 19104, USA 19. Penn Statistics in Imaging and Visualization Center, Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Philadelphia, PA 19104, USA 20. Leonard Davis Institute for Health Economics, University of Pennsylvania, PA 19104, USA ABSTRACT Mapping individual differences in behavior is fundamental to personalized neuroscience. Here, we establish that statistical patterns of smartphone-based mobility features represent unique “footprints” that allow individual identification. Critically, mobility footprints exhibit varying levels of person-specific distinctiveness and are associated with individual differences in affective instability, circadian irregularity, and brain functional connectivity. Together, this work suggests that real-world mobility patterns may provide an individual-specific signature linking brain, behavior, and mood. KEYWORDS: smartphone, GPS, mobile phenotyping, fMRI, mood, affective instability INTRODUCTION Linking individual differences in behavior to brain function is a central task of behavioral neuroscience\textsuperscript{1}. However, quantifying complex human behavior in real world settings remains a challenge. One alternative to standard behavioral assessment is \textit{digital phenotyping}, which uses mobility data from personal smartphones to quantify moment-by-moment human behavior\textsuperscript{2}. Prior work has associated geolocation features to important clinical outcomes in psychiatric disorders such as bipolar disorder and schizophrenia\textsuperscript{3}, and has linked accelerometer metrics to post-surgical recovery\textsuperscript{4-5}. Furthermore, researchers have recently begun to capitalize on the substantial variability of behavior assessed with digital phenotyping to link individual differences in brain and behavior. For example, lower prefrontal activity during processing negative emotions has been associated with individual exposure to urban green space\textsuperscript{6}, while greater functional coupling of the hippocampus and striatum has been linked to location variability\textsuperscript{7}. While these studies suggest that digital phenotyping can be a powerful tool for studying individual differences, it remains unknown whether mobility patterns are in fact \textit{person-specific}. Recent high-impact work has established that individual humans have unique patterns of functional brain connectivity\textsuperscript{8,9}. The uniqueness of such brain-based “fingerprints” (also called “connectotypes”\textsuperscript{10}) have been associated with development, cognition, and psychiatric conditions\textsuperscript{11}. Establishing analogous person-specific mobility patterns – or mobility “footprints” – would constitute an important advance in behavioral neuroscience, and provide the foundation for targeted, individual-specific interventions. Accordingly, here we test the hypothesis that mobility patterns derived from personal smartphones can be used to create person-specific behavioral footprints. Furthermore, we evaluate whether the distinctiveness of these footprints was related to individual differences in mood, sleep, and brain functional connectivity. RESULTS & DISCUSSION As part of a study of trans-diagnostic affective instability in youth, we tracked 3,317 person-days of geolocation and 2,972 person-days of accelerometer data from 41 adolescents and young adults (28 females; mean [s.d.] age = 23.4 [3.5] years, range 17–30 years) – approximately 3 months per individual (\textbf{Fig. 1a}, Supplementary Fig. 1). In this sample, 93% of participants reported clinically significant affective instability in the context of psychiatric disorders (especially borderline personality disorder; see Supplementary Table 1). After applying hot-deck imputation to missing GPS data as implemented in the Smartphone Sensor Pipeline\textsuperscript{13}, we constructed the daily mobility trajectory for each participant (\textbf{Fig. 1b} & c; see Online Methods). Instead of using raw coordinates that would allow trivial individual identification (and raise privacy concerns) given a participant’s exact location, we extracted high-level summary statistics of mobility features. These features (15 geolocation-based and 7 accelerometer-based) included time spent at home, number of locations visited, and many others (\textbf{Fig. 1d}, Supplementary Table 2). Figure 1 \| Constructing personal mobility “footprints”. a) We collected 3,317 person-days of mobility sensing data via personal smartphones from 41 adolescents and young adults. Geolocation data were recorded in cycles of 2min on and 18min off. Raw geolocation coordinates were de-identified via sphere-to-2D standard space projection and were further imputed for missing data. b) For each individual, we constructed daily personal mobility trajectories, which consist of flights (movement) and pauses (stationary segments). Length of linear lines represents the duration of flights and size of circles represents the duration of pauses. Warm and cold colors indicate daytime and nighttime, respectively. c) A representative week of trajectories is shown, which demonstrates rich characteristics of personal mobility patterns formed over time. d) We extracted timeseries of mobility statistics (e.g. daily time spent at home) from geolocation and accelerometer data that parameterize movement characteristics over weeks to months. The example represented all 110 days of participants’ geolocation metrics recorded. e) For each individual, we constructed a covariance matrix from the mobility metric timeseries. Each cell of the matrix was populated by the Pearson correlation between a given pair of mobility metrics. Warm and cold colors indicate positive and negative correlations, respectively. f) We randomly divided data into two equally sized parts, called the reference and target set. Subj X from the target set was matched to the subject in the reference that had the highest correlations between their footprints (argmax(r_1, r_2, ..r_N)). The identification was considered correct when underlying data came from the same subject; otherwise, the identification was considered incorrect. We quantified individual identification accuracy as the proportion of correct identifications across the entire sample; this procedure was repeated 1,000 times across different random partitions of the data. When tracked over weeks to months, these timeseries of mobility statistics captured rich characteristics of individual mobility patterns. One illustrative example of the sensitivity of the timeseries to track mobility patterns is when COVID-19 pandemic hit the Philadelphia area towards the end of the study period. Participants who were still engaged in active data collection (n=3) exhibited dramatic shifts in mobility features (Supplementary Fig. 2). Of note, as the data points during COVID-19 represented merely 1.1% of all data, the findings reported below did not change significantly when these data were removed. Drawing on prior work of brain connectome “fingerprinting,”$^8,^{10}$ we created a covariance matrix of each participant’s fifteen geolocation-based and seven accelerometer-based mobility features timeseries to identify individuals (Fig. 1e), akin to a person-specific mobility “footprint.” Data from each individual was partitioned into two groups: the target partition and the reference partition. For each individual, the data in the target partition was separately correlated with every individual’s data in the reference partition; this procedure yielded 41 correlation values. A correct identification was declared only when the maximum correlation was from the data belonging to the same individual across the target and reference partitions (Fig. 1f). In order to ensure that the random partitioning of the data did not impact results, this matching procedure was then repeated for each individual 1000 times (Online Methods). Initial inspection across random partitions of the data revealed that it was visually apparent that there was substantially greater correlation between mobility footprints within participants rather than between participants (Fig. 2a). Permutation testing on the entire sample revealed that individuals could be successfully identified using their mobility footprints ($p < 0.001$; Fig. 2b). Across 1,000 random data partitions, the mean individual identification accuracy was 63%. Critically, this accuracy was far better than chance performance determined by a permuted null distribution (mean: 3% accuracy; see Fig 2b inset). Moving beyond aggregate measures of accuracy across the group, we next investigated whether certain individuals could be consistently identified more accurately than others. Similar to prior studies of brain connectome fingerprinting$^{8,10,11}$, we refer to this measure as an individual’s “footprint distinctiveness”. Notably, individuals exhibited a wide distribution of footprint distinctiveness, ranging from 4% to 99% (Fig. 2c). In other words, certain participants had such distinct mobility patterns that it enabled correct identification nearly every single time; other participants were difficult to identify. Nonetheless, permutation testing showed that all participants had significant footprint distinctiveness compared to the null distribution. As the group and individual level accuracy results reported thus far were based on the combination of geolocation and accelerometer features, we next examined each feature set separately. Individual footprint distinctiveness derived by geolocation was not correlated with that of accelerometer $(r = 0.18; p = 0.26)$. Interestingly, while accelerometer data alone yielded lower identification accuracy (28%) than geolocation data (55%), combining these features resulted in higher identification accuracy, suggesting that they encode complementary information (Fig. 2d). Importantly, individual identification accuracy was stable across different inclusion thresholds for data missingness and was robust to removal of individual mobility features (Supplementary Fig. 3). Figure 2 \| Identifying individuals using personal footprints. a) As an initial step, we visualized the similarity of mobility features across multiple random reference and target partitions (R & T in inset). It was readily apparent that mobility features were more highly correlated within participants (on diagonal) across data partitions than between participants (off diagonal). Note that this visualization was not used in statistical analysis or individual identification. b) Across 1,000 random partitions, mobility footprinting enabled successful individual identification (mean: 63%, S.D.: 6%). In contrast, the mean chance accuracy from 1,000 permutation was 3% (inset, $p < 0.001$). c) For each individual, we calculated the footprint distinctiveness, or the percentage of correct identification across the 1,000 random partitions of the data. Ranked in ascending order, participants’ footprint distinctiveness exhibited a wide range, from 4% to 99%. However, even the participant with the lowest identification distinctiveness was significantly higher than the null distribution. d) Individual identification based on geolocation alone had higher accuracy than accelerometer alone. However, they appeared to encode complementary features, as performance was maximal when both measures were used in footprinting. We next investigated participant factors that influenced footprint distinctiveness. We found that data quantity (i.e. number of days recorded) was associated with footprint distinctiveness (Supplementary Fig. 4). In contrast, the amount of missing data within a given day was unrelated to footprint distinctiveness. Based on this result, all subsequent analyses of individual differences related of footprint distinctiveness controlled for number of days of data. available. As a next step, we evaluated whether footprint distinctiveness was related to age or sex in our sample of adolescents and young adults. We found that geolocation-based footprints became more distinct with age across the transition from adolescence to adulthood (partial $r = 0.33$, $p < 0.05$, Supplementary Fig. 5). Furthermore, female sex was associated with higher accelerometer-based footprint distinctiveness ($Cohen’s d = 1.27$, $p < 0.001$, Supplementary Fig. 5). We next evaluated how footprint distinctiveness was related to a key domain of psychopathology: affective instability. Affective instability is a major feature of many psychiatric disorders\textsuperscript{14}, including borderline personality disorder. Affective instability is particularly prominent in youth\textsuperscript{15}, and is an important predictor of suicide\textsuperscript{16}. However, affective instability is often challenging to quantify using standard tools as it is fundamentally a dynamic measure\textsuperscript{17}. We capitalized on participant ratings of multiple mood features collected three times a day for two weeks using ecological momentary assessment in order to quantify affective instability. We hypothesized that individuals who had less predictable patterns of mobility (i.e., reduced footprint distinctiveness) would have higher levels of affective instability. While controlling for data quantity, age, sex, and the mean of mood ratings, we found that affective instability (measured by root mean square of successive differences\textsuperscript{18}) was associated with reduced footprint distinctiveness (partial $r = -0.37$, $p < 0.05$, Fig. 3a). Furthermore, given well-established links between sleep disturbance and mood disorders\textsuperscript{19}, we also evaluated whether variability in sleep duration was also associated with footprint distinctiveness. While controlling for covariates as above, we found that variability in sleep duration was similarly associated with reduced footprint distinctiveness (partial $r = -0.36$, $p < 0.05$, Fig. 3b). Figure 3 \| Individual footprint distinctiveness is associated with affective instability, sleep irregularity, and patterns of brain functional connectivity. a) Greater affective instability, measured by root mean square of successive differences in mood measures from ecological momentary assessment items acquired three times a day, was associated with reduced footprint distinctiveness ($r = -0.37$, $p < 0.05$), after controlling for data quantity, age, sex, and mean level of mood ratings. b) Similarly, we found that increased variability in sleep duration was associated with reduced footprint distinctiveness ($r = -$ 0.36, $p < 0.05$), after controlling for covariates. c) Across functional brain networks, only greater connectivity within the somatomotor network had a significant association with footprint distinctiveness ($r = 0.46, p < 0.05$, corrected for multiple comparisons with the false discovery rate). d) Patterns of brain functional connectivity significantly predicted individual footprint distinctiveness using leave-one-out cross-validation ($r = 0.29$, inset: permutation-based $p = 0.025$). e) Six network edges consistently contributed to the sparse regression model. These edges included greater connectivity within somatomotor network, reduced connectivity between left and right frontal eye fields (FEF), increased connectivity between the somatomotor network and the left orbital frontal cortex (OFC) in the limbic network, as well as increased connectivity between the vIPFC (ventrolateral prefrontal cortex in the frontoparietal network) and the dmPFC (dorsolateral prefrontal cortex in the default mode network). Cord thickness reflects the weights in the model, reflecting each edge’s contribution to the prediction; cord color indicates the sign of the weights. As a final step, we investigated whether footprint distinctiveness was related to patterns of functional connectivity. Initially, we examined associations with a simple summary measure of high-dimensional functional connectivity data: the mean connectivity within each of seven canonical large-scale functional networks\textsuperscript{20}. While controlling for covariates as above (as well as in-scanner motion) and correcting for multiple comparisons with the false discovery rate, we found that footprint distinctiveness was associated with greater connectivity within the somatomotor network ($r = 0.46, p_{\text{fdr}} = 0.03$, \textbf{Fig. 3c}). Previous work has demonstrated that somatomotor network connectivity develops over the lifespan (years)\textsuperscript{21} and is altered acutely (days) during limb disuse\textsuperscript{22}; our results further suggest that mobility patterns over weeks-months can be predicted by somatomotor network connectivity. Lastly, we moved beyond the simple summary measure of mean network connectivity and investigated whether complex multivariate patterns of functional connectivity could predict footprint distinctiveness in unseen data. Given that there were far larger number of features than participants, we used regularized regression with leave-one-out cross-validation and nested parameter tuning, followed by permutation testing to determine significance (Online Methods). We found that multivariate patterns of functional connectivity could predict footprint distinctiveness in unseen data ($r = 0.29, p = 0.025$; \textbf{Fig. 3d}). The predictive model yielded results that aligned with the mass-univariate analyses (\textbf{Fig. 3c}), suggesting that the multivariate model was driven in part by features linked to somatomotor network (67% of edges selected by the model). Moreover, this model also revealed important features beyond the motor system, including increased connectivity between the frontoparietal and default mode system (\textbf{Fig. 3e}). Taken together, these results establish that mobility patterns collected from smartphones can be used to create a person-specific footprint. Notably, the distinctiveness of this footprint increased with age, was reduced in association with both affective instability and circadian irregularity, and was related to patterns of functional brain connectivity. These results align with impactful prior work on “connectome fingerprinting”\textsuperscript{8} (or “connectotyping”\textsuperscript{10}), which have shown that individuals can be identified based on their pattern of functional connectivity. Interestingly, result from these prior studies have shown that – like the footprint distinctiveness examined here – connectome distinctiveness increases with age and is reduced in association with psychiatric symptoms\textsuperscript{11}. Our finding that footprint distinctiveness is related to data quantity recalls recent work demonstrating that the ability to delineate person-specific functional brain networks is dependent in large part on the quantity of data available\textsuperscript{23,24}. However, while accruing large amounts of functional imaging data is often difficult and expensive, passive collection of long timeseries of mobility data is both tolerable for participants and inexpensive. The high degree of scalability enabled by ubiquitous usage of personal smartphones will allow future studies to test the generalizability of these findings across different age groups and clinical samples. Moving forward, mobility-based digital biomarkers that combine objective measurement and individual-specific analysis of behavior may accelerate the advances in personalized diagnostics for diverse psychiatric illnesses. REFERENCES 1. Insel, T. R. Digital phenotyping: Technology for a new science of behavior. \textit{JAMA} \textbf{318}, 1215–1216 (2017). 2. Onnela, J. P. Opportunities and challenges in the collection and analysis of digital phenotyping data. \textit{Neuropsychopharmacology} \textbf{46}, 45–54 (2021). 3. Fraccaro, P. \textit{et al.} Digital biomarkers from geolocation data in bipolar disorder and schizophrenia: a systematic review. \textit{J. Am. Med. Informatics Assoc.} \textbf{26}, 1412–1420 (2019). 4. Cote, D. J., Barnett, I., Onnela, J.-P. & Smith, T. R. Digital Phenotyping in Patients with Spine Disease: A Novel Approach to Quantifying Mobility and Quality of Life. \textit{World Neurosurg.} \textbf{126}, e241–e249 (2019). 5. Panda, N. \textit{et al.} Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. \textit{JAMA Surg.} \textbf{155}, 123–129 (2020). 6. Tost, H. \textit{et al.} Neural correlates of individual differences in affective benefit of real-life urban green space exposure. \textit{Nat. Neurosci.} \textbf{22}, 1389–1393 (2019). 7. Heller, A. S. \textit{et al.} Association between real-world experiential diversity and positive affect relates to hippocampal–striatal functional connectivity. \textit{Nat. Neurosci.} \textbf{23}, 800–804 (2020). 8. Finn, E. S. \textit{et al.} Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. \textit{Nat. Neurosci.} \textbf{18}, 1–11 (2015). 9. Horien, C., Shen, X., Scheinost, D. & Constable, R. T. The individual functional connectome is unique and stable over months to years. \textit{Neuroimage} \textbf{189}, 676–687 (2019). 10. Miranda-Dominguez, O. \textit{et al.} Connectotyping: Model Based Fingerprinting of the Functional Connectome. \textit{PLoS One} \textbf{9}, e111048 (2014). 11. Kaufmann, T. \textit{et al.} Delayed stabilization and individualization in connectome development are related to psychiatric disorders. \textit{Nat. Neurosci.} \textbf{20}, 513–515 (2017). 12. González, M. C., Hidalgo, C. A. & Barabási, A. L. Understanding individual human mobility patterns. \textit{Nature} \textbf{453}, 779–782 (2008). 13. Barnett, I. & Onnela, J.-P. Inferring mobility measures from GPS traces with missing data. \textit{Biostatistics} \textbf{21}, e98–e112 (2018). 14. Koenigsberg, H. W. Affective Instability: Toward an Integration of Neuroscience and Psychological Perspectives. \textit{J. Pers. Disord.} \textbf{24}, 60–82 (2010). 15. Tragesser, S. L., Solhan, M., Schwartz-Mette, R. & Trull, T. J. The Role of Affective 16. Yen, S. et al. Borderline Personality Disorder Criteria Associated With Prospectively Observed Suicidal Behavior. Am. J. Psychiatry 161, 1296–1298 (2004). 17. Trull, T. J. et al. Affective instability: Measuring a core feature of borderline personality disorder with ecological momentary assessment. J. Abnorm. Psychol. 117, 647–661 (2008). 18. Jahng, S., Wood, P. K. & Trull, T. J. Analysis of affective instability in ecological momentary assessment: Indices using successive difference and group comparison via multilevel modeling. Psychol. Methods 13, 354–375 (2008). 19. Harvey, A. G. Sleep and Circadian Rhythms in Bipolar Disorder: Seeking Synchrony, Harmony, and Regulation. Am. J. Psychiatry 165, 820–829 (2008). 20. Schaefer, A. et al. Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cereb. Cortex 28, 3095–3114 (2018). 21. Power, J. D., Fair, D. A., Schlaggar, B. L. & Petersen, S. E. The Development of Human Functional Brain Networks. Neuron 67, 735–748 (2010). 22. Newbold, D. J. et al. Plasticity and Spontaneous Activity Pulses in Disused Human Brain Circuits. Neuron 107, 580-589.e6 (2020). 23. Gratton, C. et al. Functional Brain Networks Are Dominated by Stable Group and Individual Factors, Not Cognitive or Daily Variation. Neuron 98, 439-452.e5 (2018). 24. Elliott, M. L. et al. General functional connectivity: Shared features of resting-state and task fMRI drive reliable and heritable individual differences in functional brain networks. Neuroimage 189, 516–532 (2019). Mobile Footprinting: Linking Individual Distinctiveness in Mobility Patterns to Mood, Sleep and Brain Functional Connectivity ONLINE METHODS Participants A sample of 41 adolescents and young adults (28 females; mean (s.d.) age = 23.4 (3.5) years, range 17–30 years) were enrolled as part of a study of affective instability in youth. Participants were recruited via the Penn/CHOP Lifespan Brain Institute or through the Outpatient Psychiatry Clinic at the University of Pennsylvania. Of these 41 participants, 38 participants met criteria for Axis I psychiatric diagnosis based on a semi-structured clinical interview$^1$; 33 met criteria for more than one disorder (Supplementary Table 1). Additionally, 16 of the 41 participants met criteria for a personality disorder (mainly borderline personality disorder) based on assessment with the SCID-II$^1$. All participants provided informed consent to all study procedures; for minors, the parent or guardians provided informed consent and the minor assented as well. This study was approved by the University of Pennsylvania Institutional Review Board. Mobility data acquisition Global Positioning System (GPS) geolocation data were acquired via the Beiwe platform$^2$. Participants were asked to download the Beiwe application on their personal smartphone. The application recorded the location of the participant’s phone in latitude, longitude, and altitude, as well as the precision of that measure. To conserve battery and minimize degradation of the phone performance, Beiwe was designed to track participant’s geolocation in a periodic fashion. Specifically, Beiwe tracked GPS for 2 minutes every 20 minutes, resulting in 144 minutes of data recording and 1296 minutes of dormancy in a 24-hour cycle. Due to user and device related factors in the naturalistic setting, such as phone powered off, no cell signal, or airplane mode, longer periods of recording dormancy were possible. Mobility data were automatically uploaded via WiFi to a cloud-based data management system daily. In total, 3,317 days of GPS tracking across all participants were obtained (mean (s.d.) = 77 (26) days, range 14–132 days, see Supplementary Figure 1). After removing the first and last days of each participant’s study period when only partial data were recorded and days containing no data, the remaining data available for analysis had 3,156 days. Accelerometer data were also acquired via the Beiwe platform. The application recorded the participants’ acceleration in three cardinal axes (x, y, and z) in m/s$^2$. In total, 2,972 days of accelerometer data were obtained across all participants (mean (s.d.) = 74 (32) days, range 15–134 days). After removing the first and last days of each participant’s study period when only partial data were recorded, the remaining data available for analysis had 2972 days. Mobility data processing GPS data preprocessing Raw GPS data were processed using the Smartphone Sensor Pipeline$^3$, a validated pipeline specifically designed to handle GPS data while accounting for data missingness. First, each subject’s GPS longitude and latitude coordinates on the spherical Earth’s surface were transformed to a standardized two-dimensional Cartesian plane, thus deidentifying subject’s real-world locations. Second, the data were converted to a sequence of flights and pauses, where flights were defined as segments of linear movements and pauses were defined as periods of no movement. Finally, missing flights and pauses were then imputed by the hot-deck method$^4$, which resamples from observed events over each missing interval. Mobility metrics calculation Using the constructed subject mobility traces and the Smartphone Sensor Pipeline, 15 GPS-based mobility metrics were calculated for each day of recording, defined as midnight to midnight. See Barnett et al. for details$^3$. An additional seven accelerometer-based mobility metrics were calculated for each day of recording. These were implemented according to methods described in the RAPIDS pipeline$^5$. See Supplementary Table 2 for definitions of each metric. Mobility footprint construction Inspired by person-specific connectome fingerprints$^6,7$, we constructed a mobility footprint for each participant using the covariance matrix of mobility metrics. First, we extracted the mobility metric time series by concatenating the daily mobile metric output from the Smartphone Sensor Pipeline. Then we computed the pairwise Pearson correlation for all the mobility metrics to construct a covariance matrix. The nodes of the network were the mobility metrics, and the edges of the network were the Pearson correlation coefficients between metrics. We refer to the resulting covariance matrix as the “Mobility Footprint.” This procedure was carried out separately for GPS- and accelerometer-based mobility data. For the main analysis, the upper triangle of the resulting covariance matrices from GPS and accelerometer metrics were concatenated and were used as input features for the individual identification procedure. We also repeated the identification procedure using GPS or accelerometer features alone. As a sensitivity analysis to test performance of alternative features for individual identification, we also computed the mean and the stability of each measure and used these features to identify participants. Stability was defined as the root mean square of the successive differences (RMSSD)$^8$ of each measure (Supplementary Figure 6). Individual identification procedure (“footprinting”) We randomly partitioned each subject’s data into two equally sized parts, named the “reference” and the “target”, respectively$^6$. The objective of the individual identification procedure was to match the subject from the target group to the same one in the reference group. For a given subject, $S$, we computed the Pearson correlation ($r$) between that subject’s features in the target group, $S_T$, and everyone’s features in the reference group, $S_R \ 1, S_R \ 2, S_R \ N$, where $N$ is the total number of participants. Individual identification was operationalized as the maximum of the resulting $r_1, r_2, \cdots r_N$. In other words, when the subject in the reference group having the mobility features that maximally correlated with that of the target subject, these two participants were declared correctly matched: $$S_M = \arg\max(r_1, r_2, \cdots r_N).$$ The individual identification accuracy was the number of correct identifications divided by the total number of random data partitions $P$: $$\text{individual identification accuracy} = \frac{\sum_{i=1}^{N} \begin{cases} 1, & \text{if } S_T i = S_M i \\ 0, & \text{if } S_T i \neq S_M i \end{cases}}{N}.$$ The above individual identification procedure was repeated 1,000 times, each time with a new random data partition (P). We calculated the average individual identification accuracy across the 1,000 runs, which yielded a distribution of sample-wise identification accuracy. Furthermore, we also calculated the accuracy for each participant, defined as the number of correct identifications for that specific participant divided by the number of data partitions ($B$). We refer to this participant-specific identification accuracy as the individual footprint distinctiveness: $$\text{individual footprint distinctiveness} = \frac{\sum_{i=1}^{B} \begin{cases} 1, & \text{if } P_T i = P_M i \\ 0, & \text{if } P_T i \neq P_M i \end{cases}}{B},$$ where $P_T i$ is target in a partition for subject $i$, and $P_M i$ is matched subject. We conducted the individual identification procedure using the covariance matrix of the GPS data, accelerometer data, as well as the combined feature set. Sensitivity analyses examined the mean and variance of each feature. Similarity matrix construction To visualize the individual footprint distinctiveness, we constructed similarity matrices between participants’ mobility covariance features. First, we concatenated the daily mobility metrics for a participant from multiple random data partitions. Next, a similarity matrix was constructed by computing the Pearson correlation coefficients between every pair of participants. The resulting matrix was a symmetric matrix, where the nodes were each participant and the edges were the correlation coefficients between any two participant’s mobility metrics. This grouping procedure was performed solely for visualization, highlighting the within-individual, across-partition block structures on the diagonal of the matrix. This grouping was not used in any statistical analysis. Permutation testing To assess the statistical significance of individual identification accuracy, we used a permutation testing procedure to create a null distribution of accuracy. Specifically, we randomly scrambled the identity of the daily mobility metrics, thus disrupting the linkage between the mobility data and the corresponding participant. We repeated the individual identification procedure for each random permutation. The empirical $p$-value was then calculated as the proportion of times when the permuted data yielded higher accuracy than the original data: $$P_{\text{permutation}} = \frac{\sum_{i=1}^{M} \begin{cases} 1, & \text{if } A_i \geq A_{\text{original}} \\ 0, & \text{if } A_i < A_{\text{original}} \end{cases}}{M},$$ where $A$ is the individual identification accuracy, and $M$ is the total permutations. Sensitivity analysis of data missingness To understand the effect of data missingness on our ability to identify participants’ mobility footprint, we conducted sensitivity analyses that used four sets of data constructed using different thresholds for data missingness$^3,10$. Specifically, we applied four thresholds with diminishing tolerance for the number of missing samples (i.e., minutes recorded) in a day’s worth of data to be included in analysis (Supplementary Figure 1). At the 100$^{th}$ percentile level, which corresponded to retaining all available days except for those with all data missing (or 1,440 minutes), 79 recording days were removed, which resulted 3,156 days remaining for analysis. At the 90$^{th}$ percentile, a further 216 days were removed, yielding 2,940 days for analysis. At 80$^{th}$ percentile, a further 356 days were removed, resulting in 2,584 days for analysis. Finally, at 75$^{th}$ percentile, a further 171 days were removed, resulting in 2,413 days remaining for analysis. Using these four sub-samples constructed with different inclusion criteria, we then repeated the individual identification procedure and permutation testing as described above. Feature lesion analysis To further investigate the influence of any single feature’s influence on the individual identification accuracy, we conducted a feature lesion analysis. We sequentially removed one metric (out of the total 15 geolocation mobility metrics available) and constructed a new covariance matrix which had one node (and 14 edges) less than the original feature covariance matrix. Using this reduced feature set, we repeated the individual identification and permutation testing procedures as described above (Supplementary Figure 3). Ecological momentary assessment Using the Beiwe platform application on personal smartphones, participants completed daily questionnaires specifically designed to assess mood variability at three timepoints throughout the day$^11$. In each survey, participants rated on a scale from 1 (“not at all”) to 7 (“extremely”) of their endorsement of seven statements assessing mood variability, aggression, impulsivity, and self-esteem since the last time they had answered the survey to capture their mood (Supplementary Table 3). All seven items were concatenated to create an overall mood scale. Additionally, every morning, participants were also asked about their sleep patterns and quality from the night before. To quantify the variability of answers to the mood survey, we calculated the root mean square of successive differences (RMSSD) between concatenated answers. Similarly, we also calculated the RMSSD of sleep duration as a measurement of its stability. We built a generalized additive model (GAM) to investigate the association between mood and sleep duration stability while accounting for covariate effects including data quantity, sex, age, and mean levels of the measure. Age was modeled using penalized splines within GAM using restricted maximum likelihood (REML) to estimate linear and nonlinear developmental effects without over-fitting the data\textsuperscript{12,13}. Functional Connectivity Analysis Imaging Acquisition As previously described\textsuperscript{14}, structural and functional MRI scans were acquired using in a single session on a clinically-approved 3 Tesla Siemens Prisma (Erlangen, Germany) quadrature body-coil scanner and a Siemens receive-only 64-channel head coil at the Hospital of the University of Pennsylvania. Prior to functional MRI acquisitions, a 5-min magnetization-prepared, rapid acquisition gradient-echo T1-weighted (MPRAGE) image (TR = 1810 ms; TE = 3.45 ms; TI = 1100 ms, FOV = 180 × 240 mm\textsuperscript{2}, matrix = 192 × 256, 160 slices, effective voxel resolution = 0.9375 × 0.9375 × 1 mm\textsuperscript{3}) was acquired. We used one resting-state (1200 volumes) scan as part of this study. All fMRI images were acquired with the same multi-band, interleaved multi-slice, gradient-echo, echo planar imaging (GE-EPI) sequence sensitive to BOLD contrast with the following parameters: TR = 500 ms; TE = 25 ms; multiband acceleration factor = 6, flip angle = 30°; FOV = 192 × 192 mm\textsuperscript{2}; matrix = 64 × 64; 48 slices; slice thickness/gap = 3/0 mm, effective voxel resolution = 3.0 × 3.0 × 3.0 mm\textsuperscript{3}. Image Processing All preprocessing was performed using fMRIPrep 20.0.7\textsuperscript{15}, which is based on Nipype 1.4.2\textsuperscript{16}, and XCP Engine\textsuperscript{17,18} (PennBBL/xcpEngine: atlas in MNI2009 Version 1.2.3; Zenodo: http://doi.org/10.5281/zenodo.4010846). The T1-weighted (T1w) image was corrected for intensity non-uniformity (INU) with N4BiasFieldCorrection\textsuperscript{19}, distributed with ANTs 2.2.0\textsuperscript{20}, and used as T1w-reference throughout the workflow. The T1w-reference was then skull-stripped with a Nipype implementation of the antsBrainExtraction.sh workflow (from ANTs), using OASIS30ANTS as target template. Brain tissue segmentation of cerebrospinal fluid (CSF), white-matter (WM) and gray-matter (GM) was performed on the brain-extracted T1w using FAST in FSL 5.0.9\textsuperscript{21}. Volume-based spatial normalization to MNI2009c standard space was performed through nonlinear registration with antsRegistration (ANTs 2.2.0), using brain-extracted versions of both the T1w reference and the T1w template. BOLD runs were first slice-time corrected using 3dTshift from AFNI 20160207\textsuperscript{22} and then motion corrected using mcflirt (FSL 5.0.9)\textsuperscript{21}. A fieldmap was estimated based on a phase-difference map calculated with a dual-echo GRE sequence, processed with a custom workflow of SDCFlows inspired by the epidewarp.fsl script and further improvements in HCP Pipelines\textsuperscript{23}. The fieldmap was then co-registered to the target EPI reference run and converted to a displacement field map with FSL’s fugue and other SDCflows tools. Based on the estimated susceptibility distortion, a corrected BOLD reference was calculated for a more accurate co-registration with the anatomical reference. The BOLD reference was then co-registered to the T1w reference using bbregister (FreeSurfer) which implements boundary-based registration\textsuperscript{24}. Co-registration was configured with nine degrees of freedom to account for distortions remaining in the BOLD reference. Six head-motion parameters (corresponding rotation and translation parameters) were estimated before any spatiotemporal filtering using mcflirt. Finally, the motion correcting transformations, field distortion correcting warp, BOLD-to-T1w transformation and T1w-to-template (MNI) warp were concatenated and applied to the BOLD timeseries in a single step using antsApplyTransforms (ANTS) with Lanczos interpolation. After pre-processing with fMRIPrep, confound regression was carried out in XCP Engine. Preprocessed timeseries were despiked and then de-noised using a 36-parameter confound regression model that has been shown to minimize the impact of motion artifact\textsuperscript{25}. Specifically, the confound regression model included the six framewise estimates of motion, the mean signal extracted from eroded white matter and cerebrospinal fluid compartments, the global signal, the derivatives of each of these nine parameters, and quadratic terms of each of the nine parameters as well as their derivatives. Both the BOLD-weighted time series and the confound regressor timeseries were temporally filtered simultaneously using a first-order Butterworth filter with a passband between 0.01 and 0.08 Hz to avoid mismatch in the temporal domain\textsuperscript{26}. Confound regression was performed using AFNI’s 3dTproject. Note that in-scanner head motion was also included as a covariate in all regression models (see below). \textit{Functional network and community connectivity} Functional connectivity between each pair of brain regions was quantified as the Fisher-transformed Pearson correlation coefficient between the mean regional BOLD time series. For each participant, a $200 \times 200$ weighted adjacency matrix encoding the connectome was constructed\textsuperscript{27}. Each node was assigned to one of seven canonical functional brain modules or communities defined by Yeo et al\textsuperscript{28}. The within-community connectivity is defined as $$\frac{\sum_{j,j' \in C_k} A^i_{jj'}}{\|C_k\| \times (\|C_k\| - 1)}$$, where $A^i_{jj'}$ is the weighted edge strength between the node $j$ and node $j'$, both of which belong to the same community $C_k$, for the $i$-th subject. The cardinality of the community assignment vector, $C_k$, represents the number of nodes in the $k$-th community\textsuperscript{29}. \textit{Mass-univariate analysis} For each of the seven canonical networks, we fit generalized additive model (GAM) to investigate the relationship between within-network connectivity and footprint distinctiveness, while controlling for in-scanner motion, mobility data quantity, sex, and age. Specifically, we used penalized splines using restricted maximum likelihood (REML) within GAM to estimate linear and nonlinear age-related changes\textsuperscript{12,13}. We controlled for multiple comparisons using the False Discovery Rate ($Q<0.05$). We fit a penalized regression model to predict footprint distinctiveness using brain functional connectivity\textsuperscript{6}. In each iteration of leave-one-out cross-validation, one subject was left out as the testing set and the rest the training set. Using the training set, we computed residualized footprint distinctiveness from a GAM model with covariates as above (linear terms for in-scanner motion, data quantity, sex; age was modeled with as a penalized spline). Then we fit a lasso regression model to predict the residualized footprint distinctiveness using a sparse collection of functional connectivity edges. L1 lasso hyperparameter was tuned in a nested leave-one-out fashion. Next, we calculated the predicted footprint distinctiveness for the unseen subject in the testing set. After all iterations, we obtained predicted footprint distinctiveness for all participants and then calculated the Pearson correlation between the actual footprint distinctiveness and predicted values. Code availability and data access The code for GPS data preprocessing, mobility metric extraction, individual identification, additional analysis, and data visualization is available in R on github: \url{https://github.com/PennLINC/footprinting} Code notebook is available at: \url{https://rpubs.com/hxia/footprinting} Data is available upon request. References 1. Calkins, M. E. et al. Persistence of psychosis spectrum symptoms in the Philadelphia Neurodevelopmental Cohort: a prospective two-year follow-up. World Psychiatry 16, 62–76 (2017). 2. Torous, J., Kiang, M. V., Lorme, J. & Onnela, J.-P. New Tools for New Research in Psychiatry: A Scalable and Customizable Platform to Empower Data Driven Smartphone Research. JMIR Ment. Heal. 3, e16 (2016). 3. Barnett, I. & Onnela, J.-P. Inferring mobility measures from GPS traces with missing data. Biostatistics 21, e98–e112 (2018). 4. Andridge, R. R. & Little, R. J. A. A Review of Hot Deck Imputation for Survey Non-response. Int. Stat. Rev. 78, 40–64 (2010). 5. Vega, J. et al. RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices. doi:10.2196/preprints.23246. 6. Finn, E. S. et al. Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. Nat. Neurosci. 18, 1–11 (2015). 7. Kaufmann, T. et al. Delayed stabilization and individualization in connectome development are related to psychiatric disorders. Nat. Neurosci. 20, 513–515 (2017). 8. Shaffer, F. & Ginsberg, J. P. An Overview of Heart Rate Variability Metrics and Norms. 9. Gratton, C. et al. Functional Brain Networks Are Dominated by Stable Group and Individual Factors, Not Cognitive or Daily Variation. Neuron 98, 439-452.e5 (2018). 10. Kiang, M. V et al. Sociodemographic Characteristics of Missing Data in Digital Phenotyping. medRxiv 2020.12.29.20249002 (2021) doi:10.1101/2020.12.29.20249002. 11. Shiffman, S., Stone, A. A. & Hufford, M. R. Ecological Momentary Assessment. Annu. Rev. Clin. Psychol. 4, 1–32 (2008). 12. Wood, S. N. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J. R. Stat. Soc. Ser. B (Statistical Methodol.) 73, 3–36 (2011). 13. Wood, S. N. Stable and Efficient Multiple Smoothing Parameter Estimation for Generalized Additive Models. J. Am. Stat. Assoc. 99, 673–686 (2004). 14. Jirsaraie, R. J. et al. Accelerated cortical thinning within structural brain networks is associated with irritability in youth. Neuropsychopharmacology 44, 2254–2262 (2019). 15. Esteban, O. et al. fMRIPrep: a robust preprocessing pipeline for functional MRI. Nat. Methods 16, 111–116 (2019). 16. Gorgolewski, K. et al. NiPyPe: A Flexible, Lightweight and Extensible Neuroimaging Data Processing Framework in Python. Front. Neuroinform. 5, 13 (2011). 17. Ćirić, R. et al. PennBBL/xcpEngine: atlas in MNI2009. (2020) doi:10.5281/ZENODO.4010846. 18. Ciric, R. et al. Mitigating head motion artifact in functional connectivity MRI. Nat. Protoc. 13, 2801–2826 (2018). 19. Tustison, N. J. et al. N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging 29, 1310–20 (2010). 20. Avants, B. B., Epstein, C. L., Grossman, M. & Gee, J. C. Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain. Med. Image Anal. 12, 26–41 (2008). 21. Jenkinson, M., Bannister, P., Brady, M. & Smith, S. Improved Optimization for the Robust and Accurate Linear Registration and Motion Correction of Brain Images. Neuroimage 17, 825–841 (2002). 22. Cox, R. W. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–73 (1996). 23. Glasser, M. F. et al. The minimal preprocessing pipelines for the Human Connectome Project. Neuroimage 80, 105–24 (2013). 24. Greve, D. N. & Fischl, B. Accurate and robust brain image alignment using boundary-based registration. Neuroimage 48, 63–72 (2009). 25. Ciric, R. et al. Benchmarking of participant-level confound regression strategies for the control of motion artifact in studies of functional connectivity. Neuroimage 154, 174–187 (2017). 26. Hallquist, M. N., Hwang, K. & Luna, B. The nuisance of nuisance regression: Spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity. Neuroimage 82, 208–225 (2013). 27. Schaefer, A. et al. Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cereb. Cortex 28, 3095–3114 (2018). 28. Yeo, B. T. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011). 29. Xia, C. H. et al. Multi-scale network regression for brain-phenotype associations. Hum. Brain Mapp. 41, 2553–2566 (2020). Constructing personal mobility “footprints”. a) We collected 3,317 person-days of mobility sensing data via personal smartphones from 41 adolescents and young adults. Geolocation data were recorded in cycles of 2min on and 18min off. Raw geolocation coordinates were de-identified via sphere to-2D standard space projection and were further imputed for missing data. b) For each individual, we constructed daily personal mobility trajectories, which consist of flights (movement) and pauses (stationary segments). Length of linear lines represents the duration of flights and size of circles represents the duration of pauses. Warm and cold colors indicate daytime and nighttime, respectively. c) A representative week of trajectories is shown, which demonstrates rich characteristics of personal mobility patterns formed over time. d) We extracted timeseries of mobility statistics (e.g. daily time spent at home) from geolocation and accelerometer data that parameterize movement characteristics over weeks to months. The example represented all 110 days of participants’ geolocation metrics recorded. e) For each individual, we constructed a covariance matrix from the mobility metric timeseries. Each cell of the matrix was populated by the Pearson correlation between a given pair of mobility metrics. Warm and cold colors indicate positive and negative correlations, respectively. f) We randomly divided data into two equally sized parts, called the reference and target set. Subj X from the target set was matched to the subject in the reference that had the highest correlations between their footprints (argmax(r1, r2, ..rN)). The identification was considered correct when underlying data came from the same subject; otherwise, the identification was considered incorrect. We quantified individual identification accuracy as the proportion of correct identifications across the entire sample; this procedure was repeated 1,000 times across different random partitions of the data. Figure 2 Identifying individuals using personal footprints. a) As an initial step, we visualized the similarity of mobility features across multiple random reference and target partitions (R & T in inset). It was readily apparent that mobility features were more highly correlated within participants (on diagonal) across data partitions than between participants (off diagonal). Note that this visualization was not used in statistical analysis or individual identification. b) Across 1,000 random partitions, mobility footprinting enabled successful individual identification (mean: 63%, S.D.: 6%). In contrast, the mean chance accuracy from 1,000 permutation was 3% (inset, p < 0.001). c) For each individual, we calculated the footprint distinctiveness, or the percentage of correct identification across the 1,000 random partitions of the data. Ranked in ascending order, participants’ footprint distinctiveness exhibited a wide range, from 4% to 99%. However, even the participant with the lowest identification distinctiveness was significantly higher than the null distribution. d) Individual identification based on geolocation alone had higher accuracy than accelerometer alone. However, they appeared to encode complementary features, as performance was maximal when both measures were used in footprinting. Figure 3 Individual footprint distinctiveness is associated with affective instability, sleep irregularity, and patterns of brain functional connectivity. a) Greater affective instability, measured by root mean square of successive differences in mood measures from ecological momentary assessment items acquired three times a day, was associated with reduced footprint distinctiveness ($r = -0.37$, $p < 0.05$), after controlling for data quantity, age, sex, and mean level of mood ratings. b) Similarly, we found that increased variability in sleep duration was associated with reduced footprint distinctiveness ($r = -0.36$, $p < 0.05$), after controlling for covariates. c) Across functional brain networks, only greater connectivity within the somatomotor network had a significant association with footprint distinctiveness ($r = 0.46$, $p < 0.05$, corrected for multiple comparisons with the false discovery rate). d) Patterns of brain functional connectivity significantly predicted individual footprint distinctiveness using leave-one-out crossvalidation ($r = 0.29$, inset: permutation-based $p = 0.025$). e) Six network edges consistently contributed to the sparse regression model. These edges included greater connectivity within somatomotor network, reduced connectivity between left and right frontal eye fields (FEF), increased connectivity between the somatomotor network and the left orbital frontal cortex (OFC) in the limbic network, as well as increased connectivity between the vLPFC (ventrolateral prefrontal cortex in the frontoparietal network) and the dmPFC (dorsolateral prefrontal cortex in the default mode network). Cord thickness reflects the weights in the model, reflecting each edge’s contribution to the prediction; cord color indicates the sign of the weights. Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. - MobileFootprintingSupplementaryMaterial20210412.pdf
THE CONTRACT LABOUR (REGULATION AND ABOLITION) (GUJARAT AMENDMENT) BILL, 2020. A BILL further to amend the Contract Labour (Regulation and Abolition) Act, 1970 in its application to the State of Gujarat. सन २०२०नं गुजरात विधेयक क्रमं: १८. कन्ट्रॅक्ट मजूर (नियमन अने नाबुदी) (गुजरात सुधारा) विधेयक, २०२०. गुजरात राज्यने लागू पडतो कन्ट्रॅक्ट मजूर (नियमन अने नाबुदी) अधिनियम, १९७० वच्य सुधारवा वापस विधेयक. आधी, भारतना गायराज्यना एकोत्रेमा वार्षिमां नीयेने अधिनियम करवामां आवे छः:- १. (१) आ अधिनियम कन्ट्रॅक्ट मजूर (नियमन अने नाबुदी) (गुजरात सुधारा) अधिनियम, २०२० करवावी. (२) ते सन २०२०ना जुलाई महिनानी २०मी तारीखे अहमतमां आवेचो होवानु गायबो. सन १९७०नो अमो. २. गुजरात राज्यने लागू पडतो कन्ट्रॅक्ट मजूर (नियमन अने नाबुदी) अधिनियम, १९७० (छेनी आमां हवे पहली "मुख्य अधिनियम" तरीके उल्लेख कर्यो छ ते)मां, कलम ७मां, पेटा-कलम (४)मां,- (१) पंक (५)मां, "वीस" ओ शब्दने बदले, "पचास" ओ शब्द मुक्तवो; (२) पंक (६)मां, "वीस" ओ शब्दने बदले, "पचास" ओ शब्द मुक्तवो; (३) परंतुकमां, "वीस" ओ शब्दने बदले, "पचास" ओ शब्द मुक्तवो. सन २०२०नो गुजरात वटहकम अमो. ३. (१) कन्ट्रॅक्ट मजूर (नियमन अने नाबुदी) (गुजरात सुधारा) वटहकम, २०२०, आधी, २३ करवामां आवे छः. (२) आवी रीते २३ थया छतां, सदरहु वटहकमथी सुधारेला मुख्य अधिनियम हेडल करेलु कंघपार कार्य अथवा लीघेलु क्रिया पावलु, आ अधिनियमथी सुधारेला मुख्य अधिनियम हेडल करेलु अथवा लीघेलु होवानु गायबो. ઉદેશી અને કારણો કન્ટ્રાક્ટ મજૂર (નિયમન અને નાબુધી) અધિનિયમ, ૧૯૭૦, ભારત સરકારે, હથોગોમાં કામે રાખવામાં આવેલ મજૂરની સેવાની શરતો અને ચુકવાશીયોનું નિયમન કરવા અને તે સુનિશ્ચિત કરવાના ઉપયોગ તરીકે અધિનિયમના કરયો છે. ક્રીડિક-૧૫ને તારીખ પ્રતિસ્તિત વર્તમાન પરિસ્થિતિ દરમિયાન, આંગ્લક પ્રૂણ્યતમાં વધારા કરવી જુઓ છે અને રાજ્યમાં આંગ્લક સુધારણા પેટેશન આપીને, ૩૪ સંસ્થાઓ અને કન્ટ્રાક્ટર પણ અથવા તેથી વધુ લોકોને (અગાઉ ૨૦) લોકિયે (કામે) રાખતી/રાખતા હોય તેમને આ અધિનિયમ હેઠળ આવતરી લાખ શકાય એમ છે. એમ પણ જાણાયું છે કે વિદ્યમાન અધિકતમ મધ્યદાનના કરારો, નાના એકમોને અધિનિયમ હેઠળ આપયારિકતામાં સુનિશ્ચિત કરવામાં મુખ્યત્વી પણતી હોવાયું, મુખ્ય માલિકો માટે કામ્યારીવર્ગના કામે રાખવામાં અથવા ચીજવસ્તુઓ ઉપલભ્ય કરાવતી વધારે કરારેણુ પાલતન કરવું મુખ્યત્વે અને છે. એવું રહયાને આદાયું છે કે નીચેથી મધ્યદાનને કારણો કાંતી અપાલતનને ઉટેજણ માટે છે અથવા માંગ મુજબ જુઓ મજૂરને કામે રાખવા પર નિયંત્રણ કરે છે. તેથી, એવું સ્થિત કરવામાં આદાયું છે કે વિદ્યમાન અધિકતમ મધ્યદાનમાં ઉપર સુધારો કરવાની જુઓ છે જ્યાં કરીને રાજગારની વધુ તાક પૂરી પાડી શકાય અને નાના એકમોને ધાંચી કરવામાં સરળતા રહે. ગુજરાત રાજ્યની વિદ્યાનસ્વાનું સત્ત તે સમયે ચાલુ ન હોવાયું, ઉપદ્યુક્ત ઉદેશ સિલ્સ કરવા માટે, કન્ટ્રાક્ટ મજૂર (નિયમન અને નાબુધી) (ગુજરાત સુપારા) વટફુકમ, ૨૦૨૦ પુસિલ્સ કરવામાં આવ્યો હતી. આ વિદ્યયક્ષી, રાજ્ય વિદ્યાનમંડળના અધિનિયમ દ્વારા સદરહુ વટફુકમ બદલવા ધાર્ય છું તારીખ: ૧૧મી સપ્ટેમ્બર, ૨૦૨૦. હિલ્લીપુંકુમાર નાશેર. પુરવશી કન્ટ્રોક્ટ મજૂર (નિયમન અને નાબૂડી) અધિનિયમ, ૧૬૭૦ (સન ૧૬૭૦ના ૩૫મા)માંથી ઉતારા. \| ત્રૈકી સંખ્યા, વ્યાપ્તિ, આરેક અને લાગુ પરિવા બાબત. \| ૧. (૧) શી (૩) (૪) તે- (૫) કન્ટ્રોક્ટ મજૂર તરીકે પહેલાના બાર મહિનાના કોઈ દિવસે વીસ અથવા વધુ કામદારોને જેમાં કામે રાખયા હોય અથવા કામે રાખયા હતા તેવા એરેક સંસ્થાને; (૬) વીસ અથવા વધુ કામદારોને પહેલાના બાર મહિનાના કોઈ દિવસે જે કામે રાખે અથવા જેથી કામે રાખયા હતા તેવા એરેક કોન્ટ્રોક્ટરને, લાગુ પડે છે: પરણુ સમૃદ્ધત સરકાર, તેવું કરવાના પોતાના ધરાદાની બે ઓષ્ઠમાં ઓષ્ઠ બે મહિનાની નોટિસ આદાય પણી, રાજપરંતુમાં જાહેરનામાથી, જાહેરનામાથી નિદિષ્ટ કરવામાં આવે તેવા ઓષ્ઠમાં ઓષ્ઠ વીસ કામદારોની તરતી સંખ્યાને કામે રાખનાર કોઈ સંખ્યા અથવા કોન્ટ્રોક્ટરને આ અધિનિયમની જોગવાયાયો લાગુ પડે. \| XXX \| XXX \| XXX \| [सन २०२०नं गुजरात विधेयक क्रमांक : १८.] गुजरात राज्यने लागु प्रतीक डन्ट्रेक मजूर (नियमन अने नापूर्ती) अधिनियम, १६७० वर्ष सुधारवा वाभत विधेयक. [श्री डिल्लीपुमार चाकर, श्रम अने रोजगार मंत्रीश्री.] (सन २०२०ना सप्टेंबर महिनानी १४मी तारीखे गुजरात सरकारी राजपत्रमां प्रसिद्ध कर्यां मुख्य) ड. एम. पटेल, सारिव, गुजरात विधानसभा. THE CONTRACT LABOUR (REGULATION AND ABOLITION) (GUJARAT AMENDMENT) BILL, 2020. GUJARAT BILL NO. 18 OF 2020. A BILL further to amend the Contract Labour (Regulation and Abolition) Act, 1970 in its application to the State of Gujarat. It is hereby enacted in the Seventy-first Year of the Republic of India as follows:- 1. (1) This Act may be called the Contract Labour (Regulation and Abolition) (Gujarat Amendment) Act, 2020. (2) It shall be deemed to have been come into force on the 20th July, 2020. 2. In the Contract Labour (Regulation and Abolition) Act, 1970, in its application to the State of Gujarat (hereinafter referred to as “the principal Act”), in section 1, in sub-section (4),- (i) in clause (a), for the word “twenty”, the word “fifty” shall be substituted; (ii) in clause (b), for the word “twenty”, the word “fifty” shall be substituted; (iii) in the proviso, for the word “twenty”, the word “fifty” shall be substituted. 3. (1) The Contract Labour (Regulation and Abolition) (Gujarat Amendment) Ordinance, 2020 is hereby repealed. (2) Notwithstanding such repeal, anything done or any action taken under the principal Act as amended by the said Ordinance, shall be deemed to have been done or taken under the principal Act as amended by this Act. STATEMENT OF OBJECTS AND REASONS The Contract Labour (Regulation and Abolition) Act, 1970 has been enacted by the Government of India as a means of regulating and ensuring the conditions of service and the payments to the labour employed by the industries. During the present situation prevailing due to the COVID – 19, it is necessary to boost the economic activity and by way of offering an economic revival package in the State, it is felt that the establishments and contractors which employ more 50 people or more (previously 20) may be covered under the Act. It is also felt that because of the existing threshold limit, principal employers while hiring personnel or procuring commodities find it difficult to execute contracts, as the small units face hardship in ensuring formalities under the Act. It has been observed that the lower limit either encourages non-compliance or restricts the engagement of required labour as per demand. It is therefore, proposed that the existing threshold limit needs upward revision so as to provide more opportunity of employment and facilitate business in small units. As the Legislative Assembly of the State of Gujarat was not in session, at that time, the Contract Labour (Regulation and Abolition) (Gujarat Amendment) Ordinance, 2020 was promulgated to achieve the aforesaid object. This Bill seeks to replace the said Ordinance by an Act of the State Legislature. Dated the 11th September, 2020. DILIPKUMAR THAKOR. 1. (1) to (3) XXX XXX XXX (4) It applies- (a) to every establishment in which twenty or more workmen are employed or were employed on any day of the preceding twelve months as contract labour; (b) to every contractor who employs or who employed on any day of the preceding twelve months twenty or more workmen: Provided that the appropriate Government may, after giving not less than two months' notice of its intention so to do, by notification in the Official Gazette, apply the provisions of this Act to any establishment or contractor employing such number of workmen less than twenty as may be specified in the notification. GUJARAT BILL NO. 18 OF 2020. A BILL further to amend the Contract Labour (Regulation and Abolition) Act, 1970 in its application to the State of Gujarat. [SHRI DILIPKUMAR THAKOR, MINISTER FOR LABOUR AND EMPLOYMENT] (As published in the Gujarat Government Gazette of 14th September, 2020) D.M.PATEL, Secretary, Gujarat Legislative Assembly.
Bootstrapping paired-object affordance learning with learned single-affordance features Emre Uğur, Sandor Szedmak, and Justus Piater Intelligent and Interactive Systems, Institute of Computer Science, University of Innsbruck Abstract—The aim of this paper is to propose a system where complex affordance learning is bootstrapped through using pre-learned basic affordances as additional inputs of the complex affordance predictors or as cues in selecting the next objects to explore during learning. In the first stage, the robot learns affordances in the form of developing classifiers that predict effect categories given object features for different discrete actions applicable to single objects. These predictions are later used to robot’s feature set as higher-level affordance features. In the second stage, the robot learns more-complex multi-object affordances using object and affordance features. We first applied our idea in an artificial interaction database which includes discrete actions, several manually coded object categories, and actions effects. Finally, we validated our bootstrapping approach in a real robot with poke and stack actions. We expected to obtain higher performance with affordance-features especially in small training datasets as the object-robot-environment dynamics should have already been partially learned and encoded in affordances. The experiment results showed that complex affordance learning significantly speeds up with predictors that are bootstrapped with affordance-features compared to predictors that use low-level features such as shape descriptors. We also showed that by actively selecting the next objects and by increasing the diversity of the training set using a distance measure based on learned single-object affordances, the effect of bootstrapping can be further increased. I. INTRODUCTION This study is part of a research effort where a robot system gradually develops skills and competencies in subsequent stages of development, similar to human infants. In our previous work, we showed that similar to human infants who learn a set of actions by the age of 7 months such as grasp, hit and drop [1], a robot could also self-discover a number of behavior primitives such as push, grasp and release by interacting with objects using its crude ‘reach’ action and grasp reflex, and observing the changes in its tactile perception [2]. Next, we showed that similar to infants who learn object dynamics after 7-9 month of age, our robot could learn affordances in an unsupervised way by first discovering the effect categories it could generate in the environment, and then by learning the mapping from the object features to the effect categories. After learning, the robot was shown make plans to achieve desired goals, emulate end states of demonstrated actions, monitor the plan execution and take corrective actions using the perceptual structures employed or discovered during learning. Finally, we showed that more complex actions that involve multiple objects (such as bring object 1 over object 2) can be taught to the robot through imitation using the structures developed in the previous stages with mechanisms inspired from parental scaffolding and motionese [3]. In the current study, we assume that a number of actions (such as push and stack) and effect categories (such as rolled and pushed), which were discovered in the previous stages as summarized above, are transferred to the next stage where complex affordances such as stackability are learned. We study how this complex affordance learning can be bootstrapped by use of learned simple affordances as (i) additional inputs in prediction, and (ii) in active selection of objects to explore next in an active learning setting. One hallmark feature of bootstrapped learning is that learning problems stack in the sense that higher-level learners use as input attributes concepts produced by lower-level learners. These higher-level attributes should allow faster learning than if the higher-level concepts had to be learned from the lower-level attributes alone. The aim of this paper is to propose a learning system where a developmental robotic system benefits from bootstrapping where learned simpler structures (affordances) that encode robot’s interaction dynamics with the world are used in learning of complex affordances. In detail, our robot learns the affordances of single objects and uses these affordances as additional features in the next stages of development where paired-object affordances are discovered. The use of learned similarities in the form of affordances are expected to bootstrap the learning in the next stages. Our approach can be explained by the following intuitive example: Let us assume that the robot learned rollability affordances of the objects in the first development stage, and can now predict the rollability based on object shape properties. In the next stage, robot learns a more complex affordance such as stackability from two sample interactions where it observes that stacked two balls tumble over and stacked two boxes pile up. The robot, trained only with those two stacking interactions, can find a correspondence between stackability and rollability. Then, even if the robot does not have any stacking experience with cylindrical objects, it can make better predictions for stackability depending on the roll orientation (and affordance) of the cylinders. In the context of robot affordance learning research, paired-object affordance learning has not been studied extensively with exceptions of [3] where ‘tool objects’ are interacting with other objects, and [4] where two-object relational interaction models were directly learned. However none of these studies attempted to bootstrap their robot-object-object interaction dynamics with previously obtained skills and affordance knowledge. Our method summarized in the next section is validated with an artificial interaction dataset that includes rich set of objects and interactions in Section III, in real world with robot experiments in Section IV. II. METHOD Learning of affordances corresponds to learning the relations between objects, actions and effects [5]. In this study, affordances are acquired through learning to predict what type of effects, i.e. discrete effect categories, can be generated given discrete robot actions and continuous object properties. To achieve this, we simply train a classifier for each action, which takes object features as input and predicts the effect category. A. Object features Here, we distinguish two different sets of object features. The first set includes hand-coded basic general-purpose features, computed from visual perception$^1$ with no explicit link to robot’s actions. These may include standard features used in literature related to size, shape and local distance properties of the objects. The second set of features are acquired through interaction and they correspond to the higher-level learned ones that are computed from basic features. They encode the dynamics between robot actions $(A)$ and object response (effect, $e$). The first set of features is called basic-features whereas the second that is learned through interaction is called affordance-features as the latter includes the relations between objects, actions and effects. The straightforward approach to learn effect prediction is to train a classifier $c$ for each action $a$ that takes basic-features as input: \[ c^a_{\text{basic}}(\text{basic-feat}) \rightarrow \text{effect} \] whereas we propose to speed up learning of complex effect prediction using affordance-features that are computed using the learned basic effect prediction: \[ c^a_{\text{complex}}(\text{basic-feat}, c_{\text{basic}}(.)) \rightarrow \text{effect} \] which, in a flat form, corresponds to: \[ c^a_{\text{complex}}(\text{basic-feat}, \text{affordance-feat}) \rightarrow \text{effect} \] Our approach is summarized in Fig. 1. The features shown with blue and red solid lines correspond to basic-features and action predictions based on these features give rise to affordance-features. The dashed lines correspond to affordance-features, that are learned in previous stages. The learning and prediction of complex affordances benefit from previously learned affordance features as shown in ‘Predict effect of action k’ predictor. Note that action k is considered to be a complex action as two objects are involved in execution. In particular affordance features are represented as a vector of categorical variables, i.e. the list of the effect categories, predicted to be generated by single-object actions: \[ \text{affordance-feat} = (\varepsilon^o_{a_1}, \varepsilon^o_{a_2}, ...) \] where \[ \varepsilon^o_{a_i} = c^a_{\text{basic}}(\text{basic-feat}) \] Complex affordance learning can be realized in different ways. In this paper, the action possibilities that are provided by two (or more) objects are considered to be complex. For instance, the effects created by a stack action (where the object is grasped and released over another one) is determined by the properties of both objects. We will use affordance-features (such as rollability, pushability, etc) and basic-features to learn and predict stackability affordances, and show that this learning significantly speeds up with predictors that are bootstrapped with affordance-features. B. Active object selection based on affordances We claim that the bootstrapping effect can be further increased if the objects to be explored (and learned next) are selected intelligently. A learner which is provided with a rich set of qualitatively different objects in its initial phases of development can perform better compared to the ones trained with complete random objects. Thus, an online learning system actively selects the next object to maximize the diversity of the training set, and the learned single-object affordances will be used as ‘high-level’ similarity measures between objects in computing this diversity. --- $^1$In this paper, we limit ourselves with the features that can be captured by vision only. However, object properties such as object friction or weight plays an important role on object-robot interaction dynamics. Thus, exploratory actions that can be used to perceive such properties should be implemented in a full-fledged scenario. The next object is selected from the set of possible objects (PossObjs) by maximizing the total distance from the next object to the already explored objects (ExpObjs) as follows: \[ nextObj = \arg\max_{o_1 \in PossObjs} \sum_{o_2 \in ExpObjs} dist_t(o_1, o_2) \] where $dist_t(o_1, o_2)$ is the distance between two objects in the space of affordances. \[ dist_t(o_1, o_2) = \sum_a (1 - \delta_{\varepsilon^{o_1}_a, \varepsilon^{o_2}_a}) \] where $a \in A_{\text{Basic}}$, and $\delta_{i,j}$ is Kronecker delta function. ### III. BOOTSTRAPPING IN ARTIFICIAL DATA In this section, we report our bootstrapping results obtained from a manually prepared artificial database of objects and interactions. The set of objects include cylinders, boxes, spheres and triangular prisms in different orientations and with/without holes as shown in Fig. 2. The set of manually encoded actions and their effects are as follows: - Actions: \{side-poke, top-poke, front-poke, stack\} - Poke-effects: \{pushed, rolled, toppled, resisted, nothing\} - Stack-effects: \{piled-up, inserted-in, covered, tumbled-over\} When poked from different directions, hypothetically, different effects can be generated with these objects. For example, when poked from side, lying cylinders would roll away, boxes would be pushed, objects with holes in poke direction would not be affected as finger would go through the hole without any interaction, and the tall objects would topple down. The effect of stacking objects on top of each other depends not only on their shape but also on their relative size as well. For example, while ‘inserted-in’ effect would generated when a small box is stacked on a hollow cylinder, ‘piled-up’ effect would be observed when the box is larger than the opening on top of the cylinder. Based on these assumptions, we manually created a hypothetical set of rules that give the effect based on object categories and their relative sizes. #### A. Basic and affordance features The classifier trained with basic-features uses the following features for training (and prediction later): \[ TS_{\text{basic}} = \{(shape^{o_1}, shape^{o_2}, dim^{o_1}, dim^{o_2})\} \] where $shape$ includes mean and variance of the normals of the lateral surfaces, and the direction of the hole if it exists; and $dim$ encodes the object size in different axes. The classifier trained with affordance-features uses the following features: \[ TS_{\text{aff}} = \{(\varepsilon^{o_1}_{\ast,\text{poke}}, \varepsilon^{o_2}_{\ast,\text{poke}}, dim^{o_1}, dim^{o_2})\} \] where $\varepsilon^o$ refers to the effects of the corresponding poke action on the object $o$. #### B. Bootstrapping Results The performances of the classifiers trained with basic-features and affordance-features are provided in Fig. 3. We evaluated the classifiers by systematically changing the number of categories used in training set. For each number of categories, we trained 10 classifiers by selecting 5 objects of random size from each training category. To test these classifiers, we created test sets with random sized object from the remaining categories. Each bar corresponds to mean performance of these 10 classifiers. As shown, the prediction performance of both basic-features and affordance-features based classifiers improve by including more categories into the training set. We also included the performance of a category based predictor (which takes category index as input) to show the baseline. Because the categories used in training set are never included into test set, category-based predictors do fail independent of the training set size. ![Fig. 2. The set of objects used in the artificial interaction database.](image) ![Fig. 3. The effect prediction performance of stack action obtained in artificial interaction database. The training of classifiers are done with the indicated number of categories with either shape features or affordance features. As shown, use of affordance features enable bootstrapping of the learning system.](image) These results show that because affordance-features already include properties related to object dynamics (pushability, rollability etc), classifiers that use these features have better performance especially for small training sets. With the increasing training set size, the effect of using high-level features is reduced as the basic-features classifier can also find the invariance related to stackability affordance with large dataset. Finding this invariance with small datasets is easier with affordance-features as they already include some properties of the agent-object-environment interactions. ### IV. BOOTSTRAPPING IN REAL WORLD This section provides the details of the real world experiments where the effect of bootstrapping is analyzed. We first present the robot setup along with the details of robot’s action control and perception. Next, we showed that single-object affordances can be learned through interaction, and use of these acquired single-object affordances bootstraps paired-object affordance learning. A. Robot system The robot system employs a 7 DOF Kuka Light Weight Robot (LWR) arm, which is placed on a vertical bar similar to human arm in Fig. 4. A 3 fingered Schunk gripper is mounted on the arm to enable manipulation. For environment perception, Kinect sensor placed over the torso is used. The objects shown in Fig. 4 are used in learning single-object affordances as well as pairwise-affordances. 1) Object features: The robot’s workspace consists of several objects and a table where the region of interest is defined as the volume over the table. First, the point cloud obtained from Kinect is transformed to robot’s task space, then table is removed from the point cloud with a filtering along z-axis (see Fig 4), and finally objects are segmented based on depth information. Point Cloud Library normal estimation software is used next to compute a normal vector for each point of the object. The projection of each normal vector along each axis is separated, and histograms of normal vectors along each axis are computed. Using 8 bins for each histogram, $3 \times 18 = 54$ sized feature vector is obtained for shape related features. Note that in these experiments, an object is represented by a feature vector composed of only shape related features. Please see [5] for more details on histogram representation of normal vectors. 2) Robot Actions: The robot is equipped with a number of manually coded actions that enable single and paired object manipulation. The robot can ‘poke’ a single object from its side, front and top with s-poke, f-poke, and t-poke actions, respectively. It can also stack one object on the other using stack behavior, where it grasps the first object, move it on top of the other one and release it. The object position in world coordinate (shown in Fig. 4) is computed using the depth image of Kinect sensor. An inverse kinematic solver is used to compute the joint angles for initial and final points defined in Cartesian space, and Reflexxes library [6] is utilized to generate smooth trajectories to achieve point-to-point movement. The action execution is as follows: - Regarding to poke actions, the robot gripper is placed on one side of the object with 5cm distance with an orientation depending on the poke type. Two of the fingers are flexed to enable only the third finger to physically interact with the object (similar to index finger poking in humans). Next, the robot hand moves in the corresponding direction for 10cm towards the object and it is retracted after the poking is completed. - Regarding to stack action, one object is grasped from above first by placing the gripper in a vertical orientation 10cm over of the object, then moving the wide-open gripper towards the object and finally enclosing it. Next, the gripper that carries the grasped object is repositioned over the second object in a vertical orientation again, and the object in the gripper is released over the first one by extending all the fingers. 3) Effect Categories: In the real world experiments, depending on the object(s) and the action executed, different effects were generated. When poke action was executed, the object was pushed, toppled over or rolled away depending on its shape. There was no effect in object state or robot’s sensors if the robot finger went through the hole on the object. Finally, for t-poke action, all solid objects created resistance and obstructed gripper’s movement that was detected using the force sensors. When stack action was executed, the objects in general piled up on top of each other if the object below provided a proper support (for example if it had a flat top surface). Depending on the existence of concave surfaces and holes, the released object was inserted in or hided the object below by encapsulating it. The released object also tumbled over due to the lack of stable support. Based on the above possibilities that we observed empirically, the sets of effect categories ($\mathcal{E}$) were set same as in previous section. B. Experiment Results 1) Learning single-object affordances: The robot executed its poke actions on the objects (Fig. 4) placed in different orientations, and it collected 24 interaction instances for each poke action. The object shape features along with generated effect categories are stored for learning affordances. Support Vector Machine classifiers are used to learn the mapping between object features and effect categories. In order to analyze if the affordances for poke action are generalizable, we divide the interaction set into training and \| Object from robot’s view \| Explanation \| Front-poke prediction \| Side-poke prediction \| Top-poke prediction \| \|--------------------------\|-------------\|-----------------------\|----------------------\|---------------------\| \| ![Image] \| Lying cylinder \| pushed \| rolled \| resisted \| \| ![Image] \| Lying mug \| rolled \| pushed \| resisted \| \| ![Image] \| Cylinder wall \| pushed \| pushed \| no-change \| \| ![Image] \| Tennis ball \| rolled \| rolled \| resisted \| \| ![Image] \| Lying mug \| pushed \| rolled \| resisted \| \| ![Image] \| Wooden block \| pushed \| pushed \| resisted \| \| ![Image] \| Lying mug \| pushed * \| rolled \| resisted \| \| ![Image] \| Tea box \| pushed \| pushed \| resisted \| \| ![Image] \| Thin wooden block \| pushed \| toppled * \| resisted \| \| ![Image] \| Thin wooden block \| toppled \| pushed \| resisted \| Fig. 6. Robot’s basic-affordance prediction on objects which are not included in the training set with the same orientations. Prediction fails in the examples with star (), which are difficult cases to predict. test sets with the deliberate purpose of distributing objects with same affordances into different sets. For each poke action, we trained a classifier using the objects given in Fig. 5. Then, we tested these classifiers by predicting the action effects on novel objects given in Fig. 6. As shown, the robot was able to detect the affordances of the object (in terms of effect prediction) correctly, except a small number of cases shown with stars (), where the prediction also required perception and learning of material properties. The trained three classifiers (for $s$-poke, $t$-poke, and $f$-poke) are transferred to the next stage and their predictions are used as high-level features to learn complex affordances. 2) Learning paired object affordances: In this section, the robot learns the paired-object affordances by exploring the two-object environments with its stack action. This learning is again achieved by training an SVM classifier that predicts the effect of the stack action given object features. Here we compare the prediction performance of the classifiers that are trained either with basic-features or affordance-features. Regarding to basic-features, normal vector histograms are used as we did in learning single-affordances in the previous subsection. Regarding to affordance-features, the list of effect predictions (provided by the classifiers transferred from the previous stage) for the poke actions are used. The robot executed stack action with 18 pairs of random objects. A number of snapshots taken during these interactions are given in Fig. 7, where all the possible effects were observed with different object pairs. In each interaction, basic-features and affordance-features of both objects are computed and stored along with the observed effect category. The classifier trained with basic-features uses the following features for training (and prediction later): $$\text{TS}_{\text{basic}} = \{(shape^{o_1}, shape^{o_2})\}$$ and the classifier trained with affordance-features uses the following features: $$\text{TS}_{\text{aff}} = \{(\varepsilon_{s\text{-poke}}^{o_1}, \varepsilon_{f\text{-poke}}^{o_1}, \varepsilon_{t\text{-poke}}^{o_1}, \varepsilon_{s\text{-poke}}^{o_2}, \varepsilon_{f\text{-poke}}^{o_2}, \varepsilon_{t\text{-poke}}^{o_2})\}$$ where $\{\}$ corresponds to the set operator. ![Sample interactions observed during stack action execution.](image) We evaluated the performance of these classifiers by systematically changing the size of the training set. For each training set size, we trained 10 classifiers using randomly selected samples. We tested each classifier using the remaining sample interactions. Fig. 8 gives these cross-validation results. Real-world experiment results are similar to the results obtained from the synthetic interaction dataset. As the affordance-features were obtained through interaction with the environment, they already encode the object-environment dynamics, which provides bootstrapping effect in learning multi-object affordances as shown. The basic-features are real valued larger sized vectors that encode object shape properties independent of robot-object dynamics. Thus they require more training data for learning. Additionally, basic-features are used in computation of affordance-features, so they contain the information to make predictions with the performance of affordance features. As shown in the figure, with increasing number of training samples, basic-features based classifier’s performance indeed approached to the bootstrapped classifier’s performance. V. ACTIVE SELECTION OF OBJECTS BASED ON AFFORDANCES We used affordance distance measure defined in Section II.B, to maximize the diversity in training of paired-object affordance learning in an online learning setting. 83 objects and their poke and stack effects are used in this experiment (please see [7] for the complete list of objects). The object features are computed from Kinect depth image as in previous section, however, we used a human expert to label stack effect categories and it was not feasible to execute $83 \times 83 = 6889$ stack actions in the real robot. Fig. 9 shows the resulting performance of the basic-features and affordance-features based effect predictors trained with randomly selected objects, and of the effect predictors trained with the proposed active object selection strategy. Each type of predictor was trained 10 times, starting from a different random set of objects, and the thick lines correspond to the average accuracy for each predictor type. As shown, active selection of objects based on single-affordances provides a further bootstrapping effect in learning paired-object affordances. The best predictor was trained with affordance-features, but we observed that a similar performance was achieved even if it was trained with basic-features. VI. CONCLUSION Single-object affordances encode characteristics related to robot-object-environment dynamics as they are learned through robot’s interaction with the objects. In this study, we showed that learned basic affordances can be used as additional features in order to bootstrap the next stage of development where complex paired-object affordances are learned. In our general model, affordances are used as ‘additional features’ for learning complex affordances, but in the experiments, in order to compare their independent performances, we used either only basic features or only affordance features. If an important basic action (such as top-poke) was unavailable, the affordance features, i.e. effect predictions for side-poke and front-poke actions, would have failed to predict insertability. Therefore, both channels should be used during learning and possibly a feature selection algorithm can filter out unnecessary channels. While this work serves as one of the proof-of-concept application of the structural bootstrapping idea, we need to adapt advanced representations and learning methods (such as knowledge propagation framework of [8]) that can truly exhibit the real potential of this idea. We showed that this bootstrapping enabled the robot to speed up its learning particularly with small training data. Recently generative models have been proved to be effective in their ability in capturing object-action-effect dynamics, and in making predictions in different directions, for example in inferring the required actions to achieve desired effects given object properties [9], [4]. Particularly, hierarchical Bayesian networks directly encodes the desired structure and allows inference in several directions [10]. We discuss that our ‘discriminative’ model still provides powerful mechanisms as it can effectively map the continuous object feature and behavior parameter spaces to the corresponding effects [11] without any initial categorization of object properties as in [9], [4]. Furthermore, while bi-directional relations are not explicitly encoded in our system, we showed that our robot was able to make predictions in different directions, and made plans that involved sequence of actions on automatically selected objects[5]. ACKNOWLEDGEMENTS This research was supported by European Community’s Seventh Framework Programme FP7/2007-2013 (Specific Programme Cooperation, Theme 3, Information and Communication Technologies) under grant agreement no. 270273, Xperience. REFERENCES [1] M. Asada, K. Hosoda, Y. Kuniyoshi, H. Ishiguro, T. Inui, Y. Yoshikawa, M. Ogino, and C. Yoshida, “Cognitive developmental robotics: a survey,” IEEE Tran. Auton. Mental Dev., vol. 1, 1-9, 2009. [2] E. Ugur, E. Ozdemir, and I. Piater, “Discovery of motor primitives and learning grasp affordances,” in IEEE/RSJ International Conference on Intelligent Robots and Systems, 2012, pp. 3260–3267. [3] J. Sinapov and A. Stoytchev, “Detecting the functional similarities between tools using a hierarchical representation of outcomes,” in Proceedings of the 2nd IEEE International Conference on Development and Learning, IEEE, 2008, pp. 91–96. [4] B. Moldovan, P. Moreno, M. van Otterlo, J. Santos-Victor, and L. De Raedt, “Learning relational affordance models for robots in multi-object manipulation tasks,” in Proc. of IEEE Int. Conf. on Robotics and Automation (ICRA), 2012, pp. 4373–4378. [5] E. Ugur, E. Ozdemir, and I. Piater, “Goal formulation and planning in robot task space using learned affordances,” Robotics and Autonomous Systems, vol. 59, no. 7–8, pp. 580–595, 2011. [6] T. Kroger, “Opening the door to new sensor-based robot applications – the reflexxes motion libraries,” in Proc. of IEEE Int. Conf. on Robotics and Automation (ICRA), 2008, pp. 1–8. [7] S. Szczudak, E. Ugur, and J. Piater, “Knowledge propagation and relation learning for predicting action effects,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014. [8] S. Szczudak and J. Piater, “An active learning based sampling design for structural bootstrapping,” Unpublished, Innsbruck, Tech. Rep., 2014. [9] L. Montesano, M. Lopez, A. Barreiro-Marin, and J. Santos-Victor, “Learning object affordances: From sensory–motor maps to imitation,” IEEE Transactions on Robotics, vol. 24, no. 1, pp. 15–26, 2008. [10] E. Gyftodimos and P. A. Flach, “Hierarchical bayesian networks: A probabilistic reasoning model for structured domains,” in ICML ’02, Workshop on Learning to Represent, 2002, pp. 26–30. [11] E. Ugur, E. Ozdemir, E. Sahin, “Going beyond the perception of affordances: Learning how to actualize them through behavioral parameters,” in IEEE International Conference on Robotics and Automation (ICRA), 2011.
1. Soybeans on Vertic Argiudolls in western Uruguay 2. Deep rooting of Eucalyptus on Typic Hapludults in northern Uruguay 3. Tillage for rice on Argaiaquic Argialbolls in eastern Uruguay 4. Gullies in Typic Hapludalfs of the low uplands of the Atlantic coast in southeastern Uruguay With a lot of sediments shown at the estuary of La Plata River (just above the ‘2’ of ‘2007’). From: Ing. Agr. (MSc) Artigas Durán, Cebollati 1706/303, 11200 Montevideo, URUGUAY Phone: (598 2) 411 71 51, email@example.com and from President and members of the WASWC Council (Miodrag Zlatic, Machito Mihara, John Laflen, Jiao Juran, Samran Sombatpanit, Mohamed Sabir, Eduardo Rienzi, Ted Napier, Ian Hannam) The International Association of Economic and Social Councils and Similar Institutions (AICESIS) and the UN NGO IRENE propose the establishment of the “MDGs Awards” to pay tribute to civil society organizations and public institutions that have undertaken noteworthy initiatives for the achievement of one of the eight Millennium Development Goals (MDGs). In 2007 the MDGs Award will be granted to the organization which demonstrates the most outstanding impact on the second MDG: “Achieve universal primary education”. During the Award Ceremony in Brasilia in October 2007, the recipient organization will be awarded a MDGs Trophy and donor recognition from the private sector. Please find attached a brochure of the MDG Awards and the Form to be filled out and returned via fax, post or e-mail to firstname.lastname@example.org no later than 30 April 2007 (please note – brochure and form also available in French) --- Application for the MDGs Awards - The form for submission is available online. It can be filled in online or send by email, fax or post: - Email: email@example.com - Fax: (212) 963 8652 - Post: NGO Section/DESA -1 UN Plaza, Room DC1-1480 - New York, NY 10017 - The form and all necessary information must be submitted by April 30, 2007. Any submission received after this date will not be taken into consideration. • Submissions can be made in all 6 languages recognized by the UN (English, Spanish, French, Russian, Chinese and Arabic), but an English translation has to be accompanied in case of a non English application. • Please enclose a recommendation letter when sending your application. • Applicants are encouraged to include the following supporting materials. a) Articles appearing in newspapers, professional journals, newsletters or other publications. b) Digital standard format videos less than 10 minutes in length. c) Photographs and/or other graphic material. d) Brochures or other promotional material • All submissions received will be acknowledged and assigned a catalogue number by the UN NGO IRENE for documentation purposes. • Applicants may contact UN NGO IRENE to get assistance to fill their submission form in compliance with the criteria and the regulations of the Award. 1. Name and acronym of the organization 2. Contact Information - Permanent Headquarter Address: - Telephone: - Fax: - Name and position of the contact person: - E-mail address : - Address of the Website : 3. When your organization has been granted consultative status with ECOSOC? 4. Summarize the aims, purposes and activities of your organization: 5. What are the roots of your project? (Initiative, motivation…) 6. For which MDGs does your organization take action? 7. Which means have been carried out to reach your goals? (human, financial, technical means) 8. Please explain the projects your organization has implemented to act toward the achievement of the second MDG “Achieve universal primary education” 9. Please detail the campaigns your organization carried to support the MDG “Achieve universal primary education” 10. a) In your opinion, what are the progresses and results accomplished by your organization to “achieve universal primary education”? b) What are the progresses planned by your organization in order to reach the MDG “Achieve universal primary education” by 2015? 11. Why should the MDG Award be presented to your organization? I assert the information above is exact Date and signature Global Energy Efficiency and Renewable Energy Fund CALL FOR PROPOSALS for CLEAN ENERGY INVESTMENT FUNDS The European Commission has committed EUR 80 million to the Global Energy Efficiency and Renewable Energy Fund ("GEEREF" or "the Fund"). Triodos Investment Management in conjunction with E Co, are appointed to structure and financially close the fund and build an investment pipeline. Triodos and E CO are now looking to receive proposals from fund management companies, financial institutions, project developers or individuals that intend to develop a clean energy investment fund or intend to expand an existing fund into the clean energy sector. Developers of clean energy projects in one or more countries are also invited to propose their plans as a basis for a clean energy investment fund. Proposals are expected to meet the following indicative criteria: - Financially sustainable business plan generating a fair return for investors - Specified environmental and socio-economic impacts - Focus on small and medium sized clean energy projects (< 30MW) and companies - Requiring long term patient investment capital - Locally grounded, professional fund management team, preferable with a track record in the clean energy sector, or at least the capacity to become qualified or to liaise with other parties for that purpose - Realistic pipeline that gives evidence of the viability of the business plan. GEEREF expects to make its first commitment by July 2007. Prior to the financial closing, the Fund may provide technical and financial support to potential fund managers to develop regional funds. Please contact: Triodos Investment Management, Bas Rekvelt, firstname.lastname@example.org, www.triodos.nl or E Co (Europe), Paul van Aalst, email@example.com, www.EandCo.net US Agency for International Development (USAID) (Kenya USAID-Nairobi) Teacher Education and Professional Development Program Modification 1 http://www.grants.gov/search/search.do?mode=VIEW&oppId=11663 Grants for 5 young scientists working in Europe to attend the 5th International Congress of ESSC in Palermo, Italy, June 25-30, 2007 (www.esscpalermocongress.it) The ESSC provides 5 grants of Euro 1,000 each to 5 young researchers (less that 35 years old) that work in a European country. The grants will support their participation to the congress. To apply for a grant, it is necessary to fill in the following grant application form and send it by e-mail to the President of the Organizing Committee (firstname.lastname@example.org) no later than January 31, 2007 together with: 1. An extended abstract of the paper that the applicant wishes to present to the congress (four pages with Introduction; Material and Methods; Results; Conclusions; Keywords); 2. A short curriculum vitae of the applicant; 3. A letter of presentation from the Institution/Department of the applicant. An ad hoc ESSC Commission will evaluate the grants requests. Applicants for grants will be notified of the Commission’s decision by February 20, 2007. The Lemelson-MIT Awards for Invention and Innovation [http://web.mit.edu/invent/a-main.html](http://web.mit.edu/invent/a-main.html) The Lemelson-MIT Awards honor both established and rising inventors for their ingenuity, creativity and contribution to invention and innovation. The awards recognize the profound impact that inventors can have on economic and social well-being. The awards include the following: - The $500,000 Lemelson-MIT Prize honors outstanding mid-career inventors dedicated to improving our world through technological invention and innovation. - The $100,000 Lemelson-MIT Award for Sustainability recognizes and supports inventors working to safeguard the well-being of our communities and planet. The Lemelson-MIT Award for Sustainability has replaced the $100,000 Lemelson-MIT Lifetime Achievement Award, which was presented by the Lemelson-MIT Program from 1995-2006. The Lifetime Achievement Award recognized outstanding individuals whose pioneering spirit and inventiveness throughout their careers have improved our society and inspired others. - The $30,000 Lemelson-MIT Student Prize honors promising young inventors studying at MIT. The Lemelson-MIT Program additionally funds the $30,000 Lemelson-Rensselaer Student Prize at Rensselaer Polytechnic Institute and the $30,000 Lemelson-Illinois Student Prize at University of Illinois at Urbana-Champaign. Modeling the Lemelson-MIT Program's Student Prize competition, each prize is administered by its respective school and recognizes outstanding students who demonstrate remarkable inventiveness. - Lemelson-MIT InvenTeams are grants that support a non-competitive, team-based approach to foster inventiveness among high school students. This initiative evolved from the Program's High School Invention Apprenticeship, granted annually from 1998 through 2002. Bestowed annually, the Lemelson-MIT Awards were first presented in 1995. They are part of a comprehensive program administered by the Lemelson-MIT Program and established by The Lemelson Foundation to raise the stature of inventors and to inspire invention among young people. Find out about previous recipients at our Winners Circle. --- APPLY FOR A ROLEX AWARD Do you have a project that advances human knowledge and well-being? Rolex could help you make it happen. For 30 years, the Rolex Awards for Enterprise have provided financial support and global recognition to visionary, dedicated individuals whose ongoing working projects - expand our knowledge of the world, - improve the quality of life on the planet, - or enhance the human condition. Anyone - of any age, nationality or educational background - can apply. Winners have come from every corner of the globe and have ranged from an engineer in Senegal to a taxi driver in Paris and a palaeontologist in Canada. Please note that if you live in Asia, the Pacific or North, Central and South America, the deadline is May 31, 2007. If you live in Europe, the Middle East or Africa, the deadline is September 30, 2007. Click here for further details and/or an application form. DISCOVER THE 2004 LAUREATES read about the winners of the last series And learn about Laureates for 2006 Rolex Awards announce on October 2006 by clicking http://www.rolexawards.com/laureates/laureates-2006/index.html Google Grants Program Provides In-Kind Advertising Services to Nonprofits Deadline: Open The Google Grants program supports organizations that share the company's philosophy of community service worldwide in areas such as science and technology, education, global public health, the environment, youth advocacy, and the arts. Designed for 501(c)(3) nonprofit organizations, Google Grants is a unique in-kind advertising program that harnesses the power of Google's flagship advertising product, Google AdWords, to nonprofits seeking to inform and engage their constituents online. Google Grants has awarded AdWords advertising to hundreds of nonprofit groups whose missions range from animal welfare to literacy, from supporting homeless children to promoting HIV education. Google Grant recipients use their award of free AdWords advertising on Google.com to raise awareness and increase traffic. Each organization awarded a Google Grant receives at least three months of in-kind advertising. In the United States, applicant organizations must have current 501(c)(3) status as assigned by the Internal Revenue Service to be considered for a Google Grant. Outside the United States, Google is currently accepting applications from eligible charitable organizations based in Australia, Canada, France, Germany, India, Ireland, Italy, Japan, the Netherlands, Spain, Switzerland, and the United Kingdom. Google Grants recipients are selected every quarter. Visit the program's Web site for complete program information, application procedures, and an FAQ. RFP Link: http://fconline.foundationcenter.org/pnd/10003588/google FORESTRY FUNDING NEWS ALERT # 10: June 2006, Welcome to the June issue of our forestry funding news alert. This electronic newsletter is compiled by the FAO Collaborative Partnership on Forests (CPF) team. It is a compilation of news related to forestry funding for fund seekers worldwide. All back issues of the newsletter can be found on the CPF Sourcebook’s discussion platform at http://www.fao.org/forestry/foris/community/main/listthreads?forum=1. USAID (US Agency for International Development) Promoting Transformation: Linking Natural Resources, Economic Growth, and Governance Grant, http://www.grants.gov/search/search.do?mode=VIEW&oppId=10083 Echoing Green (a funding resource), http://www.echoinggreen.org/ Launched in 1987, Echoing Green's mission is to spark social change by identifying, investing and supporting the world's most exceptional emerging leaders and the organizations they launch. Through a two-year fellowship program, we help our network of visionaries develop new solutions to society’s most difficult problems. These social entrepreneurs and their organizations work to close deeply-rooted social, economic and political inequities to ensure equal access and to help all individuals reach his/her potential. To date, Echoing Green has invested nearly $25 million in seed and start up grants to over 400 social entrepreneurs and their innovative organizations. FELLOWSHIPS The Kinship Conservation Fellows program Is now accepting applications for the 2007 program. The program will take place from June 21- July 21, 2007 on the award-winning Western Washington University campus in Bellingham, Washington. Kinship's intensive, paid fellowship equips select career conservationists with effective tools for solving environmental problems through market-based principles. Kinship offers a unique opportunity to join a dynamic community and spend a month interacting with a highly experienced faculty while exploring social and economic methods to develop solutions to environmental challenges. The Kinship program provides leadership, planning and analytical skills that will dramatically increase practitioners' ability to achieve their conservation goals. The program provides a small stipend plus paid expenses for selected Fellows, and is open to professionals with at least five years of experience in a conservation field. To learn more, or to apply, please visit our Web site at www.kinshipfellows.org or contact us at 400 Skokie Boulevard, Suite 300, Northbrook, IL 60062. The deadline for applications is January 10, 2007 and applicants will be notified of acceptance by February 28, 2007. National Endowment for the Humanities Grants for Collaborative Research in Eastern Europe and Eurasia Deadline: February 15, 2007 American Councils for International Education: ACTR/ACCELS (http://www.americancouncils.org/) and the National Council for Eurasian and East European Research (http://www.nceeer.org/) are accepting applications for the 2007-08 National Endowment for the Humanities Collaborative Humanities Fellowship. Full and partial fellowships are available through American Councils and NCEER with NEH grant support. Fellowships provide up to $40,000 each for four to nine months of research in Eastern Europe and Eurasia. Proposals must include plans to work with at least one collaborator in the field. Applications with a strong regional focus and the potential to broaden and strengthen international academic linkages beyond the traditional centers such as Moscow, St. Petersburg, Warsaw, and Prague are especially encouraged. This merit-based competition is open to all U.S. post-doctoral scholars in the humanities and most social sciences, including such disciplines as anthropology, modern and classical languages, history, linguistics, literature, jurisprudence, philosophy, political science, archaeology, comparative religion, sociology, and ethics. U.S. citizenship, permanent resident status, or residency in the U.S. for three years before date of application is required. For more information and an application form, visit the American Councils Web site. RFP Link: http://fconline.fdncenter.org/pnd/10005989/americancouncils Young soil scientists wanted ICSU is celebrating its 75 years and organizes a conference: “Global Scientific Challenges: Perspectives from Young Scientists”, that will be held in April 2007 in Germany. This conference will bring together approximately two hundred young scientists from across the world, and from different disciplines, to discuss and debate some of the key challenges for science in the 21st century. Participants should ideally be under 35 years old (with an absolute age limit of 40 years). Participants should have a PhD or MSc or have completed at least 2 years of post-graduate research studies. For more information click here. Please contact the Secretary General, Stephen Nortcliff, for further information. email@example.com COURSES April 2–20, 2007 Participatory Mapping and PGIS for Community-Level Spatial Planning & Management Short Course (3 weeks) ITC, Enschede, The Netherlands Course Organisers: Dr. Michael K. McCall; Drs. Jeroen Verplanke Brief Description: Participatory spatial planning & management (PSP) in land & land resources (both rural and urban) has a dynamic of empowerment and good governance imperatives. - issues in geo-information for participatory spatial planning & management. - concepts of local & indigenous spatial knowledge. - analysis of PSP & community-based management, including problem & conflict analysis, stakeholder interests, good governance, PGIS information management. - tools & techniques in local-level spatial information acquisition, analysis & application. Following tools & techniques are used in this course: - participatory spatial data acquisition, participatory mapping, PRA, PGIS. • GPS, mobile GIS (PDA such as iPaq), ArcPad, field data recording and data preparation. • visualization of spatial relationships, interactive mapping. • participatory spatial planning & management processes, and policy instruments. • Applications of participatory mapping & PGIS to e.g. conflict mapping, customary entitlements to land, NRM; community environmental mapping. Pre-requisites: Basic GIS skills (ArcView–ArcGIS) are expected, and useful to have some experience in local-level community surveys. Costs of the Course NB. This course unfortunately does not have its own Fellowships. Participants need to find their own funds. - Tuition fees for the course: €1000.00 - ITC will provide accommodation; including expected living cost this is estimated to a minimum of: € 750.00 - Additional costs regarding, insurances, registration fees, residence permit and visa are estimated to: € 250.00 Further Information: The Schedule of the Course and further information are available from Jeroen Verplanke firstname.lastname@example.org, or Mike McCall email@example.com Please check ITC website [http://www.itc.nl/education](http://www.itc.nl/education) for additional information about ITC and Enschede; or contact firstname.lastname@example.org To register directly, download the registration form: [http://www.itc.nl/education/_data/ITCAppFormSC.pdf](http://www.itc.nl/education/_data/ITCAppFormSC.pdf). You are kindly requested to then submit this form by e-mail to: email@example.com with cc. to firstname.lastname@example.org For further registration instructions check ITC website: [http://www.itc.nl/education/register/](http://www.itc.nl/education/register/) In order to process your registration in time we prefer you to submit before March 1, 2007. April 2-24, 2007 Regional training course on Low Cost Soil and Water Conservation Techniques and Watershed Management Activities Rationale: Mountain watersheds serve many ecological and hydrological functions vital to existence. They provide critical habitat for many plant and animal species, transport paths for sediment, nutrients, minerals, and water to human communities for drinking, cleaning, recreation, hydroelectric power, and industry. Experience in the Himalayan region demonstrates how the application of simple, low cost soil and water conservation measures and techniques can address the problem of biophysical vulnerability in the resource constrained environment of mountain watersheds and arrest land degradation. Building the capacity of field-level technicians in participatory implementation of low cost, simple soil and water conservation techniques and watershed management activities has thus become essential in better managing soil and water resources in mountain watersheds. This training package is based on experiences from this mountain region and elsewhere and includes tried and tested measures and techniques specifically attuned to middle mountain conditions. Course Objectives The training course aims to enhance the knowledge and skills of participants on low cost soil and water conservation techniques and watershed management activities. Specifically, the course aims to familiarise the participants with different low cost soil and water conservation techniques and watershed management activities, both theory and practice. In addition, the course aims to enable the participants to identify conditions of land degradation and appropriate soil and water conservation techniques and watershed management activities applicable to the participants’ contexts; and to analyse and design conservation measures within their work or area contexts. By the end of the course, participants will be able to: · understand different low cost soil and water conservation techniques and options and their importance · carry out basic land degradation assessment and identify, design, and implement appropriate soil and water conservation measures to address them · identify ways to improve the participants’ current soil conservation practices Content Areas - Introduction to Integrated Watershed Management and Low Cost Soil Conservation - Low Cost Soil Conservation Techniques - Watershed Management Activities - Strategies to Promote and Up-scale Low Cost Soil Conservation Techniques Training Methodologies The participatory training approach will be applied throughout the course. Classroom lectures will be complemented by field and practical exercises, a field visit and demonstration, and action planning based on the participant’s areas or field of work. Who should attend the training Field-level technicians and extension workers working with the government, or with international and local non-government organisations (INGOs and NGOs), in the areas of natural resource management, especially soil conservation and watershed management Course Pre-requisites - Participants must preferably come from a technical background (Agriculture, Forestry, Civil Engineering, others) related to soil conservation and watershed management (Minimum education attained: two years of College or an Intermediate or Certificate level course or its equivalent) - Can understand and speak English, which is the medium of instruction of the course - Minimum of 2-3 years working experience preferred. Experience in conducting engineering surveys is an advantage. Women participants are highly encouraged. To be more effective, we target to have 15-20 participants only. Please apply as soon as possible! Where: Training will be conducted in Godavari, Kathmandu, Nepal; When: 2-24 April 2007 Cost The training fee is US$1,800 per person. This cost includes tuition fee, food, lodging (shared double occupancy), and course materials. Single room accommodation may be provided for an additional fee of US$10 per day. The cost excludes travel costs from place of origin, DSA (daily subsistence allowance), laundry and other personal expenses (we suggest a minimum pocket allowance of $20 per day for incidental and personal expenses). Applicants are encouraged to secure financial support from their organisations or from donor agencies. Trainers’ Institute Profile Experienced professionals from ICIMOD and partners engaged in soil and water conservation, and watershed and disaster management Training is coordinated by the Natural Resource Management Programme of ICIMOD (www.icimod.org), which is an international, independent mountain learning and knowledge centre committed to improving the sustainable livelihoods of mountain peoples in the extended Himalayan region. Remarks: This training course involves intensive field and practical exercises and participants may have to devote time apart from the mentioned sessions For more information and to request for an application form please contact: Keshar Man Sthapit – Training Coordinator Please email: email@example.com. Deadline for application: 1 March 2007 Application Form for ICIMOD Training Courses Training title/date: ____________________________________________________________ Contact Details Mr/Ms/Dr First Name: _________________________Last Name: _______________________ Gender: M / F Job Title: _________________________________________________________ Project/Office: __________________________________________________________________ Organization: __________________________________________________________________ Tel no: ______________ Fax no: ______________ Email: _____________________________ Mailing Address: __________________________________________________________________ Nominating Official: _________________________Email: _____________________________ Funding source: __________________________________________________________________ Kindly answer the following questions in not more than 5 sentences. Thank you. 1. What are your current work activities and responsibilities? 2. What are your expectations from the training course? 3. How is the training course relevant to your work? 4. How will you apply the results after you return to work? 5. Educational Background (Degree/Institution/Year): 6. Other related trainings you have participated in: 7. How did you hear about the training course? Please return the completed form to: firstname.lastname@example.org June 22-July 14, 2007 Beahrs Environmental Leadership Program, University of California, Berkeley 2007 Summer Certificate Course: Sustainable Environmental Management We are seeking applicants for the 2007 summer certificate course in Sustainable Environmental Management at the University of California, Berkeley. The 2007 summer course brochure and application in MS Word and as a pdf may be downloaded from the web site Welcome Page, see Quick Links: http://nature.berkeley.edu/BeahrsELP. Applications must be received no later than February 1, 2007. The Beahrs ELP (ELP) links the complex local and global environmental challenges of practitioners and decision-makers around the world with state-of-the-art knowledge and interdisciplinary perspectives in environmental and natural resource science, policy and leadership at U.C. Berkeley. The core component of the ELP is an annual three-week summer course in Sustainable Environmental Management held on the Berkeley campus. Course graduates become members of the Berkeley ELP Alumni Network that supports on-going learning, peer exchange and collaborative research and policy projects through the Small Grants Initiative. Each summer 35-40 mid-career environment and sustainable development professionals participate from academia, government, NGOs and the private sector. A limited number of partial scholarships are available. There are no full scholarships. Application review begins January 1 and applications must be received no later than February 1, 2007. Please contact Program Representative, Leslie Correll, or Program Co-directors, Robin Marsh and David Zilberman, for more information: email@example.com. Please forward this email to share this information with others who may be interested. Robin Marsh & David Zilberman Co-Directors, Beahrs ELP Leslie Correll Program Representative, Beahrs ELP Beahrs Environmental Leadership Program http://cnr.berkeley.edu/beahrselp CENTER FOR SUSTAINABLE RESOURCE DEVELOPMENT College of Natural Resources, 4 Giannini Hall #3100 University of California Berkeley, CA USA 94720-3100 Tel. 510-643-4200 Fax 510-643-4483 25-30 June 2007 GDF - SSWG Short Course on Ethnoecology & Community-based Conservation Course in Port Elizabeth, South Africa (partial scholarships available), In the week leading up to the Society for Conservation Biology (SCB) 2007 annual meeting, the Global Diversity Foundation (GDF) and the SCB's Social Science Working Group (SSWG) will co-sponsor a course on 'Ethnoecology and Community-based Conservation' from 25 - 30 June 2007 in Port Elizabeth, South Africa. With financial support from the Christensen Fund (TCF), we will award a limited number of bursaries to cover course fees, bibliography, software, accommodation and meals during the course. Please note the following requirements before applying for a bursary: (1) a university degree in social or natural sciences; (2) fluency in English Preference will be given to candidates who are engaged in teaching and research on biocultural diversity. Members of indigenous groups are especially encouraged to apply. Candidates from TCF and GDF focal regions (outlined below) will be given priority. The deadline for application is 15 March 2007. An online application will be available in mid-January. In the meantime, enquiries may be directed to Gary J. Martin (firstname.lastname@example.org) or Rick Stepp (email@example.com). For a description of the activities of SCB's Social Science Working Group see: www.combio.org/WorkingGroups/SSWG/. More information on the Global Diversity Foundation and its focal regions (Mesoamerica, North Africa, Southeast Asia and Southern Africa) is available on www.globaldiversity.org.uk. Consult www.christensenfund.org for an overview of the Christensen Fund, including its priority areas: Turkey, Iran and Central Asia; Northern Australia and Melanesia; Greater American Southwest and the African Rift Valley. The course will precede the 21st SCB annual meeting, One World, One Conservation, One Partnership, held from 1 - 5 July 2007 at the Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. The Local Organising Committee is now accepting abstracts for oral and poster presentations until 8 January 2007. More information is available at: www.nmmu.ac.za/scb/callpapo.htm. We encourage potential applicants to submit an abstract for the SCB meeting, and to arrange independently funds for their travel to Port Elizabeth, conference registration and accommodation and meals during the meetings. We do not have funds for these expenses at this time. July 30-August 3, 2007 Short Course on Research Methods in Ethnoecology The Global Diversity Foundation (GDF), with funding from the Christensen Fund (TCF), will sponsor the participation of 3 developing country colleagues in the National Science Foundation supported Short Course on Research Methods in Ethnoecology from 30 July-3 August 2007. Information on the course is available on www.qualquant.net/training/scrm.htm#ethno. The travel grants will cover roundtrip travel to Beaufort, North Carolina, accommodation, meals, local transportation, course fees and a book and software allowance. Colleagues who hold a Ph.D. in social or natural sciences and can demonstrate fluency in English may apply. Preference will be given to faculty members at colleges and universities who are engaged in teaching and research on biocultural diversity. Members of indigenous groups are especially encouraged to apply. Candidates from TCF and GDF focal regions (see below) will be given priority. Interested candidates are requested to follow the instructions on www.qualquant.net/training/scrm.htm#app. The deadline for application is 15 February 2007. More information on the Global Diversity Foundation and its focal regions (Mesoamerica, North Africa, Southeast Asia and Southern Africa) is available on www.globaldiversity.org.uk. Consult www.christensenfund.org for an overview of the Christensen Fund, including its priority areas: Turkey, Iran and Central Asia; Northern Australia and Melanesia; Greater American Southwest and the African Rift Valley. Enquiries may be directed to Gary J. Martin (firstname.lastname@example.org) or Rick Stepp (email@example.com). JOBS ECOAGRICULTURE PARTNERS: Director, Program on Markets for Ecoagriculture Products and Services Ecoagriculture Partners (www.ecoagriculturepartners.org) is a small, rapidly-growing international non-profit organization working to facilitate landscape management approaches that simultaneously enhance agricultural production, support rural livelihoods and conserve ecosystem services. EP seeks a ¾ to full-time Director for our new program on Markets for Ecoagriculture Products and Services, to join our committed team by July 1, 2007 or earliest available. The Director of the Markets Program will: - Build and lead EP’s program on Markets (both products and ecosystem services), - Support market analysis and development in selected ecoagriculture landscapes, initially in Mesoamerica and East Africa; - Promote exchange of market models and innovations among ecoagriculture innovators; - Engage in international dialogues on ecoagriculture-related market issues; - Support adaptation of product and ecosystem service market assessment methods to achieve joint income and conservation goals in ecoagriculture landscapes; - Develop and maintain excellent collaborative relationships with key partners in private, public and community sectors. Required: - At least five years of relevant experience supporting market development and analysis, particularly in developing countries; - Experience working with the private sector and with farmer or community conservation organizations; - Demonstrated commitment to, and achievements in, promoting market initiatives that achieve both conservation and agricultural livelihood benefits; - Masters’ degree in relevant field; - Strong skills in team-building and participatory processes; - Excellent English writing and communication skills. Desirable: - Market-related experience in Latin America or Africa; - Fluency in reading and speaking Spanish; - Experience and competence in working in diverse, cross-cultural teams; - Expertise in analysis of markets. Salary and benefits: - Competitive salary, depending upon experience; - Health and retirement benefits, life and disability insurance, 20 days vacation, holidays, relocation support - Initial 3-year position, renewable - Location of posting negotiable. Please send your C.V., a cover letter indicating your background and experience, and names and contact information for three references in an email with subject line ‘Director, Markets Program’ by January 31, 2007 to Seth Shames, 1050 Potomac Street, NW, Washington, DC 20007, USA. Tel: 202-298-3004; Fax: 202-298-3014. firstname.lastname@example.org, www.ecoagriculturepartners.org Vacancies: ITC in the Netherlands currently has 2 Professor positions available in the Department of Urban and Regional Planning and Geoinformation Management. Details are attached for your information and circulation to potentially qualified and interested candidates. [http://www.itc.nl/about_itc/organisation/science_departments/pgm.asp](http://www.itc.nl/about_itc/organisation/science_departments/pgm.asp) Contact: Dr. Richard Sliuzas ([email@example.com](mailto:firstname.lastname@example.org)) Urban and Regional Planning and Geo-Information Management International Institute for Geo-Information Science and Earth Observation (ITC) Hengelosestraat 99, PO Box 6, 7500 AA Enschede, The Netherlands Tel: +31-53-4874236/ 4874532; Fax: +31-53-4874575 --- EXPOSITIONS AGRAnews Middle East Industry News for the Region’s Largest Agriculture Exhibition Contact: [email@example.com](mailto:firstname.lastname@example.org) --- United Nations Development Programme WELCOME to the Virtual Knowledge Fair on Water The Fair is now open! Please proceed to [www.waterfair.org](http://www.waterfair.org), where you will find access to the various exhibits (videos, experience write-ups, pictures and materials), background information on countries and water-related topics, a Who’s-Who about water-practitioners in the regions, photo gallery, a quiz and much more! --- ROYAL FLORA RATCHAPHRUEK 2006 Chiang Mai, Thailand International Horticultural Exposition for His Majesty the King Royal Flora Ratchaphruek 2006 International Horticultural Exposition for His Majesty the King of Thailand on a total area of 75 ha At the Royal Agricultural Research Center, Chiang Mai, Thailand 1 November 2006 – 31 January 2007 Information in [http://www.royalfloraexpo.com/](http://www.royalfloraexpo.com/) (More than 3 million people have visited) Hello friends and colleagues in WASWC, I am writing to you today regarding nature and the environment. Ever since I was a child, I have been very interested in wildlife, birds, nature, and the environment. I have a B.S. degree in wildlife biology, and have worked as a zookeeper, wildlife biologist, and ecologist. I am conducting a brief survey of world leaders, government officials, religious leaders, corporate CEOs, environmental groups, wildlife experts, and others regarding nature and the environment. I am also very interested in religious views, customs, and beliefs from around the world, and the interaction between religion and the environment. I would greatly appreciate it if you could answer the survey and return it to me, or have someone who you feel would be most appropriate answer it and return it to me. The results will be posted online at http://earthsurvey.blogspot.com. I apologize if you have already seen this message, but I have sent it out to multiple contact lists. I have tried to avoid duplicate mailings as much as possible. Contact information will not be included- only your name and title/organization will be posted. I am more interested in personal views, not necessarily official statements, and there is a disclaimer on the site that states that the views expressed are personal and not to be seen as official views. Completed surveys can be sent to me via email at email@example.com or by postal mail to: Earth Survey, c/o Jeremy Taylor, 25E Oakbrook Manor, Ravena, NY 12143, USA Thank you very much for your time, and I hope that you will consider filling out the survey and returning it to me by any means that you prefer. Jeremy Taylor (firstname.lastname@example.org) ----------------START OF SURVEY----------------- Name: Title/Organization: Address: Email address: Today's Date: 1. What interaction with an animal and/or nature in your life has had the biggest impact on you? 2. Did you have a favorite place in the great outdoors during your childhood? Now? 3. As a former zookeeper, I would love to know what your favorite animal is, and why? 4. What do you think is the greatest environmental challenge facing us now, and what do you think will be the greatest challenge in the future? 5. If you could give everyone one piece of advice regarding the environment and our natural resources, what would it be? Continuous Flow Monitoring Adapting high-resolution current profiling into a practical flow monitoring device - Irrigation - Pipes and Culverts - Natural Streams - Shallow Water Conditions - Water Level and Velocity - Depths from 0.3 to 5.0 meters - Automatic Level Compensation - Multiple Output Formats Argonaut-SW 25 cm (9.7") long Acoustic Doppler Instruments for Water Velocity and Level Measurement SonTek and Argonaut are registered trademarks of SonTek/YSI Inc., San Diego, CA USA [858.546.8327] www.sontek.com Note: New website = www.semeato.com.br Conference organizers may ask for a ‘Conference Package’ from WASWC (http://waswc.ait.ac.th/package.html), where among several benefits all the participants may obtain free Guest membership in WASWC for one year. This is our service to society. MEETINGS 2007 January 7-13, 2007 African Soil Science Society’s 4th International Conference (The contribution of Africa’s Soil Science to NEPAD) Impacts of climate change, global trade, urbanization and biotechnology on land use in Africa GIMPA, Accra, Ghana, Hosted by the Soil Science Society of Ghana in collaboration with IUSS, FAO, AfNet/CIAT, ICRISAT, IWMI, IFDC-Africa, UNU-INRA, organized by the African Soil Science Society (ASSS) Background: The African Soil Science Society (ASSS) is a non-profit organization founded in 1986. ASSS has organized successfully three International Conferences respectively in Kampala (Uganda, 1988), Cairo (Egypt, 1990) and Ibadan (Nigeria, 1994). These three Conferences were focused on land management in irrigated and upland zones with the participation of soil scientists from all over the world working on African lands. This 4th International Conference aims to gather specialists interested in land use and sustainable land management in Africa, in the context of serious emerging issues such as climate change, global trade and biotechnologies including GMO uses. With a potential of 2,844 million ha of lands, Africa is characterized by a high variability of soils, climate, vegetation, fauna and land uses. It is of great importance that scientists, planners, economists, policy makers and land users are aware of the status and trend of land use in the context of high demographic pressure, severe degradation of natural resources and high competitiveness of agricultural goods at international level. Furthermore, land users and the soil science community should bear in mind that: 1. The achievement of food security and Millennium Development Goals (MDG), particularly Goal 1 (poverty/hunger alleviation) and Goal 7 (environment protection), which remains a major challenge for Africa; 2. The requirements of the international trade with its binding conditions, the impact of climate change with potential increase of natural disasters (drought and floods), the prospect in the increasing use of biotechnologies (conventionally improved varieties, genetically modified organisms – GMOs) to face new constraints have undoubtedly consequences on land use systems and land management technologies and their changes; 3. At regional, sub-regional and national levels, there is lack of stocktaking and overview of the status and trend of land use changes, due to the impact of these great challenges; 4. The Soil Science community in Africa needs to be more vocal and participate actively in national, regional and international fora (e.g. UNCCD, CBD, UNFCCC, AU, NEPAD/CAADP, TerrAfrica), policy formulation, norms and standard setting, and good agricultural practices with respect to Soil Science and related issues. 5. There is also a need for the African Soil Science Society to position itself as a regional broker for land issues within the Civil Society in Africa and to develop: (i) active collaboration, particularly with international institutions and thematic networks dealing with soil and land issues; and (ii) strong lobbying for public and donors’ support for more investment in sustainable land management. It is with this background that the African Soil Science Society in collaboration with various partners is organizing this Conference. Themes: - Assessing the dynamics of land use in Africa - Impact of regional and global trade on land use changes - Impact of land use evolution on climate change - Impact of biotechnologies on farm biodiversity and land use - Impact of urbanization on land use in the urban-rural continuum Deadline for extended abstract submission 31\textsuperscript{th} July 2006 Extended abstract acceptance notification 30\textsuperscript{th} June 2006 2\textsuperscript{nd} Announcement with tentative program 30\textsuperscript{th} July 2006 Submission of full papers 30\textsuperscript{th}–August 2006 Conference fees: Conference fee per participant: US$250 and US$100 for students (evidence should be provided). After 30th September 2006 the fees will be respectively US$300 and US$150. Registration fees will cover conference documents, conference dinner, lunch and snacks, mid-conference tour, transport to and from conference site to hotels. Mid-conference tours: Participants will also have chance to participate in a mid-conference tour which will take them to the following places in the Central Region of Ghana: i. Okyereko Irrigation scheme, ii. Elmina Castle, which was the slave trading center during the slave trade era, and iii. Kakum Forest Reserve. Contact: Dr R.D. ASIAMAH, Soil Research Institute (SRI), Kwadaso-Kumasi, Ghana. Phone: +233-51-50353/4; Fax: 233-51-50308, soils_intcon@ yahoo.co.uk and email@example.com January 17-18, 2007 Workshop and Conference with Point Carbon: "North America and The Carbon Markets", Washington, D.C., USA This unique event is designed to help facilitate learning about the developing international and domestic carbon market. See [http://ealert.pewclimate.org/ctt.asp?u=4315018&l=134031](http://ealert.pewclimate.org/ctt.asp?u=4315018&l=134031) January 19-23, 2007 15\textsuperscript{th} Annual National No-Tillage Conference, Des Moines Marriot Des Moines, Iowa, Contact: The National No-Tillage Conference, P.O. Box 624, Brookfield, WI 53008-0624, USA. Phone: +1-866-839-8455 or +1-262-432-0388; Fax: +1-262-786-5564, firstname.lastname@example.org, www.no-tillfarmer.com. January 20-25, 2007 World Social Forum 2007, Nairobi, Kenya, The 7th World Social Forum - to be held from January 20 to 25, in Nairobi (Kenya) – faces now the challenge of being even more linked to concrete actions to build “another possible world”. This concern has been constantly expressed on various meetings of the WSF International Council and evaluations made about WSF process. What we have been looking for is to accomplish one of WSF aims, stated on its Charter of Principles: The World Social Forum is an open meeting place for reflective thinking, democratic debate of ideas, formulation of proposals, free exchange of experiences and interlinking for effective action. This is the why this preparatory consultation is being launched. It gives continuity to the methodological options, which started to be adopted in 2004, during WSF 2005 preparation, in order to improve process and event “architecture”. On that Forum, it resulted on the definition of 11 thematic terrains around which 2,000 activities were carried out– all of them were self-organized, corresponding to another advance compared to the events organized before. But the 2004 consultation had a thematic character. In this one, we want to move a step forward. Now what we request to you is to indicate the actions, campaigns and struggles in which your organization/network/entity is engaged. The proposal is to organize the 7th Forum around those actions, campaigns and struggles, aggregating, on the different Forum spaces, the activities linked to them, as well as the activities that have no direct link with them – to be indicated during the moment of activities registration – but related to the same by their themes. This will make more visible the concrete actions born on the Forum process! or those who gave sequence or more density to others that had been developed before or outside WSF, as well as its effective political results. This will also stimulate the interconnection among organizations in which we are engaged on the struggles for constructing a fair, egalitarian, democratic, participative and sustainable society. This proposal – and the execution of this preparatory consultation – was approved by the WSF International Council, based on a proposal made by its Methodology Commission, in its meeting held in March 2006, in Nairobi. We tried to simplify the questions on the consultation form at maximum, in order to make easier to answer to it. As soon as you do it, better conditions we will have to prepare the WSF 2007 adequately. The deadline to send your answers is July 30, 2006. In order to fill in the consultation questionnaire, please access the website: [http://consultation.wsf2007.org/](http://consultation.wsf2007.org/). Organizations who have registered on the Polycentric WSF 2006 website ([www.wsf2006.org](http://www.wsf2006.org)) are not required to fill in all the forms again. They can access the site [http://consultation.wsf2007.org/](http://consultation.wsf2007.org/) using the same username and password from the polycentric 2006 site above mentioned and then, click on the top menu “Consultation on actions, campaigns and struggles” > “Add new”. Up to now, the site is only in English. But soon it will be available also in French and Castellano/Spanish. For further information on the event in Nairobi, please visit the site: [www.socialforum.or.ke](http://www.socialforum.or.ke). WSF 2007 Organizing Committee executive office – Kenya. Physical Contacts: Masandukuni Lane, Off Gitanga Road Postal Address: P.O. Box 63125, 00619 Nairobi Kenya Telephone: 254-020-3860745 ou 254-020-3860746 January 22-26, 2007 U.S. North American Carbon Program (NACP), Investigators Meeting & Joint Canada-Mexico-USA Carbon Program Planning Meeting, Cheyenne Mountain Conference Center, Colorado Springs, CO, USA (First Announcement) U.S. North American Carbon Program (NACP) Investigators Meeting Monday, January 22 through Wednesday, January 24 U.S. North American Carbon Program participants will meet to report research results to date, plan and coordinate work to occur during the next year, and plan future activities. Early opportunities for synthesis will also be discussed. Each NACP project and activity is encouraged to provide a poster and/or a presentation at the meeting. Any principal investigator who cannot attend should designate a co-investigator to represent their project. Joint Canada-Mexico-USA Carbon Program Planning Meeting Thursday, January 25 and Friday, January 26 The United States is pleased to host a joint meeting of Canadian, Mexican, and U.S. investigators involved in research on carbon cycling within North America and adjacent ocean basins and coastal areas. With the goal of developing a scientific framework for a joint program to provide better estimates of the North American carbon budget and a better understanding of the processes and mechanisms involved, meeting participants will identify activities that are well suited to collaboration and topics for synthesis of results developed by each country's program. The meeting agenda will include posters and presentations on current projects; however, the focus will be on discussions to shape a Joint Canada-Mexico-USA Carbon Program and understand its relationship to other regional carbon studies. Mexican, Canadian, and U.S. scientists involved in carbon cycle research relevant to North America are welcome and encouraged to attend both meetings. Additional information will follow in a second announcement and be posted at: http://www.nacarbon.org/ U.S. North American Carbon Program Office (contact address) email@example.com January 31-February 3, 2007 INTERNATIONAL MEETING OF FIRE EFFECTS ON SOIL PROPERTIES Mediterranean Environmental research Group, Universitat de Barcelona, Background and invitation The aims of the meeting are to explore issues concerning effects of fire on soil properties. It is well known that fire and temperature produce changes on soil physical, chemical, organic and biological characteristics. The intensity and severity of forests fires has increased in the last years, and this is a widespread worldwide phenomenon. It is important to know the effect of high intensity fires and prescribed fire as a management tool on soils. It is hoped that the exchange of information will not only contribute to a better understanding of the effects of fires on soils but also will suggest solutions for soil amendment and management in terms of soil quality and forest regeneration. The purpose of this meeting is to bring together scientists who study the fire effects on soil properties and soil recovery after fires. New research methodologies, topics and conclusions about fire impacts on soils are very welcome. The meeting will include an excursion to a study site where prescribed fire is used as a management tool (Les Gavarres Mountains, Girona). Topics - Fire effects on hydrology and soil physical properties - Fire effects on organic matter content, soil chemical properties and soil recovery after fires - New methodologies to study fire effects on soil - Fire intensity and fire severity measurements Contact: Prof. Xavier Úbeda., University of Barcelona, firstname.lastname@example.org, www.ub.edu/gram February 7-10, 2007 3rd International Groundwater Conference (IGC-2007) on Water, Environment and Agriculture: Present Problems and Future Challenges, Tamil Nadu Agricultural University, Coimbatore, India Contact: K.Palanisami at email@example.com, firstname.lastname@example.org, email@example.com. Deadline for abstract submission: July 31, 2006, but you can try, and there won’t be any problem if you want only to attend. February 12-16, 2007 EC07 IECA’s Environmental Connection Conference Event The conference and expo will be at the Reno-Sparks Convention Center. Our host hotel is the Atlantis Hotel. Show dates will be Tuesday 2/14, 5:00-7:30 PM, Wednesday, 2/14 and Thursday, 2/15. Most conference activities will take place in the Reno-Sparks Convention Center: 4590 S. Virginia St. Reno, NV 89504, USA 775-827-7620 IECA Host Hotel: Atlantis Casino Resort 3800 S. Virginia St. Reno, NV 89502, USA IECA Room Rate: $112.00 Preliminary Schedule: Feb. 12, Monday – Full day training course, 8:30-5:00 Feb. 13, Tuesday – Full day training course, 8:30-5:00, Expo open, 5:00-7:30 Feb. 14, Wednesday – General Session, Technical Sessions, Expo Open Feb. 15, Thursday - Technical Sessions, Expo Open For additional information, contact: Kate Nowak, Director of Events and Sales Phone: 970-879-3010. ext. 15 Fax: 970-879-8563 Email: firstname.lastname@example.org, www.ieca.org February 20-22, 2007 Alberta Soil Science Workshop, Calgary, Alberta, Canada http://www.soilsworkshop.ab.ca/ Saskatchewan Soils & Crops Workshop, Mar 1-2, Saskatoon http://www.extension.usask.ca/ExtensionDivision/noncredit/AFH/index.htm Upcoming events page http://www.agric.gov.ab.ca/app55/events?navstyle=2 February 22-23, 2007 International Conference on the Modeling of Urban Water Systems The annual International Conference on the Modeling of Urban Water Systems will be held February 22-23 in Toronto, Canada. The conference addresses all aspects of urban water systems, including planning, design, construction, operation, monitoring, models, safety, and security; how to manage urban water systems including water supply; drainage, pollution control, domestic, municipal, and industrial waters and stormwater from surface water catchments or ground water; and real-time control along with education, outreach, and practical applications. http://www.computationallyhydraulics.com/Training-Conferences/conferencetoronto.html March 2-3, 2007 FINANCING OF FOREST CONSERVATION: Payments for Environmental Services in the Tropics, International Society of Tropical Foresters, Yale School of Forestry & Environmental Studies What mechanisms are currently available for forest conservation in the tropics? Conservation has always been approached by a number of sectors including biology, political science, economics, and social science. But these perspectives and their associated agencies have historically worked in separate spheres, with little collaboration. Today it is essential that a holistic approach be taken and that the different fields are incorporated together into conservation schemes, and economists comprise an important part of this holistic view. It is imperative to include economics in conservation planning in order to demonstrate the potential benefits of sustainable forest management. Payment for environmental services is one of the tools environmental economists use to support forest conservation and has great potential for the conservation of tropical forests. This term encompasses a wide scope of activities including carbon credits, bio-prospecting, watershed protection, soil conservation, tax credits and ecotourism. The Yale Chapter of the International Society for Tropical Foresters expect that this conference will serve as a forum to discuss the various types of payment for environmental services, the issues associated with each type, and the successes and failures to date. The discussion will be driven by questions such as: Can payment for environmental services mechanisms lead to conservation? Can such mechanisms be as economically viable as other uses? How can these methods be better integrated into conservation and management plans? What are the potential negative consequences from the standpoints of conservation, local livelihoods, and economic optimization? How can active trading markets for ecosystem services be developed? Are payments for avoided deforestation and reduced carbon emissions feasible? Will tax credit programs be viable in the long term? We encourage abstracts based on primary research, or personal or institutional experience. Selected participants will present full papers at the conference. Abstracts should be a maximum of 500 words, and all correspondence will be addressed to the principal author. Please send abstracts by December 15, 2006, electronically or by mail: Yale ISTF Conference c/o Tropical Resource Institute Yale School of Forestry and Environmental Studies 210 Prospect Street, New Haven, CT 06511 March 3-4, 2007 2007 Schuylkill Watershed Congress The 2007 Schuylkill Watershed Congress will be held on March 3-4, 2007 at the Montgomery County Community College West Campus in Pottstown, PA. The Schuylkill Watershed Congress features a diverse program covering a broad range of watershed topics. The Congress is organized for citizens seeking new information on watershed protection. [http://www.schuylkillriver.org/Detail.aspx?id=730](http://www.schuylkillriver.org/Detail.aspx?id=730) March 15, 2007 PA Green Building Conference The Green Building Alliance is sponsoring the Green$ense 2007: Smart Way to Build Profits conference on March 15, 2007 in Pittsburgh. Keynote speaker Susan Eastridge, developer for the Pittsburgh Cultural Trust’s riverfront project, will describe how the biggest downtown housing initiative in city history and its commitment to green building is a smart way to build profits. Breakout sessions will be geared to those new to green building and seasoned professionals. Afternoon tours of Pittsburgh-area green buildings will show green development in action. [http://www.gbapgh.org/Green$ense2007.asp](http://www.gbapgh.org/Green$ense2007.asp) March 19-21, 2007 First International Soil Moisture Sensing Technology Conference; Current and future research directions in soil moisture sensing, Waikiki, Honolulu, Hawaii, USA. This conference will focus on current and future research directions in soil moisture in-situ sensing technology with special emphasis on sensor reliability and measurement scale dependency. Details in [http://www.ctahr.hawaii.edu/faresa/Conference/](http://www.ctahr.hawaii.edu/faresa/Conference/). Abstracts deadline: January 15, 2007 NEW DEADLINE: February 18, 2007 Participants can come either early or stay late to enjoy the Paradise in MARCH with a cost of room as little as $69. Space is limited; book your rooms now. We would appreciate if you could spread the word around about this conference. Also please feel free to contact any of the organizing committee members if you have any question or suggestion or you want to help. Ali Fares, Associate Professor of Hydrology Natural Resources & Environmental Management Dept. University of Hawai’i - Manoa Ph: 808 956-6361(d) -7530(Sec) Fax:-6539, email@example.com, WWW.CTAHR.Hawaii.Edu/Faresa/ "A Nation that destroys its soil, destroys itself", President F.D. Roosevelt (1933-1945) March 21-23, 2007 Humic Science & Technology Conference X, Northeastern University, Boston, MA, USA, March 21-23, 2007. Contact Geoffrey Davies and Elham Ghabbour, [www.hagroup.neu.edu](http://www.hagroup.neu.edu). April 5-6, 2007 PA Stormwater Conference The Chesapeake Bay Foundation (CBF) is planning a 2007 Stormwater Conference on April 5-6 at the Penn Stater, State College, PA. The conference, rescheduled from February, will build upon DEP’s Stormwater BMP Manual Trainings being held throughout Pennsylvania by providing attendees with tools and strategies for implementing the manual’s progressive stormwater management concepts into site design and land use planning. [http://www.cbf.org/Confluence2007](http://www.cbf.org/Confluence2007) April 15–20, 2007 EGU Vienna 2007, European Geosciences Union’s General Assembly 2007, Vienna, ▲ Session on, SOIL EROSION ON AGRICULTURAL LAND, [http://meetings.copernicus.org/egu2007/](http://meetings.copernicus.org/egu2007/), Convener: Artemi Cerdà, Department of Geography, Univ. of Valencia. Spain, [firstname.lastname@example.org](mailto:email@example.com), [http://www.uv.es/~acerda/](http://www.uv.es/~acerda/) John Boardman, Environmental Change Institute, University of Oxford. UK, [firstname.lastname@example.org](mailto:email@example.com), [www.eci.ox.ac.uk/staff/jb.htm](http://www.eci.ox.ac.uk/staff/jb.htm) Yves le Bissonnais, Laboratoire d’étude des Interactions entre Sols, Agrosystèmes et Hydrosystèmes (LISAH). Montpellier, France, [firstname.lastname@example.org](mailto:email@example.com), and Dennis Flanagan, USDA-ARS. National Soil Erosion Research Laboratory West Lafayette, IN, USA, [firstname.lastname@example.org](mailto:email@example.com) Agriculture lands are worldwide affected by non-sustainable soil and water losses. Intense ploughing, excessive use of herbicides, heavy machinery, climate change, soil organic matter exhaustion and salinization induce the degradation of agriculture land, and result in high erosion rates. Agricultural soils are the most fundamental resource we have due to that (i) adequate food supply, (ii) water and soil resources, (ii) carbon sequestration, (iv) natural vegetation and fauna depend on productive land. Soil erosion by wind and water are threatening the agricultural soils. The special session on Soil Erosion on Agriculture Land (SEAL) will discuss topics on soil degradation, past and present erosion processes, experimental and laboratory studies, processes and factors, field measurement and quantification, modelling and mapping erosion risk assessment, prediction and soil conservation policies of agricultural land. Papers on soil erosion research on agricultural land are welcome. Pls contact any convener. ▲ Session on ANTS IN THE SOIL SYSTEM: A HYDROLOGICAL, CHEMICAL AND BIOLOGICAL APPROACH Convener: Anita Risch, Head of the Research Group Animal Ecology, Unit Community Ecology, Swiss Federal Research Institute, Department Landscape, Vegetation Dynamics, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland, [firstname.lastname@example.org](mailto:email@example.com) Co-conveners: Leena Finér, Finnish Forest, Research Institute, Joensuu, Finland, [firstname.lastname@example.org](mailto:email@example.com) Martin Jurgensen, School of Forest Resources and Environmental Science, Michigan Tech, USA, [firstname.lastname@example.org](mailto:email@example.com) Artemi Cerdà, Department of Geography, University of Valencia. Spain, [firstname.lastname@example.org](mailto:email@example.com) Ants are important components of most soil invertebrate communities. Besides their large contribution to biodiversity, they are considered ecosystem engineers that alter the flow of water, energy and nutrients through terrestrial ecosystems. However, despite the importance of these insects for terrestrial ecosystems, relatively little is known on how they influence soil hydrological, chemical and biological processes. In this session, we will review the status of soil/ant relationships and present case-studies on the impact of both mineral and organic mound-building ants on soil processes. We invite contributions on any of these soil-ant topics. If enough interest is shown, we would like to have selected papers published as a special issue in a soil, ecology, or an entomological journal. Pls contact any convener. April 15-20, 2007 Analysis and Characterization of Black Carbon in the Environment (Co-listed in Biogeosciences, Atmospheric Sciences, Hydrological Sciences, Ocean Science & Soil Science Systems), EGU General Assembly in Vienna. Call for papers will be open October 2006 until January 15, 2007, deadline for support applications: December 8, 2006. http://meetings.copernicus.org/egu2007/ Conveners: Michael W. I. Schmidt, firstname.lastname@example.org and Orjan Gustafsson, email@example.com. (Co-sponsored by Biogeosciences, Atmospheric, Hydrological and Ocean Science, and Soil Science Systems) which can be found under the Biogeosciences Programme Black carbon (BC) has received increasing attention due to its importance in a wide range of biogeochemical processes. For example, BC storage in ocean sediments represents long-term sink in the global carbon cycle, and BC aerosols in the atmosphere affect Earth’s radiative heat balance. BC can be a useful tracer for Earth’s fire history: it is a significant fraction of the carbon buried in soils and sediments; and it is an important carrier of organic pollutants. EGU meetings are very cross-disciplinary and thus attractive for scientists from diverse backgrounds, including those studying BC. One focus of the BC-meeting is to bring together the broad scientific community studying black carbon in the environment to discuss the latest research. This is the third meeting of this kind after Goldschmidt 1999 and EGU 2005. A second focus of the meeting is to discuss methodological aspects. Variations in BC chemistry, along a combustion continuum, create serious methodological problems, as every BC measurement method detects a unique window of the BC spectrum. For example, methods which rely on optical microscopes to detect charcoal particles fail to detect sub-micron soot particles, while chemical or thermal methods which rely on the refractory nature of BC fail to detect the partially charred material which can be easily degraded. Effective atmospheric methods, which measure the absorptivity of a sample, cannot be used when BC occurs within an absorptive matrix, like soils or sediments. To address these methodological problems a ring trial was held (2004-6) on 12 BC-containing samples and materials potentially creating artifacts analyzed by 17 labs worldwide from soil, atmospheric, marine and water sciences (http://www.geo.unizh.ch/phys/bc). Results, implications and potential spin-off projects stemming from this ring trial will be presented and discussed during the meeting. April 15-20, 2007 Scales and scaling in surface and subsurface hydrology, Vienna, Austria Dear Colleague, May I bring to your attention the session NP3.08 - Scales and scaling in surface and subsurface hydrology (EGU General Assembly 2007), in Vienna, Austria, 15-20 April 2007. A short-cut link to this session is: http://www.cosis.net/members/meetings/sessions/information.php?p_id=248&s_id=4244 Please note that the deadline for receipt of abstracts is 15 January 2007. João de Lima Meeting/Conference: EGU General Assembly 2007 Session Title: Scales and scaling in surface and subsurface hydrology Session: NP3.08 Conveners: João de Lima; firstname.lastname@example.org; Allen Hunt; email@example.com; Witold Krajewski; firstname.lastname@example.org Event Description: The issue of scales and nonlinear physical, chemical and biological processes is of fundamental importance in hydrology. The questions of how such processes are organized in space and time across a range of scales, how different processes interact at different scales, and how observations at one scale are related to those at another have profound implications for our ability to predict hydrologic cycle components. Answering these questions, in view of the undergoing environmental changes at all scales, requires concerted theoretical, modeling, and experimental efforts. Such efforts are undertaken by research groups around the world. This session is devoted to reporting research results on all aspects of scales and scaling in both groundwater and surface hydrology, including hydrometeorology and ecohydrology. Scaling research topics on hydrologic processes on scales ranging from laboratory to hillslope to small (e.g. urban) basins to regions and continents to entire Earth are all appropriate. Contributions of both scientific and engineering aspects of scaling research and applications are welcome. It is Convenors’ intention to approach this complex issue in a broad way in order to encompass different scales both in time and space including interactions between nonlinear hydrologic processes at different scales. Posters focusing on a specific scale, either for surface or groundwater processes and their modeling and observations, are also welcome. Prof. Dr. João Pedroso de Lima Department of Civil Engineering Faculty of Science and Technology Polo 2 - University of Coimbra 3030-290 Coimbra Portugal +351-239-797-183 (Department of Civil Engineering Office phone) +351-239-797-179 (Department of Civil Engineering Office fax) +351-239-797-123 (Department of Civil Engineering fax) April 29-May 2, 2007 Join Us at the Conference! April 29 - May 2, 2007 Terrigal, New South Wales, Australia Join the International Agrichar Initiative for a conference on Agrichar Science, Production and Utilization, being held in coastal New South Wales, Australia. The International Agrichar Initiative, a program of Renew the Earth, is a new consortium of research and development interests devoted to the sustainability of the world’s soils, and to sustainable bioenergy production. What is the International Agrichar Initiative? The International Agrichar Initiative is an informal, newly-formed coalition of research, commercial and policy-oriented people and organizations devoted to the sustainability of the world’s soils, and to sustainable bio-energy production. Agrichar production and utilization can renew the world’s soils through the addition of organic carbon, which can help solve the pressing problem of global climate change. The Agrichar production process also converts agricultural waste into valuable bio-fuels. History of the Agrichar Initiative During the 18th World Congress of Soil Science (WCSS) in July 2006 in Philadelphia, Pennsylvania, a group of scientists, business interests, policy experts and others met to discuss the research priorities and challenges of this important area. The result is the International Agrichar Initiative, a movement to pursue a more organized research, development and commercialization effort to further the promise of Agrichar. For information on the July 2006 meeting in Philadelphia and some current Agrichar-related projects and activities, click here. What is the ‘Agrichar process’? Agricultural feedstocks such as animal manure, rice hulls, peanut shells, corn stover or forest waste are pyrolyzed at low temperatures to produce a char product (“Agrichar” or “biochar”) and separate bio-energy streams, in the form of oils and/or gases. The biochar captures about 50% of the carbon in the feedstock, and can be used as a soil amendment to improve soil fertility, stability, and productivity, and to store carbon in the soils, as a means of mitigating global warming. The use of Agrichar in soils mimics the Terra Preta (“dark earth”) soils of the Amazon Basin, which have sequestered high quantities of carbon for thousands of years, and have dramatically improved soil fertility and sustainability without chemical inputs. The bio-energy produced, which accounts for the other 50% of feedstock carbon, can be used to fuel a variety of energy needs. For additional information on the 2007 Conference of the International Agrichar Initiative, contact one of the following: - Contact in the U.S.: Debbie Reed, International Agrichar Initiative, email@example.com Tel: 571-431-6626 - Contact in Australia: Adriana Downie, BEST Australia, firstname.lastname@example.org, Tel: +61 2 43404911 Submit an Abstract! The deadline for Abstracts for Oral and Poster Presentations is Wednesday, January 31, 2007. Click below for information on submitting an abstract. Instructions for submitting an abstract May 2007(?) COMLAND Meeting in Tanzania Tanzania, May—2007 (Postponed). Organized by Fidelis Kaihura; contact Paul F. Hudson, email@example.com. May 3-4, 5-8, 2007 International Geographical Union Commission on Biogeography and Biodiversity Seminar on Biogeography and Biodiversity, Srinagar, Uttaranchal, India About the Seminar: The IGU Commission on Biogeography and Biodiversity organizes fora for presentation and discussion (i) within the IGU structures (Regional Meetings, General Assemblies), (ii) by group meetings with national geographical bodies, and (iii) to promote biogeographical research and scholarship by geographers within other ICSU organizations, the International Association for Vegetation Science and the International Biogeography Association. Joint meetings with national biogeography groups take place regularly with help of commission’s full members. One of the prime objectives of the commission’s work is to promote biogeographical teaching, research and scholarship and to contribute to the development of biogeography as an important sub-discipline. Commission meetings address various aspects of spatial, ecological and historical biogeography, and emphasize the societal relevance of biogeography. In India, biogeography is a thriving sub-discipline. Recently, numerous studies on patterns of biodiversity, implications of global climate change, ‘Yantra’ was established on a slab here and Goddess killed the demon at a nearby place Kaliyasaur. Long ago hundreds of tantric adepts especially those belonging to the tradition of Sri Vidya did their practices here. Tantrik practices including human sacrifice prevailed until 8th Century when Adi Shankaracharya displaced this slab and hurled it into the Alaknanda river. This slab remains the most significant shrine in this area lying in the riverbed on the town outskirts. It is regarded as a living Sri Yantra, the most complex of all yantras. Here tantrics propitiated the Divine Mother in the form of Sri Vidya. The old Srinagar city was destroyed later in the Gohna Lake dam-burst which destroyed all the old relics of the town. Today this town is an important cultural and educational centre in Uttranchal. Baba Gorakhnath Cave, Shankar Math, Jain Mandir, Gurudwara, Kamleshwar Temple, Devalgarh, Khirsu Hill Station and Chauras, Dhari Devi are worth visiting in and around Srinagar (Garhwal). The seminar themes are: - Status of biogeography and biodiversity in developing and developed countries - Biodiversity hot-spots in the Himalaya - Medicinal plants and their potential for livelihood enhancement in the Himalaya - Land use and land cover change impacts on biogeography and biodiversity - Impacts of climate change and natural hazards on biodiversity - Biodiversity conservation, indigenous knowledge and community participation in the Himalaya - Political ecology of resource use conflicts in the Himalaya - Culture and civilization for environmental conservation and human development Location: Srinagar (Garhwal) is situated in Garhwal Himalaya, approximately 110 km NE of Rishikesh, at the bank of Alaknanda River. During the summer months, temperatures range between 16° and 38° C. In May (pre-monsoon season), weather is usually pleasant. Srinagar was once the capital of Garhwal Kings. Home to H N Bahuguna Garhwal University, Srinagar was the capital of Garhwal before the arrival of British rule. Srinagar received its name from Sri Yantra. Invitation: Community people, practitioners, technicians, academicians, NGOs, research and other institutions are invited to attend this seminar. The seminar will be a forum for national and international experts on the above subjects. The programme includes talks and discussions followed by deliberations for recommendations. Moreover, poster sessions will be organized during the seminar. Registration fee for foreigners: Seminar only: US$150; Field trip (w. accommodation & all meals: US$500. Contact: Dr R.B. Singh, Department of Geography, Delhi School of Economics, Delhi University, Delhi 110007, India. +91-9412079068, +91-9412922575; Fax: +91-1346252970; firstname.lastname@example.org, email@example.com. Other contacts: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com May 22-25, 2007 1st International Biodiversity Congress: Working Together for Livelihood Security, Food Security and Ecological Security for Life on Earth, Bangkok, Thailand. Contact: Ram Bhandari, firstname.lastname@example.org, email@example.com May 22-27, 2007 The world is experiencing rapid urbanization, industrialization and globalization. The pace, depth, and magnitude of these changes, have exerted severe ecological stresses on humankind living conditions and life support ecosystems across all scales - from local to regional, and global scales. Water shortages, desertification, soil degradation, greenhouse gas emissions, elevated sediment and nutrient fluxes to the coastal seas and other environmental problems are increasingly becoming the common side effects of those human activities. Sustainability can only be assured with an ecological understanding of the complex interactions between environmental, economic, political, and social/cultural factors and with careful planning and management grounded in ecological principles. Ecological complexity and sustainability are becoming a core concept and instrument for improving our common future. This Eco Summit will focus on integrative aspects of all ecological science and its application under the general theme of “Ecological Complexity and Sustainability: Challenges and Opportunities for 21st-Century’s Ecology”. The aim of this Eco Summit is to encourage a greater integration of both the natural and social sciences with the policy and decision-making community to develop a better understanding of the complex nature of ecological systems. This understanding will provide the basis for sustainable solutions to environmental problems. We expect this meeting to attract the broadest representations of ecological organizations, ecologists and practitioners on ecological sustainability issues from all over the world. We intend to show the unification and determination of our ecological community as a whole to maximize our commitment to use ecological knowledge and understanding to meet the challenges raised from the Earth Summit (1992), the World Summit on Sustainable Development (2002), and the United Nations 2005 Millennium Review Summit. Our scientific program will feature: plenary sessions, symposia, contributed oral and poster sessions, and special evening sessions. A variety of scientific field trips will also be planned. We will have 14 world known plenary speakers and more than 10 Elsevier ecology journals plan to publish special issues from this conference. The language of the conference is English. Visit the Eco Summit website at www.ecosummit2007.elsevier.com to learn more about the conference, find out about the key deadlines on submitting symposium and session proposals as well as abstracts for oral and poster presentations and sign up for Eco Summit Alerts. Abstract deadline: January 15, 2007. May 24-26, 2007 International Conference on Zinc Crops 2007: Improving Crop Production and Human Health, Istanbul, Turkey. Co-organized by the International Zinc Association (IZA) and International Fertilizer Industry Association (IFA), in cooperation with the Sabanci University and HarvestPlus (the Challenge Program on Food Fortification of the Consultative Group on International Agricultural Research). Contact: Prof. Ismail Cakmak, Sabanci University (Faculty of Engineering and Natural Sciences), 349 56 Tuzla, Turkey. Phone: 90-216-4839524, Fax: 90-216-4839550, firstname.lastname@example.org. Regular updates on Zinc Crops 2007 will be posted on: www.zncrops2007.info. June 11-15, 2007 Second Conference: Sharing Indigenous Wisdom: An International Dialogue on Sustainable Development. www.sharingindigenouswisdom.org Abstract/Proposal Deadline: 12/1/2006 Venue: Radisson Hotel & Conference Center, Green Bay, WI, U.S.A. The second conference is being hosted by College of Menominee Nation’s Sustainable Development Institute to foster dialogue on traditional indigenous knowledge being utilized and incorporated as models and methods of sustainable practices. Traditional or Indigenous Knowledge refers to the wisdom, embodied in indigenous communities or groups, being utilized to preserve and protect resources vital to the continuity of indigenous communities or groups. This year the conference will focus on Natural Environment foundational element of the Menominee model of sustainable development. In collaboration with the USDA Forest Service and the International Union of Forest Research Organization’s Task Force on traditional forest knowledge, papers or presentations are invited for the following topics, consistent with the focus on Natural Environment: - History of the relationship of formal forest science and traditional knowledge in forest management - Application of traditional knowledge in forest ecosystem assessments and management - Indigenous community priorities for forest research and collaborative forest management - Case studies on joint initiatives and utilization of traditional and formal scientific knowledge in forest management activities - Development of good practices for including both traditional knowledge and formal science in forestry education, research and forest management activities If you wish to present a paper, give a poster presentation, or organize a workshop session, send your proposal, abstract or description, maximum 500 words single spaced, no later than December 1, 2006 to: Dale Kakkak, Sustainable Development Institute, College of Menominee Nation, P.O. Box 1179, Keshena, WI 54135, U.S.A. Or send by e-mail to email@example.com Acceptance letters will be mailed no later than January 31, 2007. Priority will be given to papers or presentations done jointly between indigenous community members or practitioners and researchers. For more information about the call for papers visit our website: www.sharingindigenouswisdom.org Deadline for submissions of abstracts or proposals - December 1, 2006 Notification of acceptance – January 31, 2007 Final Papers/Presentations Due - April 30, 2007 Registration fee: US$300; Student rate: US$150; early registration through April 1, 2007. After that time, the registration fee will be US$375. Conference registration includes scheduled meals/breaks and materials. Individuals are responsible for their own travel and lodging reservations and expenses. Radisson Hotel & Conference Center Green Bay 2040, Airport Drive Green Bay, WI 54313, U.S.A. Phone: +1-(920) 494-7300; Fax: (920) 494-9599; www.radisson.com/greenbaywi For more information contact: Dale Kakkak, firstname.lastname@example.org and/or Nathan Fregien, email@example.com, www.SustainableDevelopmentInstitute.org June 11-15, 2007 The 10th International Symposium on Soil and Plant Analysis will be held in Budapest, Hungary. For further information, please visit: http://www.spcouncil.com/symposium.htm. Contact Yash Kalra at firstname.lastname@example.org June 13-17, 2007 5th International Water History Association Conference, "Past and Futures of Water", Tampere, Finland The International Water History Association (IWHA) will hold its fifth biennial conference in Tampere, Finland, in June 2007. The conference program addresses diverse topics related to water history and futures, and will provide an excellent opportunity for scholars and practitioners from a variety of disciplines and different parts of the world to meet and discuss the many fascinating aspects of water history. Major themes are: Water and the city; Water, health and sanitation; Water, food and economy; Water history research methodologies; and Water history and its relevance. Other relevant session topics are welcome. The first Call for Sessions and papers by 15 Oct is now out. For more information please visit: [5th IWHA Conference Web site](http://www.iwha2007.tavicon.fi), or download: [First Call for Sessions and Papers](http://www.iwha2007.tavicon.fi). You may also contact: email@example.com. June 25-30, 2007 --- 5th International Congress of ESSC of the EUROPEAN SOCIETY FOR SOIL CONSERVATION Changing Soils in a Changing World: the Soils of Tomorrow June, 25 -30 2007 Palermo, Italy Third Announcement BACKGROUND Soil is the thin layer of the Earth that fulfils fundamental functions in terms of life in general and particularly in terms of need and well being of human societies. More and more demands are being imposed upon the soil by different human activities. These activities are growing and competing with each other, without taking account of the diversity of soils, of their function and of their potentials. As a consequence soils are degraded in different ways depending on their vulnerability. The second half of the twentieth century was particularly disastrous: erosion, urbanisation, landslide and flooding, local and diffuse contamination, salinization, entisolization, were the main problems linked to an unbalanced ratio Man/Soil. In both the western and the eastern countries of Europe, the headlong rush into agricultural, industrial and urban development wreaked destruction upon soils and their functions. Little has been done in reality to mitigate soil degradation and to improve the condition of soil already heavily degraded. This is largely due to ignorance, in all spheres of society, of what the soil really is and of why it is necessary to preserve its functions. Such ignorance has its roots in the absence of any soil awareness. OBJECTIVES The main objectives of the 5th ESSC International Congress is to promote exchange and discussion about the problems that affect the soils due to the pressure of Man on Soils and Landscape that year after year is becoming more and more evident and to stimulate the soil awareness in the civil society. The Congress is open for soil scientists, educators, policymakers. It will consist of invited lectures, scientific sessions with oral and poster presentations and scientific and cultural excursions. DEADLINES \| Date \| Event \| \|--------------------\|--------------------------------------------\| \| January, 31 2007 \| Deadline for abstract submission \| \| February, 28 2007 \| Deadline for registration at reduced fee \| \| March 31 2007 \| Notice of acceptance of abstracts \| \| April, 30 2007 \| Last announcement and final program \| For registration and abstract submission, please refer to website: [www.escepalermocongress.it](http://www.escepalermocongress.it) The Organizing Committee will take into account the possibility of publishing the congress proceeding in a volume printed by CATENA VERLAG. All participants are kindly requested to submit their papers/presentation, in compliance with the instructions indicate on web site. TOPICS AND INVITED SPEAKERS** \| Topic \| Speaker \| Location \| Country \| \|--------------------------------\|--------------------\|----------\|---------\| \| Lectio Magistralis \| Ahmet MERMUT \| Saskatchewan \| Canada \| \| Topic 1: Soils and Society \| Victor TARGULIAN \| Moscow \| Russia \| \| Topic 2: Soil Erosion \| Eric ROOSE \| Montpellier \| France \| \| Topic 3: Soil Organic Matter \| Nicola SENESI \| Bari \| Italy \| \| Topic 4: Soil Degradation and Desertification \| Marcello PAGLIAI \| Florence \| Italy \| \| Topic 5: Soil Pollution and Contamination \| Steve McGRATH \| Rothamsted \| United Kingdom \| \| Topic 6: Soil Conservation and Soil Quality \| Lars BERGSTRÖM, \| Uppsala \| Sweden \| \| Topic 7: Policies for Environmental Conservation in a Global Society \| Winfried BLUM \| Wien \| Austria \| \| Topic 8: New Approaches and Technologies for Soil Assessment \| Paolo SEQUI \| Rome \| Italy \| ORGANIZING COMMITTEE Carmelo Dazzi, Vito Ferro, Vincenzo Bagarello, Salvatore Monteleone, Ignazio Poma, Edoardo Costantini, Lucio Gristina, Giuseppe Lo Papa Università di Palermo Viale delle Scienze, 90128 Palermo, Italy Tel: +39 0916650247; fax: +39 0916650229 PRELIMINARY PROGRAMME 25th June, Monday Morning: Arrival of participants and registration 16:00–18:00 ESSC Council Meeting 18:00–19:00 Opening ceremony 19:00–20:00 Lectio Magistralis 20:00 Welcome party 26th June, Tuesday (Topic 1 & 2) 08:30 – 13:00 Registration 09:00 – 09:45 Opening lecture 09:45 – 10:00 Coffee-break 10:00 – 12:30 Thematic session (Topic 1) 12:30 – 14:30 Lunch 14:30 – 15:15 Opening lecture 15:15 – 17:45 Thematic session (Topic 2) 17:45 – 18:00 Coffee-break 18:00 – 19:00 Poster sessions (Topic 1 & 2) 27th June, Wednesday (Topic 3 & 4) 09:00 – 09:45 Opening lecture 09:45 – 10:00 Coffee-break 10:00 – 12:30 Thematic session (Topic 3) 12:30 – 14:30 Lunch 14:30 – 15:15 Opening lecture 15:15 – 17:45 Thematic session (Topic 4) 17:45 – 18:00 Coffee-break 18:00 – 19:00 Poster sessions (Topic 3 & 4) 19:00 – 20:00 ESSC General Assembly 28th June, Thursday (Topic 5 & 6) 09:00 – 09:45 Opening lecture 09:45 – 10:00 Coffee-break 10:00 – 12:30 Thematic session (Topic 5) 12:30 – 14:30 Lunch 14:30 – 15:15 Opening lecture 15:15 – 17:45 Thematic session (Topic 6) 17:45 – 18:00 Coffee-break 18:00 – 19:00 Poster sessions (Topic 5 & 6) 21:00 Congress dinner 29th June, Friday (Topic 7 & 8) 09:00 – 09:45 Opening lecture (invited speaker) 09:45 – 10:00 Coffee-break 10:00 – 12:30 Thematic session (Topic 7) 12:30 – 14:30 Lunch 14:30 – 15:15 Opening lecture (invited speaker) 15:15 – 17:45 Thematic session (Topic 8) 17:45 – 18:00 Coffee-break 18:00 – 19:00 Poster sessions (Topic 7 & 8) 19:00 – 19:30 Conclusions of the Congress 30th June, Saturday 07:30 – 19:45 One day scientific and cultural excursion (2 choices) 1st choice scientific and cultural excursion to Sparacia farm and Temple valley of Agrigento; 2nd choice scientific and cultural excursion to Belice area and Selinunte archaeological area; SCIENTIFIC AND CULTURAL EXCURSION 1st option SPARACIA FARM AND TEMPLE VALLEY IN AGRIGENTO Cultural and pedo-agronomical aspects of the Mediterranean environment. Vertic soils of the Mediterranean environment. Experimental installations for measuring soil water erosion at microplot, plot and basin scale in a hilly Sicilian area. The trip goes on to Agrigento with the visit of the TEMPLES VALLEY, where there is one of the best preserved Greek archaeological zones in the world. An English speaking guide will show the Concordia, Venus, Zeus and Hercules temples. At the end of the visit return to Palermo 2nd option BELICE AREA AND SELINUENTE ARCHAEOLOGICAL PARK Cultural and pedo-agronomical aspects of the Mediterranean environment. Problems of land management due to land managers wrong decisions. Environmental problems due to soil consumption. The trip goes on with the visit of the important ARCHAEOLOGICAL PARK OF SELINUENTE. It consists of various temples, set in a semi-deserted zone, since the old town is not inhabited nowadays. There are the remains of the old city, destroyed by an earthquake in the antiquity. At the end of the visit return to Palermo WEBSITE: www.esscpalermocoongress.it CALL FOR PAPERS: Sustainability is a word that is used in many contexts, and increasingly so when rural and agricultural issues are under discussion. It is a concept that suggests that the continuity of economic, social, and institutional aspects of our society, as well as the non-human environment, is far from assured, and that careful thought and management is needed to sustain them. The EFITA/WCCA 2007 conference will examine the role and use of information and communication technologies in three key areas of rural and agricultural sustainability: * ICT SUPPORTING ON-FARM BUSINESS What does rural industry want/need from an ICT system to help it maintain itself, meet targets, meet legislative requirements and secure competitive edge? Theme co-ordinator: Prof Gerhard Schiefer, University of Bonn Papers under this topic heading include, but are not limited to: - Farm management systems - Decision support tools - Aquatic Resource Management - Dairy and Animal Production Systems - e-AgBusiness and Production Chain Management - Field Data Acquisition and Recording - Food Safety Control/Tracking-Tracing - GIS and Precision Agriculture - Grid Applications - Information Systems and Databases - Instrumentation and Control - Portal / Internet Services - Wireless and Sensor Networks - e-collaboration/e-communities * ICT SUPPORTING RURAL SUSTAINABILITY Can ICT support rural sustainability? What is the evidence that it can help? What have we learned from recent implementations and what new technologies are on the way that may take us further? Theme co-ordinator: Dr Sarah Skerratt, Scottish Agricultural College Papers under this topic heading include, but are not limited to: - Broadband applications for rural communities - SMEs and digital communication - The emergence of the rural e-citizen - Education/Training and Distance Learning/Professional Accreditation - Rural youth and ICTs - Internet supported communities - Rural living labs - ICTs role in capacity-building and social capital development in rural communities? - Cultural evaluation of ICTs in rural development - Transferability of 'solutions' across territories and countries * ICT SUPPORTING ENVIRONMENTAL SUSTAINABILITY Can ICT support the achievement and maintenance of environmental sustainability in the rural and agricultural sectors? How can ICT be used to support the development of environmental policies and how can it help farmers and agronomists and other end users to deliver them? Theme co-ordinator: Andy Offer, ADAS Papers under this topic heading include, but are not limited to: - Predicting the effect of policy changes - Answering policy maker's questions - Spatial analysis, GIS and datasets - Design and evaluation of mitigation measures - Modelling ecological consequences of land management practice - Mitigating and adapting to multiple stressors - Environmental monitoring systems - Using ICT to deliver the policy objectives - Increasing uptake/understanding of legislation - Tracking uptake of schemes and policies - e-Government - Environmental knowledge transfer using ICT - DSS for farmers and consultants You are invited to submit an extended abstract for one of these themes by Nov. 1, 2006. Instructions for authors, and other information about the conference, can be found on our website [www.efitaglasgow.org](http://www.efitaglasgow.org). July 2-6, 2007 --- Durban International Convention Centre 2–6 July 2007 --- 4TH WORLD ENVIRONMENTAL EDUCATION CONGRESS LEARNING IN A CHANGING WORLD --- Invitation from Chair Committees & Congress Organiser Event Format Location & Venue Programme Theme & Sessions Papers/Authors/Proceedings Technical Tours Invitation From Chair On behalf of the WEEC 2007 Committee and the Environmental Education Association of Southern Africa, I invite you to join us in Durban, South Africa, for an exciting and innovative congress. This is the first time that the World Environmental Education Congress comes to Africa, and it comes at a time when there are calls for revisiting our teaching and learning beyond the classroom. As researchers, practitioners and policy makers we need to ensure that we contribute to the values and goals of sustainability as found in the UN Decade of Education for Sustainable Development, the UN Development Millennium Goals; Education for All and many other international directives. In this world of dynamic social, economic and environmental influences, the challenge that faces all practitioners in all spheres of development and governance is not only what to teach but also how to effect learning. We trust that the Congress theme LEARNING IN A CHANGING WORLD, opens the opportunity for critical reflection on the role of education - practice and theory - in effecting these international goals for a better future for all! The World Environmental Education Congress Secretariat aims to launch the World Environmental Education Association during the Congress. An exciting 2007 Congress programme includes not only the daily keynote papers, research-based papers, poster presentations, workshops and round tables, social marketing of programmes in a market place (exhibitions and soap boxes), but also a number of exciting cultural events and scientific tours. The Congress also opens opportunity for optional pre and post tours for delegates and accompanying persons to explore the rich natural and cultural heritage of Southern Africa. We are determined to create rich opportunities for LEARNING from each other, and to enable CULTURAL EXCHANGE to strengthen education for sustainability, theory, practice and new ways of living together. I invite you to Shosholoza all the way to Msawawa (Southern Africa) to experience it for yourself. Ms Mumsie Gumede Chairperson WEEC 2007 http://www.weec2007.com/index.php Contact: firstname.lastname@example.org / email@example.com July 2-13, 2007 The IAHS Workshop on Impact of Environmental Change on Sediment Sources and Sediment Delivery, Perugia, Italy There is increasing concern for the potential impact of environmental change, including both climate change and land use change, on sediment fluxes in catchments and river basins. Increased sediment loads and concentrations can result in significant environmental problems associated with sedimentation in river channels, reservoirs, canals and related hydraulic structures, increased water treatment costs and degradation of water quality and aquatic habitats. Such impacts relate to both the physical presence of the sediment and its biogeochemical impact. Any attempt to develop an improved understanding of the potential impact of climate and land use change on sediment fluxes must consider their interaction with both sediment sources and sediment delivery mechanisms. Sediment source exerts a key influence on the sensitivity of mobilization and delivery processes to environmental change and on the biogeochemical properties of sediment (e.g. nutrient and contaminant content). Similarly, small changes in catchment hydrology could generate important changes in sediment delivery dynamics, which increase slope-channel connectivity and greatly increase the proportion of the sediment mobilized from the catchment surface that reaches the channel network. The workshop will review the existing understanding of sediment sources and sediment delivery in catchments and river basins and their likely sensitivity to the impacts of environmental change. Deadline for receiving abstracts is January 31, 2007. All abstracts should be submitted electronically through the General Assembly Website http://www.iugg2007perugia.it/. Convenor: Des Walling, University of Exeter, School of Geography and Archaeology, Amory Building, Rennes Drive, Exeter EX4 4RJ, UK. Tel: +44 1392 263345; Fax: +44 1392 263342, firstname.lastname@example.org Contact: Dirk de Boer, ICCE Secretary, Centre for Hydrology, Department of Geography, University of Saskatchewan, 9 Campus Drive, Saskatoon, Saskatchewan, S7N 5A5 Canada. Phone: +1 306 966 5671; Fax: +1 306 966 5680; email@example.com July 12-15, 2007 3rd International Conference on Social and Organizational Informatics and Cybernetics: SOIC 2007 Orlando, Florida, USA As a response to the many requests we have received, the Organizing Committee of The 3rd International Conference on Social and Organizational Informatics and Cybernetics: SOIC 2007 has decided to extend the deadline for papers/abstracts submissions, as well as for Invited Sessions proposals. Deadlines: Abstract/paper draft submissions and Invited Session Proposals: February 22, 2007 Authors Notification: March 29, 2007; Camera ready, full papers: April 19, 2007 The registration fee of effective invited sessions organizers will be waived and they will receive at the registration desk, for free, a package of 4 DVDs and one CD containing the 6-hour tutorial "Fundamentals and History of Cybernetics: Development of the Theory of Complex Adaptive Systems". The market price of this package is US $295. Twelve more benefits for invited session organizers are listed at SOIC 2007 web page. For submissions or Invited Sessions Proposals, please go to the web site: http://www.cyber-inf.org/saic2007 Authors of the best 10%-20% of the papers presented at the conference will be invited to adapt their papers for their publication in the Journal of Systemics, Cybernetics and Informatics. SOIC 2007 Secretariat. Torre Profesional La California, Av. Francisco de Miranda, Caracas, Venezuela firstname.lastname@example.org July 16-19, 2007 I invite you to participate in the international Symposium on Organic Matter Dynamics in AgroEcosystem which will be organized on July 16-19 2007 in Poitiers, France. Acquiring knowledge about soil organic matter is essential for sustainable use and conservation of resources, for which climate change and increasing demands for bio-energy become crucial global challenges. As you will see, the symposium programme will feature a variety of panel discussions and keynote addresses touching on a range of issues concerning soil organic matter research in this system. Detailed information regarding the symposium (sessions, invited speakers preliminary schedules) and other practical information can be found on the symposium website: http://www.inra.fr/Symposium_OMD_2007 I kindly invite you to register and submit your extended abstract right now. See you in Poitiers in July 2007. Abad Chabbi email@example.com Dr. Abad Chabbi Coordinateur Nationale du l’Observatoire de Recherche en Environnement- Prairies Cycle Biogéochimique et Biodiversité (PCBB) Institut National de la Recherche Agronomique INRA-UEFE Les Verrines 86 600 Lusignan, FRANCE Tel. +33 (0) 5 49 55 61 78, +33 (0) 5 49 55 60 24 (secrétariat) Mobil: +33 (0) 6 82 80 02 85 Fax +33 (0) 5 49 55 60 66 Email: firstname.lastname@example.org, http://pcbb.ore.fr/ July 21-25, 2007 International Conference of SWCS for 2007 Don’t miss next year’s conference, already in the works, at the Saddlebrook Resort in Tampa, Florida. See details in http://www.swcs.org/en/swcs_international_conferences/2006_annual_conference/ Contact: Suzi Case, Administrative Assistant, email@example.com, phone: 515-289-2331, www.swcs.org July 22 to 26, 2007 Conference on Coastal Zone 2007, Portland, Oregon, www.csc.noaa.gov/cz/ August 28-31, 2007 Royal Geographical Society (with IBG) Annual International Conference 2007: ‘Sustainability and Quality of Life’, 28th to 31st August 2007, London, UK. Paper session sponsored by the Political Geography Research Group. Convenors: Daniel Hammett (University of Edinburgh), Libby Porter (University of Sheffield) Abstract: The International Decade of the World’s Indigenous Peoples, seeking the protection and promotion of indigenous peoples’ rights, ended in 2004. Yet, overt and covert abuses of these rights continue. Indigenous peoples often find themselves marginalised in the political and social structures of states, dispossessed of land and access to resources. Globalisation and the spread of transnational corporations bring capital and industry into conflict with Indigenous groups. In many states, such as Botswana, Namibia, Australia and Brazil, conflicts persist over access to land, mineral wealth, natural resource management, as well as broader rights questions. Yet Indigenous people continue to struggle for their rights, and those stories of struggle (and the gains won) are equally crucial to our understanding of contemporary expressions and conflicts concerning Indigenous rights. Identity, politics, space and place are vital to the assertion, and denial, of Indigenous rights. We invite papers concerning such themes, from all continents, to engage in a dialogue about the contemporary Indigenous rights issues. Papers are particularly welcome on the following themes: - Indigenous peoples and land-rights, - Resource access and conflict - mineral extraction, access to water, land management - Political rights and social justice, - The exploitation of knowledge and natural resources, - Policy and governance in the Indigenous domain Please send abstracts of no more than 200 words to firstname.lastname@example.org and email@example.com by 31st January 2007. September 1-5, 2007 Global Forum on Soils, Society & Global Change Celebrating the Centenary of Soil Conservation and Land Restoration in Iceland, Selfoss, Iceland The Icelandic Soil Conservation Service is organizing a Global Forum to highlight the crucial role of soil conservation and land restoration in addressing key environmental and socio-economic challenges for a sustainable future. Most particularly, it marks the celebration by Iceland of a century of organized soil conservation and land restoration operation in 2007 - one of the longest standing in the world. The Forum is organized in partnership with several universities and international agencies, including the United Nations University (UNU) and Ohio State University (OSU), and Icelandic bodies. It is to be hosted by the Icelandic Government under the patronage of the President of Iceland, His Excellency Ólafur Ragnar Grímsson. Land degradation, manifesting in the form of soil erosion and desertification is a global problem of major proportions. The Millennium Ecosystem Assessment ranks land degradation as among the world’s greatest environmental challenges, affecting climate and biological diversity, reducing environmental security, destabilizing societies, worsening food security, and increasing poverty. The Forum highlights innovative thinking to collectively tackle the inter-related facets of these problems through sustainable land management. The Forum will emphasize the synergistic role of sustainable soil management in achieving the Millennium Development Goals. In this context, it will also explore interlinkages in successful implementation of the UN conventions focusing on climate change, combating desertification and biological diversity. The Forum will bring together a group of key stakeholders from around the world, representing a broad spectrum of disciplines. The aim is to use the meeting as a venue for cross-fertilization of ideas in order to fashion a new global consensus on the above-mentioned crucial linkages and develop a common strategy as well as action plans to help achieve sustainable land management during the 21st century. The main emphasis of the Forum will be on discussions and dialogue; this will be supported through selected seminar presentations. Field excursions will demonstrate the century-long Icelandic expertise. and experience, which potentially has a great value to other nations facing similar challenges. The Forum will conclude with a special centennial commemoration event hosted by the President of Iceland. For more information please contact the forum organizers at firstname.lastname@example.org. September 10-14, 2007 The Fourth International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction, and Assessment, International Convention Center, Chengdu, Sichuan, China Organized by the Institute of Mountain Hazards and Environment, CAS, China The first three International Conferences on Debris-Flow Hazards Mitigation were held in 1997 at San Francisco, in 2000 at Taipei, and in 2003 at Davos. Chengdu, China was selected as the venue of the Fourth International Conference on the same themes. China has suffered heavily from debris-flow hazards in about 1,000 counties and cities. Among the most serious recent debris-flow disasters is one that killed 57 people and was triggered by a torrential storm in Yunnan Province in July 2004. In China, efforts have been increasingly made to describe and understand debris-flow processes, thus enabling development of methodologies for debris-flow hazards prediction and assessment. For example, the pre-conference field trip will examine engineering structures that have been installed to counter debris-flow hazards behind one of the World Natural Heritage sites in Jiuzhaigou Valley. The post-conference field trip will visit the Dongchuan Debris Flow Observation and Research Station (DDFORS) in Jiangjiagou Valley (Yunnan Province), which was established by the Institute of Mountain Hazards and Environment, the Chinese Academy of Sciences (CAS), to enhance the understanding of debris-flow mechanics and foster the development of cost-effective countermeasures. The DDFORS has observed, measured, and monitored debris flows for more than 40 years. On the post-conference field trip, visitors will gain an insight into every stage of the debris-flow process. The likelihood to witness a debris flow during the post-conference field trip is high, for more than 10 debris flows occur annually in Jiangjiagou Valley during the rainy season. Pre- and Post-Conference excursions will be organized. Full information, including online contact and registration, is available in http://4thdfhm.imde.ac.cn. Or contact: Dr. Peng Cui, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, P.O. Box 417, Chengdu, Sichuan 610041, China. Phone: +86-28-85214421; Fax: +86-28-85222258, email@example.com September 10-17, 2007 International Workshop on Environmental Changes and Sustainable Development in Arid and Semi-arid Regions, Alashan Left Banner - Long Xin Hotel - Inner Mongolia, China Co-organizers and Co-sponsors: - COMLAND (Commission on land degradation and desertification, International Geographical Union) - IAG (International Association of Geomorphologists, Working group on interactions between aeolian, fluvial and lacustrine processes) - IGCP 500 - INQUA (Project number: INQUA 0503) - GSC (Geographical Society of China) - National Natural Sciences Foundation of China - PAGES Local Organizer: Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing Language of the workshop: English Introduction: The workshop will start on September 10 (registration and a reception in the evening). Paper sessions (normal oral presentation for 15 to 20 minutes, keynote 30 to 40 minutes and posters) are planned for the following three full days including a half day field trip to see rural places around the conference venue. Participants will see irrigation fields facing land degradation, the life of rural people, rehabilitated vegetation against desertification and a field with small sand dunes. The paper sessions will include oral and poster presentations, and there will be evening lectures. An optional pre-conference field trip (maximum participants 35): Highlights of the trip will be the reactivation of the stable dunes and rehabilitation measurements (mainly by vegetation) north of Beijing. Meeting point in Beijing and we shall travel northwards to Hunshandake Sandy Land by bus for three days, starting in the morning of September 7 and returning via the hilly landscape around the Great Wall to Beijing in the afternoon of September 9. This is also the area where considerable effort has been made to reduce soil erosion and deflation as well as to reduce the threat of dust storms in northern China. Xiaoping Yang, having done a little research here, will look forward to leading this trip. An optional post-conference field trip (maximum participants 30): Highlights of the trip will be the megadunes and lakes in the Badain Jaran Desert. The field trip will start on September 14 by bus from Left Banner and then by jeep in the sandy areas. Participants will see high sand dunes (often ca 300 m, maximum 460 m in this sand sea) and permanent lakes in the inter-dune basins. Palaeo-shorelines are distributed around the lakes, illustrating late Quaternary climatic changes. The forms of the dunes are quite diversified and were probably created due to a) different formations related to climatic changes and b) basement rocks. In the last two years the sand sea has been visited by an increased number of tourists. Therefore, the jeep drivers are well experienced to drive in the dune areas (it is really exciting and quite safe). We shall stay two nights in tents in the field and stay in a nice hotel in the Right Banner (Ekenhudge) when returning from the dune fields. We shall travel back to the Left Banner on September 17 (ca 500 km, good road). Along the road between left and right banners we shall visit places where a lot of effort has been practised to plant grasses and trees in the arid environment. Xiaoping Yang, having worked in this area since 1988, will look forward to leading this trip. Alternative trips: It is possible to participate in normally comfortable (and sometimes less expensive) sightseeing tours offered by various domestic and overseas tourist agencies. Tourist agencies organize trips ranging one day to more than two weeks around China, also from Yinchuan. The conference city is relatively close to Xi’an, one of the most famous places for tourists, largely because of the terracotta soldiers buried ca 2000 years ago. THEMES - Case studies and theories relating to land degradation and sustainable development in arid and semi-arid as well as sub-humid regions of various continents; - Comparisons of regional-scale reconstruction of Late Quaternary changes in the deserts of various climate zones (monsoon regions, subtropics and westerlies); - Natural and human impacts on the landscape in various climate zones; - Interactions amongst aeolian, fluvial and lacustrine processes in desert margins. Registration fee: US$600; see more details in http://www.iggcas.ac.cn/iw07/index.htm Deadlines: Early bird payment (US$500) up to April 1, 2007 Final payment for regular registration: August 10, 2007 Receipt of abstracts: June 1, 2007 Receipt of full papers for consideration in the special issues of the international journals (see conference publications below): at the workshop. Contact persons: - Xiaoping YANG, Steering Committee member, COMLAND and local organiser of the Workshop (E-mail: firstname.lastname@example.org; email@example.com). Phone: +86 (0)10 62008389. Fax: +86 (0)10 62032495. Postal address: Institute of Geology and Geophysics, Chinese Academy of Sciences. P.O. Box 9825, Beijing 100029, P.R. China) - Guðrún Gísladóttir, Chair, COMLAND (firstname.lastname@example.org), Department of Geology and Geography, University of Iceland, Askja, Sturlugata 7, IS 101 Reykjavík, Iceland) - Paul F. Hudson, Secretary, COMLAND (email@example.com), Department of Geography and the Environment, University of Texas, Austin, TX 78712-1098, USA) - Arthur Conacher, Steering Committee member, COMLAND (firstname.lastname@example.org), School of Earth and Geographical Sciences, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia) Conference publications: A special theme issue on land degradation and desertification in Geographical Research (Journal of the Institute of Australian Geographers, www.blackwellpublishing.com/journals/ages) and a special theme issue on geomorphology and palaeoclimatic changes in Quaternary Research (http://ees.elsevier.com/qyres/) will be the official venues to publish papers from the workshop. Both oral and poster presenters are welcome to submit their papers to these special issues and the deadline for submission will be September 10, 2007. The guidelines for writing the papers for each journal can be found in the abovementioned websites. Of course, all papers will be subject to the normal reviewing process. September 17-19, 2007 IV International Symposium on Gully Erosion, Pamplona, Spain The Department of Projects and Rural Engineering of the Public University of Navarre (Spain) is pleased to invite you to participate in the IV International Symposium on Gully Erosion, to be held at Pamplona (Spain). This symposium aims to review current understanding and information, and to report progress in the following topics on gully erosion: Importance: e.g. on/off-site gullying impact on soil properties, landforms, and landscape at different times and space scales. Characterization: e.g. experimental investigation on different kinds of gullies, monitoring techniques of gully erosion, historical reconstruction of processes. Controlling factors: e.g. effect of anthropogenic activities on gullying, interaction between gully erosion and other land degradation processes, role of vegetation. Control measures: e.g. prevention of gullying, gully restoration methods. Processes and modelling: last but not least, we shall greatly appreciate receiving contributions on this important topic: e.g. gully headcut mechanics, channel network evolution, gully erosion thresholds, and so on. For more details, please visit the webpage of the symposium: http://www.unavarra.es/erosion/index.htm, which will be periodically updated. Kindly tell your colleagues about this symposium. Dr. Javier Casalí Dr. Rafael Giménez Program Committee Chairs Members Contact: Symposium Gully email@example.com September 23-26, 2007 3rd International Conference on Mechanisms of Organic Matter Stabilisation and Destabilisation in Soils and Sediments, Stamford Grand Hotel, Glenelg, Adelaide, South Australia, Australia As with the two previous conferences, the main goal is to bring together scientists from a range of backgrounds to share ideas and present recent research results. We welcome participation by researchers working in organic geochemistry, soil science, ecology and other disciplines related to the conference themes. With this conference, the themes have been expanded to include natural organic matter (NOM) in aquatic systems and processes that link stabilisation and destabilisation mechanisms in terrestrial and aquatic ecosystems. The aims of the conference are to facilitate the exchange of ideas pertaining to NOM stability in soils and sediments, and encourage the development of integrative and multi-disciplinary collaborative projects. Themes include: Formation, characterisation and fate of NOM including Black Carbon Movement of NOM within and between terrestrial/aquatic ecosystems Application of isotopic techniques in NOM studies Stabilisation/destabilisation mechanisms for NOM Biological significance of NOM pools at various time scales Modelling NOM dynamics. Contact information: Jan Skjemstad, www.clw.csiro.au/conferences/organic/ September 25-28, 2007 First Announcement and Call for papers INTERNATIONAL CONFERENCE “EROSION AND TORRENT CONTROL AS A FACTOR IN SUSTAINABLE RIVER BASIN MANAGEMENT” 25-28 September 2007 Belgrade – Serbia Initiated by: Faculty of Forestry, Belgrade University - marking 100 YEARS OF EXPERIENCE EROSION AND TORRENT CONTROL IN SERBIA Organised jointly with: World Association of Soil and Water Conservation (WASWC) World Association for Sedimentation and Erosion Research (WASER) Conveied in line with the objectives of the: International Sediment Initiative Ibi of UNESCO’s International Hydrological Programme Sponsored by: Public Water Management Company SRBLJAVODE Ministry of Science and Environmental Protection of the Republic of Serbia Ministry of Agriculture, Forestry and Water Management of the Republic of Serbia, Directorate of Water Contact: firstname.lastname@example.org, email@example.com October 17-19, 2007 Bottlenecks, Solutions, and Priorities in the Context of Functions of Forest Resources, Faculty of Forestry, Istanbul University, Istanbul, TURKEY An international symposium will be held by Istanbul University, Faculty of Forestry in Istanbul between October 17-19, 2007 to celebrate the 150th Anniversary of Forestry Education in Turkey with support of Rectorate of Istanbul University, Ministry of Environment and Forestry and International Union of Forest Research Organizations. The symposium aims to provide an international scientific atmosphere with participation of experts from universities, research institutes, public and private sectors and non-governmental organizations in areas relating to the functions of forest resources. Forest resources have a lot of functions such as protecting biological diversity, climate regulation, water production, soil and water conservation, providing natural food, supporting urban life, rural development, creating recreational opportunities, developing game and wildlife, protecting visual and cultural values, strengthening social stability, supporting society’s health, producing wood and non-wood forest products, creating employment and income, providing possibilities for export etc., which have effects on the ecological environmental, social-cultural and economical-financial fields. Subjects of the presentations for the symposium should be thought in the context of the above-mentioned functions of forest resources. It is recommended that authors explain causes of bottlenecks and failures taking place during providing of these functions of forest resources; deal with analyses and solutions toward overcoming them from point of view of one or multidimensional perspective, that is, biological, ecological, economical, organizational, political etc. sides rather than contenting with only inventorial aspects. Symposium Sessions: Bottlenecks in the context of functions of forest resources and analyses relating to them; biological, ecological, economical, organizational, political etc. solutions concerning these bottlenecks (http://www.orman.istanbul.edu.tr/fored150, http://orman.istanbul.edu.tr/fored150, http://184.108.40.206/fored150) Presentation: Oral and poster; Deadlines: Abstract (800-1,000 words): December 20, 2006 Full Paper: June 1, 2007; Accommodation and Registration details: Will be announced later. Symposium Secretariat: Assoc. Prof. Dr. Ferhat GÖKBULAK, Istanbul University, Faculty of Forestry, 34473 Bahçeköy, Istanbul TURKEY. Phone: +90 212 226 11 11, Fax: +90 212 226 11 13, firstname.lastname@example.org. October 18-23, 2007 4th International Conference on Soils of Urban, Industrial, Traffic, Mining and Military Areas (SUITMA4), Nanjing, China, www.issas.ac.cn/suitma4.html/ Dear colleagues, The 4th International Conference on Soils in Urban, Industrial, Traffic, Mining and Military Areas (SUITMA) is scheduled to be held at October, 2007 in Nanjing. The conference will discuss soil changes, problems, remediation, utilization, planning, in relation to urban ecological, environmental and biological health issues. The previous events were in Essen (Germany), Nancy (France) and Cairo (Egypt), respectively. We appreciate your continuous interest in SUITMA and welcome scientists, planners, policymakers, to join the forthcoming event. The world is becoming more and more urbanized, and by which serious problems related to land resource use and ecosystem protection are emerged. Like the rest many parts of the world, China is also experiencing rapid urbanization and industrialization. We are honored to organize such a meeting in Nanjing, a beautiful, friendly ancient city, and believe that SUITMA4 in Nanjing may generate new knowledge and enrich our future collaboration possibilities. Therefore, on behalf of the organizing committee, I sincerely invite you to take part in the important event. See all details on the website. Modern human activities have ever-increasing impacts on soil resources while soils of urban, industrial, traffic, mining and military sites are the very cases showing these impacts. How these soils are formed and how they will affect our ecosystem are big concerns of soil scientists worldwide. Globally, more and more attention is being paid to soils under urbanization and industrialization, with regard to their changes and interactions with urban ecosystem. SUITMA WG (Working Group), IUSS, has taken a leading role in coordinating international efforts to advancing our understanding about the special soil system, by providing a forum for international scientific community of the related fields, since its setup in 1998. The establishment of SUITMA also symbolizes the increasing importance of urban soil study within modern soil science in general. With its previous highly successful academic events in Essen, Nancy and Cairo, SUITMA has clearly defined its scopes and is exploring its new horizons. The scientific committee of SUITMA4 sets the following major areas as its themes and opens to more topics and interests. A) Methodologies 1. Soil survey, description, terminology standardization and sampling in urban areas 2. Laboratory methods for urban soils 3. Soil classification and mapping in urban areas B) Impacts of urbanization on soil resources 4. Land use change under urbanization 5. Pedogenesis and quality changes of urban soils 6. Physical aspects of urban soil changes 7. Chemical aspects of urban soil changes 8. Biological aspects of urban soil changes 9. Historical aspects of urban soil formation and characterization C) Urban soils in relation to urban ecosystem 10. Ecological impact of urbanization in regional and global scales 11. Urban soils and greenery plants 12. Urban soils in relation to water environment 13. Biogeochemical cycling of life-dependent materials in urban ecosystem 14. Pollution status and control of urban soils 15. Remediation of contaminated SUITMA 16. Ecological assessment of urban soil quality 17. Ecology-based urban land use planning D) Socio-economic aspects of soil management in urban areas 18. Urban soil management and legislation E) Other related topics Registration for Conference Regular participants: 250 USD/200 EUR (before August 31st, 2007) Students/Young participants (age below 30): 125 USD/100EUR (with valid certificate) Accompany person: 100 USD/80 EUR. Registration fee covers conference abstract book, reception party, lunches during conference, tea/coffee, closing banquet. Registration payment after August 31st, 2007 will incur 100 USD/80 EUR surcharge. Accommodation: Hanting Hotel (room charge for participants: $25 and $27), 10-min walk from the venue Scientific contact: Prof. Dr. Gan-Lin Zhang, Chair, SUITMA4 Institute of Soil Science, Chinese Academy of Sciences 71 Beijing Donglu, Nanjing 210008, China November 6-10, 2007 International Conference on Integrated Watershed Management for “Bharat Nirman”, New Delhi, India Bharat Nirman (Building up India) has to be based upon a firm foundation. The most firm foundation for the purpose would be the scientific Care, Conservation, Development and Management of the Land — the Mother of the Natural Resources. The productive base of Land is the Soil. World Soil Charter, issued by the FAO in 1981 emphasized the importance of Soil in these words: “Recognizing the paramount importance of land resources for the survival and welfare of people, economic independence of the country and also rapidly increasing needs for more food production, it is imperative to give high priority to promoting optimum land use, maintaining and improving soil productivity and conserving soil resources”. Increasing globalization, liberalization and privatization regimes have not benefited all sections of society in the developing countries. Poverty is still associated with agriculture and rural living. The concerns of the small and marginal farmers, the landless and weaker and disadvantaged sections of the society have to be addressed. Population is rising and per capita availability of land and other natural resources is shrinking. How to produce the diverse needs of humanity today, how to add value to the produce land to benefit the small and marginal farmers, and how to accomplish all these without adversely importing the environment and the productive base of the soil and other natural resources, is today’s challenge. This challenge can be met by converging all the programmes of conservation and ecology, sustainable production, processing of the bio-produce and marketing, infrastructure and a policy umbrella within the natural boundaries of a watershed. Many countries of the world have gathered valuable knowledge and experience in watershed management. All will benefit from an exchange of this pool of knowledge, which should facilitate formulating sound policies and programmes for the new millennium. To delegate and discuss the various aspects related to the issues mentioned, the Soil Conservation Society of India is holding the International Conference on Integrated Watershed Management for Bharat Nirman, from November 6 to 10, 2007 at New Delhi, India with the following objectives: Objectives - Conservation and Management of Land, Water and Forest Resources for the People; and sustainable production. - Promotion of Horticulture to the separates Freedom from Hidden Hunger (of Nutrition). - Care of the Livestock and the Common Property Resources - Mainstay of the Rural Landless and “Antodaya” - the Poorest of the Poor. - Biomass Processing and Marketing for Rural Employment and Livelihoods. - Achieving India’s “Bharat Nirman” Programme. - Achieving United Nations Millennium Development Goals. - Evolving Integrated Watershed Management Systems for Ecology, Economics, Employment and Equity. Major Themes 1. Land Resource Management 2. Water Resource Management 3. Plant Resource Management 4. Animal Resource Management 5. Human Resource Management — Gender, Equity and Rural Institutions. 6. Integrated Watershed Management for Biomass Processing, — Agri-business, Rural Employment and Livelihoods. 7. Rural Investments through Public-Private Partnership. 8. Science and Technology — Research, Extension and Capacity Building. Venue: NASC Complex, New Delhi Duration: (i) The Conference: November, 6-9, 2007 (ii) Field Visits for Delegates: November 10-11, 2007 Important Dates: Submission of Abstract: 31-11-2006 Receipt of full-length paper: 31-05-2007 with electronic copy Registration Fee Foreign Delegates: US$500 (With late fee: US$600) - Students/Research Scholars: US$250 (With late fee: US$300) - Accompanying members: US$100 Indian Delegates: Rs. 4,000.00 (With late fee Rs. 5,000) * Students/Research Scholars: Rs. 2,000 (With late fee Rs. 2,500) Accompanying member: Rs. 1,000 (Indian) * Students/Research Scholars have to provide a certificate from their Head of the Institution/Department for availing of this special rate. Call for Abstracts: Abstracts of unpublished research related to one of the Symposium Themes are invited for presentation as the poster papers. Abstracts are to be submitted as an electronic copy in MS WORD, preferably accompanied by a set of hard copy (complete in all respects) on A4 size paper, typed single space and should not exceed one page (about 300 words) with 25 mm margins all around. The title should be followed by the names of author/authors (with the name of the person presenting underlined) and their affiliation. Please include the e-mail address of the corresponding author for expediting communication. Steering Committee: Chairman: Prof. M.S. Swaminathan; Convenor: Dr. Suraj Bhan Organizing Committee: Chairman: Shri Mohan Kanda; Convenor: Shri B. Rath Contact: Dr Suraj Bhan, President, Soil Conservation Society of India, email@example.com, firstname.lastname@example.org, www.soilcsi.org --- MEETINGS 2008 ++ February 2008 IV World Congress on Conservation Agriculture, India. Contact: Raj Gupta, email@example.com May 18-23, 2008. 15th International Soil Conservation Organization (ISCO) Conference, Soil and Water Conservation, Climate Change and Environmental Sensitivity Budapest, Hungary. May 18-23, 2008. Contact: Professor Ádám Kertész, ISCO President, Head of Department, Sustainable development, global change and ecosystems, Hungarian Academy of Sciences, Geographical Research Institute, Budapest 1112, Budaorsi ut 45, Hungary. Phone: +36-1-309-2686; Fax: +36-1-309-2686, firstname.lastname@example.org More info about ISCO at www.isco.org. July 14-18, 2008 Second International Conference on Ground Bio- and Eco-engineering: The Use of Vegetation to Improve Slope Stability, Beijing, China. http://liama.ia.ac.cn/wiki/doku.php?id=projects:envmodeling:iceco2 Contact: Alexia Stokes LIAMA-CASIA Zhongguancun Dong Lu 95 Haidian District PO Box 2728 Beijing 100080 P.R. China Tel: +86-10-82614528; Fax: +86-10-62647458 email@example.com, http://liama.ia.ac.cn August 25-29, 2008 Eurosoil 2008: Soil - Society - Environment Vienna, Austria, Venue: Vienna University of Technology, Vienna, AUSTRIA Organizers: The national soil science societies of Austria, Croatia, the Czech Republic, Hungary, Slovenia, Slovakia and Switzerland, supported by the Vienna Medical Academy Excursions: Pre- and post-congress excursions in Austria, Hungary, Croatia and Slovenia, the Czech Republic, the Slovak Republic and in Switzerland (together with Austria) Further information can be obtained from the website of the European Confederation of Soil Science Societies (ECSSS) www.ecsss.net. The first complete set of information including the scientific programme, technical excursions and the registration form will be put into the website and/or sent to interested participants in December 2006. Contact: Winfried Blum. firstname.lastname@example.org September 18-22, 2008 Dr. Pandi Zdruli, Project Manager, MEDCOASTLAND Thematic Network sent to us info for the 5th International Conference on Land Degradation to take place at Valenzano, Bari, Italy, at http://www.iamb.it/5ICLD/. Contact Dr Zdruli at CIHEAM-Mediterranean Agronomic Institute of Bari, Via Ceglie 9, 70010 Valenzano (BA), ITALY. Tel: 39 080 4606 253, Fax: 39 080 4606 274, email@example.com, http://medcoastland.iamb.it. Late 2008 6th International Hani-Akha Culture Conference, Luchun County, Yunnan Province, China, near the border area with Northern Vietnam and Lao P.D.R. Mr Bai Bibo welcomes you to the conference organized in his hometown. Trust our members will be interested in presenting academic papers regarding Akh-a-Hani Land and Water Conservation in the Mekong River Basin. Contact Bai Bibo, firstname.lastname@example.org August 1-6, 2010 The 19th World Congress of Soil Science will be held in Brisbane, Australia, 1-6 August 2010, Contact Neil McKenzie at email@example.com. The 20th Congress will be held in Seoul, Korea in 2014. CHANGE OF ADDRESS The new address of Science Publishers Inc., our publishing partner, is at: 113-B Shahpur Jat, 2nd Floor, Behind JP House New Delhi 110 049, INDIA Tel: 41745356, 41745358, 41517046, 41745490; Fax: (91-11) 41517559; firstname.lastname@example.org The WASWC Council highly appreciates the generous help from the following enterprises for sponsoring the publishing of the WASWC Special Publication III “No-Tillage Agriculture” Syngenta AG, Basel, Switzerland www.syngenta.com/en/index.aspx SEMEATO Farm Machinery Co., Passo Fundo, Brazil www.semeato.com.br Eijkelkamp Agrisearch Equipment, Giesbeek, The Netherlands www.eijkelkamp.com Donald Fryrear Custom Products and Consultants, Big Spring TX, USA www.cstlars.usda.gov/wewc/bfryrear.htm The publication will be available in 2007.
SUMMARY AGENDA: CHAMBER 11.30am Prayers Afterwards Oral Questions: Northern Ireland 12 noon Oral Questions: Prime Minister Afterwards Ministerial Statements, including on: - Update on the Northern Ireland Protocol (Secretary of State for Northern Ireland) - NHS update (Minister for Care) No debate Presentation of Bills Up to 20 minutes Ten Minute Rule Motion: United Kingdom Atomic Energy Authority Pension Transfers (Parliamentary and Health Service Ombudsman Investigation) (David Johnston) Until 7.00pm Building Safety Bill: Second Reading Followed by Motions without separate debate: - Programme - Money - Ways and Means At 7.00pm Ministerial Statements, including on: - Strategy for Tackling Violence Against Women and Girls (Minister for Safeguarding) No debate Electoral Commission (Motion) No debate after 7.00pm Draft Online Safety Bill (Joint Committee) (Motion) No debate Presentation of Public Petitions Until 7.30pm or for half an hour Adjournment Debate: Forced repatriation of Chinese seamen from Liverpool after World War Two (Kim Johnson) WESTMINSTER HALL \| Time \| Topic \| \|--------\|----------------------------------------------------------------------\| \| 9.25am \| Privatisation of Channel 4 \| \| 11.00am\| Early years education funding \| \| \| (The sitting will be suspended from 11.30am to 2.30pm) \| \| 2.30pm \| The role of the Trade and Agriculture Commission in international \| \| \| trade deals \| \| 4.05pm \| Guidelines for Do Not Attempt Resuscitation orders \| \| 4.50pm \| Social Justice and Fairness Commission and implications for \| \| \| Government policy \| ## CONTENTS ### PART 1: BUSINESS TODAY - 4 Chamber - 9 Westminster Hall - 10 Written Statements - 11 Committees Meeting Today - 14 Committee Reports Published Today - 15 Announcements - 18 Further Information ### PART 2: FUTURE BUSINESS - 20 A. Calendar of Business - 43 B. Remaining Orders and Notices Notes: Item marked [R] indicates that a member has declared a relevant interest. Virtual participation in proceedings will commence after Prayers. 11.30am Prayers Followed by QUESTIONS 1. Northern Ireland 2. Prime Minister The call list for Members participating is available on the House of Commons business papers pages. STATEMENTS Ministerial Statements, including on: - Update on the Northern Ireland Protocol (Secretary of State for Northern Ireland) - NHS update (Minister for Care) The call list for Members participating is available on the House of Commons business papers pages. PRESENTATION OF BILLS No debate (Standing Order No. 57) - Judicial Review and Courts Secretary Robert Buckland Bill to make provision about the provision that may be made by, and the effects of, quashing orders; to make provision restricting judicial review of certain decisions of the Upper Tribunal; to make provision about the use of written and electronic procedures in courts and tribunals; to make other provision about procedure in, and the organisation of, courts and tribunals; and for connected purposes. - Button Batteries (Safety) Jo Gideon Bill to make provision about the safety of button batteries; and for connected purposes. Local Electricity David Johnston Bill to enable electricity generators to become local electricity suppliers; and for connected purposes. BUSINESS OF THE DAY 1. UNITED KINGDOM ATOMIC ENERGY AUTHORITY PENSION TRANSFERS (PARLIAMENTARY AND HEALTH SERVICE OMBUDSMAN INVESTIGATION): TEN MINUTE RULE MOTION Up to 20 minutes (Standing Order No. 23) David Johnston That leave be given to bring in a Bill to make provision to enable the Parliamentary and Health Service Ombudsman to investigate advice and information given by the Secretary of State and the Government Actuary relating to transfers of pensions from the United Kingdom Atomic Energy Authority pension schemes to the AEA Technology pension scheme; and for connected purposes. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. 2. BUILDING SAFETY BILL: SECOND READING Until 7.00pm (Standing Order No. 9(3)) Notes: Queen’s consent to be signified on Third Reading. Prince of Wales’s consent to be signified on Third Reading. Relevant Documents: Seventh Report of the Housing, Communities and Local Government Committee, Session 2019-21, Cladding remediation—Follow-up, HC 1249 Fifth Report of the Housing, Communities and Local Government Committee, Session 2019-21, Pre-legislative scrutiny of the Building Safety Bill, HC 466, and the Government Response, CP 473 Second Report of the Housing, Communities and Local Government Committee, Session 2019-21, Cladding: progress of remediation, HC 172, and the Government Response, CP 281 Letter from the Chair of the Housing, Communities and Local Government Committee to the Minister for Building Safety and Communities regarding the Government’s response to the Committee’s pre-legislative scrutiny of the Building Safety Bill, dated 19 July 2021 The call list for Members participating is available on the House of Commons business papers pages. 3. BUILDING SAFETY BILL: PROGRAMME No debate (Standing Order No. 83A(7)) Secretary Robert Jenrick That the following provisions shall apply to the Building Safety Bill: Committal (1) The Bill shall be committed to a Public Bill Committee. Proceedings in Public Bill Committee (2) Proceedings in the Public Bill Committee shall (so far as not previously concluded) be brought to a conclusion on Tuesday 26 October 2021. (3) The Public Bill Committee shall have leave to sit twice on the first day on which it meets. Proceedings on Consideration and Third Reading (4) Proceedings on Consideration shall (so far as not previously concluded) be brought to a conclusion one hour before the moment of interruption on the day on which proceedings on Consideration are commenced. (5) Proceedings on Third Reading shall (so far as not previously concluded) be brought to a conclusion at the moment of interruption on that day. (6) Standing Order No. 83B (Programming committees) shall not apply to proceedings on Consideration and Third Reading. Other proceedings (7) Any other proceedings on the Bill may be programmed. 4. BUILDING SAFETY BILL: MONEY No debate (Standing Order No. 52(1)(a)) Jesse Norman That, for the purposes of any Act resulting from the Building Safety Bill, it is expedient to authorise the payment out of money provided by Parliament of: (a) any expenditure incurred under or by virtue of the Act by the Secretary of State, and (b) any increase attributable to the Act in the sums payable under any other Act out of money so provided. Notes: Queen’s Recommendation signified. 5. BUILDING SAFETY BILL: WAYS AND MEANS No debate (Standing Order No. 52(1)(a)) Jesse Norman That, for the purposes of any Act resulting from the Building Safety Bill, it is expedient to authorise: (1) the charging of fees, charges and levies under or by virtue of the Act; and (2) the payment of sums into the Consolidated Fund. STATEMENTS Ministerial Statements, including on: - Strategy for Tackling Violence Against Women and Girls (Minister for Safeguarding) The call list for Members participating is available on the House of Commons business papers pages. BUSINESS OF THE DAY 6. ELECTORAL COMMISSION No debate (Standing Order No. 118(6) and Order of 13 July) Mr Jacob Rees-Mogg That an humble Address be presented to Her Majesty, praying that Her Majesty will appoint Dr Katy Radford as an Electoral Commissioner with effect from 1 September 2021 for the period ending on 31 August 2025; and that Her Majesty will re-appoint Sarah Chambers as an Electoral Commissioner with effect from 31 March 2022 for the period ending on 30 March 2026. Note: If this item is opposed after 7.00pm, the division will be deferred. Relevant Documents: Speaker’s Committee on the Electoral Commission, Second Report of 2021, Appointment of an Electoral Commissioner with responsibility for Northern Ireland, HC 522 Speaker’s Committee on the Electoral Commission, Third Report of 2021, Re-appointment of an Electoral Commissioner, HC 523 7. DRAFT ONLINE SAFETY BILL (JOINT COMMITTEE) No debate after 7.00pm (Standing Order No. 9(6)) Mr Jacob Rees-Mogg That this House concurs with the Lords Message of 19 July, that it is expedient that a Joint Committee of Lords and Commons be appointed to consider and report on the draft Online Safety Bill (CP 405) presented to both Houses on 12 May. That a Select Committee of six Members be appointed to join with a Committee appointed by the Lords to consider the draft Online Safety Bill. That the Committee should report by 10 December. That the Committee shall have power— (i) to send for persons, papers and records; (ii) to sit notwithstanding any adjournment of the House; (iii) to report from time to time; (iv) to appoint specialist advisers; (v) to adjourn from place to place within the United Kingdom; and That Debbie Abrahams, Damian Collins, Darren Jones, John Nicolson, Dean Russell and Suzanne Webb be members of the Committee. PRESENTATION OF PUBLIC PETITIONS No debate or decision (Standing Order No. 153) - Post Office counter services in North East Fife: Wendy Chamberlain - VAT on sunscreen: Patricia Gibson - Proposals for review of tax rates to reduce poverty and inequality: Richard Burgon - Proposed loss of Deangate Ridge: Kelly Tolhurst ADJOURNMENT DEBATE Until 7.30pm or for half an hour (whichever is later) (Standing Order No. 9(7)) - Forced repatriation of Chinese seamen from Liverpool after World War Two: Kim Johnson The morning sitting will start at 9.25am and finish at 11.30am with a five minute suspension at 10.55am. The afternoon sitting will start at 2.30pm and finish at 5.50pm with a five minute suspension at 4.00pm and a 15 minute suspension at 4.35pm (Standing Order No. 10(1) and Orders of 25 March and 16 June). The call list for Members participating is available on the House of Commons business papers pages. 9.25am - That this House has considered privatisation of Channel 4: Dame Angela Eagle Notes: The debate at 9.25am will last for up to 90 minutes. 11.00am - That this House has considered early years education funding: Wera Hobhouse Notes: The debate at 11.00am will last for up to 30 minutes. The sitting will be suspended from 11.30am to 2.30pm. 2.30pm - That this House has considered the role of the Trade and Agriculture Commission in international trade deals: Neil Parish Notes: The debate at 2.30pm will last for up to 90 minutes. Relevant Documents: Oral evidence taken before the Environment, Food and Rural Affairs Committee on 27 April 2021, on the Trade and Agriculture Commission, HC 1346 4.05pm - That this House has considered guidelines for Do Not Attempt Resuscitation orders: Martin Vickers Notes: The debate at 4.05pm will last for up to 30 minutes. 4.50pm - That this House has considered the Social Justice and Fairness Commission and implications for Government policy: Kirsten Oswald Notes: The debate at 4.50pm will last for up to an hour. The sitting will be suspended and time added if divisions take place in the main Chamber (Standing Order No. 10(3)). SECRETARY OF STATE FOR BUSINESS, ENERGY AND INDUSTRIAL STRATEGY 1. Use of the Industrial Development Act 1982 for Coronavirus related assistance CHANCELLOR OF THE DUCHY OF LANCASTER AND MINISTER FOR THE CABINET OFFICE 2. Notification to Parliament of a Contingent Liability: Indemnification to UNFCCC 3. Response to the 2021 Senior Salaries Review Body Report SECRETARY OF STATE FOR DIGITAL, CULTURE, MEDIA AND SPORT 4. Update on the Government response to the loot box call for evidence SECRETARY OF STATE FOR EDUCATION 5. Teachers update SECRETARY OF STATE FOR HEALTH AND SOCIAL CARE 6. Government Responses to the CQC ‘Out of Sight - Who Cares?’ Report and to Baroness Hollins’ Interim Report on Seclusion and Restrictive Practice for People with a Learning Disability and Autistic People 7. Publication of the national strategy for autistic children, young people and adults: 2021 to 2026 8. Publication of the Government response to the Report of the Independent Medicines and Medical Devices Safety Review (IMMDS Review) 9. DHSC update SECRETARY OF STATE FOR THE HOME DEPARTMENT 10. Police Pay 11. NCA Remuneration Review Body Report 2021 SECRETARY OF STATE FOR HOUSING, COMMUNITIES AND LOCAL GOVERNMENT 12. Housing update 13. Local government update Notes: Texts of Written Statements are available from the Vote Office and on the internet at https://questions-statements.parliament.uk/. Broadcasts of proceedings can be found at https://www.parliamentlive.tv/Commons Some committee members and witnesses might now physically attend meetings, however, there is no public access at present. SELECT COMMITTEES - Science and Technology Committee Subject: The role of hydrogen in achieving Net Zero Witnesses: 9.30am: Guy Newey, Strategy and Performance Director, Energy Systems Catapult; Mark Neller, Director, Energy Business Leader UKIMEA, Arup 10.30am: Rt Hon Kwasi Kwarteng MP, Secretary of State, and Professor Paul Monks, Chief Scientific Adviser, Department for Business, Energy & Industrial Strategy; Sir Patrick Vallance, Government Chief Scientific Adviser Room 8 9.00am (private), 9.30am (public) - Work and Pensions Committee Subject: Protecting pension savers – five years on from the Pension Freedoms: Accessing pension savings Witnesses: 9.30am: Laura Myers, Partner and Head of DC, Lane, Clark & Peacock; Matthew Arends, Head of UK Retirement Policy, Aon; Rachel Vahey, Senior Technical Consultant, AJ Bell; Stephen Lowe, Group Communications Director, Just Group 10.30am: Peter Glancy, Head of Pensions Policy, Scottish Widows; Philip Brown, Director Policy and External Affairs at B&CE; Colin Clarke, Head of Product Policy Strategy for the Workplace DC Pensions business, Legal and General Room 5 9.15am (private), 9.30am (public) - Education Committee Room 15 9.30am (private) - Justice Committee Virtual meeting 9.30am (private) Transport Committee Subject: Work of the DVLA Witnesses: 9.30am: Mark Serwotka, General Secretary, and Sarah Evans, DVLA Branch Chair, Public and Commercial Services Union (PCS) 10.00am: Julie Lennard, Chief Executive Officer, Driver and Vehicle Licensing Agency; The Baroness Vere of Norbiton, Minister for Roads, Buses and Places, Department for Transport Room 16 9.30am (public) Home Affairs Committee Subject: The work of the Home Secretary Witnesses: 10.30am: Rt Hon Priti Patel MP, Home Secretary, Matthew Rycroft CBE, Permanent Secretary, and Paul Lincoln OBE, Director General, Border Force, Home Office The Grimond Room, Portcullis House 9.45am (private), 10.30am (public) Digital, Culture, Media and Sport Committee Virtual meeting 9.30am (private) Public Administration and Constitutional Affairs Committee The Wilson Room, Portcullis House 12.30pm (private) Environmental Audit Committee Subject: Mapping the path to net zero Witnesses: 2.30pm: Chris Stark, Chief Executive, Climate Change Committee; Tom Sasse, Associate Director, Institute for Government 3.30pm: Rachel Maclean MP, Parliamentary Under-Secretary of State, and Dr Bob Moran, Deputy Director, Environment Strategy, Department for Transport; Eddie Hughes MP, Parliamentary Under-Secretary of State, and Charlotte Baker, Director of Net Zero and Greener Buildings, Ministry of Housing, Communities and Local Government Virtual meeting 2.00pm (private), 2.30pm (public) Treasury Committee Subject: Office for Budget Responsibility Fiscal Risks Report Witnesses: 2.30pm: Richard Hughes, Chair, Andy King, Member of Budget Responsibility Committee, and Professor Sir Charles Bean, Member of Budget Responsibility Committee, Office for Budget Responsibility Virtual meeting 2.00pm (private), 2.30pm (public) Statutory Instruments Virtual meeting 3.40pm (private) Committee of Selection Virtual meeting 4.30pm (private) JOINT COMMITTEES Human Rights Virtual meeting 2.40pm (private) Statutory Instruments Virtual meeting As soon as convenient after 3.40pm (private) DELEGATED LEGISLATION COMMITTEES Sixth Delegated Legislation Committee To consider the draft Pensions Regulator (Employer Resources Test) Regulations 2021 Room 14 9.25am (public) Seventh Delegated Legislation Committee To consider the draft Environmental Authorisations (Scotland) Regulation 2018 (Consequential Modifications) Order 2021 Room 14 2.30pm (public) COMMITTEE REPORTS PUBLISHED TODAY INTERNATIONAL DEVELOPMENT - 1st Report: Assessing DFID’s results in nutrition Review HC 103 Time of publication: 00.01am TREASURY - 7th Report: Appointment of Sarah Breeden to the Financial Policy Committee HC 571 Time of publication: 00.01am - 8th Report: Appointment of Catherine Mann to the Monetary Policy Committee HC 572 Time of publication: 00.01am VIRTUAL PARTICIPATION IN PROCEEDINGS On 30 December 2020 the House agreed an order extending virtual participation to debate and certain other proceedings in accordance with a scheme prepared by the Speaker. On 16 June the House extended these provisions until 22 July. The scheme for virtual participation in proceedings and the revised Members’ Guide to Chamber proceedings are available on the intranet. The Members’ Guide to Westminster Hall proceedings is available on the intranet. PARTICIPATION IN WESTMINSTER HALL On 25 February the House agreed a motion to resume sittings in Westminster Hall on Monday 8 March. The Order agreed on 16 June extends provision for Members to participate virtually until the House rises for summer recess. These sittings continue to take place in the Boothroyd Room in Portcullis House. Applying to speak Applications to speak should be made via email to the Speaker’s Office (email@example.com). The deadlines are as follows: 1.00pm on the previous Friday for debates on a Monday; and 1.00pm on the previous day for debates on a Tuesday, Wednesday and Thursday. Call lists will be published at around 12.30pm on a Monday for debates that day and 6.30pm the previous day for debates on a Tuesday, Wednesday and Thursday. Participation Members are able to participate physically or virtually, but should note Mr Speaker’s request for Members to participate virtually in proceedings wherever possible. Members who wish to participate physically should note this when they contact the Speaker’s Office. For 60-minute and 90-minute debates, and debates scheduled by the Petitions Committee, only Members who are on the call list are permitted to attend. Members will be called in the order they appear on the call list, unless there are technical problems which prevent this. For 30-minute debates, there will not be a call list. Members wishing to contribute should follow existing conventions about contacting the Member in charge of the debate, the Speaker’s Office and the Minister. Arrangements for social distancing On the horseshoe in the Boothroyd Room, there are a maximum of 16 seats. Of these seats, there are 9 available from which backbenchers can speak (others are reserved for the frontbenchers, the PPS or Whip, the Member in charge, the Chair and the Clerk). There are 11 additional seats in the public gallery and a further 6 seats at the side table for Members in the latter part of the call list from which they can listen to debates and move onto the horseshoe when spaces become available. Members should clean the microphone and the desk around them before and after use. Cleaning materials will be available for this purpose. Guidance for Members on hybrid arrangements in Westminster Hall can be found on the intranet. FORTHCOMING END OF DAY ADJOURNMENT DEBATES - Monday 6 September to Monday 13 September (deadline 21 July 7.00pm or the rise of the House, whichever is earlier) Applications should be made in writing to the Table Office. Members must submit their application from their own email account. Application forms are available on the Table Office page on the Parliamentary intranet. The ballot will take place on Thursday 22 July. - Tuesday 14 September to Monday 20 September (deadline 8 September 7.00pm or the rise of the House, whichever is earlier) Applications should be made in writing to the Table Office. Members must submit their application from their own email account. Application forms are available on the Table Office page on the Parliamentary intranet. The ballot will take place on Thursday 9 September. FORTHCOMING WESTMINSTER HALL DEBATES - Tuesday 14 and Wednesday 15 September (deadline 6 September 10.00pm or the rise of the House, whichever is earlier) Details of Department’s answering in Westminster Hall will be announced as soon as possible. DEADLINE FOR CALL LISTS FOR SUBSTANTIVE PROCEEDINGS \| Deadline \| Date of proceeding \| Proceeding \| \|---------------------------\|--------------------\|------------------------------------------------------------------------------------------------------------------------------------------\| \| 1.00pm Wednesday 21 July \| Thursday 22 July \| Backbench business (1): motion relating to the Fifth Report of the Public Administration and Constitutional Affairs Committee, A Public Inquiry into the Government’s response to the Covid-19 pandemic \| \| 1.00pm Wednesday 21 July \| Thursday 22 July \| Backbench business (2): matters to be raised before the forthcoming adjournment \| Members wishing to speak in substantive proceedings should apply to the Speaker’s Office by emailing firstname.lastname@example.org from their parliamentary email address. SUMMER ADJOURNMENT - TABLING OF PARLIAMENTARY QUESTIONS The following arrangements will apply for the tabling of Questions during the Summer Adjournment. - Questions for oral answer Under Standing Order No. 22(6), the Speaker has made the following arrangements for tabling Questions for oral answer when the House returns: \| Last date of tabling* \| Date for answer \| Departments etc. \| \|-----------------------\|-----------------------\|---------------------------\| \| Tuesday 31 August \| Monday 6 September \| Education (T) \| \| Tuesday 31 August \| Tuesday 7 September \| Treasury (T) \| \| Tuesday 31 August \| Wednesday 8 September \| Scotland \| \| Tuesday 31 August \| Wednesday 8 September \| Prime Minister \| \| Tuesday 31 August \| Thursday 9 September \| Transport (T) \| \| Tuesday 7 September \| Monday 13 September \| Work and Pensions (T) \| \| Tuesday 7 September \| Wednesday 15 September\| Wales \| The results of the shuffles on 31 August will be published on 1 September. They will be available on the internet at http://www.publications.parliament.uk/pa/cm/cmfutoral/futoral.pdf For further details of last tabling days for other departments and answering bodies, see the Order of Questions rota available from the Vote Office and on the internet at http://www.parliament.uk/documents/commons-table-office/Oral-questions-rota.pdf Notes:** * The latest time for tabling is 12.30pm on each of these days. (T) indicates that there are also Topical questions for that Department. - Questions for written answer on a named day The latest time for tabling a Question for written answer on a named day before the Summer Adjournment is Monday 19 July (for answer on Thursday 22 July). \| Time of tabling \| Earliest date for named day answer \| \|------------------------------------------------------\|------------------------------------\| \| Monday 19 July \| Thursday 22 July \| \| Tuesday 20 July until Thursday 22 July \| Monday 6 September \| \| From 5.30pm on Thursday 22 July to 2.30pm on Friday 3 September \| Wednesday 8 September (Each Member may table five named day questions during this period.) \| \| Monday 6 September \| Thursday 9 September \| FURTHER INFORMATION MEMBERS’ GUIDE TO CHAMBER PROCEEDINGS The Members’ Guide to Chamber proceedings is available on the Parliamentary website. BUSINESS OF THE DAY Documents and reports relating to the business being held in the Chamber are available on the Commons Business Briefings webpage: www.parliament.uk/business/publications/research/commons-business-briefings/ WRITTEN STATEMENTS Text of today’s Written Statements: https://questions-statements.parliament.uk/ SELECT COMMITTEES Select Committees Webpage: https://committees.parliament.uk/ STANDING ORDERS RELATING TO PUBLIC BUSINESS Text of Standing Orders relating to public business: www.parliament.uk/business/publications.parliament.uk/pa/cm201919/cmstords/341/body.html EUROPEAN BUSINESS European Business Referrals and Motion documents for consideration by European Committees or on the Floor of the House are available on the European Business webpage: https://old.parliament.uk/business/publications/business-papers/commons/european-business11/ DIGITAL ENGAGEMENT Information about digital engagement opportunities for debates is available on the parliamentary website: https://www.parliament.uk/get-involved/have-your-say-on-laws/digital-engagement/. All business papers are available via the HousePapers app on mobile devices. FUTURE BUSINESS A. CALENDAR OF BUSINESS Business in either Chamber may be changed, and further business added, up to the rising of the House on the day before it is to be taken, and is therefore provisional. Government items of business in this section have nominally been set down for today, but are expected to be taken on the dates stated. B. REMAINING ORDERS AND NOTICES Business in this section has not yet been scheduled for a specific date. It has been nominally set down for today but is not expected to be taken today. A. CALENDAR OF BUSINESS Business in either Chamber may be changed, and further business added, up to the rising of the House on the day before it is to be taken, and is therefore provisional. THURSDAY 22 JULY CHAMBER QUESTIONS - 9.30am Questions to the Secretary of State for Environment, Food and Rural Affairs - 10.00am Topical Questions to the Secretary of State for Environment, Food and Rural Affairs - 10.10am Questions to Church Commissioners, House of Commons Commission, Parliamentary Works Sponsor Body, Public Accounts Commission and Speaker’s Committee on the Electoral Commission - Select Committee statement on the publication of the Third Report of the Business, Energy and Industrial Strategy Committee, Post-Pandemic Economic Growth: Levelling Up, HC 566 Darren Jones BACKBENCH BUSINESS - Fifth Report of the Public Administration and Constitutional Affairs Committee of Session 2019-21, A Public Inquiry into the Government’s response to the Covid-19 pandemic, HC 541; and the Government’s response, HC 995 Mr William Wragg Debate on a motion on the Fifth Report of the Public Administration and Constitutional Affairs Committee of Session 2019-21, A Public Inquiry into the Government’s response to the Covid-19 pandemic, HC 541; and the Government’s response, HC 995. Notes: The subject for this debate was determined by the Backbench Business Committee on the recommendation of the Liaison Committee. Relevant Documents: e-petition 302576, Hold a Public Inquiry into the handling of the Covid-19 crisis, and e-petition 577292, Launch a judge-led public inquiry into the Covid-19 pandemic Matters to be raised before the forthcoming adjournment Ian Mearns, Bob Blackman, on behalf of the Backbench Business Committee That this House has considered matters to be raised before the forthcoming adjournment. Notes: The subject for this debate was determined by the Backbench Business Committee ADJOURNMENT DEBATE Review of the Gambling Act 2005: Scott Benton WESTMINSTER HALL 1.30pm That this House has considered the priorities for the COP26 conference: Mr Simon Clarke, Darren Jones, Wera Hobhouse 3.15pm That this House has considered support for unpaid carers and Carers Week 2021: Ed Davey, Wera Hobhouse Relevant Documents: e-petition 579692, Increase Carers Allowance to equal 35hrs at National Minimum Wage, and e-petition 300032, Pay Carers an allowance equivalent to a fulltime job at the National Living wage Notes: The subjects for these debates were determined by the Backbench Business Committee. The sitting will be suspended and time added if divisions take place in the main Chamber (Standing Order No. 10(3)). MONDAY 6 SEPTEMBER CHAMBER QUESTIONS 2.30pm Questions to the Secretary of State for Education 3.15pm Topical Questions to the Secretary of State for Education Afterwards NATIONAL INSURANCE CONTRIBUTIONS BILL: REMAINING STAGES Not amended in Public Bill Committee, to be considered. TUESDAY 7 SEPTEMBER CHAMBER QUESTIONS - 11.30am Questions to the Chancellor of the Exchequer - 12.15pm Topical Questions to the Chancellor of the Exchequer Afterwards TRANSPORT (DISABLED PASSENGER CHARTER): TEN MINUTE RULE MOTION Charlotte Nichols That leave be given to bring in a Bill to establish a passenger charter for disabled land transport passengers setting out their rights, the legal obligations of transport operators, complaints procedures, passenger assistance schemes and accessibility requirements; and for connected purposes. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. WESTMINSTER HALL - 9.25am That this House has considered the future of the East Midlands economy: Ben Bradley Notes: The Chairman of Ways and Means appointed this debate on the recommendation of the Backbench Business Committee. - 11.00am That this House has considered impact of floods in Westminster North: Ms Karen Buck Notes: The sitting will be suspended from 11.30am to 2.30pm. - 2.30pm That this House has considered continued nuclear fuel manufacturing in the UK: Mark Menzies - 4.00pm That this House has considered the supply of affordable, good quality housing in the South West: David Warburton - 4.30pm That this House has considered Global Britain, human rights and climate change: Hywel Williams Notes: The second part of the sitting will be suspended and time added if divisions take place in the main Chamber (Standing Order No. 10(3)). The debate at 4.30pm will last for up to an hour. WEDNESDAY 8 SEPTEMBER CHAMBER QUESTIONS - 11.30am Questions to the Secretary of State for Scotland - 12 noon Questions to the Prime Minister Afterwards MULTI-ACADEMY TRUSTS (OFSTED INSPECTION): TEN MINUTE RULE MOTION Jonathan Gullis That leave be given to bring in a Bill to amend section 5 of the Education Act 2005 to provide that Ofsted may inspect the governing bodies of Multi-Academy Trusts. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. WESTMINSTER HALL - 9.30am That this House has considered the humanitarian situation in Tigray: Sarah Champion - 11.00am That this House has considered co-operative purchase of companies by employee groups at risk of redundancy: Christina Rees Notes: The sitting will be suspended from 11.30am to 2.30pm. - 2.30pm That this House has considered ending rough sleeping: Layla Moran - 4.00pm That this House has considered British Council closures: Wendy Chamberlain - 4.30pm That this House has considered reducing plastic waste: Elliot Colburn Notes: The second part of the sitting will be suspended and time added if divisions take place in the main Chamber (Standing Order No. 10(3)). The debate at 4.30pm will last for up to an hour. THURSDAY 9 SEPTEMBER CHAMBER QUESTIONS - 9.30am Questions to the Secretary of State for Transport - 10.15am Topical Questions to the Secretary of State for Transport FRIDAY 10 SEPTEMBER CHAMBER - Education (Careers Guidance in Schools) Bill: Second Reading Member in Charge: Mark Jenkinson Notes: Bill not yet printed. - Taxis and Private Hire Vehicles (Safeguarding and Road Safety) Bill: Second Reading Member in Charge: Peter Gibson Notes: Bill not yet printed. - Cultural Objects (Protection from Seizure) Bill: Second Reading Member in Charge: Mel Stride Notes: Bill not yet printed. - Miniature Mobile Phones (Prohibition of Sale) Bill: Second Reading Member in Charge: Sarah Atherton Notes: Bill not yet printed. - Covid-19 Vaccine Damage Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: ■ Employment Bill: Second Reading Member in Charge: Martin Docherty-Hughes Notes: Bill not yet printed. ■ Public Advocate Bill: Second Reading Member in Charge: Maria Eagle Notes: A money resolution is required for this to be proceeded with in Committee. ■ Wellbeing of Future Generations (No. 2) Bill: Second Reading Member in Charge: Simon Fell Notes: A money resolution is required for this Bill to be proceeded with in Committee. ■ Climate and Ecology Bill: Second Reading Member in Charge: Caroline Lucas Notes: Bill not yet printed. ■ Tips Bill: Second Reading Member in Charge: Dean Russell Notes: Bill not yet printed. ■ Hospitals (Parking Charges and Business Rates) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Voter Registration Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Asylum Seekers (Return to Safe Countries) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Motor Vehicles (Compulsory Insurance) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. Workers (Employment Security and Definition) Bill: Second Reading Member in Charge: Martin Docherty-Hughes Notes: Bill not yet printed. Workers (Rights) (No. 2) Bill: Second Reading Member in Charge: Anum Qaisar-Javed Notes: Bill not yet printed. Automated External Defibrillators (Public Access) Bill: Second Reading Member in Charge: Jim Shannon Notes: Bill not yet printed. Fire and Building Safety (Public Inquiry) Bill: Second Reading Member in Charge: Daisy Cooper Notes: Bill not yet printed. Israel Arms Trade (Prohibition) Bill: Second Reading Member in Charge: Richard Burgon Notes: A money resolution is required for this Bill to be proceeded with in Committee. Queen’s consent to be signified on Third Reading. TUESDAY 14 SEPTEMBER CHAMBER NON-DISCLOSURE AGREEMENTS: TEN MINUTE RULE MOTION Mrs Maria Miller That leave be given to bring in a Bill to restrict the use of non-disclosure agreements; and for connected purposes. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. WEDNESDAY 15 SEPTEMBER CHAMBER ABUSE OF PUBLIC-FACING WORKERS (OFFENCES): TEN MINUTE RULE MOTION Olivia Blake That leave be given to bring in a Bill to make provision for and in connection with offences relating to verbal and physical abuse of public-facing workers in the course of their employment. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. WEDNESDAY 22 SEPTEMBER CHAMBER TEN MINUTE RULE MOTION Henry Smith That leave be given to bring in a Bill under SO No. 23 [details to be provided]. Notes: The Member moving and a Member opposing this Motion may each speak for up to 10 minutes. FRIDAY 22 OCTOBER CHAMBER - Employment and Trade Union Rights (Dismissal and Re-engagement) Bill: Second Reading Member in Charge: Barry Gardiner Notes: Bill not yet printed. - Local Government (Disqualification) Bill: Second Reading Member in Charge: Sir Paul Beresford Notes: Bill not yet printed. - Corporate Homicide Bill: Second Reading Member in Charge: Chris Stephens ■ Education Employment (Accompaniment to Hearings) Bill: Second Reading Member in Charge: Brendan Clarke-Smith Notes: Bill not yet printed. ■ Energy Pricing Bill: Second Reading Member in Charge: Martyn Day Notes: Bill not yet printed. ■ Electric Vehicle Charging Points (New Buildings) Bill: Second Reading Member in Charge: Felicity Buchan Notes: Bill not yet printed. ■ General Election (Leaders’ Debates) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Homeless People (Current Accounts) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Electoral Candidates (Age) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Workers (Rights and Definition) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. ■ Immigration (Health and Social Care Staff) Bill: Second Reading Member in Charge: Christine Jardine Notes: Channel 4 (Privatisation) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. British Broadcasting Corporation (Privatisation) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Public Sector Exit Payments (Limitation) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Workers (Rights) Bill: Second Reading Member in Charge: Gavin Newlands Notes: Bill not yet printed. Planning and Local Representation Bill: Second Reading Member in Charge: Rachel Hopkins Notes: Bill not yet printed. FRIDAY 29 OCTOBER CHAMBER Menopause (Support and Services) Bill: Second Reading Member in Charge: Carolyn Harris Notes: Bill not yet printed. Childcare Bill: Second Reading Member in Charge: Matt Rodda Notes: - Animals (Penalty Notices) Bill: Second Reading Member in Charge: Andrew Rosindell Notes: Bill not yet printed. - Approved Premises (Substance Testing) Bill: Second Reading Member in Charge: Rob Butler Notes: Bill not yet printed. - Paternity (Leave and Pay) Bill: Second Reading Member in Charge: Gareth Davies Notes: Bill not yet printed. - Education (16 to 19 Academies) (Religious Character) Bill: Second Reading Member in Charge: Scott Benton Notes: Bill not yet printed. - Crown Estate (Devolution to Wales) Bill: Second Reading Member in Charge: Liz Saville Roberts Notes: Bill not yet printed. - BBC Licence Fee (Abolition) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. - Public Health (Control of Disease) Act 1984 (Amendment) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. - Caravan Site Licensing (Exemptions of Motor Homes) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: BBC Licence Fee Non-Payment (Decriminalisation for Over-75s) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Caravan Sites Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Green Belt (Protection) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Bereavement (Leave and Pay) Bill: Second Reading Member in Charge: Patricia Gibson Notes: Bill not yet printed. FRIDAY 19 NOVEMBER CHAMBER Marriage and Civil Partnership (Minimum Age) Bill: Second Reading Member in Charge: Mrs Pauline Latham in place of Secretary Savid Javid Notes: Bill not yet printed. Planning (Enforcement) Bill: Second Reading Member in Charge: Dr Ben Spencer Notes: Bill not yet printed. Glue Traps (Offences) Bill: Second Reading Member in Charge: Jane Stevenson Notes: Bill not yet printed. Domestic Building Works (Consumer Protection) Bill: Second Reading Member in Charge: Mark Garnier Pedicabs (London) Bill: Second Reading Member in Charge: Nickie Aiken Notes: Bill not yet printed. Leader of the House of Commons (Election) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. NHS England (Alternative Treatment) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Children’s Clothing (Value Added Tax) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Flexible Working Bill: Second Reading Member in Charge: Tulip Siddiq Notes: Bill not yet printed. Children (Access to Treatment) Bill: Second Reading Member in Charge: Bambos Charalambous Notes: Bill not yet printed. FRIDAY 26 NOVEMBER CHAMBER Down Syndrome Bill: Second Reading Member in Charge: Dr Liam Fox Notes: - Pension Schemes (Conversion of Guaranteed Minimum Pensions) Bill: Second Reading Member in Charge: Margaret Ferrier Notes: Bill not yet printed. - Registers of Births and Deaths Bill: Second Reading Member in Charge: Saqib Bhatti Notes: Bill not yet printed. - Public Bodies (Representation from Devolved Nations) Bill: Second Reading Member in Charge: Wendy Chamberlain Notes: Bill not yet printed. - Electoral Commission (Abolition) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. - Illegal Immigration (Offences) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. - Full Employment Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. - Health and Safety at Work Bill: Second Reading Member in Charge: Chris Stephens Notes: FRIDAY 3 DECEMBER CHAMBER - Copyright (Rights and Remuneration of Musicians, Etc.) Bill: Second Reading Member in Charge: Kevin Brennan [R] Notes: Bill not yet printed. - Acquired Brain Injury Bill: Second Reading Member in Charge: Chris Bryant Notes: Bill not yet printed. - Asylum Seekers (Permission to Work) Bill: Second Reading Member in Charge: Carol Monaghan Notes: Bill not yet printed. - Non-Disclosure Agreements Bill: Second Reading Member in Charge: Mrs Maria Miller Notes: Bill not yet printed. - Miscarriage Leave Bill: Second Reading Member in Charge: Angela Crawley Notes: Bill not yet printed. - Asylum Seekers (Accommodation Eviction Procedures) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. - Disability Benefit Assessments (Recording) Bill: Second Reading** Member in Charge: Chris Stephens Notes: Intimate Images (Offences) Bill: Second Reading Member in Charge: Mrs Maria Miller Notes: Bill not yet printed. Pregnancy and Maternity (Redundancy Protection) Bill: Second Reading Member in Charge: Mrs Maria Miller Notes: Bill not yet printed. Workers (Rights and Definition) (No. 2) Bill: Second Reading Member in Charge: Angela Crawley Notes: Bill not yet printed. FRIDAY 10 DECEMBER CHAMBER Medical Cannabis (Access) Bill: Second Reading Member in Charge: Jeff Smith Notes: Bill not yet printed. Climate Change Bill: Second Reading Member in Charge: Colum Eastwood Notes: Bill not yet printed. British Goods (Public Sector Purchasing Duty) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Clean Air Targets (World Health Organization Guidelines) Bill: Second Reading Member in Charge: Christine Jardine Notes: Consumer Pricing Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Broadcasting (Listed Sporting Events) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Puppy Import (Prohibition) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Employment (Application Requirements) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Public Sector Website Impersonation Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Hunting Trophy Import (Prohibition) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. Armenian Genocide (Recognition) Bill: Second Reading Member in Charge: John Spellar Notes: Bill not yet printed. House of Lords (Hereditary Peers) (Abolition of By-Elections) (No. 2) Bill: Second Reading Member in Charge: John Spellar Notes: Benefit Sanctions (Warnings) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. Universal Credit Sanctions (Zero Hours Contracts) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. FRIDAY 14 JANUARY 2022 CHAMBER Taxis and Private Hire Vehicles (Disabled Persons) Bill: Second Reading Member in Charge: Jeremy Wright Notes: Bill not yet printed. Prime Minister (Temporary Replacement) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. Kinship Care (Parental Leave) Bill: Second Reading Member in Charge: Katherine Fletcher Notes: Bill not yet printed. Regulatory Impact Assessments Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Parliamentary and Health Service Ombudsman (Powers) Bill: Second Reading Member in Charge: Chris Stephens Notes: Under-Occupancy Penalty (Report) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. Social Security Benefits (Healthy Eating) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. Planning (Proper Maintenance of Land) Bill: Second Reading Member in Charge: Jonathan Gullis Notes: Bill not yet printed. FRIDAY 21 JANUARY 2022 CHAMBER Hare Coursing Bill: Second Reading Member in Charge: Richard Fuller Notes: Bill not yet printed. Human Trafficking (Child Protection) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. Barnett Formula (Replacement) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Housing Standards (Refugees and Asylum Seekers) Bill: Second Reading Member in Charge: Chris Stephens Notes: Asylum Seekers (Permission to Work) (No. 2) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. FRIDAY 28 JANUARY 2022 CHAMBER British Sign Language Bill: Second Reading Member in Charge: Rosie Cooper Notes: Bill not yet printed. Business of the House Commission Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. Mobile Homes Act 1983 (Amendment) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. Evictions (Universal Credit) Bill: Second Reading Member in Charge: Chris Stephens Notes: Bill not yet printed. FRIDAY 4 FEBRUARY 2022 CHAMBER Unsolicited Explicit Images and Deepfake Pornography Bill: Second Reading Member in Charge: Angela Richardson Notes: ■ Employment (Caring Leave) Bill: Second Reading Member in Charge: Jack Brereton Notes: Bill not yet printed. ■ Prime Minister (Accountability to House of Commons) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ NHS (Prohibition of Data Transfer) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. ■ National Health Service Co-Funding and Co-Payment Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. ■ Plastic Pollution Bill: Second Reading Member in Charge: Mr Alistair Carmichael Notes: Bill not yet printed. FRIDAY 25 FEBRUARY 2022 CHAMBER ■ Local Authority Boundaries (Referendums) Bill: Second Reading Member in Charge: Robbie Moore Notes: Bill not yet printed. ■ City of Bradford (Referendum on Shipley and Keighley) Bill: Second Reading Member in Charge: Philip Davies Notes: Human Trafficking (Sentencing) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. Rule of Law (Enforcement by Public Authorities) Bill: Second Reading Member in Charge: Sir Christopher Chope Notes: Bill not yet printed. FRIDAY 18 MARCH 2022 CHAMBER Mental Health Provision (Children and Young People) Bill: Second Reading Member in Charge: Munira Wilson Notes: Bill not yet printed. Goods Delivery Services Bill: Second Reading Member in Charge: Jamie Stone Notes: Bill not yet printed. Commercial Rent (Prohibition of Upward-Only Reviews) Bill: Second Reading Member in Charge: Sarah Olney Notes: Bill not yet printed. Conversion Therapy (Prohibition) Bill: Second Reading Member in Charge: Wera Hobhouse Notes: Bill not yet printed. Palestine Statehood (Recognition) Bill: Second Reading Member in Charge: Layla Moran Notes: ■ North Northamptonshire (Urgent Care Facilities) Bill: Second Reading Member in Charge: Mr Peter Bone Notes: Bill not yet printed. ■ Dogs (DNA Databases) Bill: Second Reading Member in Charge: Andrew Griffith Notes: Bill not yet printed. ■ Tibet and Xinjiang (Reciprocal Access) Bill: Second Reading Member in Charge: Tim Loughton B. REMAINING ORDERS AND NOTICES Business in this section has not yet been scheduled for a specific date. It has therefore been set down formally to be taken in the Chamber today but is not expected to be taken today. 1. EXITING THE EUROPEAN UNION (TRADE) Secretary Elizabeth Truss That the draft Cat and Dog Fur (Control of Movement etc.) (EU Exit) Regulations 2021, which were laid before this House on 12 January, in the last Session of Parliament, be approved. Notes: The Joint Committee on Statutory Instruments has drawn the special attention of both Houses to the Instrument in its Forty-Third report of 2019-21 (HC 75-xliii). 2. EXITING THE EUROPEAN UNION (CUSTOMS) Secretary George Eustice That the draft International Waste Shipments (Amendment) (EU Exit) Regulations 2021, which were laid before this House on 17 December, in the last Session of Parliament, be approved. 3. EXITING THE EUROPEAN UNION (CULTURAL OBJECTS) Caroline Dinenage That the draft Introduction and the Import of Cultural Goods (Revocation) Regulations 2021, which were laid before this House on 19 May, be approved. 4. ANIMAL WELFARE (KEPT ANIMALS) BILL: SECOND READING Notes: Queen’s consent to be signified on Third Reading. 5. ANIMAL WELFARE (KEPT ANIMALS) BILL: MONEY Jesse Norman That, for the purposes of any Act resulting from the Animal Welfare (Kept Animals) Bill, it is expedient to authorise: (1) the payment out of money provided by Parliament of any increase in expenditure in the sums payable under other Acts out of money so provided, where that increase is attributable to: (a) any provision of the Act relating to primates; (b) any power in the Act to apply such provision to other animals; and (2) the payment of sums into the Consolidated Fund. Notes: Queen’s Recommendation signified. 6. LICENCES AND LICENSING Secretary Priti Patel That the draft Alcohol Licensing (Coronavirus) (Regulatory Easements) (Amendment) Regulations 2021, which were laid before this House on 8 June, be approved. 7. CONSTITUTIONAL LAW Secretary Alister Jack That the draft Environmental Authorisations (Scotland) Regulation 2018 (Consequential Modifications) Order 2021, which was laid before this House on 8 June, be approved. 8. COMPENSATION (LONDON CAPITAL & FINANCE PLC AND FRAUD COMPENSATION FUND) BILL: REMAINING STAGES Not amended in Public Bill Committee, to be considered. Notes: The Northern Ireland Assembly has approved a Legislative Consent Resolution relating to this Bill. Copies of the Resolution are available in the Vote Office (also available on the documents webpage for the Bill). 9. PENSIONS Guy Opperman That the draft Occupational Pension Schemes (Administration, Investment, Charges and Governance) (Amendment) Regulations 2021, which were laid before this House on 21 June, be approved. 10. INSOLVENCY Secretary Kwasi Kwarteng That the Corporate Insolvency and Governance Act 2020 (Coronavirus) (Extension of the Relevant Period) (No. 2) Regulations 2021 (SI 2021, No. 718), a copy of which was laid before this House on 21 June, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 11. PENSIONS Guy Opperman That the draft Pensions Regulator (Employer Resources Test) Regulations 2021, which were laid before this House on 28 June, be approved. 12. SUBSIDY CONTROL BILL: SECOND READING Notes: Queen’s consent to be signified on Third Reading. 13. ENERGY CONSERVATION Secretary Kwasi Kwarteng That the draft Ecodesign for Energy-Related Products and Energy Information (Lighting Products) Regulations 2021, which were laid before this House on 1 July, be approved. 14. ENERGY CONSERVATION Secretary Kwasi Kwarteng That the draft Ecodesign for Energy-Related Products and Energy Information (Amendment) Regulations 2021, which were laid before this House on 5 July, be approved. 15. ELECTIONS BILL: SECOND READING 16. ELECTIONS BILL: MONEY Jesse Norman That, for the purposes of any Act resulting from the Elections Bill, it is expedient to authorise— (1) the payment out of money provided by Parliament of— (a) any expenditure incurred by a Minister of the Crown under or by virtue of the Act, and (b) any increase attributable to the Act in the sums payable under any other Act out of money so provided, (2) the payment out of the Consolidated Fund of any increase attributable to the Act in the sums payable under any other Act out of that Fund, and (3) the payment of sums into the Consolidated Fund. Notes: Queen’s Recommendation signified. 17. MARINE POLLUTION Secretary Grant Shapps That the draft Merchant Shipping (Prevention of Air Pollution from Ships) (Amendment) Regulations 2021, which were laid before this House on 6 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 18. AGRICULTURE Secretary George Eustice That the draft Organics (Equivalence and Control Bodies Listing) (Amendment) Regulations 2021, which were laid before this House on 6 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 19. NORTHERN IRELAND (MINISTERS, ELECTIONS AND PETITIONS OF CONCERN) BILL: REMAINING STAGES As amended in Public Bill Committee, to be considered. 20. DISSOLUTION AND CALLING OF PARLIAMENT BILL: COMMITTEE Notes: Remaining stages may also be taken (Order, 6 July). Queen’s consent to be signified on Third Reading. 21. INTERNATIONAL IMMUNITIES AND PRIVILEGES Nigel Adams That the draft Conference of the Parties to the United Nations Framework Convention on Climate Change (Immunities and Privileges) Order 2021, which was laid before this House on 8 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 22. RATING (CORONAVIRUS) AND DIRECTORS DISQUALIFICATION (DISSOLVED COMPANIES) BILL: REMAINING STAGES Not amended in Public Bill Committee, to be considered. Notes: The Northern Ireland Assembly has approved a Legislative Consent Resolution relating to this Bill. Copies of the Resolution are available in the Vote Office (also available on the [documents webpage](#) for the Bill). 23. FINANCIAL SERVICES AND MARKETS John Glen That the draft Capital Requirements Regulation (Amendment) Regulations 2021, which were laid before this House on 12 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 24. FINANCIAL SERVICES Jesse Norman That the Money Laundering and Terrorist Financing (Amendment) (No. 2) (High-Risk Countries) Regulations 2021 (SI, 2021, No. 827), dated 12 July 2021, a copy of which was laid before this House on 12 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 25. PUBLIC HEALTH Secretary Sajid Javid That the Health Protection (Coronavirus, Restrictions) (Self-Isolation) (England) (Amendment) Regulations 2021 (SI 2021, No. 851), dated 15 July 2021, a copy of which was laid before this House on 15 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 26. PUBLIC HEALTH Secretary Sajid Javid That the Health Protection (Coronavirus, Restrictions) (Steps etc.) (England) (Revocation and Amendment) Regulations 2021 (SI 2021, No. 848), dated 14 July 2021, a copy of which was laid before this House on 15 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 27. PUBLIC HEALTH Jo Churchill That the Health Protection (Coronavirus, Restrictions) (Self-Isolation) (England) (Amendment) (No. 2) Regulations 2021 (SI 2021, No. 864), dated 17 July 2021, a copy of which was laid before this House on 19 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 28. CIVIL AVIATION Secretary Grant Shapps That the draft Airports Slot Allocation (Alleviation of Usage Requirements) (No. 2) Regulations 2021, which were laid before this House on 19 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 29. ELECTRONIC INFORMATION Secretary Kwasi Kwarteng That the draft Internal Markets Information System Regulation (Amendment etc.) Regulations 2021, which were laid before this House on 20 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 30. WATER INDUSTRY Secretary George Eustice That the draft Water and Sewerage Undertakers (Exit from Non-household Retail Market) (Consequential Provision) Regulations 2021, which were laid before this House on 20 July, be approved. Notes: The instrument has not yet been considered by the Joint Committee on Statutory Instruments. 31. INDEPENDENT EXPERT PANEL RECOMMENDATIONS FOR SANCTIONS AND THE RECALL OF MPS ACT 2015 Mr Jacob Rees-Mogg That— (1) the following Standing Order be made: “IEP recommendations for sanctions and the Recall of MPs Act 2015 (1) The Chair of the IEP shall send to the Chair and Members of the Committee on Standards and to the Clerk of that Committee any report from a sub-panel of the IEP which he has referred to the Clerk of the House under subparagraph (5)(d) of Standing Order No. 150A (Independent Expert Panel) and which contains a determination for a sanction that would, if made by the Committee on Standards, engage the provisions of the Recall of MPs Act 2015. (2) Where a report has been sent to the Committee on Standards in accordance with paragraph (1) of this Order the Committee of Standards shall make a report to the House in relation to the Member named in that report, setting out a recommendation for a suspension equal to that recommended by the sub-panel to run concurrently with any sanction imposed as a result of the sub-panel’s determination. (3) Reports under paragraph (2) must be made no later than on the third sitting day after the report of the IEP sub-panel is sent to members of the Committee on Standards, save that the day on which the report is sent shall not be counted in calculating this period. (4) If the Committee on Standards is unable to meet within 3 sitting days, the Chair shall, if satisfied that the report from the IEP sub-panel has been sent to all members of the Committee, make the report to the House from the Committee required under paragraph (2).” (2) The following amendments to Standing Orders be made: (a) In Standing Order No. 149 (Committee on Standards), paragraph (1), at the end insert “(c) in accordance with Standing Order No. (IEP recommendations for sanctions and the Recall of MPs Act 2015) to report to the House recommendations for sanctions to run concurrently with sanctions determined by a sub-panel of the IEP and implemented by the House.” (b) In Standing Order No. 150D (Motions consequent on the ICGS), paragraph (1), at the end, insert “( ) a motion under paragraph (1) of this Order includes a motion to implement a sanction recommended by the Committee on Standards under Standing Order No. (IEP recommendations for sanctions and the Recall of MPs Act 2015), or a motion to implement both such a sanction and a sanction determined by a sub-panel of the IEP”.
Effects of stress on parental care are sexually dimorphic in prairie voles Karen L. Bales\textsuperscript{a,b,}, Kristin M. Kramer\textsuperscript{a,c}, Antonia D. Lewis-Reese\textsuperscript{a}, C. Sue Carter\textsuperscript{a} \textsuperscript{a} Department of Psychiatry and Brain-Body Center, University of Illinois at Chicago, Chicago IL 60612, United States \textsuperscript{b} Department of Psychology, One Shields Ave, University of California, Davis, CA 95616, United States \textsuperscript{c} Department of Biology, University of Memphis, Memphis TN 38152, United States Received 6 May 2005; received in revised form 30 July 2005; accepted 7 November 2005 Abstract The effects of stress on parental care are poorly understood, especially in biparental species where males also display care. Data from previous studies in prairie voles, as well as parallels with pair-bonding behavior, suggest the hypothesis that a stressful experience might facilitate parental care in males but not in females. In the present study, male and female prairie voles were exposed to either a 3-min swim stressor or no stressor; 45 min later each animal was tested in a parental care paradigm. Following the parental care test, blood samples were collected and assayed for corticosterone (CORT). After the stressor males, but not females, showed significant changes in parental behavior including significantly more time in kyphosis (arched-back huddling), and a tendency to spend more time licking and grooming pups. In males, CORT levels measured following the parental care test were inversely related to licking and grooming but positively correlated with retrievals. These findings support earlier studies suggesting that the neuroendocrine substrates of parental behavior, as well as the effects of stressors, are sexually dimorphic in this species. © 2005 Elsevier Inc. All rights reserved. Keywords: Stress; Parental care; Corticosterone; Sex differences The effects of stress and hormones of the hypothalamic-pituitary-adrenal (HPA) axis on parental behavior are not well understood. For example, injection of corticotropin-releasing factor (CRF) reduced maternal aggression, although not other maternal behaviors, in mice [1]. In contrast, lactating female rats exposed to a noise stress showed an increase in both locomotor activity and licking and grooming of the pups. In this same study, virgin females also showed a stress-related increase in activity, including rearing and locomotion around the home cage [2]. It also has been reported that a tail pinch can facilitate maternal behavior in virgin female rats [3]. Human mothers with higher postpartum cortisol were more attracted to, and better able to discriminate, their own infant’s odors [4]; however in captive gorillas [5] and baboons [6] high levels of postpartum cortisol were associated with lower levels of maternal behavior. Thus in female mammals, activation of the HPA axis can either disrupt or facilitate maternal behavior, depending on the reproductive status of the female and the measure of maternal behavior used. The effects of stressful experiences on male parental care have not to our knowledge been previously examined. Male parental care is most commonly observed in socially monogamous species, which make up a small minority (3%) of mammals [7]. The physiological conditions underlying paternal care have only recently become the subject of analysis. Cortisol in parental males varies across the mate’s pregnancy [8–11]. In male black tufted-ear marmosets, males with higher CORT carried infants less [12]. The neuropeptides oxytocin (OT) and arginine vasopressin (AVP), hormones associated with stress in rats and mice [13,14] also have been linked to male paternal care [15–18]. Prairie voles are socially monogamous rodents that typically show spontaneous male parental behavior [19,20]. This species also forms pair bonds, indexed by partner preference behavior. In prairie voles the effects of stress are sexually dimorphic, with stress facilitating the formation of a partner preference in males and inhibiting the development of a partner preference in females [21–23]. In a previous study, we observed that males that had undergone a surgical stressor spent more time in kyphosis (arched-back huddling over pups) than males that had not undergone surgery [18]. Kyphosis is a major component of parental behavior [24] and we therefore hypothesized that administration of a mild stressor (3 min of forced swimming) would facilitate parental care in males. In contrast, because stress can be disruptive to certain aspects of female social behavior, we hypothesized that a prior stress would disrupt parental care in females. The forced swim test has also been used to measure “depressive-like behavior” in voles and other rodents [25]. We also examined plasma levels of the “stress hormone” corticosterone (CORT) for its relationship to parental behavior. To avoid disrupting behavior, blood samples were taken after the parental care test. We have recently observed that exposure to a pup is associated with a rapid (15 min) reduction in CORT in males, but not in females [26]. For this reason we hypothesized that CORT might be related to behavior, either causally or as a consequence of the behavioral experience of the animal. Finally, we also examined the behaviors displayed during the swim test/stressor (swimming, struggling, and floating) for correlations with parental behavior or CORT. 1. Methods Subjects were laboratory-bred male and female prairie voles (Microtus ochrogaster), descendants of a wild stock originally caught near Champaign, Illinois. Prairie voles were maintained on a 14:10 light:dark cycle and allowed food (Purina high-fiber rabbit chow) and water ad libitum. Breeding pairs were maintained in large polycarbonate cages (44 cm long × 22 cm wide × 16 cm high) and provided with cotton for nesting material. At 21 days of age, voles were weaned and housed in same-sex sibling pairs in standard mouse cages (27 cm long × 16 cm wide × 13 cm high). The same-sex sibling pairs were then kept in a single-sex colony room. All test subjects were sexually naive, had never been exposed to pups, and had not been used in any previous experiments. Females do not experience a spontaneous estrus cycle [27], and therefore stage of estrus is not an issue in this species. At approximately 60 days of age, voles underwent a parental care test, preceded either by a swim stressor or by no stress. The swim stressor consisted of 3 min of swimming in lukewarm water in a large plastic cage (20 × 25 × 45 cm). At this depth the voles can neither touch the bottom nor climb out. The parameters of the response to this stressor in voles have been previously characterized [21,23,28]. Normal animals show approximately a doubling of corticosterone (CORT) with a maximum increase within approximately 45 min following swimming [23], and levels remain elevated for several hours. In a previous study performed by the first author on this paper, control animals showed an average of an 82% rise in CORT following a swim stressor. Swim tests were scored for three behaviors: swim (vole is quietly swimming); struggle (vole is violently swimming against the side and trying to get out of the cage) and float/immobility (vole is not paddling at all). Scoring was performed using Behavior Tracker 1.0 (www.behaviortracker.com). Immediately following the swim stressor voles were placed into an alloparental care testing arena. Unstressed voles were placed directly into the arena. Procedures for testing parental behavior were modeled on previous work from this lab [29,30]. Voles were given 45 min following introduction to acclimate to the empty testing arena; this time period also allowed maximal changes in CORT to occur. The arena consisted of two standard mouse cages connected by a short, clear tube. At the end of the acclimation period, the subject was removed from the testing arena and an unrelated 1–3 day old pup placed in the front cage. Each pup was used as a stimulus no more than twice. The subject then was placed into the arena, and pup-directed behavior was recorded for 10 min on videotape. If the test subject displayed aggression towards the pup, the test was stopped immediately, and the pup was removed and returned to its parents. The incidence of pup-directed aggression was higher in females, as has been previously shown [29,31], but was not affected by stress. In most cases, the pup was not injured, and this protocol resulted in a full recovery for the pup. Tapes were scored by an experimentally blind observer using Behavior Tracker 1.0 software. Subjects were initially categorized as either parental or nonparental. A subject was said to be parental if it displayed kyphosis over or retrieved an pup (without later attacking it; in about 50% of attacks, an extremely short retrieval was a prelude to immediate attack [29]). The term “parental” therefore only refers to those animals that displayed full parental behavior (i.e. in this characterization, those that just investigated the infant were not categorized as parental). Individual components of behavior were then further analyzed, including location in the testing arena (cage with pup, far cage, or the connecting tube), huddling or “kyphosis” which consisted of sitting over the pup with an arched-back posture [32]; licking/grooming; sniffing; non-kyphotic contact (sitting flat on a pup or next to it, while touching, but without an arched back); and retrievals. These parental behaviors as scored were mutually exclusive. Following the alloparental care test, voles were anesthetized using a combination of Ketamine (67 mg/kg) and Xylazine (13 mg/kg) and a blood sample taken via a heparinized capillary tube from the supraorbital sinus. Blood samples were centrifuged and serum stored at −20 °C until assay. One week later, voles underwent an alloparental care test preceded by the opposite condition (i.e. if they were stressed in week 1, they were not stressed in week 2). Following the second test, another blood sample was collected. 1.1. Corticosterone assay Corticosterone was assayed using a commercially available radioimmunoassay (MP Biomedicals, Irvine, CA) previously validated for the prairie vole [28]. Samples were assayed following a dilution of 1:2000. Intra-assay CV’s averaged 2.19%. No interassay CV is applicable because all samples were assayed at once. 1.2. Data analysis Exposure to an pup has been shown to affect behavior during a subsequent encounter [33]. In the present data set, not only were males more likely to display huddling behavior in the second test, but the effect was sexually dimorphic, with females not showing this order effect (see Results section). Therefore, for the analysis of stress effects on parental care only the first parental care test was used. Logistic regression calculated using a likelihood ratio $\chi^2$ was used to predict parental behavior based on whether or not the test animal was stressed. Individual components of parental behavior were analyzed according to treatment by non-parametric Kruskal–Wallis tests [34] because residuals in many cases were non-transformable to normality. Correlations between swim test, parental care test, and hormonal variables were performed using Spearman correlations on the entire dataset (both parental care tests for each animal) so that each animal would have a swim test represented in the dataset. All statistical tests were performed in SAS 8.2 (SAS Institute, Cary, NC). Tests for effects of treatment on parental behavior were one-tailed due to a priori predictions of direction, while correlations with swim test and hormonal variables were two-tailed. Values of $p < 0.05$ were considered significant. ### Table 1 \| Variable \| Non-stressed \| Stressed \| \|---------------------------\|--------------\|------------\| \| Kyphosis \| 6.5 ± 5.73 (s) \| 8.65 ± 5.25 \| \| Non-kyphotic contact \| 73.0 ± 25.83 (s) \| 46.45 ± 15.93 \| \| Licking and grooming \| 102.15 ± 32.01 (s) \| 92.55 ± 32.60 \| \| Sniffing \| 25.95 ± 4.51 (s) \| 24.0 ± 5.15 \| \| Retrievals \| 4.0 ± 0.82 \| 3.95 ± 1.26 \| ### 2. Results #### 2.1. Effects of testing order During the second parental care test, males spent more time engaged in kyphosis ($n = 20$, Wilcoxon signed-rank (paired) test, $S = 19.5$, $p = 0.019$; two-tailed) than during the first test, while females did not show an order effect ($n = 20$, Wilcoxon signed-rank test, $S = 2.5$, $p = 0.625$). Data for females are presented in Table 1. There were no test order differences in males or females in non-kyphotic contact, licking and grooming, retrievals, or CORT values. #### 2.2. Effects of stress Males were significantly more likely to display parental behavior following a stressor (Fig. 1; $n = 20$, logistic regression, likelihood ratio $\chi^2 = 3.4522$, $p = 0.032$; one-tailed). However, stress did not significantly affect the proportion of females that displayed parental care towards the pup ($n = 20$, logistic regression, likelihood ratio $\chi^2 = 0.2024$, $p = 0.326$; one-tailed). Following stress, males spent a significantly longer duration engaged in kyphosis (Fig. 2; $n = 20$, Kruskal–Wallis test, $\chi^2 = 3.327$, $p = 0.034$; one-tailed) and tended to spend more time in licking and grooming ($n = 20$, $\chi^2 = 2.069$, $p = 0.075$; one-tailed). Stressed versus nonstressed males did not differ in non-kyphotic contact ($n = 20$, $\chi^2 = 0.824$, $p = 0.182$; one-tailed) or retrievals ($n = 20$, $\chi^2 = 0.054$, $p = 0.408$; one-tailed). Females showed no significant differences in any of these behaviors following stress. Following the parental care test, CORT levels were no longer elevated in the stressed groups compared to the non-stressed groups (see Discussion). Female values were 1501.73 ± 133.35 ng/ml for the unstressed group and 1736.82 ± 211.33 ng/ml for the stressed group, while male values were 1030.83 202 ± 61.10 ng/ml for the unstressed group and 1111.17 ± 60.94 ng/ml for the stressed group. ![Fig. 1. Proportion of animals displaying parental behavior after a swim stressor or no swim stressor ($n = 10$ animals/group). Males were more likely to display parental behavior following stress (logistic regression, likelihood ratio $\chi^2 = 3.4522$, $p = 0.032$; one-tailed), while females were not.](image) ![Fig. 2. Time spent (seconds) by males in parental care behaviors. $p < 0.05$, $p < 0.1$.](image) 2.3. Relationships between CORT and parental behavior In males (all tests combined), CORT levels were negatively correlated with time spent licking and grooming the pup (Spearman correlation, $n=40$, $r=-0.354$, $p=0.025$, two-tailed; Fig. 3a) and tended to be positively correlated with the number of retrievals ($r=0.288$, $p=0.071$, two-tailed, Fig. 3b). A median split by CORT level (Fig. 4) also showed differences in male behavior by CORT. Males with high CORT (above the median) tended to lick and groom pups less ($n=40$, Kruskal–Wallis test, $\chi^2=3.142$, $p=0.076$, two-tailed) but retrieved pups more ($n=40$, Kruskal–Wallis test, $\chi^2=7.357$, $p=0.007$, two-tailed). In females, no measures of parental care were significantly correlated with CORT. 2.4. Relationships between swim test behavior and other variables When all tests were combined, time spent struggling during the swim test was positively correlated with time spent in kyphosis in the parental care test (Spearman correlation, $n=40$, $r=0.326$, $p=0.04$; Fig. 5a), while struggling tended to be negatively correlated with retrievals ($n=40$, $r=-0.265$, $p=0.098$; Fig. 5b). Females tended to float more than males (Kruskal–Wallis test, $n=40$, $\chi^2=3.40$, $p=0.065$). CORT was not correlated with any swim test variables. 3. Discussion In prairie voles, a biparental and relatively monomorphic species, stress affects parental care in reproductively naïve animals in a sexually dimorphic manner. As with pair bond formation, males became more parental following a stressor. In particular, in males the behavior of kyphosis (huddling) was significantly increased following stress. It is also notable that male kyphosis increased during a second parental care test (also potentially a stressor). In females neither a prior stressor nor repeated testing significantly influenced the overall tendency to show parental behavior or kyphosis. Although the swim stressor is a relatively mild stress and the subsequent changes in behavior were not large, we believe that these reflect important sex differences and an area for future research. It is possible that the effects of stress on parental care in males are partially mediated by the type of stress; i.e., a swim stressor produces a group of wet animals. In order to dry off, voles subjected to a swim stressor spend significant time auto-grooming, which may induce hormonal changes potentially including changes in both oxytocin [35–37] and dopamine [38] systems. These same hormonal systems have been implicated in parental behavior in both males and females [18,39–41]. However, as many types of stress may induce a rise in auto-grooming, this does not rule out a general effect of stress on parental care that is not dependent on becoming wet. The relationship between the hormonal measure (CORT) and parental behavior in this study is complicated because this hormone might not only be affected by stress but also by exposure to an pup [26]. Post-hoc measurements, which did not differ between stressed and non-stressed groups, therefore may be reflective of the effects of pup exposure or the parental behavior in which animals engaged [26]. In fact, it appears quite likely that males used the pup to regulate their stress response. Although swim stress reliably increases CORT, and males were more parental following swim stress, male licking and grooming of pups was associated with a lower post-testing level of CORT. We hypothesize that stressed males are more motivated to interact with pups, and therefore are actively engaged in lowering their own stress response and reducing the difference in CORT between stressed and non-stressed groups. It is also possible that the 45-min exposure to the testing arena, which was chosen in order to get the maximal response of CORT resulting from the swim test, partially masked the differences between groups due to the stress of exposure to a novel environment. Male retrieval of pups, as opposed to licking and grooming, was associated with higher post-testing CORT. These behaviors are also distinguished by the fact that while licking and grooming tended to be affected by prior stress, retrieval behavior was not. In general, the findings of this study are consistent with the hypothesis that male prairie voles may use social contact to modulate stressful experiences; in addition, exposure to a stressor may have different behavioral consequences in males versus females [21,26,42]. It is possible that retrieval, while beneficial to the pup, does not carry the same anti-stress benefit for the adult as a behavior which is less active and more fully engages the olfactory system. It is quite probable that the effects of stress on parental care, and of pup exposure on CORT, are both mediated by neuropeptides such as oxytocin (OT) and vasopressin (AVP). Both OT and AVP may be released by stress, but the response of OT and AVP to stress appears to vary by species and by stressor [43]. For example, OT is released by stress in rats [43], but not in humans following exercise stress [44]. In contrast, AVP in rats is not released by a swim stressor [14]. In female rats, kyphosis as well as lordosis (a similar arched-back posture) is dependent on OT acting in the periaqueductal gray of the midbrain [45–49]. Both OT and AVP are also involved in male parental care in voles [18,50,51]. Further examination of the effects of stress on neuropeptides in voles, as well as differential effects of neuropeptides on components of parental behavior (kyphosis and licking/grooming vs. more active behaviors such as retrievals), seems indicated. Behavior during the swim test has not been previously examined in this species. Swim test variables did not correlate with many of the other variables measured; however struggling in the swim test did correlate positively with kyphosis in the parental care test, and negatively with retrievals. This once again suggests possible different hormonal bases for the two behaviors, as well as validating the relevance of swim test measures in this species. The sexual dimorphism shown here in the effects of stress on parental care parallels that seen in the effects of stress on partner preference in this species. Males that received exogenous CORT, or an optimal dosage of corticotropin-releasing factor, pair-bonded faster than controls [21,52]. On the other hand, females pair-bonded more slowly under stress [21]. The current results suggest that this relationship may be relevant to other forms of social behavior in this species. Acknowledgments This research was supported by NIH PO1 HD38490 to C. S. Carter, NIH NRSA F32 HD08702 and NSF 0437523 to K.L. Bales and NIH NRSA F32 HD 41293 to K.M. Kramer. Additional support was provided by IRUL. We thank Tim Hill and Dr. James Artwohl for animal and veterinary care. We also thank Sheryl Katta and Julie Hazelton for research assistance. References [1] Gammie SC, Negron A, Newman SM, Rhodes JS. Corticotropin-releasing factor inhibits maternal aggression in mice. Behav Neurosci 2004;118:805–14. [2] Windle RJ, Wood S, Shanks N, Perks P, Conde GL, da Costa APC, et al. Endocrine and behavioral responses to noise stress: comparison of virgin and lactating female rats during non-disrupted maternal activity. J Neuroendocrinol 1997;9:407–14. [3] Soechting HF, Siegel HI, Rosenblatt JS, Komisaruk BR. Tail pinch facilitates onset of maternal behavior in rats. Physiol Behav 1977;19:807–9. [4] Fleming AS, Steiner M, Corter C. Cortisol, hedonics, and maternal responsiveness in human mothers. Horm Behav 1997;32:85–98. [5] Bair NI, Pryce CR, Dobeli M, Martin RD. Evidence from urinary cortisol that maternal behavior is related to stress in gorillas. Physiol Behav 1998;64:429–37. [6] Bardi M, French JA, Ramirez SM, Brent L. The role of the endocrine system in baboon maternal behavior. Biol Psychiatry 2004;55:724–32. [7] Kleiman DG. Monotremes in mammals. Q Rev Biol 1977;52:39–69. [8] Staats AB, Rissman EF, Quinton RL, Wynne-Edwards KE. Hormonal correlates of paternal responsiveness in new and expectant fathers. Evol Hum Behav 2000;21:79–95. [9] Ziegler TE, Washabaugh FK, Snowdon CT. Responsiveness of expectant male cotton-top tamarins, Saguinus oedipus, to mate’s pregnancy. Horm Behav 2004;45:84–92. [10] Ziegler TE, Wegner FH, Snowdon CT. Hormonal responses to parental and nonparental conditions in male cotton-top tamarins, Saguinus oedipus, a new world primate. Horm Behav 1996;30:287–97. [11] Rebum CJ, Wynne-Edwards KE. Hormonal changes in males of a naturally biparental and a uniparental mammal. Horm Behav 2000;33:163–76. [12] Nunes S, Fite JE, Patera KJ, French JA. Interactions among paternal behavior, steroid hormones, and parental experience in male marmosets (Callithrix kuhlii). Horm Behav 2001;39:70–82. [13] Bale TL, Piccirri R, Contarino A, Koob GF, Vale WW, Lee KF. Mice deficient for both corticotropin-releasing factor receptor 1 (CRF1) and CRFR2 have an impaired stress response and display sexually dichotomous anxiety-like behavior. J Neurosci 2002;22:193–9. [14] Neumann ID. Involvement of the brain oxytocin system in stress coping: interactions with the hypothalamo-pituitary-adrenal axis. Prog Brain Res 2002;139:147–62. [15] Wang ZX, Ferris CF, Devries GJ. Role of septal vasopressin innervation in paternal behavior in prairie voles (Microtus ochrogaster). Proc Natl Acad Sci U S A 1994;91:400–4. [16] Wang ZX, Young LJ, De Vries GJ, Insel TR. Voles and vasopressin: a review of molecular, cellular, and behavioral studies of pair bonding and paternal behaviors. Prog Brain Res 1998;119:483–99. [17] Bester-Meredith JK, Marler CA. Vasopressin and the transmission of paternal behavior across generations in mated, cross-fostered Peromyscus mice. Behav Neurosci 2003;117:455–63. [18] Bales KL, Kim AJ, Lewis-Reece AD, Carter CS. Both oxytocin and vasopressin may influence alloparental behavior in male prairie voles. Horm Behav 2002;42:145-154. [19] Gerl LL, Carter CS, Gavish L. The mating system of the prairie vole Microtus ochrogaster: field and laboratory evidence for pair-bonding. Behav Ecol Sociobiol 1981;8:189–94. [20] Carter CS, Getz LL. Monogamy and the prairie vole. Sci Am 1993;268:100–6. [21] DeVries AC, DeVries MB, Taymans SE, Carter CS. The effects of stress on social preferences are sexually dimorphic in prairie voles. Proc Natl Acad Sci 1996;93:11980–4. [22] DeVries AC, Taymans SE, Carter CS. Social modulation of corticosterone responses in male prairie voles. Integr Neurobiol Affil 1997;80:404–7. [23] DeVries AC, DeVries MB, Taymans S, Carter CS. Modulation of pair bonding in female prairie voles (Role of septal vasopressin innervation in paternal behavior in prairie voles (Microtus ochrogaster) by corticosterone. Proc Natl Acad Sci U S A 1995;92:7744–8. [24] Stern JM, Lonstein JS. Neural mediation of nursing and related maternal behaviors. In: Russell JA, Douglas AJ, Windle RJ, Ingram CD, editors. The maternal brain: neurobiological and neuroendocrine adaptation and disorders in pregnancy and postpartum. New York: Elsevier Press; 2001. p. 263–78. [25] Galvez LM, Lee TT, Kostaras X, Sidhu JA, Barr AM. High levels of estradiol impair spatial performance in the Morris water maze and increase “depressive-like” behaviors in the female meadow vole. Physiol Behav 2002;77:217–25. [26] Carter CS, Epperson PM, Kramer KM, Ruscio MG, Bales KL. The infant as a neuroendocrine manipulation in monogamous prairie voles. Abstr—Soc Neurosci 2004;423:13. [27] Carter CS, Witt DM, Schneider J, Harris ZL, Volkening D. Male stimuli are necessary for female sexual behavior and uterine growth in prairie voles (Microtus ochrogaster). Horm Behav 1987;21:74–82. [28] Taymans SE, DeVries AC, DeVries MB, Nelson RJ, Friedman TC, Detert-Wadleigh S, et al. The hypothalamic-pituitary-adrenal axis of prairie voles (Microtus ochrogaster): evidence for target tissue glucocorticoid resistance. Gen Comp Endocrinol 1997;106:48–61. [29] Bales KL, Pfeifer LA, Carter CS. Sex differences and effects of manipulations of oxytocin on alloparenting and anxiety in prairie voles. Dev Psychobiol 2004;44:123–31. [30] Roberts RL, Williams JR, Wang AK, Carter CS. Cooperative breeding and monogamy in prairie voles: influence of the sire and geographical variation. Anim Behav 1998;55:1131–40. [31] Lonstein JS, De Vries GJ. Sex differences in the parental behavior of rodents. Neurosci Biobehav Rev 2000;24:669–86. [32] Stern JM, Johnson SK. Ventral somatosensory determinants of nursing behavior in Norway Rats: 1. Effects of variations in the quality and quantity of pup stimuli. Physiol Behav 1990;47:993–1011. [33] Roberts RL, Miller AK, Taymans SE, Carter CS. Role of social and endocrine factors in alloparental behavior of prairie voles (Microtus ochrogaster). Can J Zool-Revue Canadienne de Zoologie 1998;76:1862–8. [34] Sokal RR, Rohlf FJ. Biometry. New York: W.H. Freeman and Company; 1981. [35] Kocsis GL, Samayai Z, Szabo G. Oxytocin and addiction: a review. Psychoneuroendocrinology 1998;23:945–62. [36] Witt DM, Winslow JT, Insel TR. Enhanced social interactions in rats following chronic, centrally infused oxytocin. Pharmacol Biochem Behav 1992;43:855–61. [37] Amico JA, Vollmer RR, Karam JR, Lee PR, Li X, Koenig JL, et al. Centrally administered oxytocin elicits exaggerated grooming in oxytocin null mice. Pharmacol Biochem Behav 2004;78:333–9. [38] Bertrand KC, Aldridge JW, Houchard KR, Zhuang X. Sequential super-stereotypy of an instinctive fixed action pattern in hyper-dopaminergic mutants: a model of obsessive–compulsive disorder and Tourette’s. BMC Biol 2005;3:4. [39] Pedersen CA, Prange Jr AJ. Induction of maternal behavior in virgin rats after intracerebroventricular administration of oxytocin. Proc Natl Acad Sci U S A 1979;76:6661–5. [40] Pedersen CA. Brain oxytocin systems and maternal behavior. Biol Psychiatry 1997;41:192. [41] Lonstein JS. Effects of dopamine receptor antagonism with haloperidol on nurturing behavior in the biparental prairie vole. Pharmacol Biochem Behav 2002;73:1–9. [42] DeVries AC, Glaser ER, Detillion CE. Social modulation of stress responses. Physiol Behav 2003;79:399–407. [43] Onaka T. Neural pathways controlling central and peripheral oxytocin release during stress. J Neuroendocrinol 2004;16:308–12. [44] Altemus M, Deuster PA, Galliven E, Carter CS, Gold PW. Suppression of hypothalamic-pituitary-adrenal axis responses to stress in lactating women. J Clin Endocrinol Metab 1995;80:2954–9. [45] Ogawa S, Kow L-M, Pfaff DW. Effects of lordosis-relevant neuropeptides on midbrain periaqueductal gray neuronal activity in vitro. Peptides 1992;13:965–75. [46] Lonstein JS, Stern JM. Somatosensoary contributions to c-fos activation within the caudal periaqueductal gray of lactating rats: effects of perivocal rooting, and suckling stimuli from pups. Horm Behav 1997;32:155–66. [47] Lonstein JS, Stern JM. Role of the midbrain periaqueductal gray in maternal nurturance and aggression: c-fos and electrolytic lesion studies in lactating rats. J Neurosci 1997;17:3364–78. [48] Lonstein JS, Stern JM. Site and behavioral specificity of periaqueductal gray lesions on postpartum sexual, maternal, and aggressive behaviors in rats. Brain Res 1998;804:21–35. [49] Witt DM. Mechanisms of oxytocin-mediated sociosexual behavior. Ann N Y Acad Sci 1997;807:298–301. [50] Wang ZX, Ferris CF, Devries GJ. Role of septal vasopressin innervation in paternal behavior in prairie voles (Microtus ochrogaster). Proc Natl Acad Sci U S A 1994;91:400–4. [51] Wang ZX, Young LJ, De Vries GJ, Insel TR. Voles and vasopressin: a review of molecular, cellular, and behavioral studies of pair bonding and paternal behaviors. Prog Brain Res 1998;119:483–99. [52] DeVries AC, Gupta T, Cardillo S, Cho M, Carter CS. Corticotropin-releasing factor induces social preferences in male prairie voles. Psychoneuroendocrinology 2002;27:705–14.
510(k) Summary Altatec GmbH CONELOG® Implant System K113779 November 20, 2012 ADMINISTRATIVE INFORMATION Manufacturer Name: Altatec GmbH Maybachstrasse 5 D-71299 Wimsheim, Germany Telephone: +49 7044 9445 0 Fax: +49 7044-9445 723 Official Contact: Tina Steffanie-Oak Senior Regulatory Affairs Specialist, CAMLOG USA Representative/Consultant: Linda K. Schulz, BSDH, RDH Kevin A Thomas, PhD PaxMed International, LLC 11234 El Camino Real, Suite 200 San Diego, CA 92130 Telephone: +1 (858) 792-1235 Fax: +1 (858) 792-1236 Email: email@example.com firstname.lastname@example.org DEVICE NAME AND CLASSIFICATION Trade/Proprietary Name: CONELOG® Implant System Common Names: Dental implants and dental implant abutments Classification Name: Implant, Endosseous, Root-form Abutment, Implant, Dental, Endosseous Classification Regulations: 21 CFR 872.3640, 21 CFR 872.3630, Class II Product Codes: DZE, NHA Classification Panel: Dental Products Panel Reviewing Branch: Dental Devices Branch INTENDED USE Implant Indications for Use CONELOG® Implant System Implants are intended for immediate or delayed placement in the bone of the maxillary or mandibular arch. CONELOG® Implant System Abutments are intended for use as support for crowns, bridges or overdentures. When a one-stage surgical approach is applied, the implant may be immediately loaded when good primary stability is achieved and the functional load is appropriate. CONELOG® Implants with 3.3 mm diameter have the following additional specific indications: These are an alternative in cases where the alveolar ridge width is only 5 – 6 mm. Because of their lower mechanical strength compared with larger diameter implants, they should only be used under the following conditions: - As single implants, they should be used only to replace mandibular incisors and/or maxillary lateral incisors. - An edentulous arch can only be restored with a bar retained superstructure with at least four implants of 3.3 mm diameter without distal extensions. - Implants of Ø 3.3 mm are suitable for a partially edentulous arch when combined with implants of larger diameter for splinted superstructures. However, the limited strength of the implants with Ø 3.3 mm must be taken into account. - Avoid excessive mechanical stressing of the implants when using ball abutments in combination with Ø 3.3 mm implants. - The healing time for Ø 3.3 mm implants is at least 12 weeks. CONELOG® Implants with 7 mm length have the following additional specific indications: CONELOG® SCREW-LINE Implants should only be used when there is not enough space for a longer implant. Delayed loading in single tooth replacement is indicated with these implants. If the ratio of crown length to implant length is unfavorable the biomechanical risk factors have to be considered and appropriate measures have to be taken by the dental professional. Abutment Indications for Use CONELOG® Abutments are intended to be used to fabricate prosthetic restorations in conjunction with CONELOG implants to support prostheses in the maxillary and/or mandibular arch. Vario SR Indications for Use CONELOG® Vario SR components for crown and bridge restorations: - Occlusal screw-retained crown, bridge and bar constructions on CONELOG® implants in the anterior and posterior region of the maxilla and mandible CONELOG® Vario SR components for bar restorations: - Anchorage of implant-supported full dentures for the edentulous maxilla in conjunction with 4 or more CONELOG® implants and in the edentulous mandible in conjunction with 2 or 4 or more CONELOG® implants. DEVICE DESCRIPTION The purpose of this submission is to add an alternative to the CAMLOG® Implant System by the introduction of an implant line with a tapered implant abutment interface. The new line of implants and abutments is named the CONELOG® Implant System. The primary features of the current CAMLOG Implant System are a change from the tube-in-tube interface with a square cam design for anti-rotation to the tapered implant/abutment interface connection with three positioning cams and the addition of a 7.0 mm long implant. The three positioning cams of the CONELOG abutments are placed more apically on the abutment shaft but are comparable in design principle to the three positioning cams of the CAMLOG Implant System. All other design features of the system remain the same. The CONELOG implant system features an implant/abutment interface that includes a 7.5° taper with anti-rotation cams at the base of the connection. Components available with the CONELOG implant/abutment interface include the SCREW-LINE Promote® plus implant and cover screw, healing caps, Universal abutments, Telescope abutments, Esthomic® abutments, the Esthomic® abutment inset, temporary abutments, Gold-plastic abutments, Logfit® abutments, ball abutments, bar abutments and Vario SR abutments. All components correspond to previously cleared CAMLOG components with the tube-in-tube interface. EQUIVALENCE TO MARKETED DEVICES Altatec GmbH has demonstrated that, for the purposes of FDA’s regulation of medical devices, the CONELOG® Implant System is substantially equivalent in indications and design principles to the following predicate devices: Altatec GmbH, CAMLOG Implant System Modified Implants and Abutments cleared under K083496; Altatec GmbH, CAMLOG® VarioSR Abutments cleared under K103252; Altatec GmbH, Altatec CAMLOG Abutments and Healing Caps cleared under K051636; Astra Tech, Inc., OsseoSpeed™ 4.0S - 6 mm cleared under K063779, Sulzer Dental, Inc., Screw-Vent Dental Implant System cleared under K011028, Implant Innovations, Inc., OSSEOTITE® Certain® Dental Implants cleared under K063341, Bicon, LLC., Ø3.0 x 8 mm and Ø3.0 x 11 mm Bicon Implant 3.0mm cleared under K101849. The CONELOG Implant System is an additional implant line to the CAMLOG Implant System using a tapered implant/abutment interface. A tapered implant/abutment interface is used for the Astra Tech implant system cleared under K063779 and many other implants that are cleared for marketing in the United States. Other than the interface and the addition of a 7.0 mm length implant, all the components of the CONELOG system have the same design and material as those cleared under, K083496, K103252 and K051636. The 7.0 mm implant is similar in design and length to the Astra Tech OsseoSpeed 4.0S - 6 mm, having a tapered connection and a treated surface. Other design features of the 7.0 mm implant are the same as the other implants in the CONELOG Implant System. Mechanical testing, including static and dynamic compression-bending testing, has been performed according to ISO 14801 Dentistry - Implants - Dynamic fatigue test for endosseous dental implants. In conclusion, the data included in this submission demonstrates substantial equivalence to the predicate devices listed above. Overall, the CONELOG Implant System has the following similarities to the predicate devices: - has the same intended use, - uses the same operating principle, - incorporates the same basic design, - incorporates the same or very similar materials, and - has similar packaging and is sterilized using the same materials and processes. December 14, 2012 Altatec GmbH C/O Ms. Linda K. Schulz, BSDH, RDH PaxMed International, Limited Liability Company 11234 El Camino Real, Suite 200 SAN DIEGO CA 92130 Re: K113779 Trade/Device Name: Conelog® Implant System Regulation Number: 21 CFR 872.3640 Regulation Name: Endosseous Dental Implant Regulatory Class: II Product Code: DZE, NHA Dated: December 7, 2012 Received: December 10, 2012 Dear Ms. Schulz: We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading. If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register. Please be advised that FDA’s issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act’s requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting (reporting of medical device-related adverse events) (21 CFR 803); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050. If you desire specific advice for your device on our labeling regulation (21 CFR Part 801), please go to http://www.fda.gov/AboutFDA/CentersOffices/CDRH/CDRHOffices/ucm115809.htm for the Center for Devices and Radiological Health’s (CDRH’s) Office of Compliance. Also, please note the regulation entitled, “Misbranding by reference to premarket notification” (21CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm for the CDRH’s Office of Surveillance and Biometrics/Division of Postmarket Surveillance. You may obtain other general information on your responsibilities under the Act from the Division of Small Manufacturers, International and Consumer Assistance at its toll-free number (800) 638-2041 or (301) 796-7100 or at its Internet address http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/default.htm. Sincerely yours, Anthony D. Watson, B.S., M.S., M.B.A. Director Division of Anesthesiology, General Hospital, Respiratory, Infection Control and Dental Devices Office of Device Evaluation Center for Devices and Radiological Health Enclosure Indications for Use 510(k) Number: K113779 Device Name: CONELOG® Implant System Implant Indications for Use CONELOG® Implant System Implants are intended for immediate or delayed placement in the bone of the maxillary or mandibular arch. CONELOG® Implant System Abutments are intended for use as support for crowns, bridges or overdentures. When a one-stage surgical approach is applied, the implant may be immediately loaded when good primary stability is achieved and the functional load is appropriate. CONELOG® Implants with 3.3 mm diameter have the following additional specific indications: These are an alternative in cases where the alveolar ridge width is only 5 – 6 mm. Because of their lower mechanical strength compared with larger diameter implants, they should only be used under the following conditions: • As single implants, they should be used only to replace mandibular incisors and/or maxillary lateral incisors. • An edentulous arch can only be restored with a bar retained superstructure with at least four implants of 3.3 mm diameter without distal extensions. • Implants of Ø 3.3 mm are suitable for a partially edentulous arch when combined with implants of larger diameter for splinted superstructures. However, the limited strength of the implants with Ø 3.3 mm must be taken into account. • Avoid excessive mechanical stressing of the implants when using ball abutments in combination with Ø 3.3 mm implants. • The healing time for Ø 3.3 mm implants is at least 12 weeks. CONELOG® Implants with 7 mm length have the following additional specific indications: CONELOG® SCREW-LINE Implants should only be used when there is not enough space for a longer implant. Delayed loading in single tooth replacement is indicated with these implants. If the ratio of crown length to implant length is unfavorable the biomechanical risk factors have to be considered and appropriate measures have to be taken by the dental professional. Prescription Use X AND/OR Over-The-Counter Use (Part 21 CFR 801 Subpart D) (21 CFR 801 Subpart C) (PLEASE DO NOT WRITE BELOW THIS LINE-CONTINUE ON ANOTHER PAGE IF NEEDED) Concurrence of CDRH, Office of Device Evaluation (ODE) Abutment Indications for Use CONELOG® Abutments are intended to be used to fabricate prosthetic restorations in conjunction with CONELOG implants to support prostheses in the maxillary and/or mandibular arch. Vario SR Indications for Use CONELOG® Vario SR components for crown and bridge restorations: - Occlusal screw-retained crown, bridge and bar constructions on CONELOG® implants in the anterior and posterior region of the maxilla and mandible CONELOG® Vario SR components for bar restorations: - Anchorage of implant-supported full dentures for the edentulous maxilla in conjunction with 4 or more CONELOG® implants and in the edentulous mandible in conjunction with 2 or 4 or more CONELOG® implants. Prescription Use X (Part 21 CFR 801 Subpart D) AND/OR Over-The-Counter Use (21 CFR 801 Subpart C) (PLEASE DO NOT WRITE BELOW THIS LINE-CONTINUE ON ANOTHER PAGE IF NEEDED) Concurrence of CDRH, Office of Device Evaluation (ODE)
METHODS OF WARFARE THAT CAUSE UNNECESSARY SUFFERING OR ARE INHERENTLY INDISCRIMINATE: A MEMORIAL TRIBUTE TO HOWARD BERMAN† ROGER S. CLARK* I. INTRODUCTION—HOWARD Howard Berman was a card-carrying member of that esteemed professional community, the “Invisible College of International Lawyers.”1 I really do not remember where we first met, although I am fairly sure that we were introduced by a friend from my Columbia Law School days, Tim Coulter of the Indian Law Resource Center. Howard worked with him between 1978 and 1981. We were introduced again by Virgina Leary of the Law School at SUNY Buffalo, where he had his first teaching appointment as an Adjunct Assistant Professor. She was eager to show off her colleague who taught Indian Law. He and I got together and talked, very informally, at all sorts of meetings over the next two decades. We exchanged a lot of phone calls and reprints. Alas, like most good professional relationships, ours did not generate a publishable correspondence. So this small tribute will have to suffice as a token of our friendship. For the most part, we talked and shared documents for those twenty years about issues involving indigenous and colonized (or formerly colonized) peoples. He brought to the table not only a deep understanding of matters affecting Native Americans, but also a detailed knowledge of the ef- † Editor’s Note: Howard Berman was a Professor at California Western School of Law from 1987 until his death on June 18, 1997. The rest of the memorial tribute to Professor Howard Berman is contained in the introductory pages to this issue. * Board of Governors Professor, Rutgers State University School of Law, Camden, New Jersey; B.A., LL.B., LL.M., LL.D. Victoria Univ. of Wellington; LL.M., J.S.D. Columbia Univ. The author represents the government of Samoa at meetings of the Preparatory Committee on the Establishment of an International Criminal Court. Any views expressed here should not be attributed to that government. 1. See Oscar Schachter, The Invisible College of International Lawyers, 72 NW. L. REV. 217 (1977). forts of the International Labor Organization and the United Nations to draft international standards for the rights of such peoples. I brought some experience with the Maori, the tangata whenua, or people of the land, of my native New Zealand. We talked also about many other issues of contemporary public international law. It is easy to like people who we think are similar to us in some major respect. I liked him because he combined a scholar's regard for both principle and detail with the activist's urge to get involved with the real problems of real people, even if the issues he worked on were a little esoteric for some. The last time we met was at the Annual Meeting of the American Society of International Law in April of 1997. He was very upbeat about his medical prognosis. We spoke for a long time about the ongoing negotiations to create an International Criminal Court ("ICC"). I have been representing the government of Samoa at the deliberations of the Preparatory Committee on the Establishment of an International Criminal Court, which have been taking place mostly in New York.$^2$ Those deliberations will culminate in a Diplomatic Conference to complete the drafting that will take place in Rome in June to July 1998. I also represented Samoa before the International Court of Justice in the advisory proceedings on the legality of the use or threat of use of nuclear weapons.$^3$ A major focus of the conversation Howard and I had last April was on the connections between these two exercises, both of which, as I told him, have plumbed the depths of just about everything I ever learned about international law and international organizations. We talked about some broad issues, like whether the ICC will be a good thing, and whether its powers will be illusory because of interference by the Security Council. We also talked, in particular, about the way in which the substantive provisions of the proposed ICC statute might deal with weapons of mass destruction, including nuclear weapons. This is only one of dozens of issues that have arisen in the course of the drafting, but it seemed to Howard that it was especially interesting both intrinsically and because it raises many general issues about the nature of international diplomatic drafting. When the ICC preparatory exercise began, many of us were hopeful that --- 2. The Preparatory Committee began drafting on the basis of the International Law Commission's Draft Statute for an International Criminal Court, Report of the International Law Commission on the Work of Its Forty-Sixth Session, May 2-July 22, 1994, U.N. GAOR, 49th Sess., Supp. No. 10, at 43, U.N. Doc. A/49/10 (1994). The debate has ranged much wider. Over a hundred governments have been participating in this open-ended Committee. At the time of this writing, the current Draft Statute for the Court (with many issues unsettled) is contained in Preparatory Committee on the Establishment of an International Criminal Court, Report of the Interessional Meeting from Jan. 19-30, 1998, in Zutphen, the Netherlands, U.N. Doc. A/AC. 249/1998/L.13 (1998) [hereinafter Zutphen Draft]. Many informal meetings have taken place throughout the process. Particularly valuable have been the series of meetings in Siracusa under the auspices of the International Institute of High Studies in Criminal Sciences, which have produced numerous useful texts. 3. Legality of the Threat or Use of Nuclear Weapons, 1996 I.C.J. [hereinafter Nuclear Weapons Advisory Opinion]. On the role of Samoa and other Pacific nations in that proceeding, see infra note 25. this exercise was an opportunity for a comprehensive codification of the general and special parts of international criminal law.$^4$ It soon became clear that few governments wanted to go that far and that what we would produce is a relatively cautious document that addresses some of the issues, but which is essentially an instrument giving jurisdiction to a court over selected crimes that are already well established in general international law. Efforts have been made to draft some general savings clause that tries to ensure that the existence of this particular jurisdictional statement will not be seen as backsliding on current achievements or as freezing efforts in other forums to engage in progressive development of the rules.$^5$ Traditional war crimes were the least controversial aspect of the Nuremberg and Tokyo trials. As one might expect, therefore, a significant part of the ICC's subject matter jurisdiction will be over offenses against the laws of armed conflict.$^6$ There also seems to be a consensus among those participating that the offenses will be spelled out in rather more detail than was done at Nuremberg$^7$ or Tokyo$^8$ and in the Statutes for the Tribunals for Former Yugoslavia$^9$ and Rwanda.$^10$ It was in this spirit of seeking more detail that the Preparatory Committee had taken a first cut at the war crimes issue in general and the weapons issue in particular in a Working Group that met in February 1997.$^11$ --- 4. This was the approach taken in the very useful "Siracusa Draft" produced by a number of NGOs as 1994 ILC Draft Statute for an International Criminal Court with Suggested Modifications, at Siracusa/Freiburg/Chicago (1996). 5. See currently, Zutphen Draft, \textit{supra} note 2, at 16, art. Y: Without prejudice to the application of the provisions of this Statute, nothing in this part of the Statute shall be interpreted as limiting or prejudicing in any way existing or developing rules of international law. 6. Whatever else is ultimately placed within the subject matter jurisdiction of the ICC, it is now clear that it will at least deal with genocide, "war crimes," and crimes against humanity. Still undecided is whether it will have jurisdiction over aggression, terrors, drug offenses, and offenses against United Nations and associated personnel. \textit{See} Zutphen Draft, \textit{supra} note 2, art. 5. Personally, I think some term such as "offenses against the laws of armed conflict" would be more accurate than "war crimes," since many of the offenses, derived from the law of the Hague and Geneva, will apply in cases of non-international armed conflict, as well as in the cases of international conflict for which the term "war" is more traditional. The term "war crimes" is, however, used in the Draft. 7. Charter of the International Military Tribunal, Aug. 8, 1945, 82 U.N.T.S. 279. 8. Charter of the International Military Tribunal for the Far East, established at Tokyo, Jan. 19, 1946, U.S. Dep't of State Pub. No. 2675. 9. Report of the Secretary-General Pursuant to Paragraph 2 of Security Council Resolution 808 (1993), including the Statute of the Tribunal, U.N. Doc. S/25704 & Add. 1 (1993). 10. S.C. Res. 955, U.N. SCOR, 49th Year, 3453d mtg. at 1, U.N. Doc. S/RES/955 (1994). 11. Decisions Taken by the Preparatory Committee at its Session held from Feb. 11-21, 1997, at 10-11, U.N. Doc. A/AC.249/1997/L.5 (1997). The draft produced on that occasion was based on the very general draft prepared by the International Law Commission, along with submissions made particularly by the United States, Switzerland, and New Zealand. The Swiss/New Zealand material was heavily influenced by the positions of the InterAfter clarifying my thoughts on the weapons aspects of this first draft by talking to Howard, I wrote an Aide-Memoire, a revised version of which was ultimately circulated at the December meeting of the Committee on behalf of Samoa, as well as Solomon Islands and Marshall Islands.\textsuperscript{12} In what follows, I have now worked the Aide-Memoire back into this very edited memoir of our last conversation as a tribute to Howard. I think he would be gratified by what follows. He helped me shape the issues, and the purpose that we were trying to serve is in line with what he tried to do all of his professional life—to protect human dignity. II. THE DRAFTING ISSUE In a major drafting exercise like the creation of an ICC, everything turns on how everything else is resolved; thus, at any one moment, responding to each particular proposal is much like aiming at a moving target. A major feature of the product of the Working Group’s efforts in February 1997 (and still of much of the text) was numerous square brackets indicating choices that had been postponed. Between then and the meeting of the Preparatory Committee in the following December, there was a gathering of what are known in diplomatic-speak as “interested delegations,” all of them, as it happened, NATO members. They produced another version that ultimately became the focus of the debate in December. It is convenient to explore the issues by examining this text. As officially circulated, it offered two options: \textit{Option I} (o) employing the following weapons, projectiles and material and methods of warfare which are calculated to cause superfluous injury or unnecessary suffering: (i) poison or poisoned weapons, (ii) asphyxiating, poisonous or other gases, and all analogous liquids, materials or devices, (iii) bullets which expand or flatten easily in the human body, such as bullets with a hard envelope which does not entirely cover the core or is pierced with incisions, (iv) bacteriological (biological) agents or toxins for hostile purposes or \textsuperscript{12} Aide-Memoire by Samoa, Marshall Islands and Solomon Islands, Laws of War: Why We Prefer Option II of Draft Paragraph B (o) Concerning Methods of Warfare That Cause Unnecessary Suffering or Are Inherently Indiscriminate (1997). Samoa, Marshall Islands, and Solomon Islands are three small island states located in the Pacific Ocean. They have taken a particular interest since their independence in issues relating to the laws of armed conflict, in particular to matters relating to weapons that kill and maim civilians and to weapons of mass destruction. Two of their territories were occupied and became venues for bloody battles during the Second World War. Much debris from that conflict is still with them. They all live in an area that was used by three of the nuclear weapons states for testing their material. The Marshall Islands was itself the scene of one series of those tests. in armed conflict, (v) chemical weapons as defined in and prohibited by the 1993 Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and On Their Destruction; Option II (o) employing weapons, projectiles and material and methods of warfare which are of a nature to cause superfluous injury or unnecessary suffering or which are inherently indiscriminate. The Aide-Memoire argued strongly in favor of Option II. I turn now to the rationale behind that preference. III. SOME HISTORY AND JURISPRUDENCE At the outset, one must note the difference between the words of the chapeau in Option I, “calculated to cause,” and the corresponding language of Option II, “of a nature to cause.” The Option I wording seems to require that the prosecution prove a more specific intent than the alternative requires.$^{13}$ I accept the statement by the International Committee of the Red Cross (“ICRC”) that the language in Option II is an appropriate translation of the relevant provisions of the Hague Conventions of 1899 and 1907.$^{14}$ The laws of armed conflict concerning weapons that, for the sake of --- $^{13}$ Jordan Paust, The Preparatory Committee’s “Definition of Crimes”—War Crimes, 8 CRIM. L. F. (1997) (arguing for a “wanton or reckless disregard” standard for the mens rea for this offense). This drafting point also highlights a probably intractable problem with the current drafting. While the general part, Zutphen Draft, supra note 2, art. 23, defines the mens rea terms “intent,” “knowledge,” and “recklessness,” the special part where the crimes are defined often uses words like “wilful,” “calculated,” “wanton,” or “unlawful” to describe the mens rea element. It is probably not possible in the ICC negotiations to rework such formulations, since they are taken directly from existing treaty language, and one basic assumption within the group is that it is almost impossible to improve upon previously agreed text. The United States delegation to the Preparatory Committee has proposed addressing this problem, at least in part, by including detailed “elements of offenses” in the Court’s Rules of Procedure and Evidence. An informal draft is available. $^{14}$ ICRC, COMMENTS ON INFORMAL WORKING PAPER ON WAR CRIMES OF OCT. 13, 1997, at 3 (New York, Dec. 2, 1997). The ICRC noted: The French text in 1899 and 1907 used the same wording: “propre a causer des maux superflus.” The French text of these Hague Conventions is the only authentic text. In 1899 the terminology was correctly translated into English, that is “to employ arms . . . of a nature to cause superfluous injury,” but incorrectly translated in 1907 into “calculated to cause superfluous injury.” This has been rectified in the wording of article 35 (2) of 1977 Protocol I (“of a nature to cause”), in article 3 of Protocol II (amended 3 May 1996) of the 1980 Convention (“designed or of a nature to cause”) and recently, in the preamble of the Ottawa Convention on the prohibition of anti-personnel mines which provides for the “principle that prohibits the employment in armed conflicts of weapons, projectiles and materials and methods of warfare of a nature to cause superfluous injury or unnecessary suffering.” simplicity, we shall describe as causing unnecessary suffering or that are inherently indiscriminate, have developed at two levels of abstraction, what one might call the level of "principles" on the one hand and of "rules" on the other. Since at least the Declaration of St. Petersburg in 1868,\textsuperscript{15} states have accepted that the means of killing the enemy are not unlimited and that general principles establish limitations on broad categories of weapons and methods. In tandem with these principles, states have from time to time agreed to rules to the effect that certain named weapons are forbidden per se. Thus, in St. Petersburg, the parties banned any projectile of less than 400 grams that was explosive or was charged with fulminating or inflammable substances.\textsuperscript{16} At the Hague in 1899 and 1907, it was agreed that poison or poisoned weapons\textsuperscript{17} and dum-dum bullets were forbidden.\textsuperscript{18} In the 1925 Geneva Protocol,\textsuperscript{19} asphyxiating gases were banned (although many thought they were already illegal under customary law or under the Hague Conventions), and biological weapons were added to the list. There have, of course, been later additions to the list of absolutely forbidden items, notably weapons that injure primarily by non-detectable fragments and blinding laser weapons.\textsuperscript{20} The Ottawa Land Mines Convention,\textsuperscript{21} signed on behalf of the three Pacific states in December 1997, is the most recent example. Now, the most important point about those weapons banned per se is that it is never lawful to use them—never, under any circumstances. No considerations of military necessity can ever authorize their use. None of the justifications or excuses for an individual's acts coming within the rubrics of self-defense and necessity—over which the Preparatory Committee also agonized at great length\textsuperscript{22}—can ever apply. An absolute rule is just that: absolute. Absolute, per se, prohibitions are good examples of the point made \begin{itemize} \item \textsuperscript{15} Declaration Renouncing the Use, in Time of War, of Explosive Projectiles Under 400 Grammes Weight, Dec. 11, 1868, reprinted in 1 Am.J. Int’l L. 95 (Supp. 1907). \item \textsuperscript{16} The general language of the Declaration spoke of the “employment of arms which uselessly aggravate the sufferings of disabled men, or render their death inevitable.” \item \textsuperscript{17} 1899 Hague Declaration 3 Concerning Expanding Bullets, July 29, 1899, 1907 Gr. Brit. T.S. No. 32 (Cmd. 3751), 1 A.J.I.L. Supp. 155. \item \textsuperscript{18} Annex to Hague Convention IV Respecting the Laws and Customs of War on Land, Regulations, Oct. 18, 1907 36 Stat. 2277, 205 Consol. 277, art. 23 (a) (poison or poisoned weapons). \item \textsuperscript{19} Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare, June 17, 1925, 26 U.S.T. 571, 94 L.N.T.S. 65. \item \textsuperscript{20} Protocol I to the Convention on Prohibition or Restrictions on the Use of Certain Conventional Weapons Which May Be Deemed to Be Excessively Injurious or Have Indiscriminate Effects, Oct. 10, 1980, 1342 U.N.T.S. 7 (“Convention on Conventional Weapons”) (non-detectable fragments); Protocol IV to the Convention on Conventional Weapons of 1980, Oct. 13, 1995, U.N. Doc. CCW/CONF.II/7 (blinding laser weapons). \item \textsuperscript{21} Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-personnel Mines and on Their Destruction, 36 I.L.M. (1997). \item \textsuperscript{22} See Zutphen Draft, supra note 2, art. 25, paras. (c)-(e) (grounds for excluding criminal responsibility). \end{itemize} by a United States Military Tribunal at Nuremberg in 1948 in convicting a German military leader, General List: "[T]he rules of international law must be followed even if this results in the loss of a battle or even a war. Expediency or necessity cannot warrant their violation."23 But, of course, the existence of the particular absolute rules does not exhaust the meaning of the general principle. All those rules do is create cases where there is no room for argument about whether the general rule applies to the particular instance.24 Marshall Islands, Samoa, and Solomon Islands had combined their resources on a previous occasion and sought to apply this understanding of humanitarian law to nuclear weapons in a joint presentation to the International Court of Justice in the advisory proceedings concerning the use, or threat of use, of such weapons.25 It was their view that the laws of armed conflict, the law of human rights, and the law on the environment point inexorably, both separately and in combination, towards a per se rule that bans the use of nuclear weapons in all instances. They also contended before the Court that, at the very least, any use of nuclear weapons must be allowed only so far as it is compatible with provisions of the law of the Hague, such as those reiterated in Option II. The four nuclear powers and their allies who participated in the proceedings conceded this much. The Court, apparently unanimously, endorsed the position adopted by the Nuremberg Tribunal, that while some of the norms in the law of the Hague might have been only conventional in origin, all of them have since become customary law.26 Three of the fourteen judges accepted the anti- 23. U.S. v. List, Feb. 19, 1948, in 11 TRIALS OF WAR CRIMINALS BEFORE THE NUREMBERG MILITARY TRIBUNALS UNDER CONTROL COUNCIL LAW NO. 10, 1946-1949, at 1230, 1272. 24. Oliver Wendell Holmes made a similar point about the common law of negligence: From the time of Alfred to the present day, statutes and decisions have busied themselves with defining the precautions to be taken in certain familiar cases; that is, with substituting for the vague test of the care exercised by a prudent man, a precise one of specific acts or omissions. . . . It will be observed that the existence of the external tests of liability . . . while it illustrates the tendency of the law of tort to become more and more concrete by judicial decision and statute, does not interfere with the general doctrine maintained as to the grounds of liability. OLIVER WENDELL HOLMES, THE COMMON LAW 112 (1881) (discussing rules about falling on spears, the rules of the road, and shipping regulations). 25. The oral and written pleadings of the three Pacific Island States are contained in THE CASE AGAINST THE BOMB: MARSHALL ISLANDS, SAMOA AND SOLOMON ISLANDS BEFORE THE INTERNATIONAL COURT OF JUSTICE IN PROCEEDINGS ON THE LEGALITY OF THE THREAT OR USE OF NUCLEAR WEAPONS (Roger S. Clark & Madeleine Samm eds., 1996). Other helpful discussions are contained in INTERNATIONAL REVIEW OF THE RED CROSS, No. 316, Special Issue (Jan.-Feb. 1997); JOHN BURRUSOHS, THE LEGALITY OF THE THREAT OR USE OF NUCLEAR WEAPONS: A GUIDE TO THE HISTORIC OPINION OF THE INTERNATIONAL COURT OF JUSTICE (1997). 26. Nuclear Weapons Advisory Opinion, supra note 3, paras. 79-82. nuclear position about a per se rule on nuclear weapons having emerged by now. Seven of the judges (who with the casting vote of the President constituted a majority) came within a hairsbreadth of this position. They went so far as to say that "the threat or use of nuclear weapons would generally be contrary to the rules of international law applicable in armed conflict, and in particular the principles and rules of humanitarian law."27 They added that they were "unable to conclude definitively whether the threat or use of nuclear weapons would be lawful or unlawful in an extreme circumstance of self-defense, in which the very survival of a State would be at stake."28 Fourteen of the judges—all who sat—held, as the anti-nuclear powers had contended and the nuclear powers conceded, that "a threat or use of nuclear weapons should also be compatible with the requirements of international law applicable in armed conflict, particularly those of the principles and rules of humanitarian law, as well as with specific obligations under treaties and other undertakings which deal specifically with nuclear weapons."29 The Court's unanimous opinion on this point is an emphatic endorsement of the application of the general principles contained in Option II to nuclear weapons, and by obvious implication, to other weapons that are not somewhere proscribed by name. In a particular (extremely rare) case involving nuclear weapons, since the majority did not go so far as to adopt a per se rule, it might be possible to argue that the requirements of the principle had been met. But the absence of a definitive per se rule did not mean that the general principle was not of great force. There are then, in settled existing law, both general principles and specialized rules in the area in question. What would Option I do to this structure? It would say that the list of proscribed weapons therein represents the whole content of the general principle. The general principle has gone. All we are left with is a list of proscribed weapons, and a limited list at that. Let me give an example of what this means. The list includes chemical and bacteriological weapons. It also includes dum-dum bullets and poison arrows. It does not include nuclear weapons. Think about it: It is absolutely forbidden to kill someone with a poisoned arrow, and the ICC could have jurisdiction. If, however, one were to incinerate a hundred thousand people with an atomic weapon, the ICC would have no jurisdiction. A nuclear weapon is not one of the "following" means of which the provision speaks! And there is no point in appealing to the general principle established by 1899; it has been emasculated except to the extent that it applies to the particular cases on the list in Option I. And the same is true of other weapons of mass destruction and other weapons that cause excessive suffering. If it is not on the list, there is no case—standing the law of armed conflict on its 27. Id., dispositive para. 105 (2) E. 28. Id. The views of the seven covered a spectrum on this issue. See THE CASE AGAINST THE BOMB, supra note 25, at 24. 29. Nuclear Weapons Advisory Opinion, supra note 3, dispositive para. 105 (2) D. head. It ought to be feasible to state a case in most such circumstances pursuant to the general principles. Now, it may be possible for a creative prosecutor to repackage the atomic example (and other cases not expressly mentioned) under some other provisions in other articles of the draft text.$^{30}$ As is the case with domestic criminal laws, there will be some overlap among the offenses within the jurisdiction of the ICC, enabling prosecutors to have some discretion over what alternatives to charge. But is there not something intellectually unsatisfying to have a law that speaks to poisoned arrows but not to the most terrible weapons of mass destruction in our century, not addressing them even in terms of general principle? Is it really the case, in the words attributed to Honoré de Balzac, that "the laws are spider webs through which the big flies pass and the little ones get caught"$^{31}$? Are we really prepared to accept that the supposedly noble preparatory drafting exercise of creating an ICC deals with the small flies but not the big ones? In the words of Judge Weeramantry (albeit arguing for a per se rule) in the Nuclear Weapons Case: At least, it would seem passing strange that the expansion within a single soldier of a single bullet is an excessive cruelty which international law has been unable to tolerate since 1899, and that the incineration of a hundred thousand civilians is not. This astonishment would be compounded when that weapon has the capability, through multiple use, of endangering the entire human species and all civilization with it.$^{32}$ "Passing strange," indeed! Fortunately, the existing general law is not so strange. It contains the general statement in Option II, a statement that is appropriate to deal with the real cases. The Statute of the International Criminal Court should do no worse than to restate the existing law. In the best of all worlds, restating the existing law would mean re-stating both the general principle and, in a separate article, a list of the specific per se items. The problem with that approach, in the present setting, is that delegations might never agree on what to include on the list. I would want to include such items as anti-personnel mines, and, like the ICRC,$^{33}$ weapons that primarily injure by non-detectable fragments,$^{34}$ as well as $^{30}$ See, for example, Zutphen Draft, \textit{supra} note 2, art. 5 (War Crimes section), Part A (grave breaches of Geneva Conventions), "any of the following acts against persons or property protected under the provisions of the relevant Geneva Conventions: (a) wilful killing, . . . (c) willfully causing great suffering, or serious injury to body or health; (d) extensive destruction and appropriation of property, not justified by military necessity and carried out unlawfully and wantonly." $^{31}$ Philippe Sands, \textit{Oral Presentation for Solomon Islands}, in \textit{The Case Against The Bomb}, \textit{supra} note 25, at 279. $^{32}$ Nuclear Weapons Advisory Opinion, \textit{supra} note 3, dissent of Judge Weeramantry, at 8 (mimeo edition). $^{33}$ \textit{COMMENTS ON INFORMAL WORKING PAPER}, \textit{supra} note 14. $^{34}$ \textit{Supra} note 20. blinding laser weapons.\textsuperscript{35} Others might argue that the conventional prohibitions of these particular engines of inhumanity are too recent and not widely enough ratified to have found their way into the general law. Some could also be tempted into insisting that we should accept in these negotiations the advice of the three judges in the International Court who believe that there is a \textit{per se} rule against nuclear weapons.\textsuperscript{36} \section*{IV. WHAT HAPPENED} The result of the December debate was not, as might have been expected, the triumph of reason. Instead, it was the creation of more options, so that we now have Options 1 to 4, with the fourth option having two alternatives.\textsuperscript{37} All the possibilities are now on the record, and it looks as though the issues will not be resolved until the Diplomatic Conference. Option 1, discussed above, has now become Option 1; Option II has now become Option 3. A new Option 2 contains the language “which are of a nature to cause” in its chapeau, rather than the words “which are calculated to cause.” It then has as its exclusive list the same weapons as in Option 1 (and the old Option I), with this addition: (vi) such other weapons or weapons systems as become the subject of a comprehensive prohibition pursuant to customary or conventional international law.\textsuperscript{38} While this version introduces a welcome note of flexibility that allows for subsequent developments, it seems to be too open-ended for some of the major powers, while at the same time it is unacceptable to many small ones because of the exclusive nature of the list. Option 4 in its first incarnation would have an exclusive chapeau, drafted in the same language as Option 2. In its second version, the chapeau would read: employing weapons, projectiles and material and methods of warfare which are of a nature to cause superfluous injury or unnecessary suffer- \textsuperscript{35} \textit{Id.} \textsuperscript{36} That is one way to interpret the reference to nuclear weapons in para. B 4. (k) (v) at 11 of U.N. Doc. A/AC.249/1997/L.5 from the Preparatory Committee’s Feb. 1997 Session. \textsuperscript{37} Decisions Taken by the Preparatory Committee at its Session Held from Dec. 1-12 1997, at 5-6, U.N. Doc. A/AC.249/1997/L.9/Rev.1 (1997), Zutphen Draft, \textit{supra} note 2, art. 5, War Crimes, B (o). \textsuperscript{38} In one respect, this formulation follows the wisdom of Holmes, \textit{supra} note 24. Holmes envisaged the creation of new \textit{per se} rules either by the legislature or by the courts. This formulation (produced by the Canadian delegation, building on a suggestion from the ICRC) contemplates development either by treaty or by custom. Where it departs from Holmes is that it is part of a list of rules that now swallows up and completely exhausts the general principle. The (almost successful) strategy of those who promoted the I.C.J. nuclear proceeding, \textit{supra} notes 3, 25, was to invite the Court to declare a further \textit{per se} category, while retaining the general principle. ing or which are inherently indiscriminate, such as\textsuperscript{39} but not limited to: What follows the chapeau in each instance is the list contained in Option 2, with the addition of nuclear weapons, anti-personnel mines, and blinding laser weapons. As one who pushed the old Option II, now Option 3, last December without forging a consensus, I am probably the wrong person to predict where we might go. I still think that this option probably has the best chance as most delegates' second choice. It has the beauty (and the liability) of being just a little vague. But I leave you, where Howard was forced to leave us, with much work undone. V. FAREWELL Vale Howard! But stay tuned. \textsuperscript{39}. A helpful NGO proposal, circulated at the meeting as Proposal of the Weapons System Caucus, suggested the words "including, but not limited to." California Western International Law Journal, Vol. 28, No. 2 [1998], Art. 4 https://scholarlycommons.law.cwsl.edu/cwilj/vol28/iss2/4
CERTIFICATE OF INCORPORATION OF A PRIVATE LIMITED COMPANY Company No. 4868351 The Registrar of Companies for England and Wales hereby certifies that QUID NOVI LIMITED is this day incorporated under the Companies Act 1985 as a private company and that the company is limited. Given at Companies House, Cardiff, the 15th August 2003 N04868351M Declaration on application for registration Company Name in full QUID NOVI LIMITED I, GRANT SECRETARIES LIMITED of 2nd Floor, Mountbarrow House, 12 Elizabeth Street, London, SW1W 9RB do solemnly and sincerely declare that I am a [ ] person named as director or secretary of the company in the statement delivered to the Registrar under section 10 of the Companies Act 1985 and that all the requirements of the Companies Act 1985 in respect of the registration of the above company and of matters precedent and incidental to it have been complied with. And I make this solemn Declaration conscientiously believing the same to be true and by virtue of the Statutory Declarations Act 1835. Declarant's signature AUTHORISED SIGNATORY Declared at 45 Lower Belgrave Street, London, SW1W 0LS Day Month Year On 13 08 2003 HUTCHINSON MAINPRICE SOLICITORS 45 Lower Belgrave Street Belgravia London SW1W 0LS Signed Date 13th August 2003 † A Commissioner for Oaths or Notary Public or Justice of the Peace or Solicitor Please give the name, address, telephone number and, if available, a DX number and Exchange of the person Companies House should contact if there is any query. Moneypenny Management Limited, 12 Elizabeth Street, London SW1W 9RB Tel 020 7730 0883 DX number DX exchange When you have completed and signed the form please send it to the Registrar of Companies at: Companies House, Crown Way, Cardiff, CF14 3UZ DX 33050 Cardiff for companies registered in England and Wales or Companies House, 37 Castle Terrace, Edinburgh, EH1 2EB DX 235 Edinburgh for companies registered in Scotland First directors and secretary and intended situation of registered office Company Name in full: QUID NOVI LIMITED Proposed Registered Office: 2nd Floor, Mountbarrow House 12 Elizabeth Street London County / Region: [ ] Postcode: SW1W 9RB If the memorandum is delivered by an agent for the subscriber(s) of the memorandum mark the box opposite and give the agent's name and address. Agent's Name: [ ] Address: [ ] Post town: [ ] County / Region: [ ] Postcode: [ ] Number of continuation sheets attached: [ ] Please give the name, address, telephone number and, if available, a DX number and Exchange of the person Companies House should contact if there is any query. Moneypenny Management Limited, 12 Elizabeth Street, London SW1W 9RB Tel 020 7730 0883 DX number: [ ] DX exchange: [ ] When you have completed and signed the form please send it to the Registrar of Companies at: Companies House, Crown Way, Cardiff, CF14 3UZ DX 33050 Cardiff for companies registered in England and Wales or Companies House, 37 Castle Terrace, Edinburgh, EH1 2EB for companies registered in Scotland DX 235 Edinburgh Company Secretary (see notes 1-5) Company name NAME Style / Title Honours etc * Voluntary details Forename(s) Surname GRANT SECRETARIES LIMITED Previous forename(s) Previous surname(s) Address Usual residential address For a corporation, give the registered or principal office address. 2nd Floor, Mountbarrow House, 12 Elizabeth Street, London County / Region Postcode SW1W 9RB Country England I consent to act as secretary of the company named on page 1 Consent signature For Grant Secretaries Limited Date 13th August 2003 Directors (see notes 1-5) Please list directors in alphabetical order NAME Style / Title Honours etc Forename(s) Surname GRANT DIRECTORS LIMITED Previous forename(s) Previous surname(s) Address Usual residential address For a corporation, give the registered or principal office address. 2nd Floor, Mountbarrow House, 12 Elizabeth Street, London County / Region Postcode SW1W 9RB Country England Day Month Year Date of birth 10 12 1993 Nationality British Business occupation Company Formation Agent Other directorships No other directorships I consent to act as director of the company named on page 1 Consent signature For Grant Directors Limited Date 13th August 2003 Directors (continued) (see notes 1-5) \| NAME \| Style / Title \| Honours etc \| \|------\|----------------\|--------------\| \| \| Forename(s) \| \| \| \| Surname \| \| \| \| Previous forename(s) \| \| \| \| Previous surname(s) \| \| Address Usual residential address For a corporation, give the registered or principal office address. \| Post town \| County / Region \| Postcode \| \|-----------\|-----------------\|----------\| \| \| \| \| \| Country \| Day \| Month \| Year \| \|---------\|-----\|-------\|------\| \| \| \| \| \| Date of birth Business occupation Other directorships I consent to act as director of the company named on page 1 Consent signature --- This section must be signed by Either an agent on behalf of all subscribers Or the subscribers (i.e those who signed as members on the memorandum of association). \| Signed \| Date \| \|--------\|------\| \| \| \| For Grant Directors Limited Authorised Signatory Date For Grant Secretaries Limited Authorised Signatory Date Signed Date Signed Date Signed Date Signed Date Signed Date 1. The Company's name is QUID NOVI LIMITED 2. The registered office of the company will be situate in England. 3. The objects for which the company is established are:- (1) To carry on business as a general commercial company and any trade or business whatsoever and any lawful purpose pursuant to the Companies Act 1985 (hereinafter called "the Act") as amended, extended or applied by or under any other enactment or as re-enacted. (2.) To carry on any other business, which, in the opinion of the company, may be capable of being conveniently carried on in conjunction with or subsidiary to any other business of the company and is calculated to enhance the value of the company's property or further its objects or any of them. (3.) To purchase or by any other means, acquire freehold, leasehold or any other property for any estate or interest whatever, movable or immovable, or any interest in such property, and to sell, lease, let on hire, develop such property, or otherwise turn the same to the advantage of the company. (4.) To apply for, register, purchase, or by any other means acquire and protect, prolong and renew, whether in the United Kingdom or elsewhere any patents, patent rights, brevets d'invention, licences, secret processes, trade marks, designs, protections and concessions and to disclaim, alter, modify, use and turn to account and to manufacture under or grant licences or privileges in respect of the same, and to expend money in experimenting upon, testing and improving any patents, inventions or rights which the company may acquire or propose to acquire. (5.) To build, re-construct or generally maintain buildings and works of all kinds, whether or not these are situate on the property of the company. (6.) To invest and deal with the monies of the company in such shares or upon such securities and in such manner as from time to time may be determined. (7.) To acquire or undertake the whole or any part of the business, goodwill, and assets of any person, firm or company carrying on or proposing to carry on any of the businesses which the company is authorised to carry on and as part of the consideration for such acquisition to undertake all or any of the liabilities of such person, firm or company, or to acquire an interest in, amalgamate with, or enter into partnership, or into any agreement for sharing profits, or for co-operation, or for mutual assistance with any such person, firm or company, or for subsidising or otherwise assisting any such person, firm or company, and to give or accept, by way of consideration for any of the acts or things aforesaid or property acquired, any shares, debentures, debenture stock or securities that may be agreed upon, and to hold and retain, or sell, mortgage and deal with any shares, debentures, debenture stock or security so received. (8.) To enter into any arrangements with any government or authority (Supreme, Municipal, Local or otherwise) that may seem conducive to the attainment of the company's objects or any of them, and to obtain from any government or authority any charters, decrees, rights, privileges or concessions which the company may think desirable and to carry out, exercise, and to comply with any such charters, decree, rights, privileges and concessions. (9.) To sell or otherwise dispose of the whole or any part of the business or property of the company, for any consideration, shares or debentures and the company may think fit. (10)a) To lend and advance money or give credit on such terms as may seem expedient and with or without security to customers and others, to enter into guarantees, contract of indemnity and surety ships of all kinds, to receive money on deposit or loan upon any terms and to secure or guarantee the payment of any sums of money or the performance of any obligation by any company, firm or person including any holding company, subsidiary or fellow subsidiary company in any manner. b) Either with or without the company receiving any consideration or advantage, direct or indirect, from giving such guarantee and so as to be an independent object of the company, to guarantee the performance of the obligations of others including the payment of capital or principal together with any premium of any and dividends or interest on or other payment in respect of loans, credits, stocks, shares, or securities or other obligations of any nature whatsoever and without limiting the generality of the foregoing obligations for the repayment of money and/or discharge of liabilities both present and future, actual or contingent and insofar and the same is not prohibited by law, obligations and liabilities incurred in connection with or for the purpose of the acquisition of shares in the company or in any company which is for the time being the company's holding company as defined by Section 736 of the Companies Act 1985, due, owing or incurred to Bankers or any other person of any other company, firm or person, and in particular (but not by way of limitation) of the company's holding company or any company which is contemplated to become the company's holding company as a subsidiary, as defined in Section 736 of the Companies Act 1985 of the company or of the company's holding company or otherwise associated with the company in business, or any company, firm or person which the directors of the company shall think appropriate and to create mortgages charges or liens upon all or any of the property or assets of the company (both present and future) including its uncalled capital in support of such guarantees or otherwise as security for any obligations and liabilities of others. (11.) To borrow and raise money in such manner and to secure the repayment of any money borrowed, raised or owing by mortgage, charge, standard security, lien or other security upon the whole or any part of the company's property or assets (whether present or future), including its uncalled capital, and also by a similar mortgage, charge, standard security, lien or security to secure and guarantee the performance by the company of any obligation or liability it may undertake or which may become binding upon it. (12.) To pay or remunerate any person, firm or company for rendering services to the company in the promotion of the company or the placing and issue of shares, debentures, debenture stock or other securities of the company. (13.) To support and subscribe to any charitable or public object and to support and subscribe to any institution, society, or club which may be for the benefit of the company or its directors or employees, or may be connected with any town or place where the company carries on business; to give or award pensions, annuities, gratuities, and superannuation or other allowances or benefits or charitable aid and generally to provide advantages, facilities and services for any persons who are or who have been employed by or who are serving or who have served the company or the holding company of the company or a fellow subsidiary of the company or the predecessors in business of the company or of any such subsidiary, holding or fellow subsidiary company and to the wives, widows, children and other relatives and dependants of such persons; to make payments towards insurance; and to set up, establish, support and maintain superannuation and other funds or schemes (whether contributory or non-contributory) for the benefit of any of such persons and of their wives, widows, children and other relatives and dependants; and to set up, establish, support and maintain profit sharing or share purchase schemes for the benefit of any of the employees of the company or of any subsidiary, holding or fellow subsidiary company and to lend money to any such employees or to trustees on their behalf to enable any such purchase schemes to be established or maintained. (14.) To draw, make, accept, endorse, discount and execute bills, warrants, notes or other negotiable or transferable instruments. (15.) To assist in the promotion of or promote any company or undertaking which may appear likely to assist or benefit the company and to place or guarantee the placing of, subscribe or underwrite or otherwise acquire any part of the stock, debentures, debenture stock or other obligations of such company. (16.) To promote by way of advertising the company's products and services in any manner and to reward customers or potential customers and to promote or take part in any scheme likely to benefit the company. (17.) To distribute in specie any of the shares, debentures, or securities of the company between the members of the company in accordance with their rights. (18.) To do all such other things as may be deemed incidental or conducive to the attainment of the above objects or any of them. The objects set forth in each clause shall not be restrictively construed but the widest interpretation shall be given thereto, and they shall not, except where the context expressly so requires, be in any way limited or restricted by reference to or inference from any other object or objects set forth in such clause or from the terms of any clause or from the name of the company. None of such clauses or the object or objects therein specified or the powers thereby conferred shall be deemed subsidiary or ancillary to the objects or powers mentioned in any other clause, but the company shall have as full a power to exercise all or any of the objects conferred by and provided in each of the said clauses as if each clause contained the objects of a separate company. The word "company" in this clause except where used in reference to the word company, shall be deemed to include any partnership or other body or persons, whether incorporated or unincorporated and whether domiciled in the United Kingdom or elsewhere. 4. The liability of the members is limited. 5. The share capital of the company is £1,000 divided into 1,000 shares of £1 each. We, the subscribers to this Memorandum of Association wish to be formed into a Company pursuant to this Memorandum, and we agree to take the number of Shares shown opposite our respective names:- \| NAMES and ADDRESSES of SUBSCRIBERS \| Number of Shares taken by each Subscriber \| \|-----------------------------------\|------------------------------------------\| \| Grant Directors Limited \| ONE \| \| 2nd Floor, Mountbarrow House \| \| \| 12 Elizabeth Street \| \| \| London \| \| \| SW1W 9RB \| \| \| Company Director \| \| \| Grant Secretaries Limited \| ONE \| \| 2nd Floor, Mountbarrow House \| \| \| 12 Elizabeth Street \| \| \| London \| \| \| SW1W 9RB \| \| \| Company Secretary \| \| Dated this day Wednesday, 13 August 2003 WITNESS to the above signatures Evan Jones 373 Southborough Lane Bromley Kent BR2 8BQ THE COMPANIES ACT 1985 PRIVATE COMPANY LIMITED BY SHARES ARTICLES OF ASSOCIATION of QUID NOVI LIMITED PRELIMINARY 1. The Company is a private company and accordingly no shares or debentures may be offered to the public. 2. The regulation contained or incorporated in Table A set out in the Schedule to the Companies (Tables A to F) Regulations 1985 shall apply to the company save insofar as they are excluded or varied hereby and such regulations (save as so excluded or varied) and the Articles hereinafter contained shall be the regulations of the company. 3. In these Articles the expression "the Act" means the Companies Act 1985 including any statutory modification or re-enactment thereof for the time being in force. 4. The following regulations of Table A shall not apply to the company videlicet 8, 24, 40, 41, 60, 64, 73, 74, 75, 81, 89, 94 and 95. 5. In regulation 1 of Table A between the words "regulations" and "the Act" the words "and in any regulations adopting the same" shall be inserted. ALLOTMENT OF SHARES 6.i) The directors are unconditionally authorised for the purpose of Section 80 of the Act, to exercise any power of the company to allot shares up to the amount of the original or any increased share capital of the company at any time or times during the period of five years from the date of incorporation. ii) The directors are also unconditionally authorised for the purposes of Section 80 of the Act to allot redeemable shares up to the amount of the original or any increased redeemable share capital of the company at any time or times during the period of five years from the date of incorporation. Any shares so allotted shall be subject to such terms as to redemption and premium on redemption, participation in profits and as to voting as the members shall from time to time by special resolution decide but shall conform to the provisions of Section 159 to 161 and 170 to 172 of the Act. iii) In accordance with Section 91 (1) of the Act, Sections 89(1) and 90(1) to (6) inclusive of the Act shall not apply to the company. iv) Subject to Section 80 of the Act after the initial allotment any shares proposed to be issued shall first be offered to the members in proportion as nearly as may be to the nominal value of the existing shares held by them respectively unless the company shall by special resolution otherwise direct. The offer shall be made by notice specifying the number of shares offered, and limiting a period (not being less than 14 days) within which the offer, if not accepted, will be deemed to be declined. After the expiration of that period, those shares so deemed to be declined shall be offered in the proportion aforesaid to the persons who have, within the said period accepted all shares offered to them; such further offer shall be made in like terms in the same manner and limited by a like period as the original offer. Any shares not accepted pursuant to such offer or further offer as aforesaid or not capable of being offered as aforesaid except by way of fractions and any shares released from the provisions of this article by such special resolution as aforesaid shall be under the control of the directors, who may allot, grant options over or otherwise dispose of the same to such persons, on such terms, and in such manner as they think fit provided that, in the case of shares not accepted as aforesaid, such shares shall not be disposed of on terms which are more favourable to the subscribers thereof than the terms on which they were offered to the members. PURCHASE OF OWN SHARES 7. Subject to the provisions of Section 162 of the Act the company may with the Sanction of an Ordinary Resolution purchase its own shares (including any redeemable shares) on such terms as the directors may think fit and make a payment in respect of the redemption or purchase on such shares otherwise than out of the distributable profits of the company or the proceeds of a fresh issue of shares and subject to the provisions of Section 173 to 175 of the Act. TRANSFER OF SHARES 8(i) A transfer of a fully paid share need not be executed by or on behalf of the transferee and clause 23 of Table A shall be modified accordingly. (ii) A/ No share in the company shall be transferred except in accordance with the provisions of this clause which (a) shall apply to renunciations or nominations of shares as it applies to transfers thereof and (b) may be waived with the written agreement of all members of the company (in relation to either proposed transfers on sale or proposed transfers of any other kind whatsoever) B/ (a) any member who desires to sell transfer or otherwise part with any share or shares or any interest therein (the vendor) shall give to the company notice in writing of such desire (a Transfer Notice) which shall constitute the company the Vendor's agent for the sale of the share or shares specified therein (the Transfer Shares) in one or more lots at the discretion of the Directors at the prescribed price (as hereinafter defined) and which may, (except in the case of a Transfer Notice given or deemed to have been given under Articles 8(ii)B(b)) contain a provision that unless all the Transfer Shares are sold pursuant to this Article none shall be so sold and any such provision shall be binding upon the Vendor and any applicant for Transfer Shares. (b) If any member shall die or become bankrupt or go into liquidation or being an employee of the company shall cease to be so employed for any reason he or it shall on the happening of that event be deemed to have given a Transfer Notice in respect of the whole of his or its shares in the Company to which the provisions of this clause shall apply and be deemed the Vendor in respect thereof C/ The "prescribed price" shall be such sum per share as shall be agreed between the Vendor and the Company failing which it shall be determined and certified in writing by a chartered accountant as being in his opinion the fair value thereof as between a willing buyer and a willing seller valuing the Company on a going concern basis such accountant to be nominated by agreement between the Vendor and the Company or in default of such agreement by the President for the time being of the Institute of Chartered Accountants in England and Wales and if so nominated the said chartered accountant when determining and certifying the fair value of the Transfer Shares as aforesaid shall act as an expert and not as an arbitrator but without incurring liability to the Vendor or any Member and his certificate shall be final and binding on the Vendor and the other Members. D/ Upon the prescribed price being either agreed upon or determined in accordance with Article 8(ii)C/ (as the case may be) the Company shall forthwith upon receipt of written notice inform the Vendor and other Members of the number of the Transfer Shares as specified in the Transfer Notice and the prescribed price thereof and invite each such Member to apply in writing to the Company within ten working days of the date of that notice (the Application Period) for such maximum number of the Transfer Shares (being all or any thereof) as he shall specify in such application. E/ With the Application Period the Vendor may by written notice to the Company (save where the prescribed price has been agreed by the Vendor or a Transfer Notice has been given or is deemed to have been given under Article 8(ii) B/(b) withdraw the Transfer Notice. F/ Immediately after the Application Period the Company shall allocate the Transfer Shares (or so many of them as shall be applied for as aforesaid) to and amongst those Members who have made applications as aforesaid and in the case of competition pro rata between them according to the number of shares of which they are registered or unconditionally entitled to be registered as holders save that no applicant shall be obliged to take more than the maximum number of shares applied for by him as aforesaid and the Company shall within five working days after the Application Period give notice of such allocations (the Allocation Notice) to the Vendor and to those Members to whom the shares have been allocated and shall specify in such notice the place and time (being not earlier than five working days and not later than fifteen working days after the date of the notice) at which the sale of the shares allocated shall be completed. G/ The Company shall also be entitled to sell to any person or company of whom or which in its absolute discretion it shall approve within the said period of five working days after the Application Period at the prescribed price any of the Transfer shares for which Members shall not have applied as aforesaid and any shares so sold shall for the purposes of this Article be deemed to have been included in an Allocation Notice. H/ The Vendor shall be bound to transfer the shares comprised in the Allocation Notice to the purchasers named therein at the time and place therein specified and if he shall fail to do so the Chairman of the Company or some other person appointed by the Directors shall be deemed to have been appointed attorney of the Vendor with full power to execute complete and deliver in the name and on behalf of the Vendor transfers of the shares to the purchasers thereof against payment of the prescribed price to the Company and on payment of the prescribed price to the Company and execution and delivery of the transfer the purchaser shall be entitled to require that his name be entered in the register of members as the holder by transfer of the same and the prescribed price shall be paid forthwith into a separate bank account in the company's name and held in trust for the vendor but without any obligation to invest the same. I/ During a period of ninety days after the expiry of the time for service of an Allocation Notice the Vendor shall be at liberty to transfer to any person at any price (not being less than the prescribed price) any of the Transfer Shares which he has not become obliged to sell under the foregoing provisions LIEN 9. The company shall have a first and paramount lien on every share for all monies (whether presently payable or not) called or payable at a fixed time in respect of that share and the company shall also have a first and paramount lien on all shares standing registered in the name of a single person for all monies presently payable by him or his estate to the company, but the directors may at any time declare any share to be wholly or in any part exempt from the provisions of this regulation. The company's lien (if any) on a share shall extend to all dividends payable thereon. NOTICE OF GENERAL MEETING 10. Every notice convening a general meeting shall comply with the provisions of Section 372 (3) of the Act as to giving information to the members in regard to their right to appoint proxies; and notice of all other communications relating to any general meeting which any member is entitled to receive shall be sent to the directors and to the auditors for the time being of the company. QUORUM OF MEMBERS 11.i) No business shall be transacted at any general meeting unless a quorum is present when the meeting proceeds to business. Two persons entitled to vote upon the business to be transacted, each being a member or a proxy for a member or a duly authorised representative of a corporation, shall be a quorum. ii) If a quorum is not present within half an hour for the time appointed for the meeting the meeting shall stand adjourned to the same day in the next week at the same time and place or such time and place as the directors may determine and if at the adjourned meeting a quorum is not present within half an hour from the time appointed for such adjourned meeting, it shall be dissolved. PROCEEDINGS AT GENERAL MEETINGS 12.i) At any general meeting a resolution put to the vote of the meeting shall be decided on a show of hands unless a poll is (before or on the declaration of the result of the show of hands) demanded by the Chairman or by any member present in person or by proxy. Unless a poll be so demanded a declaration by the Chairman that a resolution has on a show of hands been carried or carried unanimously, or by particular majority, or lost or not carried by a particular majority and an entry to that effect containing the minutes of the proceedings of the company shall be conclusive evidence of the fact without proof of the number or proportion of votes recorded in favour of or against such resolution. The demand for a poll may be withdrawn. ii) Subject to the provisions of the Act any resolution in writing signed by all the members for the time being entitled to receive notice of and attend and vote at general meetings, or by their duly appointed attorneys, shall be as valid and effective as if the same had been passed at a general meeting of the company duly convened and held. Any such resolution may consist of several instruments in the like form each executed by or on behalf of one or more of the members or their attorneys. VOTES OF MEMBERS 13. Subject to any special rights or restrictions for the time being attached to any special class of shares in the capital of the company, on a show of hands every member personally present shall have one vote only and in the case of a poll every person who is present in person or by proxy shall have one vote for each ordinary share held by him. PROXY 14. The instrument appointing a proxy shall be in writing in any usual common form, or such other form as may be approved by the directors, and shall be signed by the appointer or his attorney, duly authorised in writing, or if the appointer is a corporation shall be either under its common seal or under the hand of an officer or attorney so authorised. An instrument of proxy need not be witnessed. DIRECTORS 15.i) The number of directors shall be determined by the company in general meeting but unless and until so fixed the minimum number of directors shall be one and there shall be no maximum number. ii) A director shall not require any share qualification, but shall nevertheless be entitled to attend and speak at any general meeting of the company or at any separate meeting of the holders of any class of shares of the company. iii) Any person may be appointed or elected as a director irrespective of whether or not he has attained the age of seventy years or any other age, and no director shall be required to vacate his office by reason of his attaining or having attained the age of seventy years or any other age. iv) In the event of the minimum number of directors fixed by or pursuant of these articles or Table A, being one, a sole director shall have authority to exercise all the powers and discretions by Table A or these articles expressed to be vested in the directors generally. v) The directors may, by power of attorney or otherwise appoint any person to be the agent of the company for such purposes and on such conditions as they determine, including authority for the agent to delegate all or any of his powers. ALTERNATE DIRECTORS 16.i) Each director shall have the power to nominate any other director or any person approved for that purpose by resolution of the Board to act as alternate director in his place during his absence, and at his discretion to revoke such nomination, and on such appointment being made, each alternate director whilst so acting shall be entitled to exercise or discharge all the functions, powers and duties and undertake all the liabilities and obligations of the director he represents but shall not be entitled to receive any remuneration from the company. An alternate director shall have one vote for each director he represents, in addition to his own vote if he is a director, but shall not be counted more than once in the quorum. A nomination as an alternate director shall ipso facto be revoked if the appointer ceases for any reason to be a director. ii) Notice of all Board Meetings shall be sent to every alternate director as if he were a director of the company until revocation of his appointment. iii) The appointment of an alternate director shall be revoked and the alternate director shall cease to hold office whenever the director who appointed such alternate director shall give notice in writing to the secretary of the company that he revokes such appointment. POWERS AND DUTIES OF DIRECTORS 17. The directors shall cause minutes to be made in books provided for the purpose: (a) Of the names of the directors present at each meeting of the directors and any meeting of any committees of the directors. (b) Of all resolutions and proceedings at all meetings of the company and of the directors and of any committee of the directors. (c) Of all appointments of officers made by the directors. (d) Of all documents sealed with the common seal of the company. PROCEEDINGS OF DIRECTORS 18.i) The directors may meet together for the despatch of business, adjourn and otherwise regulate their meetings as they think fit. Questions arising at any meeting shall be decided by a majority of votes. A director may, and the secretary on the requisition of a director shall, at any time summon a meeting of the directors. ii) A resolution in writing signed by all the directors shall be as valid and effectual as if it had been passed at a meeting of the directors duly convened and held. QUORUM OF DIRECTORS 19. The directors may from time to time fix a quorum necessary for the transaction of the business at meetings of the directors and unless so fixed the quorum shall be two except: When the number of directors is without contravention of regulation 15 hereof, one: or when the only business of the meeting is to convene a meeting of the members; When the quorum shall be one. DIRECTORS BORROWING POWERS 20. The directors may exercise all the powers of the company to borrow money and to mortgage or charge its undertaking, property and uncalled capital, or any part thereof, and subject to Section 80 of the Act, to issue and create mortgages, charges, memoranda of deposits, debentures, debenture stock and other securities whether outright or as security for any debts, liability or obligation of the company or any third party. INTEREST OF DIRECTORS 21. A director may, notwithstanding his interest, vote in respect of any contract or arrangement with the company in which he is interested, directly or indirectly, and be taken into account for the purposes of a quorum at a meeting at which such contracts or arrangement is considered, and retain for his own absolute use and benefit all profits and advantages accruing to him therefrom. ROTATION OF DIRECTORS 22. Directors shall not be liable to retire by rotation. DIRECTORS GRATUITIES AND PENSIONS 23. The directors on behalf of the company may exercise the powers of the company conferred by clause 20 of the Memorandum of Association of the Company and Regulation 87 of Table A to provide benefits with regard to gratuities, pensions and insurances for any director or member of his family. DISQUALIFICATION OF DIRECTORS 24. The office of director shall be vacated if the director: i) Ceases to be a director by virtue of any provision of the Act or he becomes prohibited by law from being a director. ii) Becomes bankrupt or makes any arrangement or composition with his creditors generally; or becomes of unsound mind. iii) Resigns his office by notice in writing to the company. iv) Is absent from directors meetings for six calendar months without reasonable excuse and without the consent of the other directors and they resolve that he vacate office. THE SEAL 25. The Company may have a Seal if it so wishes. If the Company has a Seal, the Directors may determine who shall sign any instrument to which the seal is affixed and unless otherwise so determined, it shall be signed by a Director and by the Company Secretary or by a second Director. The obligation under Clause 6 of Table A relating to the sealing of share certificates shall apply only if the company has a Seal. INDEMNITY 26. Every director or other officer or auditor of the company for the time being shall be entitled to be indemnified out of the assets of the company against all costs, charges, expenses, losses or liabilities which he may sustain or incur in or about the execution of the duties of his office or otherwise in relation thereto, including any liability incurred by him in defending any proceedings, whether civil or criminal, in which judgement is given in his favour or in which he is acquitted, or in connection with any application under Section 144 or Section 727 of the Act in which relief is granted to him by the Court, and no director or other officer shall be liable for any loss, damage or misfortune which may happen to or be incurred by the company in the execution of the duties of his office or in relation thereto, but this article shall only have effect insofar as its provision are not avoided by Section 310 of the Act. NAMES and ADDRESSES of SUBSCRIBERS Grant Directors Limited 2nd Floor, Mountbarrow House 12 Elizabeth Street London SW1W 9RB Company Director Grant Secretaries Limited 2nd Floor, Mountbarrow House 12 Elizabeth Street London SW1W 9RB Company Secretary Dated this day Wednesday, 13 August 2003 WITNESS to the above signatures:- Evan Jones 373 Southborough Lane Bromley Kent BR2 8BQ
## CONTENTS \| Title \| Page \| \|----------------------------------------------------------------------\|------\| \| Editorial \| 59 \| \| The Philippine Coconut Industry 1934-1950 \| 61 \| \| Charles O. Houston, Jr. \| \| \| PHILCUSA-FOA Pump Irrigation Program in the Philippines \| 91 \| \| Dominador Z. Rosell \| \| \| Pulp and Paper Project \| 95 \| \| Ernesto P. Villareal \| \| \| Climate of Manila \| 101 \| \| Severino L. Koh \| \| \| Marble Development of the Cebu Portland Cement Company \| 116 \| \| Filomeno Duterte, Jr. \| \| \| Typhoons as a Retarding Influence on the East Coast of Samar \| 122 \| \| Mike McIntyre \| \| \| Papers on Philippine Geology and Geography \| 127 \| \| Compiled by Warren D. Smith and Listed by Charles O. Houston, Jr. \| \| \| Geographical Notes: \| \| \| Rice Production Contest \| 134 \| \| Maria Cristina Falls Hydroelectric Program \| 135 \| \| The Eighth Pacific Science Congress \| 135 \| \| One Year of the Irrigation Service Unit \| 136 \| \| Colombo Plan Trainees \| 137 \| ₱3.00 per year ₱1.00 per copy PHILIPPINE GEOGRAPHICAL SOCIETY Soil Conservation Building, Florida St. Manila, Philippines Charles O. Houston, Jr. — Director, Graduate Studies, University of Manila .......................................................... President Porfirio San Buenaventura — Forester and Chief, Administrative Division, Bureau of Forestry ........................................ Vice-President Dominador Z. Rosell — Administrator, Irrigation Service Unit, Department of Agriculture and Natural Resources ................ Secretary Arturo Alcaraz — Geophysicist, Commission of Volcanology, University of the Philippines, Diliman, Quezon City ........ Treasurer PHILIPPINE GEOGRAPHICAL JOURNAL Official Organ of the Philippine Geographical Society Soil Conservation Building, Florida St. Manila, Philippines Charles O. Houston, Jr. ................................................ Editor Mitos Sison .............................................................. Associate Editor Ramon Samaniego ..................................................... " " Arturo Alcaraz ......................................................... " " Dominador Z. Rosell ................................................ Business Manager EDITORIAL CONSULTANTS Mr. W. R. Daly .................................................... Manager of Sales, International Harvester Company of the Philippines Dr. Leopoldo B. Uichanco ............................... Dean, College of Agriculture, U. P. College, Laguna Dr. Jose M. Feliciano ................................. Head, Department of Geology & Geography, University of the Philippines, Diliman, Quezon City Mr. Cayetano C. Pineda .......................... Administrator, Fertilizer Administration, Department of Agriculture and Natural Resources. Mr. Porfirio San Buenaventura ....... Forester and Chief, Administrative Division, Bureau of Forestry. ADVERTISING CONSULTANT Mr. Martin V. Arroyo .......................... Advertising Manager, Standard Vacuum Oil Company of the Philippines The Philippine Geographical Journal is the Quarterly Official Organ of the Philippine Geographical Society and is sent to all members. Non-members' subscription is three pesos a year in the Philippines and three dollars gold in foreign countries payable to the Business Manager, Philippine Geographical Journal. PUBLICATION OFFICE Philippine Geographical Society Soil Conservation Building, Florida St. Manila, Philippines Business correspondence and subscriptions may be sent to the Publication Office or to the Secretary of the Society. For Your Safe Travel TAKE THE M/S GEN. LIM M/S GEN. LUKBAN M/S GEN. MOJICA M/S GEN. SEGUNDO M/S GEN. MALVAR M/S GEN. DEL PILAR M/S GEN. ROXAS M/S GEN. LUNA GENERAL SHIPPING COMPANY, INC. (100% FILIPINO CAPITAL) 401 MUELLE DE LA INDUSTRIA TEL. 27177 & 26892 Accepting Passengers & Cargo MORE water DELIVERED AT REDUCED COSTS! Wherever large volumes of liquids are to be moved . . . whether for irrigation water supply or for pumping water away from flood control and drainage projects . . . there is a JOHNSTON PROPELLER PUMP to meet the condition with great economy. Available in either water-lubricated or oil-lubricated units. Electric, Water or Belt drive. Write us for detailed information. Marsman & Co., Inc. — TRADING DIVISION — Anda and Sta. Lucia, Intramuros, Manila Tel. 3-20-41 Branches: CEBU • ILOILO • DAVAO • BACOLOD Facts About Leyte Rock Asphalt - It is tough and durable—Its density is high, its hardness is uniform. It becomes tougher, harder with constant traffic use. - It is heat and water-resistant—Unlike commercial bituminous asphalt, it does not bleed even if subjected to the heat of the tropical sun. Water can't wear it off. - It is rigid—Unlike other paving material it contains no artificial mixture of gravel or other matter that makes for shifty, bumpy pavements. - It is cheaper than commercial bituminous asphalt or other paving materials. - It requires no cooking or curing—Nature-processed and cured, it is ready for use. - Easy to lay—Crushed and prepared in the mine, it is ready for use. Just lay it on the road or street foundation, pass the roller over it and you have a non-skid, hard surface, ready for heavy traffic. - Road maintenance is economical—Because of its consistent toughness and durability, repair and maintenance are minimized. OTHER FACTS ABOUT LEYTE ROCK ASPHALT It is a natural hard-rock mineral deposit containing asphalt in proper proportions for road paving. Rock asphalt is extensively used in the Western and Middle States, U.S.A. and in the Bicol and Visayan provinces, Philippines. Rock asphalt is found in Villaba, Leyte, the only mineral deposit of its kind in the Far East. The Villaba deposits, according to recent survey, can pave all the roads and streets in the Philippines for an indefinite period. Compliments of MANILA ROCK ASPHALT COMPANY Derham Bldg., Port Area, Manila — Tel. 2-98-11 this symbol contributes more to your life than you know! Where things grow — farms, groves, gardens... where things are made — factories, laboratories... where the people go — by road, air, waterways; Shell quality products, experience, and skill make vital contributions to your life, in some way, everyday. Application for Membership in the PHILIPPINE GEOGRAPHICAL SOCIETY Members are entitled to receive all communications and publications of the society and to participate in all its activities. Secretary, Philippine Geographical Society Soil Conservation Building, Florida St., Manila Please enter my name for membership in the Philippine Geographical Society. Enclosed is five (₱5.00) to cover the entrance fee and annual due for 1953. Name (Please Print) .................................................................. Address .......................................................................................... Occupation ....................................................................................... CUT HERE CHAMBER OF MINES OF THE PHILIPPINES 214 REGINA BUILDING • P. O. BOX 1165 • TEL. 3-22-31, LOCAL 50 MANILA, PHILIPPINES An Association of Mining Companies for the Promotion and Protection of the Mining Industry of the Philippines OFFICERS — 1953 MR. JAN H. MARSMAN MR. B. G. BRANDT President Vice-President MR. T. M. JORDAN MR. NESTORIO N. LIM Vice-President Sec.Treasurer DIRECTORS Mr. Jan H. Marsman Mr. Henry A. Brimo Mr. B. G. Brandt Mr. Marino Olondriz Mr. T. M. Jordan Mr. R. W. Crosby Sustaining Members: BAGUIO GOLD MINING CO. BALATOC MINING CO. BENGUET CONSOLIDATED MINING CO. COCO GROVE, INC. CONSOLIDATED MINES, INC. ITOGON MINING CO. LUZON STEVEDORING CO., INC. MINDANAO MOTHER LODE MINES, INC. PHILIPPINE IRON MINES, INC. SURIGAO CONSOLIDATED MINING CO., INC. Contributing Members: Santa Rosa Mining Co. Antamok Goldfields Mining Co. Demonstration Gold Mines, Ltd. East Mindanao Mining Co., Inc. I. X. L. Mining Co. Masbate Consolidated Mining Co. Paracale Mining Corporation Paracale Gumans Consolidated Mining Co. San Mauricio Mining Co. Suyoc Consolidated Mining Co. United Paracale Mining Co. Antonio M. Rosado (Associate Member) EDITORIAL A Critical Situation Since 1900, various administrations of government in the Philippines have been attempting to solve socio-economic problems which have hindered the development of the Philippines. To date, none of the programs established for this purpose can be said to have had unqualified success and, indeed, the great majority have been failures. Since 1946, the Republic has bent every energy to this problem. Yet, the road is still rough and questionable ahead. Probably as significant a reason as any for this situation has been the indifference exhibited at all levels toward a serious study of geography and related disciplines. The major source for a study of Philippine geology and geography is the work by Warren D. Smith, Geology and Mineral Resources of the Philippine Islands, printed in 1924. From its publication to date, nothing correlative to it has appeared in the country; as of this month, a revised edition, prepared by Mr. Juan Teves of the Bureau of Mines, is due to appear—twenty-nine years after the Smith work. The literature on the principal aspects of geography is scanty indeed, and the program for the Eighth Pacific Science Congress, to be held in Manila during November, does not contain a single paper on Philippine geography, with the exception of monographs to be delivered during symposia connected with Pacific fisheries. The text on geography, currently used in the Philippine school system, is wholly inadequate—and actually misinformative—for the purpose. Post-war numbers of the Social Sciences and Humanities Review of the University of the Philippines lack a single paper on Philippine geography. In five years of teaching in the Philippines, this writer has been unable to discover a single student able to name the provinces of the country, or even the chartered cities. A graduate student, appearing for his orals, and examined on his thesis entitled "Philippine Geopolitics", was unable to supply the coordinates for the Philippines, the name of the longest river in the Philippines, the major ports and their annual tonnage, or even the most basic information on the subject. Official publications of the government—including the Census—have located incorrectly various municipalities, even placing them in provinces other than the one in which they are located. Articles appearing in "magazine" sections of Sunday editions of metropolitan newspapers are superficial in the extreme and rather misleading. The Philippine Geographical Society has expressed its deep concern over this situation from time to time but has so far failed to arouse any interest on the part of individuals most concerned. At a monthly meeting of the Society a resolution was approved requesting Malacañan to take under consideration the necessity of careful study of the problem of the change of geographic names in the country. Dr. Leopoldo Uíchance, of the College of Agriculture, the University of the Philippines, pointed out that this frequent and ill-advised alteration of names causes great confusion in botanical and zoological studies. He was supported by Mr. Arturo Alcaraz, geophysicist of the Philippine Weather Bureau, who mentioned the fact that certain weather stations in the country are known by name to world meteorological stations and societies and when changed require a world-wide revision of data. That the change of names has been capricious, hasty and ill-advised may be seen readily by even a hasty perusal of acts passed by the Commonwealth and Republic governments. The situation in the city of Manila is confusion compounded, with identically-named streets appearing in almost every section of the city. A motion was introduced recently in the City Council to remedy this situation but so far has been sidetracked by other issues. The Society recommends the creation of a board in Malacañan, under the Office of the President, to be composed of one historian, one geographer, one zoologist, and one representative from the Weather Bureau, to advise the President on bills presented for his signature containing changes of names in the Philippines. Each bill of such a nature, the Society recommends, should be sent to the Society for scrutiny, after which it would send the bill accompanied by its recommendation to the committee suggested, who would then forward this to the President for action. The Society recommends that a full year's course on Philippine geography be included in the curriculum of secondary schools, such a course not to be taught as an "integrated social studies" subject. However, in order for this necessary step to be taken, it will first be necessary to provide an adequate text on Philippine geography. This the Society has undertaken as one of its projects. Since the members of the Society are all employed full-time and many are immersed in university work of a very demanding nature, it is questionable whether or not this may be completed within two years. The Society is appalled at the absence of geographers from the Philippine scene, and the problem of securing text-book writers is a frustrating one. A good geographical background or a thorough understanding of geography should be a prerequisite for administrators or others charged with developing and carrying out programs designed to solve the problems facing the people of the Philippines. Geographical place names should be standardized, since at the present many different spellings for a name can be found on several maps available, rivers will have several names for different sections of the same system, spellings vary with dialectical change or the language background of the individual naming a locality and confusion exists as to the language to be used for Philippine maps. Some names are used with the Spanish rendering, some with English and others with the particular language of the map-maker. Regional and area studies should be undertaken to gather together available knowledge of the country, its inhabitants and resource patterns. The very strong regional diversity and heterogeneity so characteristic of the Philippines has never been fully explained or explored. If plans are drawn up without reference to these factors it seems like flying in the face of Providence to suppose that they will ever become effective. The situation is now critical, since plans and programs are in constant preparation and if the present situation is allowed to go unchanged much time, energy and money will have been spent in vain. (C. O. H.) THE PHILIPPINE COCONUT INDUSTRY 1934 - 1950 CHARLES O. HOUSTON, JR.* A major element of the Philippine economy is the important industries derived from the coconut. The coconut palm (Cocos nucifera) is found in abundance throughout the tropics all over the world and finds a favorable home in the Philippines. Every portion of the tree has its utility — the nut, the leaves, the trunk — and it has occupied a position of importance in the native economy of all the tropical Asiatic countries. In the Philippines, it has been of primary importance since man first appeared there. The nuts supply food, with many different methods of preparation, and several kinds of drinks, from the pleasant unfermented water (erroneously called “milk” by Westerners) taken directly from the nut, to less pleasant beverages produced by fermenting and distilling.$^1$ The juice is drawn from the unopened flowers and is boiled down to sugar or is fermented and distilled producing a drink called “arrack” in certain parts of the world and “tuba,” in the Philippines. The young bud, cut from the top of the tree, produces a “cabbage” highly esteemed by the people. The trunk yields soft lumber much used throughout the Islands for furniture and firewood. The leaves are woven into a great variety of useful products, fans, baskets, receptacles of one sort or another, and finds a further use as roofing. The shell of the nut is made into household vessels, utensils and a few implements. The external husk is excellent for polishing floors, being cut in half, turned faced down, and then briskly rubbed back and forth. The husk also provides coir from which ropes, cordage, brushes, door-mats, and many other articles are produced. The meat of the nut is the source of oil widely used for soap and margarine making. When broken into small pieces and dried (either in the sun or in ovens) it becomes copra. A general rule of the thumb is that 1,000 nuts will produce around five hundred pounds of copra, from which is extracted about twenty-five gallons of oil.$^2$ In the Philippines, at least 4,000,000 people are wholly or partially dependent upon the coconut industry. The average annual value of coconut products exported in the period, 1927-1936, was $34,832,455, or 27 per cent of the total value of all exports, and represented some 34 per cent of the world’s production of copra.$^3$ In 1936, the Philippines, being the second largest producer of coconuts in the world, had a total of 600,000 hectares containing over 115,000,000 trees, of which 75,500,000 were bearing. Of the world’s total of over three million hectares, the Philippines thus had nearly 20 per cent. The most important producing provinces were, in that year, Tayabas, Laguna, Cebu, Oriental and Occidental Misamis, Albay, Samar, Leyte, Negros, Bohol, Camarines Sur, Romblon and Pangasinan. The total of the provinces of Laguna and Tayabas alone was about 210,000 hectares, or 35 per cent of the national total. Director, Graduate Studies, University of Manila; President, Philippine Geographical Society $^1$“Milk” is expressed from the nut and is not the fluid found in the hollow of the kernel. $^2$The Encyclopaedia Britannica, (Chicago: The Encyclopaedia Britannica, Inc., 1947 edition), Vol. 5, p. 950. $^3$For information with regard to the industry and production practices in the Philippines, see: The Coconut Industry in the Philippines, Department of Agriculture and Commerce, (Manila: Bureau of Printing, 1939), 19 pp. There are listed over one hundred variety names for the coconut in the Philippines many of which undoubtedly refer to the same variety in various areas of slightly different dialectical change. There are probably no more than thirty distinct varieties, of which the most important for copra are the Romano and the Laguna. Other varieties serve various local purposes, some like the Macapuno, being used for ice cream and other delicacies. Costs of production varied widely throughout the Philippines, depending, naturally, upon local conditions. In Tayabas, the leading producer, the total average cost, in 1936, of producing 100 kilos of copra was P5.60. This resulted in a net average income of P69.58 per hectare. The situation of the coconut planters in 1933 and 1934, when prices dropped to P4.48 and P3.98 per 100 kilos is better imagined than described. For domestic purposes, the Philippines manufactured four important products from the coconut: oil, shredded coconut, soaps, and vegetable lard. As a by-product in the manufacture of oil, copra cake and meal were of some importance at various periods. This item ranked third in the list of coconut export products. Other minor products were fatty acids, glycerine, charcoal and coir. Margarine, during the early Commonwealth, was rather inferior in quality, and had a difficult time competing with foreign products. Fatty acids were the subject of much discouragement and agitation in the United States, and glycerine and charcoal had only been in demand during the First World War. The coir industry, while capable of expansion, was believed only able to succeed when costs were reduced to levels prevailing in India. A writer, in 1937, believed that the domestic consumption of coconut products was "so tremendous that evidently, it has virtually reached the saturation point." Therefore, for it to survive, he believed, "it must at least maintain its export trade." From the vantage point of later years, there seems little reason to believe that domestic consumption had reached the saturation point. Since the coconut industry was particularly geared to American economy, its fate was determined by far-away men whose irresponsibility often brought confusion, frustration and destruction to many. The coconut industry was most sensitive to stimuli from abroad. The value of the product in 1929 was P89,093,620; it declined to P27,146,650, in 1934. It then rose to P92,126,490, in 1938, the greatest rise taking place in homemade oil and the utilization of nuts for food, and then dropped to P28,013,002 in 1939, a fall of over P64,000,000 in one year. In that year, the production of copra fell off P56,000,000, home-made oil some P200,000, and the utilization of nuts for food some eight million, while the production of tuba increased almost P1,000,000. The first blow struck at the Philippine coconut producer was delivered by the American Congress (at the instigation of domestic oil producers) through the Revenue Act of 1934 (H.R. 7835). Aside from the selfish motives which inspired such a step, it was disastrous to the economy of the Islands, contrasting sharply with the "altruistic" motive which supposedly inspired the Tydings-McDuffie Act. Observers have --- 4 Domingo B. Paguirigan: "Philippine Coconut Industry," The Commercial and Industrial Manual of the Philippines, 1937-1938, p. 246. 5 Ibid. 6 Ibid. 7 Yearbook of Philippine Statistics: 1946, (Manila: Bureau of the Census and Statistics, 1947), p. 153. 8 Grayson L. Kirk, Philippine Independence, (New York: Farrar and Rinehardt, 1936), Chapter V, pp. 102-135. pointed to the fact that the processing tax provided by the Act was to be returned to the Philippine Commonwealth Government, thus providing that Government with a handsome revenue. It is interesting to note, as Professor Kirk points out, a provision in that amendment which stipulated that such payments were to cease if the Philippine Government provided any subsidy to be paid the producers of copra, coconut oil, or allied products.\textsuperscript{9} A writer heatedly termed this act "The Congressional vendetta," calling the legislation senseless and unjust.\textsuperscript{10} "Even when considering the economic unwisdom of such a tax," he said, "and questioning the legal right of Congress to impose it, the question of the morality of such an action stands foremost.... Never in the 35 years of American sovereignty over the Philippines has the United States Government ever perpetrated an act of unmitigated injustice.... But of recent years, the United States Government, particularly the Congressional branch, has passed or has sought to pass laws in utter disregard of the interests of the people over whom or against whom these laws apply."\textsuperscript{11} Under the circumstances, this must be considered a temperate statement. Governor-General Murphy was moved to wire, on February 24, 1934: "Intimate contact with the situation locally forces me to the conclusion that the unlimited application of the tax will provoke a near disaster in the economy of the Philippines. The general feeling is pronouncedly against the moral right of the United States to legislate so severely against a territory under the flag as practically to destroy an industry on which more than 3,000,000 people are dependent."\textsuperscript{12} But, in 1934, the United States Congress was not interested in morality nor in the future of a people, of whom, President McKinley stating the reason for acquiring the country in 1898, said: "...there was nothing left for us to do but to take them all, and to educate the Filipinos, and uplift and civilize and Christianize them, and by God's grace do the very best we could for them, as our fellow-men for whom Christ also died."\textsuperscript{13} In January, 1934, the copra market was noted for its instability and most crushers were unwilling to enter into further contracts due to the unfavorable world market conditions. Spain had placed restrictions upon copra imports and it was rumored that France would take similar action. The excise tax by the United States, capped the climax and forced the exporters and oil millers to curtail purchases while they awaited further developments. The continuing uncertainty was greatly influenced by developments in Washington. One bill after another was introduced into Congress, providing for Philippine independence, which generally was welcomed by Filipinos, but all these bills were distinguished by their vagueness with regard to the future of Philippine economy or their stringent provisions with regard to it. This state of affairs was not conducive to calmness in the Philippines and brought about a general instability in those exports products most sensitive to conditions in America. The remarks of men like Rep. Harold Knutson of Minnesota and Rep. E. P. Burke of Nebraska, upon the independence question and the restriction of coconut oil created panic in business circles in the Philippines. For example, Rep. Knutson opposed the Hawes bill for independence, because it did not offer sufficient protection for the dairy interests in the \textsuperscript{9}Ibid., p. 133 \textsuperscript{10}The Congressional Vendetta", Editorial, Philippine Magazine, XXXI, No.3 (March, 1934), p. 108. \textsuperscript{11}Ibid., \textsuperscript{12}Ibid. \textsuperscript{13}Kirk, op. cit., p. 17. United States. Rep. A. C. Schallenberger of Nebraska stated that coconut oil was largely replacing beef and pork products and should be greatly restricted in any independence measure. The movement for granting independence became a stampede, and Representative Magnus Johnson introduced a bill calling for independence within thirty or forty months but saying nothing definitive with regard to the economy of the country.\textsuperscript{14} Mr. H. M. Cavender, President of the American Chamber of Commerce in Manila, declared that the Chamber should maintain an aggressive policy in opposing the mimical measures that were being considered in Washington.\textsuperscript{15} The Philippine Chamber of Commerce went on record against the excise tax, stating that it would cause hardship to some 4,000,000 people, affecting an even larger number than "unfair" restrictions on the sugar industry. Governor-General Murphy proposed that either all Philippine oil and copra used for non-edible purposes be exempted from the tax, or that a quota of 200,000 long tons of oil be exempted. The Secretary of War, George C. Dern, addressed a letter to the House Ways and Means Committee, objecting strongly to the tax on oil and copra.\textsuperscript{16} On February 1, the directorate of the Coconut Planters Association adopted a resolution which declared that while a solution of the Philippine-American relationship was pending, the Filipinos were entitled to the "full enjoyment of economic opportunity under American sovereignty" and that the proposed tax "would be a flagrant violation of this fundamental right."\textsuperscript{17} In spite of these objections and proposals, the House Ways and Means Committee refused to reconsider its action and endorsed the proposal to levy an excise tax on copra and oil. On February 4, Dern again protested against the tax, stating that it would defeat its purpose of raising revenue by destroying Filipino purchasing power. The Committee, however, countered with the statement that it was motivated as much by the desire to remove competition from the field of animal and vegetable fats as by the idea of raising revenue.\textsuperscript{18} As pointed out by Professor Kirk, this argument was without adequate basis and had been repeatedly refuted by experts who appeared before the Committee.\textsuperscript{19} The Governor-General again, on February 6 and 9, protested the tax stating that it was equal to 200 per cent of the current price and would work "incalculable harm to the Philippines without advantage to continental United States...." It meant, he said, financially, the "bankruptcy of eight important provinces... questionable solvency for ten others.... Socially it will entail widespread distress and disaffection among the people." He asked for a two or four years' trial of the limitation plan previously proposed by him. The House Committee, at that time, had voted three times to sustain the tax. On February 15, President Quezon stated that he had received the promises of Tydings and McDuffie to oppose the excise tax and Representative R. R. Eltse, of California, sharply attacked the measure, stating accurately that "American farmers and all our people have been misled by false propaganda...." "The tax," he continued, "would be detrimental to business generally on the Pacific coast and also to the dairy industry in all parts of the country since the Philippines is the largest \textsuperscript{14} \textit{Philippine Magazine}, XXXI, No. 3 (March, 1934), pp. 90-92. \textsuperscript{15} \textit{Ibid}. \textsuperscript{16} \textit{Ibid}. \textsuperscript{17} \textit{Ibid}. \textsuperscript{18} \textit{Ibid}. \textsuperscript{19} Note 8 above. export market for American canned milk.... It seems strange to me that the predominantly what Democratic Ways and Means Committee should have given birth to this renegade high tariff measure."20 In the Philippines, Under Secretary Vargas appointed a committee to study the possibilities of utilizing coconut by-products in local industries, realizing that there was little chance that the tax would be defeated. Rumors were circulated, following the approval of the Tydings-McDuffie Independence Act, that Congress would fail to enact the excise tax. The rumor was started, it was reported, to bring about a rise in prices, which, however, failed to materialize. Two days after the passage of the bill, the Senate finance committee voted to retain the excise tax but to reduce it from five to three cents a pound. Local producers saw no relief in this, however, and predicted the destruction of the industry. Major William B. Anderson, Manila businessman, stated that ten or twelve years of slow economic strangulation was worse than a quicker death in three years. He predicted that invested capital would liquidate and that new capital would not enter, and that the only certainty was that the Japanese would soon control the interior trade held by the Chinese and also the import and export trade of the Islands. "It will probably be found highly inimical for Americans to continue."21 Mr. S. Dazai, manager of the Yokohama Specie Bank and head of the Japanese Chamber of Commerce, in Manila, stated happily: "I am glad that the Philippines is now to obtain her independence. No, I don't think Japanese capital will fly away; on the contrary, there will be greater inducement for it to come into the Philippines. I am afraid that it is the American capital which will fly away. I am glad of independence because I believe that the Orient should be for Orientals. We are like one people. Oriental capital will come to the Philippines. Japan will play a great part in the economic development of this country."22 No prophecy was ever more accurate and none received less attention on the part of those most concerned. How short the years were until fulfillment arrived with the checks of Japanese Government subsidies and the bayonets shining under a December sun. The excitement over the excise tax, which manifested itself largely in restricted buying and which had brought a gradual decline in the market, gradually subsided at the end of 1934. The lowest ebb was reached in May, June, and July, of 1934, after which the market gradually recovered. During this lowest period, the industry might well have come to a complete halt had it not been for the development of factors outside the immediate Philippine economy. The most important of these, and the one which permitted the industry to regain its feet, was the advantage derived from an inflated currency. This enabled the Philippines to undersell all other producers. An American economist in the Philippines at the time stated that with the former gold currency, the Philippines would have been unable to continue production, as European price equivalents would have been lower than the production cost.23 Despite the slight advantage gained by inflated currency, the price was so low that many planters in areas where transportation and marketing costs were high, found themselves unable to sell at cost and fed --- 20Philippine Magazine, XXXI, No. 3 (March, 1934), p. 94. 21Ibid., XXXI, No. 5 (May, 1934), p. 179. 22Ibid. Since this bank was subsidized by the Japanese Government, Mr. Dazai's comments conceivably represented the official position on the subject. 23Norbert W. Schmelkes: "The 1935 Prospects for the Coconut Grower," Philippine Magazine, XXXII, No. 3 (March, 1935), p. 136. the nuts to their hogs or extracted oil for their own consumption. In so doing they withdrew from the market a considerable quantity of copra. Since trees increase in the Philippines by from two to three million annually, the amount of copra which normally would have been shipped was considerable, one estimate placing it at 60,000 tons. At this point, fate stepped in. In 1934-35, the American farmer faced one of the worst droughts in history. In addition, the United States was experimenting with crop reduction programs under the AAA. Until this time, the United States had provided but a small market for Philippine copra meal. But with the farmers unable to feed their cattle with home-produced feeds and facing the possible loss of entire herds they frantically turned to copra meal. The tremendous increase in consumption of a commodity which had been selling in the Philippines at prices barely above its value as fuel, increased prices some 250 per cent to the delight of the starving coconut producers. This naturally increased the value of copra and as soon as the shortage in fats became evident and previous stocks of oil were consumed, the market boomed.$^{24}$ This would have been enough to bring a certain prosperity to the producer, but fate again intervened in a typhoon which swept through the major coconut-producing provinces, causing prices to soar again. By the beginning of 1935, prices were high enough to be considered "spectacular" and caused considerable speculation whether the level could be maintained. In addition, the two-cent excise tax differential in favor of the Philippines had the effect of shutting out other copra producing countries and undoubtedly brought benefits to Philippine producers. Had the excise tax not been imposed, there is little doubt that the crisis of 1934-1935 in oil and copra meal could have been avoided, which would have meant a considerable saving to American farmers—one of the major groups eager for the imposition of the tax. The Joint Preparatory Committee received a brief, in 1937, on the state of the coconut industry, setting forth its nature, history, activity, importance to Philippine economy, extent of dependence upon the American market, and the probable fate of the industry as a result of the Tydings-McDuffie Act and other possible future legislation leading to Philippine Independence. Unless favorable amendments were added to the Independence Law, the excise and export taxes as provided therein would be disastrous to the industry, it was believed. It would result, they said, in the withdrawal of three-fourths of hectarage from cultivation, adversely affecting some three million people and lowering the revenues of the Philippine Government from the P7,000,000, of 1935, to a probable P1,750,000.$^{25}$ The industry believed that it was mutually beneficial for the status quo to be maintained, since the trade in oils and their derivatives was supplemented by certain advantages to America, such as freight, insurance, continued purchases by the industry in America, and interest payments on rather extensive American investments. And since the United States had to import seeds and oils from other countries because of the excise taxes reducing imports of the Philippine oil into America, this showed that the United States had to import these items because of industrial demands. Since this was so, and because of the long association of the two countries, coupled with the mutual trade advantage "and the responsibility shared by American capital in the promotion of the Philip- $^{24}$Ibid. $^{25}$"Brief," edited by D. B. Paguirigan, The Commercial and Industrial Manual of the Philippines, 1937-1938, p. 249. pine coconut industry," there should exist sufficient motive to waive any discrimination against an industry so vital to the Philippines and its people.\textsuperscript{26} During the early years of the Commonwealth, unit prices of copra continued to rise until, by 1938, they had reached the highest annual level since 1930. The quantity of copra exported was lower, but the total shipped to the United States was greater than at almost any time in a decade. The United States, however, continued to legislate against Philippine oil, altering its competitive position adversely, with the Revenue Acts of 1935 and 1936. According to the Joint Preparatory Committee Report: The Revenue Act of 1935 amended that of 1934 so as to place a compensatory tax on imported articles manufactured or produced in chief value from taxable oils. The rates were substantially the equivalent of the excise taxes which would have been collected had the oil ingredients been imported into the United States in the form of oil. The principal effect of this law on the Philippines was to subject Philippine-made fatty acids, vegetable lard, soap and other products made from coconut oil to the equivalent of the excise tax imposed on coconut oil. The Revenue Act of 1936 amended both of the preceding revenue acts. The most important changes, from the standpoint of the Philippines, were the extension of the list of taxable oils and the increases in rates on some of the oils already taxed. The rate on coconut oil was not changed.\textsuperscript{27} The Joint Preparatory Committee also pointed to the fact that the so-called preferential position accorded Philippine coconut oil did not work "as much to its advantage as might appear to be the case" since "it is still obliged to sell in the world market...."\textsuperscript{28} Although the Revenue Act of 1934 did not change the status of Philippine copra, experts believed that its effect had been favorable because it tended to equalize the competitive position of copra and oil "relative to other articles on which new or higher taxes have been imposed."\textsuperscript{29} During 1936, for the first time in some years, the European buyers were very active and offered better prices than the United States' market. However, uncertainty prevailed and the market fluctuated widely, with speculative buying and selling controlling the situation to a large degree. After May, a general rise in price took place reaching levels which had not been considered possible earlier in the year, with a peak in October of a monthly average price of P14.25 for resecada (dried). The market for coconut oil was much quieter with little business for bulk shipments. Price fluctuations occurred over a narrow range, with quotations at P0.20 for January and P0.36 for December. Copra cake and meal showed substantial increases, with the continued drought in the United States driving prices for foodstuffs to new highs. While this normally would have meant a boom in prices, due to the sanctions applied against Italy, and the consequent heavy demand for Philippine oils, many \textsuperscript{26}Ibid. \textsuperscript{27}Joint Preparatory Committee on Philippine Affairs, \textit{Report of May 20, 1938}, Vol. I, pp. 63-64. The change in the act of 1935 went into effect September 30, 1935, and the second, August 21, 1936. \textsuperscript{28}Ibid. \textsuperscript{29}"U.S. Revenue Act Boosts P.I. Oil," \textit{Philippine Journal of Commerce}, XIII, No. 1 (January, 1937), p. 13. The subtitle of this article read: "Revision Apparently Gives Advantage to a Number of Products." The reservation was well taken in view of later developments. See also: "Review of Coconut Products for 1938," \textit{The Commercial and Industrial Manual of the Philippines, 1937-1938}, p. 250. planters, in order to take advantage of the high prices while they lasted, harvested all the nuts possible, including green ones which they converted into low quality copra. This practice usually results in injuries to the trees and lowers the bearing capacity.\textsuperscript{20} In addition, a severe typhoon in December swept across Southeastern Luzon causing extensive damage to trees in Albay, Sorsogon, and Camarines. These areas had barely recovered from the severe typhoon of 1934. Also, Pacific coast crushers, in the United States, faced difficult days because of the great shipping strike which paralyzed both current and future operations. In an effort to aid the shaky coconut industry, the National Assembly, on October 14, 1936, passed Commonwealth Act No. 50, "directing the Secretary of Agriculture and Commerce, the Philippine National Bank and the National Development Company to establish, operate and maintain warehouses for copra and other marketable products." Any municipality wishing such a warehouse erected was directed to apply to the Government through these agencies which would examine the request and, if justified, invest funds in the enterprise. The municipality would make the proper arrangements with the Bank or the NDC for reimbursement of the money invested. The problem of adequate storage plagued the coconut producer throughout the period. It had always been a source of dissatisfaction, since so much of the planter's profits were eaten away by storage charges, and with the unstable conditions facing planters during the Commonwealth period, it added fuel to discontent. The problem was never solved. In 1937, the better prices the industry hoped for were not forthcoming because of a shipping strike on the Pacific Coast. The bulk of the copra exports to the United States normally went to these Pacific Coast ports. It so happened that at that time stocks on the coast were low and prices were gradually rising. The benefits to be derived from the situation, however, were denied the Philippine producers as they were unable to get their shipments to the buyers. Hemp and cordage were little affected by the strike since the demand for lower grades of fiber generally came from Europe and from the eastern United States. It is difficult in the United States to appreciate the tremendous importance of a single crop in the life of the people of the Philippines. Americans could not understand why such an outcry was raised in the Philippines over the control of coconut oil and copra. It should be realized that at the time of the beginning of the Commonwealth, there were some 115,312,400 coconut trees in the Philippines, of which about 75,414,200 were bearing and were distributed over 608,360 hectares or about 15 per cent of the total cultivated area of Philippine lands. When we say that about four million people were wholly or in part dependent upon the coconut industry it is difficult to grasp just what this means to the country. It may be better appreciated if it is realized that over 75 per cent of the taxes of the province of Tayabas (now Quezon) came from the assessment on coconut lands and trees; between 50 and 75 per cent for Laguna, Marinduque, Masbate, Oriental Misamis, Romblon and Zamboanga; and between 15 and 50 per cent for Agusan, Bohol, Camarines Norte, Camarines Sur, Capiz, Cotabato, Lanao, Mindoro, Palawan and Surigao. The government assessment, in 1933 amounted to P327,099,255, from which the government derived a direct land tax of P2,872,599, including revenues from industries resulting therefrom. It was estimated that over 4 million pesos was derived from the coconut industry annually by the govern- \textsuperscript{20}\textit{Ibid.} ment. To a large extent, interisland shipping depended upon the coconut industry.\textsuperscript{31} Imagine, for a rough comparison, seventeen of the States, of the United States, containing some 33,000,000 of the population of the country in 1940, with all rail, truck, bus and water lines of transportation between New York and Los Angeles, wholly or in part dependent upon one crop. Imagine further what the fate of these states, transportation facilities, and people would be were their future dependent upon Europe as the major buyer for the crop (some 96.5 per cent). Imagine still further that while this crop, corn for example, was selling in the American export centers at ten to fifteen cents a bushel, a tax was imposed in Europe of eight cents a bushel, this tax then being remitted to the United States Government with the proviso that it was not to be used, in any way, to aid the corn producers or those engaged in the corn business, directly or indirectly. This, in effect, was the fate of the coconut producers in the Philippines. Coconut producers, in 1937, were making a profit of $0.0006 per pound on coconut oil. A five per cent export tax was provided by the Tydings-McDuffie Law (Section 6), to be levied during the sixth year following the inauguration of the Commonwealth Government, which amounted to $0.001 per pound. It was further provided that, in the tenth year, the export tax would amount to 25 per cent or $0.005 per pound. The reason for the panic in the coconut-growing regions when these provisions were announced should be self-evident. They believed that, with the imposition of this export tax, their industry was faced with extinction. The situation was sadly ironic in that the industry had been developed largely through the initiative of buyers in the United States who were now attempting to destroy it—each of which actions taking place with no thought of resultant effects upon Philippine economy, either in the United States or in the Philippines. In order to meet a highly uncertain future, the old Philippine Coconut Planters' Association was reorganized into the Philippine Coconut Association, including for the first time exporters, millers and industrialists. The new organization petitioned the Government for the reservation of P1,000,000, from the proceeds of the excise tax fund, for research, educational promotion of the industry and publicity in the United States. Their hopes, however, were thwarted. This Association was formed at the initiative of the Bureau of Plant Industry who had arranged a meeting of interested parties at the time of the 1937 Philippine Exposition. Its aims were to improve generally the industry through seed selection, fertilization, cover-cropping, research into methods of improving the quality of copra and to encourage industrialization and commercialization of the numerous by-products. Another purpose was to keep members informed about developments in the industry and, of course, one of its major objectives was to fight and eliminate the objectionable provisions in the Independence Law and other American tax legislation. The Director of the Bureau, Hilarion Silayan, instructed field men in the Bureau to aid in all possible ways in the organization and direction of local chapters of the Association. On October 18, 1937, President Quezon addressed a message to the National Assembly. In the course of this famous speech, dealing primarily with the demand for independence in 1939, he touched upon the subject of social justice under the Commonwealth and the means of carrying \textsuperscript{31} "Philippine-American Trade Relations," Bureau of Commerce, \textit{PJC}, XIII, No. 2 (February, 1937), pp. 12-13. out this program. Since the main problem was one of finance, he mentioned that "fortunately for us a new source of income has come to our hands that will facilitate the carrying out of our program of social justice and economic readjustment." This was the proceeds of the excise tax (amounting to P95,507,237.30 at the end of June, 1937) in the American Federal Treasury, "the transfer of which amount to the Treasury of the Philippines I had secured before I left America on my last trip." He stated that the sum was then ready for appropriation. The final decision as to how the fund shall be spent is, of course, yours. But in the exercise of my constitutional prerogatives I shall take the liberty of making some suggestions regarding the purposes for which this money should be spent. The first thing that we must bear in mind is that this fund does not constitute an ordinary income of the Government upon which we may depend for recurring obligations. When Independence shall have been granted, this source of our income will cease. Were we to defray from this fund services that we cannot maintain once this income is terminated we have thrown away this money thus spent. We must therefore limit the use of this fund for what might be termed capital investments or for self-supporting enterprise. Above all we should use this fund for national objectives, for purposes where the greatest good may be derived by the Filipino people. Concretely, I recommend that this fund be devoted to the following purposes: 1. To improve the sanitary condition of centers of population by constructing water systems or artesian wells. 2. For combating malaria where there is assurance that it can be done at reasonable expense. 3. For the prevention of tuberculosis and establishment of more sanatoriums, as it is well-known the white plague is the worst scourge afflicting our race. 4. For the building of leprosariums.... 5. For extending free dispensary service to the poor not only in centers of population but also in outlying barrios.... 6. For building public schools in every barrio where there is a sufficient number of children justifying the opening of the school. 7. For opening national highways and helping in the construction of provincial and even barrio roads whenever the respective provinces and municipalities pledge themselves to maintain the roads constructed, and in the case of barrio roads where the volume of traffic on said roads also justifies their construction. 8. For construction of office buildings for the National Government so as to reduce, if not eliminate, the continued expense in rents. 9. For the purchase of large landed estates and their resale in small lots to the actual occupants thereof. 10. For the development of water power, the reforestation of denuded areas, the colonization and development of Mindanao; and 11. For the financing of a long range program of economic adjustment necessary to prepare the country for the new industries which at the same time will give work to the unemployed. The disposition of these funds was to provide many heated discussions for the remainder of the Commonwealth period. Quezon's eleven --- 32 PJC, XIII, Nos. 10-11, (October-November, 1937), p. 15. This was not accurate. The transfer of funds was automatic. The President had nothing to do with the transaction. 33 Ibid. Italics supplied. We may consider that the money was thrown away since the services organized were unable to continue in operation without extensive appropriations by the Government. A great portion of such services, begun with these funds, have long since disappeared. objectives were never reached, nor, with the exception of a few enterprises touching immediately upon the benefits to a few branches of the government and the construction of one highway in Mindanao, were they ever seriously started. There was much objection to using the funds for the redemption of the nation's bonded indebtedness as had been suggested by many in the Government. Secretary of Finance Antonio de las Alas, Auditor General Jaime Hernandez, and Executive Secretary Jorge B. Vargas, suggested that the funds be set aside for the creation of a merchant marine. The Auditor General stated that all Philippine bond issues had fixed dates of maturity, and were provided with sinking funds adequate for the purpose. Suggestions were made by the others also that the funds to establish the capital stock of the National Development Company would probably come from the excise tax funds. The Company was to be capitalized at P20,000,000, half of which had been provided. However, even the entire amount would be insufficient, some stated, to meet the demands of a merchant marine, necessitating application of excise tax funds.\textsuperscript{34} A Filipino columnist in America, Vicente Villamin, claimed that the application of the excise tax upon Philippine copra was beneficial and would prove a godsend to the industry and the Government.\textsuperscript{35} The Secretary of Finance, however, felt differently, and immediately began laying plans to initiate strengthening measures for other industries as well as remedial measures for the coconut industry itself. His first plan was the establishment of abaca and coconut centrals which would be charged with research and other activities leading to a general improvement in the respective industries. He instructed Mr. M. L. Roxas, agricultural adviser in Malacañan, to proceed to the Bicol and Tayabas to select localities. The \textit{Bulletin}, in an editorial of May 8, 1937, suggested that a large portion of the funds be used to strengthen and rehabilitate the government pension system to reward faithful and suffering employees.\textsuperscript{36} President Quezon never considered this suggestion. High Commissioner McNutt counselled the wise spending of the funds whatever was done with them, and Assemblyman Felipe Jose and former Representative Francisco Varona suggested that a portion of the funds be utilized for the development of agricultural colonies in Mindanao. This suggestion, too, received short shrift. The Department of Interior announced that it would submit a plan to the cabinet suggesting the use of P10,000,000, of the fund, for the improvement of sanitary facilities in various communities throughout the nation. This worthy plan was also ignored. Secretary Vargas, replying to requests of planters, stated definitely that the President would not approve the use of any portion of the funds for aiding the coconut industry, either directly or indirectly. One definite plan, he stated, would be the extension of the railroad lines to Legaspi, Albay; since this line would extend rail service for planters, he believed it would be of benefit to them. In addition, he suggested that some way might be found to use a portion of the funds for the introduction of new drying methods for copra. If this proved impossible, he thought that perhaps a portion of the funds provided by the New Industries Act could be made available for that purpose.\textsuperscript{37} \textsuperscript{34}\textit{The Commercial and Industrial Manual of the Philippines, 1937-1938}, p. 270. \textsuperscript{35}\textit{Ibid.} \textsuperscript{36}\textit{Ibid.}, p. 271. \textsuperscript{37}\textit{Ibid.}, p. 272. Manuel de la Fuente, president of the Manila city board, advanced an excellent suggestion for the disposition of the funds. This was to provide a revolving calamity fund of P10,000,000, for the improvement of housing conditions in Manila, the extension of small loans to laborers and low salaried employees, with preference to be given to victims of calamities. Councilor Hermenegildo Atienza declared that the proposal was "one of the most progressive measures to ameliorate conditions of masses" yet advanced. "We cannot talk of social justice and reform," he said, "without taking the first step essential to any program of this nature — the improvement of the housing conditions of the masses." Councilor Jose Advancula declared that such a fund would be "providential" and Councilor Celestino Ramos said that it was "a right step in the right direction." He was supported by Councilor Vicente Alindada who declared that it was an excellent investment from which "not only the government but the nation may profit considerably in the form of a healthy and contented populace." The city councilors, however, spoke in vain. Secretary Eulogio Rodriguez, of the Department of Agriculture and Commerce, stated that the funds accruing from the excise tax belonged to the people at large and would be spent for the economic development of the country as a whole, the funds being apportioned to the different provinces. He believed that the plan to use P23,657,000, for the realization of a five-year road and highway construction program in Mindanao and Sulu, was excellent, and should be supplemented by aid to the fishing industry and other infant industries. The Philippines Herald* pointed to a danger not thoroughly understood by most observers of the time. This fund would accrue to the Government only so long as Philippine copra and oil were sold in the American market. As early as 1937, signs were apparent that exports to the United States were decreasing in value. If this trend were to continue, naturally the amounts accruing to the fund would decrease. Since the end result would be the crippling of the coconut industry as well as bringing serious injury to the financial structure of the government, the paper counselled full support to the Filipino members of the Joint Preparatory Committee in their struggle to adjust satisfactorily the economic relations between the two nations. In September, 1937, the Philippine-American Joint Preparatory Committee returned to Manila after a tour of inspection and investigation in the Bicol, Visayan and Mindanao provinces. It then conducted its third and last series of public hearings. Various briefs, prepared by the different chambers of commerce, were submitted, supplemented by oral arguments delivered by the authorized representatives of these organizations. The general tenor of the pleas was for an indefinite continuation of the free-trade relations. If this proved unacceptable, they advocated a reasonable period of time for readjustment before the free-trade period was terminated. The first day of hearings was given over to the coconut industry, the representatives of which pleaded for a fairer treatment through the abolition of the excise tax and the continuation of existing trade arrangements between the two countries, even after independence. The Philippine Coconut Association recommended that copra should continue on the free list; that no excise tax which did not apply equally --- 38 Ibid. 39 Ibid., pp. 272-273. 40 Ibid., pp. 273-274. 41 Ibid., p. 274. to Philippine and foreign products be levied; and that a duty-free quota of coconut oil and desiccated coconut, after independence, be established to continue as long as the United States needed to import these commodities. Mr. Maximo M. Kalaw, acting president of the Association, appealed to the committee, saying: "We do not wish to infringe upon the American farmers or upon the profits derived by them, but in view of the fact that the United States does not and cannot supply its own industrial oils and fats, the Philippines... should be entitled to even more advantageous position than that of any competing oils and fats produced in other countries not under the American flag." Mr. Kenneth B. Day, spokesman for the industrialists, said: The Philippine coconut oil industry is before you today to fight for its life. Ours is not a question of increased advantages—we ask none. It is not a question of special privileges. It is rather a question of non-discrimination. Our business is already limited in volume by the Tydings-McDuffie Act. All we are asking is the right to continue to operate on this limited basis, not only through the period of the Commonwealth but thereafter. The coconut oil mills recommended the annual admittance of two hundred thousand long tons of coconut oil to the United States "under the same conditions and terms as those given to oil made from duty-free copra shipped... to the United States." This meant that such oil would be free of export taxes and import duties. They also suggested "a readjustment of the excise taxes to permit Philippine coconut oil to recover its position as American oil in the American market, which position had been adversely affected by the products of American agriculture rather than by competing vegetable oils and fats imported from foreign countries." They concluded by advocating "a continuation of this arrangement not only during the Commonwealth period, but also indefinitely thereafter." 3 The condition of the local market during the last quarter of 1937 was very quiet the Bureau of Commerce stated, "as traders' ideas drifted far apart under the influence of two factors: the absence of encouragement from the American oil market which prevented the mills from raising bids materially, and the higher copra market in Europe which enabled exporters to offer better prices than the mills." Europe continued to outbid America, but the industry gained little advantage thereby because of the great lack of shipping. The market was a source of disappointment to those who had believed that year-end trends would be better. They drew their idea from market activities in 1936 and did not consider the operation of entirely different factors. The previous year had seen a shortage of fats and oils in the United States arising from a larger demand due to the industrial revival and greater purchasing power. This was also true in Europe. At the same time, the local supply was materially restricted aiding in the rise of prices. While this was temporarily beneficial, it had the ultimate effect of encouraging buyers in the States to look elsewhere for such cheaper oils as palm, kernel, and babasu from Brazil, Europe, and European dependencies. By the middle of the year, consumers had bought coconut oil, as well as cheaper oils, heavily, to last them through the year. So adequate were the stocks that large soap manufacturers held aloof from the Philippine market from --- 42 Op. cit., p. 61. 43 Ibid. 44 Ibid. 45 PJC, XIII, Nos. 10-11 (October-November, 1937), p. 54. July forward. In addition, large cotton and corn crops in the United States hit Philippine oil heavily, with cotton seed oil in abundance and its price dropping considerably below coconut oil. The outlook, then, was far from bright, despite the rosy picture painted by Bureau of Commerce in its review of business conditions for the year.\textsuperscript{46} The Bureau of Plant Industry pointed out to the industry the importance of coir (coir is the dried outer husk of the coconut) as an additional source of income, able to bring in annually over three million pesos.\textsuperscript{47} It was important to consider such other products, E. E. Cruz, the Bureau spokesman said, "in the face of the impending ruin of the coconut industry which is brought about by world competition, recurring typhoons, pests and diseases, excise tax, plus the present limitation and forthcoming export taxes that will be levied...."\textsuperscript{48} Some coir was consumed locally by furniture factories and automobile industries. The foreign market, if properly developed, would undoubtedly take the remainder of the annual production, or so the Bureau predicted. The many uses of coir, if the industry had developed them, would have provided a considerable cushion for the tired industry. Actually, despite the abundance of coir in the Philippines, the country imported coir mats from other countries. The rising discontent with the condition of the industry was brought to a head in the convention held by the hemp and coconut producers in February, 1938. President Quezon urged an educational program to acquaint the public, both in the Philippines and the United States, with the problems facing the industry thus removing the indifference or hostility directed against it. He went on to say, however, that although the government was prepared to give as much aid as possible, the main task was the industry's and the salvation of the industry lay within itself.\textsuperscript{49} Cornelio Balmaceda, at that time acting manager of the National Produce Exchange, told the convention, that the elimination of the middlemen "would tend to increase the prices which the local producers will get if the plan for an organized system for handling local commodities presented to the convention is carried out."\textsuperscript{50} He proposed a system which would place the producers, large or small, close to the exporters or manufacturers; the small farmers being organized into effective groups for the cooperative selling of their products. A system of farm credit would enable these small producers to get loans at low interest enabling them to meet the demands of the market. The system would necessitate a chain of warehouses for the convenience of the small producers. At the top, the National Produce Exchange would act as an organized central market. While not enthusiastic, the convention was interested. It adopted a policy of wait-and-see, and the program was never effectively carried out. The Coalition Platform, in 1938, had promised that "we shall support the organization of the producers of abaca, coconut, tobacco, rice, and other articles for the defense and promotion of their interests.... We shall help industries based on the coconut and its derivatives, and we shall continue working for the elimination of unjust burdens imposed in the United States upon these products. We shall exert our utmost with \textsuperscript{46}Anastacio de Castro: "Review of Business Conditions for 1937," \textit{PJC}, Vol. XIV, No. 1 (January, 1938), pp. 7-8. \textsuperscript{47}Eugenio E. Cruz: "Coir Industry of the Philippines," \textit{PJC}, XIV, No. 1 (January, 1938), p. 9. \textsuperscript{48}\textit{Ibid.} \textsuperscript{49}\textit{M.O.P.}, Vol. 5, Pt. 1, 1941, pp. 40-45. \textsuperscript{50}\textit{PJC}, XIV, No. 3 (March, 1938), p. 38. a view to finding a sure and profitable market for... coconut... and other important products... and if necessary with the financial assistance of the government."51 In his budget message to the National Assembly, in 1938, President Quezon stated, however, (referring to the excise tax): "This means that not only the proceeds of the said excise tax, but even the funds derived from the local revenue of this Government cannot be used for aiding the coconut industry as long as we continue receiving the benefits of the coconut oil excise tax."52 He concluded by saying that he was transferring, from the proceeds of the excise tax, P26,840,000, for the purposes of "replenishing the current surplus," to certain "extra-ordinary purposes." These were: Subscription of stock of the National Development Company, P10 million; subscription of stock of the Manila Railroad Company, P1 million; for the acquisition of land and construction of buildings for laborers, P250,000; a loan to the Manila Railroad Co., P9,990,000; for the purchase of homesteads, P1 million; for reforestation and afforestation, P250,000; for organizing the National Power Corp., P250,000; a revolving fund for the construction of water works, P2 million; for stabilizing prices of buntal fibers, P500,000; for a survey and subdivision of public agricultural lands, P100,000, and, for the new census P1,500,000.53 He concluded, by emphasizing the need for continued economy. The excitement and anger of the coconut producers, expressed in their convention, in February, 1938, can be readily understood. They saw the appropriation of twenty-six million pesos, for enterprises many of which were considered needless, and they were unable to touch a centavo of it, either directly or indirectly. President Quezon addressed a tea party of coconut planters and municipal mayors at Malacañan, on February 19. Here he discussed the resolution passed by the planters asking for the abolition of the excise tax. He did not believe that Congress would act on the resolution but continued: "I am willing, nevertheless, to support you in your request.... I have not changed my attitude in this respect although I confess that I am not as positive today as I was before that the tax had done any harm to the industry, or that if it did, the harm was not so much as I had thought it to be."54 He stated that he thought it was unfair to those who were taxed not to derive benefits from their taxation and promised to support a movement aimed at the abolition of the proviso prohibiting the use of excise tax funds to aid the coconut industry, were they unable to secure the abolition of the entire tax. But, he continued, "as long as the law exists, I will continue complying with the law honestly and strictly."55 The storm of indignation and discontent that arose throughout the country grew so intense that President Quezon went to some pains to explain the situation, particularly in his home province of Tayabas, a large coconut-producing region. He told his people of the difficulties facing him in this problem, saying frankly that there was little he could do. The solution, he said, lay in part with the planters. He was particularly interested in seeing that the people learned new uses for the coconut and that they increased their consumption of its products. The President believed there were great possibilities for the coconut industry in utilizing coconut for bukayo (the meat of the nut cooked with sugar) and matamis sa bao (coconut jelly). "The trouble with us Filipinos," the President 51 M.O.P., Vol. I (rev. ed.), pp. 250-251. Italics supplied. 52 M.O.P., Vol. 3, Pt. 1, p. 234. Italics supplied. 53 Ibid., pp. 249-251. 54 M.O.P., Vol. 4, Pt. 1, p. 27. 55 Ibid., p. 30. said, "is that we forget what is truly our own. We have so many needs and uses for our coconut. Why don't we rediscover them so as to increase not only our export but also our local consumption of coconuts? The National Development Company has several expert chemists studying the different uses of the coconut; more important, however, is to have our coconut planters take full initiative in this matter." 56 During the Japanese occupation, with food a major problem, many Filipinos rediscovered the coconut, and the consumption of its fats and oils stayed off starvation for countless thousands living in areas where the nuts were available. 57 In April, 1938, the National Assembly, fearful that the coconut industry would suffer complete destruction, created a committee to study means of improving it, with Assemblyman Lavidea of Tayabas as Chairman. The committee was charged with conducting a survey and recommending measures for increasing production. The Assembly hoped that immediate action would be taken. It was pointed out that the industry had but one good year (1936) since 1930, and the warning of H. B. Pond, in 1934, was recalled; he had stated that under the provisions of the Independence Act and the Revenue Act of 1934, coconut profits would be eliminated after the fifth year of the Commonwealth or surely after the sixth year. 58 The Philippine Coconut Association, in May, called a meeting of two hundred planters in Siniloan, Laguna. Governor Bonifacio extolled the advantages of the association and urged the planters to take united action to meet the crisis. He stated that the problem could be solved only if the planters pledged, before adjourning, a contribution of two centavos per one hundred kilos of copra sold, for a working fund. The same month it was reported that Governor Bonifacio was starting a campaign to educate the people of his province to look for a new industry to replace the coconut, the major source of their income. With representatives from the Bureau of Commerce, Bureau of Plant Industry and the Bureau of Forestry, he visited some twenty-eight municipalities for this purpose and at the same time initiated a study of the economic conditions existing in the province. This was believed essential since the continuing poverty produced a rising trend in discontent and lawlessness and weakened the Government's general economic program. On May 20, the Joint Preparatory Committee released its report, in which it made the following recommendations: Coconut oil should be exempt from Philippine export taxes but, in lieu thereof should be subject to annually declining duty-free quotas commencing with 200,000 long tons (of United States imports) for the calendar year 1940. This amount should be reduced by 5 per cent for each succeeding calendar year until it equals 150,000 long tons for the calendar year 1945. For the period January 1, 1946, through July 3, 1946, the quota should be one-half of the quota for the preceding year, or 75,000 long tons. Shipments in excess of the aforementioned quotas should be subject to whatever United States duty may be in force at the time. The Committee also recommends that the imposition of full United States duties should be postponed from July 4, 1946, to January 1, 1961, during which interim United States imports of coconut products, except coconut oil, should be subject to preferential tariff rates, in force on July 4, 1946 and increasing on each subsequent January 1, by 5 per cent of the then- 56 Ibid., p. 213. 57 That the industry is capable of great extension is unquestioned but Filipino capital will have to lose its timidity and its desire for high returns on investment before the industry can stand steadily on its own feet. 58 "The Future of Our Basic Industries," The Philippines Herald Yearbook, 1934-1935, pp. 101, 110. existing rates. Coconut oil should be subject to the above duties only on those amounts in excess of declining, duty-free quotas, fixed as follows: For the period July 4, 1946, through December 31, 1946, the quota should be 75,000 long tons; for the following calendar year, it should be 140,000 long tons; and thereafter it should be reduced annually by 10,000 long tons (5 per cent of 200,000 long tons) until 1961.... The Committee believes that if the above recommendations were adopted, the Philippine Coconut industry would have an adequate opportunity to adjust itself to a position independent of the preferential tariff and excise tax treatment in the United States market. The Report met with an unenthusiastic reception in the Philippines. By October, 1938, some P128 million in coconut oil tax money had been placed to the credit of the Philippine Commonwealth in the United States Treasury, a sum which excited the politicos in the Philippines to a fever pitch. Elaborate plans were laid for spending it, but little of permanent value resulted. The money was used to develop the future capital site, Quezon City (at the end of the war, a wasteland and no money for development), a new legislative building, a Jai Alai palace, the hemp industry, the NARIC, cooperative enterprises, the purchase of the Buena-vista estate, the Koronadal Valley settlements, a fruit and fish cannery, a textile mill, two sugar refineries, and others. In 1950, the Quezon City capital site was still largely on paper, the war-ruined legislative building was being rebuilt, the Jai Alai palace (in sharp contrast to the rest) had been the first repaired and started in operation, the hemp business was almost prostrate, the NARIC a practical failure, cooperatives in their fetal stage, the Buena-vista estate still in the courts, Koronadal settlements barely begun, the fish and fruit canneries but fond memories, the textile mill at a standstill facing disposal, and the sugar refineries producing spasmodically, and likewise facing disposal. A sharp critic of the Commonwealth at that time had this to say of the oil money: In the opinion of some critics, the money should be used to pay off all the bonds of the government-owned Manila Railroad, and all the Commonwealth's public debt—which is not large.... However, Manuel Quezon is a politician, and he may want to be President again after this term is over. It is normal for politicians, in any country, to use available funds in the showiest possible way. Building an army and paying off the public debt are --- 59 Op. cit., pp. 66-67. 60 In a message to the Second National Assembly, on the budget for the fiscal year ending June 30, 1940, President Quezon referred to the fund, in a special section. He stated that, at the end of 1937, there existed a cash surplus in this fund of P82,669,-S60.82, and with additions and interest, the total amount available for 1938 amounted to over P108 million. The unexpended cash balance for 1938 was over P71 million, with total outstanding balances of authorized appropriations made against the fund as of that date being P110 million, which was P39 million greater than the available balance in the fund. However, no deficit was incurred, he said, because he had authorized the release of only some P44 million. The total income expected for the six months ending June 30, 1939, was P21 million, the total amount available for release at that time being P48 million. The estimated income from the fund for the fiscal year ending June 30, 1940 would amount to P25 million. With authorized expenditures, in the budget for that year, there would remain, he said, only an unappropriated surplus of P6 million. Therefore, he concluded, there could be no further appropriations made from the fund, especially when it was realized that the fund income would not be available until it was actually accredited to the Philippine Treasury. For this reason, he stated, to maintain the balanced budget presented, it was necessary to effect substantial economies, "limiting salary increases as well as creation of new positions to only those found to be absolutely necessary to insure the efficient operation of the government." M.O.P., Vol. 5, Pt. 1, pp. 274-275, passim. 61 By 1950, a steady money-loser. gestures that do not impress the electorate half as much as public-works funds scattered liberally in the proper geographical places... When the President of the Commonwealth is asked about the probable shrinkage in the government budget after 1946, he coolly answers that it won't shrink. He or his successor will soak the rich by heavy taxation. (The rich, unhappily, will be fewer and less rich). Quezon also says that, beginning in 1946, he (or another) can make his own tariff, unhindered by the U.S. He will then put heavy import duties on all sorts of things... and thereby tax his people in other ways. He seems to have extremely optimistic ideas about the capacity of his people to pay... Although there was some truth in what Miss Horn had to say, few of these predictions were validated by events. The rich increased in numbers and wealth and the budget continued to expand. The question of taxation remained unresolved. On May 17, 1939, the President requested the National Assembly to send an observer abroad to study foreign coconut industries. The Assembly approved Resolution No. 23, and Maximo M. Kalaw was designated as observer. Five days later, the President discussed briefly the excise tax. He repeated, by implication, what he had to say earlier, on February 22, and then gave his opinion with regard to the repeal of the excise tax. The proposal was that the tax be collected only for coconut oil to be used for edible purposes. He expressed the opinion that this would complicate the situation for the large American buyers would purchase only non-edible oils and "pocket the difference in prices." He continued: "In other words, the benefit to be derived from the elimination of this tax, instead of aiding the producers of copra, will only go to the pocket of the purchaser in the United States. The benefit may go to the soap distributor in America and, in turn, to the consumer; but I am not sure that it will be of any good to the producer in the Philippines." The President, it is evident, had now lost his enthusiasm for abolishing the excise tax, despite his promises to producers that he would support any movement to do away with it. The returns from the United States were too great to be thrown away carelessly. He presented the other side of the picture skillfully: On the other hand, what will be the result if that tax is removed? Our income will be reduced annually by over P20,000,000, an amount that comes from the excise tax. This yearly income, after 1940... or after seven years, will mean a loss of P140,000,000 in our income. How will such reduction affect our finances? In the first place, I will be forced to veto a large proportion of appropriations for public works, and I may have to suspend many of the constructions under way which are for the improvement of our means of communication and the economic development of the country. And I want you to remember, gentlemen, that I will not leave my successor in office emptyhanded, much less leave the government of the Philippines... with a problem in its hands: the lack of funds in the Treasury. No matter how much it may affect the success of my administration, even if it should mean the difference between a successful President and a failure, I would prefer to leave my office and be pointed out as a man who had failed rather than have the future of the Philippines exposed to bankruptcy. (Applause) --- 62 Florence Horn: Orphans of the Pacific, (New York, 1941), pp. 230-231. See also, J. R. Hayden: The Philippines, (New York, 1947), pp. 157-160, and note 23, pp. 888-889. 63 Supra. 64 Supra. 65 M.O.P., Vol. 5, Pt. 1, pp. 129-130. 66 Ibid. As a result of this statement, the campaign for the removal of the excise tax was conducted with little vigor, except by a few die-hards in the coconut industry. The Government's position was clear: it would like to help the coconut industry but not to the detriment of the country as a whole. Twenty million pesos annually was a sum to be regarded with appreciation. This rather equivocal campaign (equivocal in the sense that the industry was promised aid but not at the expense of the excise tax fund) was continued throughout 1939. President Quezon returned to his home province to allay the fears of his constituents, showing them the benefits to be derived from the excise tax fund. He pointed to the roads being constructed, and said that it was important that such work be done for the benefit of the country at large. Nothing should be done, he believed, which would benefit a section of the nation at the expense of other sections. He suggested that it was important to seek means of stabilizing the economy. "The Filipino must be taught to be self-reliant. We must know that we can realize profits if we have exportation, and we export those products which will give us profit even if a tariff duty or excise tax is levied. That is what we should ponder upon. Hence we must look for other native products which we can profitably sell abroad, even if they are levied a tariff duty or an excise tax." Such words were cold comfort to a region dependent upon the coconut industry for its livelihood. It is difficult for the people of a certain section, even in the United States, to accept a diminution in their prosperity on the ground that it is good for the whole nation. What was wrong, then, with the country's economy? The middlemen, according to the President. "One of the reasons why the Philippines remains poor," he stated, "is the fact that the man who tills the soil does not receive his just profits from the sale of his crops. Take the copra producers, for instance. The lack of proper commercial facilities compels them to pass their copra through the hands of many middlemen before they finally sell it. Each one of these middlemen engaged in the copra trade derives profit therefrom, and their profits, combined, take so much away from the amount that should go to the copra producer. Likewise, it is not only the poor sales made by the planter from his copra which gives him financial difficulties, but also the fact that he pays more for his daily purchases than city residents... because everyday commodities pass through so many middlemen.... To end this, I have caused a thorough study of the best plan by which small farmers may form cooperatives so that they can sell their products direct to the dealer without passing them through many middlemen. I have also advised the study of the means by which trading in commodities of prime necessity in the Philippines could be handled exclusively by Filipinos. (Shouts of 'Mabuhay!')." While this may have been a popular analysis of the country's economic ills, it was far from being completely accurate. If nothing more than the existence of many middlemen plagued the nation, the solution of its problems would have been immeasurably simplified. Basic ills demanded remedy, and while the middlemen contributed to the uncertain state of the economy, ills arising from antiquated traditions and --- 65Cf. telegraphic correspondence between Quezon and Roosevelt during February and March with reference to proposed legislation in the U.S. Congress leading to an increase in the excise tax. M.O.P., Vol. 5, Pt. 1, p. 432. 66M.O.P., Vol. 5, Pt. 1, pp. 164-172, 158-163, passim. 67M. O. P., Vol. 5, Pt. 1, p. 214. Speech on the occasion of the fourth annual celebration of the establishment of the Commonwealth, New Luneta, Manila November 15, 1939. techniques needed attention before much lasting good could be accomplished. There was no lack of information as to these ills. In January, 1940, Assemblyman Kalaw, who had been selected by the National Assembly to go abroad to study foreign methods of coconut production, submitted his report to the Assembly. He presented seven "salient points." He noted that "one of the greatest economic wastes of the Philippines is the poor preparation of our copra." He stated that despite the fact that the Philippines was the largest copra-producing country in the world, its quality of copra was among the worst and as a result the country lost an estimated ten million pesos every year. He pointed to the waste of over three hundred thousand tons of coconut fiber with a value of P75,000,000 each year. The Ceylon planter, he said, produced smaller nuts than his Filipino counterpart but received from nine to sixteen pesos more per thousand nuts than the Filipino planter. "The reason is that he produces very good copra and utilizes the by-products." He mentioned the well-known fact that, in 1939, the Philippines was the fifth best customer of the United States, and the United States would continue to need coconut products, thus maintaining a market for Philippine products. He suggested that "if there is to be some further reciprocal agreements effective after independence, coconut products should be given favorable treatment" in the American market. He predicted that the coconut industry would withstand the shock of independence much easier than others and that, if the necessary steps were taken by the Government, it would be "our greatest bet in the open markets of the world." Finally, he referred to the fact that the diet of the Filipinos was deficient in fats, a situation which would be remedied with an increased home consumption of coconut products. The force of this statement was illustrated by the experience during the Japanese occupation when the population, in many parts of the country, existed largely on coconut products. The solution of these problems, the Assemblyman believed, lay in industrialization, socialization and cooperation of and in the industry. "The Government must lead the way," he said, "but the cooperation of the planters themselves is indispensable." He specifically recommended that after the creation of a government body similar to the National Coconut Corporation (which was then before the Assembly for approval) six steps be taken by that or another corporation. These steps were: the establishment of drying plants and coconut centrals aided by government standardization to improve the nation's copra; the utilization of husks and shells and other by-products, through industrialization; the establishment of cooperatives with the purpose of eliminating middlemen and improving credit facilities for the benefit of planters; establishment of "regular freight and shipping service to foreign countries"; establishment of a research and experimental station devoted exclusively to problems facing the coconut industry, preferably on a government-owned plantation; and the "fostering of a greater home consumption of coconut products." If these steps were taken, he believed, the industry would "stand the shocks of political and economic separation from America" and would furnish the country's leading export. The cost of this program he estimated at P20,000,000 for the remainder of the Commonwealth period. This was to be spent on 5,500 copra driers, twenty provincial centrals, five national centrals, the coconut research institute, and loans to planters, the last item bulking the largest, amounting to P9 million. --- 70 Maximo M. Kalaw: The Coconut Industry, (Manila, 1940), pp. vii, ix, passim. On May 7, 1940, the National Assembly passed, and the President approved, Commonwealth Act No. 518, creating the National Coconut Corporation (known as the NACOCO). The corporation was to be organized within six months to exist for thirty years. It was "to establish, keep, maintain and operate or help establish, keep, maintain and operate drying plants, or copra driers, or coconut centrals with a view of adjusting the coconut industry to a position independent of trade preferences in the United States and to provide facilities for the better curing of copra products and the proper utilization of coconut by-products, provided that no subsidy, direct or indirect, shall be paid to producers, or processors of copra, coconut oil and allied products." Section 2 of the act also provided that the corporation was "to afford facilities for bona fide production loans to Philippine coconut planters and copra producers." In addition, the act provided for a "Coconut Industry Promotion Fund" which was to be derived from the coconut oil excise tax collected after January 1, 1939, two million pesos of which was appropriated with the approval of the act and the total not to exceed twenty million pesos. This fund was to be utilized by the directors as provided by the act and all moneys accruing to the corporation were to flow into it. The stipulation with regard to subsidies to producers, was included with an eye to the American Congress which had imposed the excise tax at the insistence of the American farmer. In the years following the imposition of this tax, there was, in the words of Grayson Kirk, a "steady increase in American imports of coconut oil..." which proved that the Philippine oil was "not interchangeable" with any American oil. Professor Kirk demonstrates that the result of the tax was only a charge to the American public of seventeen million dollars a year which benefitted neither the public nor the farmer. The sole benefits derived from the tax went to the Philippine Commonwealth as a "windfall," which enabled the Commonwealth to balance its budget at the expense of the American housewife. The suspension of this tax was asked by Commissioner Elizalde, in 1940, and was so provided in the bill introduced by Senator Tydings on June 11. The bill, however, was not passed until after the outbreak of the War. According to the Philippine Trade Act of 1946 (the Bell Act), to anticipate events for a moment, the amount of duty-free coconut oil exported to the United States was set at 200,000 long tons and "during the effectiveness of the agreement the United States will not reduce the preference of 2 cents per pound provided in section 2470 of the Internal Revenue Code... except that it may suspend the provisions of... such section during any period as to which the President of the United States, after consultation with the President of the Philippines, finds that adequate supplies of neither copra nor coconut oil... are readily available for processing in the United States." This hostile attitude of the United States toward the Philippine coconut industry, expressed in all legislation between 1933 and 1941 (and 1946), was directly responsible for the critical situation which arose at the outbreak of the war in the American supply of coconut oil. Since more than half of the U.S. imports of fats and vegetable oils came from the Philippines, the prices of edible coconut rose to fifteen cents a pound in December, 1940, more than twice the --- 71 The Lawyers' Journal, VIII, No. 22 (November 30, 1940), p. 882. Italics supplied. 72 Grayson Kirk: "Philippine-American Relations: Recent Trends," Political Science Quarterly, LIV, No. 3 (September, 1939), p. 336. 73 Public Law 371, 79th Congress, 2nd session, H. R. 5856, Chapter 244, Section 403, subsection (d). The tax is still collected but is, of course, not returned to the Philippine Government. price of the preceding December. The price of oil rose proportionately, while coconut meal increased almost five times during the same period.\textsuperscript{74} The outbreak of the European war in 1939-1940 dealt a serious blow to the coconut industry. The Philippines had such large stocks of coconut products that copra prices in August of 1940 dropped to an all-time low. Copra cake, normally exported for cattle feed, was used for fuel, the requirements being far below the supply. The normal American market for copra and oil was greatly reduced by an abundance of domestic oils and the export trade to Latin-America was hurt by lard exports from the United States. In addition, the great increase in ocean freight charges restricted shipments abroad. Rates, for copra, from Manila to American Pacific coast ports, increased from $9.50 to $13.50 per ton; for copra meal, from $8.00 to $11.50 per ton, and, for oil, from $15.50 to $20.00 per ton. This, naturally, cut sharply the profits of local producers. The Japanese reaped the benefit of this situation when they entered the Philippines, acquiring at one stroke a considerable portion of the oil needed for the year 1942. With the outbreak of the War with Japan, the United States, which had already felt a need for oils and fats because of defense preparations, found itself cut off from its largest and most faithful producer of these necessities. During 1941, the United States had increased its consumption of Philippine oil through the new federal regulations (the McNutt standard) which encouraged margarine producers, oil consumption increasing some three million pounds in one year.\textsuperscript{75} Had the United States taken advantage of the low prices for Philippine coconut products after 1939 (a step which was never considered by Congress) it might well have faced December 7, 1941, with more confidence than was the case. In 1946, the United States, returning to its position as the world's largest consumer of coconut products, signed an agreement with the Philippines by which it agreed to buy all the surplus copra and coconut oil for the year beginning July 1, 1946 at $103.56 per long ton for copra and 7.1 cents per pound for coconut oil.\textsuperscript{76} During the first six months of that year, the Philippines exported 190,000,000 pounds of copra in terms of oil, about two-thirds going to the United States. While this meant prosperity for the long-suffering coconut products producer, the iniquitous economic system, as usual, robbed the small producer and enriched the middlemen and the large producers. The outbreak of the War halted plans for carrying out Assemblyman Kalaw's recommendations for stabilizing the industry. As provided in the act creating the NACOCO, most of his recommendations would have had to be executed by other agencies than that provided by the Government. The end of the War gave a second chance for the stabilization of the industry which, as had been predicted, weathered the conflict more easily than the other major industries. While some damage was suffered by equipment and buildings, the trees still remained. There was a worldwide demand for coconut products at the end of the War and prices soared to unbelievable heights. During the transition period, between the end of military rule and the beginning of private trading, the American need for coconut products was illustrated by the special arrangements made by the Foreign Economic Administration for the procurement of coconut products in 1945. The \textsuperscript{74}Britannica Book of the Year, 1941, p. 178; 1942, p. 186. \textsuperscript{75}Encyclopaedia Britannica Book of the Year, 1942, p. 186. \textsuperscript{76}Ibid. FEA announced the signing of an agreement between the U.S. Commercial Company and the Copra Export Management Company, a corporation formed by representatives of five producing companies engaged in the business before the war. This agreement gave an impulse to the rehabilitation of the industry which enabled it to achieve an early prosperity not shared by other Philippine exports. Prosperity was not immediately achieved, however. International agreements, following the War, had placed the price at P12.00 per 100 kilos of resecada (which is the dried husked flakes of meat). The Copra Export Management Co. (called the CEMCO) authorized buyers to pay no more than P10.50, with the result that few producers would sell at this price claiming, with justice, that the price did not cover the cost of production. Late in 1945, buyers offered as high as P15.00 per 100 kilos, but most producers continued to hold their stocks. In February, 1946, CEMCO then announced that copra would be paid for on the basis of 35 to 40 per cent cash, 30 per cent in textiles, 20 per cent in rice, and 15 per cent in beans, canned fish, milk, sugar, flour, etc. This offer was attractive enough to bring into the market increasing quantities of copra, the price rising to around P16.00 per 100 kilos. Buyers, during this period, paid higher than the authorized price; because, according to one source, they were selling the products on the black market at prices high enough to cover the high copra costs. In August, 1946, the Copra and Coconut Oil Agreement was signed between the Philippines and the United States fixing the price at $103.50 per long ton in bulk. In October, it was announced that the copra ceiling was $116.20 per short ton at Atlantic and Gulf ports. This fixed the price at around P20.00 per 100 kilos, buyers' prices in Manila being about P16.00 per 100 kilos, delivered. This agreement also stipulated that the Philippines should sell its entire crop to the United States at these OPA ceiling prices, the Philippines to receive in addition, a loan of $2,000,000 to be used in the rehabilitation of the hemp plantations. The United States also agreed to transfer to the Philippines some $3,500,000 worth of trade goods, including textiles, rice, and sugar. The American embassy announced that the Commodity Credit Corporation in the United States would allocate shares of copra from the Philippines and Java to other foreign nations, and would appoint companies in the United States as authorized buyers of Philippine copra. This agreement was basically unfair to copra producers in the Philippines, who were thus deprived of an opportunity, quite necessary for the stabilization of their industry, to sell their stocks in the open world market where they could get higher prices than those provided in the Agreement. It is difficult to understand the motives of the President of the Philippines, who was quite conscious of the necessity of finding world markets for Philippine products to prepare the country for the day when the protected market in the United States would be closed. Producers and exporters in the Philippines raised loud outcries, but with little immediate effect. They realized that free trade would have a marked effect upon the development of their industry in the crucial years following the War, and were faced with the fact that the United States, although stating that it was working toward the establishment of free world trade, was engaged in activities exactly opposite in practice to its announced policy. The copra producers hoped to be able to sell enough stocks in the free world market to acquire sufficient strength to meet an uncertain future when prices might again tumble to the level of 1934. 77 Reported in Far Eastern Survey, XIV, No. 15 (August 1, 1945), p. 214. Evidently the Philippine Government hoped to receive concessions from the United States of a general nature, even at the expense of one of its hard-pressed industries. Its eagerness to rehabilitate the hemp industry, while laudable, should not have obscured the need of stabilizing the vital coconut industry. The Government replied, to domestic critics, by saying that it was duty-bound to share with other nations the commodities it had in abundance in order to receive such needed supplies as sugar and rice. It pointed to its membership in the International Emergency Food Council and the necessity of abiding by international agreements. The producers, however, suggested that a food emergency existed in the Philippines, and that it could partly be met by raising the prosperity of one industry, thus benefiting others and the people as a whole. One author, writing in the Philippines, pointed to hard facts as proof that the Government's argument was specious. The Emergency Food Council had allocated, in 1946, 145,000 tons of rice for the Philippines, 45,000 tons of which was to come from Siam and the remainder from the United States and South America. However, 1946 saw a serious shipping strike in the United States, the result of which was to deprive the Philippines of rice supplies for a period of over three months. In addition, the price of rice rose in South America, to a point where NARIC officials stated they would lose money by selling at the ceiling price in the Philippines. Since the country was in dire need of rice, it had to continue buying the cereal from South America necessitating either a rise in the ceiling price at home or the assumption of a loss in the operations of the NARIC: "There is no parity," this observer stated, "in the prices of the commodities we export and those that we badly need in terms of the index of the cost of living in other countries." Since the Philippines had no control over the prices in the American countries it was at their mercy, for the United States held the balance of trade through the pegged prices of copra. Thus, the Philippines continued to ship copra to the United States disadvantageously and received in exchange consumer goods which, however, remained largely on paper because of the great delay in shipping articles to the Philippines. Then, on October 29, 1946, the OPA decontrolled all fats and oils, with the result that the price of coconut oil soared from eight cents a pound to twenty cents a pound. A great rush of buyers scoured the world market for available supplies. Copra dealers demanded, and received, as high as twenty-two cents a pound. The planters, however, were bound to continue selling at prices ranging from ₱17.00 to ₱20.00 per 100 kilos, which meant a price of eight cents a pound for oil, whereas the actual market value of the product, on the basis of twenty cents a pound for oil, was ₱46.00 per 100 kilos. Thus the individuals who benefited were not the planters, but the exporters and middlemen. It was quite clear that if the stabilization of the industry was to be achieved, it only could be done through the abrogation of the Agreement. On November 22, 1946, abrogation was announced, with the result that buyers' bids rose 56 per cent over the level of mid-October. The price reached ₱30.00 per 100 kilos, and with the release of controls on fats and oils, domestic bids topped quotations in the United States. During 1931-1935, the average exports to the United States were 64.9 per cent of the total. In the period up to December 31, 1946, of the total shipment of 599,000 long tons, 74 per cent went to other countries, such as the United Kingdom, France, Poland, Belgium, Norway, the Nether- --- 78 Domingo C. Abadilla: "Confusion in the Copra Market," The Sunday Post Magazine, Vol. II, No. 8 (November 10, 1946), p. 24. lands, Canada, Sweden, Denmark, and Panama. The percentage of price increase in copra was the second highest of the four major export products, being 376 per cent.\textsuperscript{79} During the first six months of 1948, copra prices reached an all-time high of P65.00 per 100 kilos and coconut oil reached P1.10 per kilo. By the end of July, copra dropped to P10.00 and oil to P0.90. During the first half of the year export sales dropped more than 30 per cent indicating that the goose was dying. Although the value was higher than in the preceding year because of the increase in prices, it was feared that the producers and sellers had priced themselves out of the market. American importers stated that costs were too high for soap and margarine and the United States forbade the sixteen nations then using ERP funds to purchase Philippine copra. American importers, affected by the high prices, had requested the Government to take this step. In addition, a new and disturbing factor entered the copra and coconut oil picture. With the loss of copra and oil during the War, American manufacturers had turned in increasing numbers to the use of soy bean oil, cotton-seed oil, and tallow. New techniques developed by American research enabled manufacturers to utilize oils which a few years earlier they used with reluctance. This meant eventually a declining demand for coconut oil, since its price was determined by prices of other oils. Demand slackened also as a result of the American development of synthetic detergents and substitutes for soap, made previously with coconut oil. While soap-makers of the United States before the War utilized 70 per cent of imported coconut oil, the post-war period showed a drop of from 10 to 15 per cent because of synthetic oils. This was expected in 1950 to continue, thus presenting a gloomy prospect to Philippine copra producers. Complicating the situation was the activity of certain exporters who, in 1947, were found to have misrepresented the weights of their shipments and their classification. Some foreign importers had complained of shortages of as much as 25 per cent in actual received weight compared to manifests. In addition, fresh copra had been deliberately classified as resecada. When it is realized that pre-war shrinkage in weight was allowed a maximum of 5 per cent, the seriousness of the complaints can be understood. The Philippine Coconut Planters Association adopted a resolution requesting President Roxas to order the prosecution of these exporters.\textsuperscript{80} Many foreign importers expressed dissatisfaction year after year with the poor quality of Philippine copra, and the planters feared that unless the unscrupulous exporters were curbed, great harm would befall the industry. While the situation was ameliorated slightly by the threats of the Government, few individuals were actually prosecuted and the Bureau of Commerce in 1950 was still seeking means of controlling the classification of the product. The NACOCO was supposed to exert efforts with regard to this problem, but complaints of this body's inefficiency and ineffectiveness were widespread throughout 1948 and 1949. Although the NACOCO met with some success in stabilizing copra prices, it encountered sharp criticism from many quarters for its apparent inability to handle its work to the benefit of the industry and the nation. In January, 1948, for instance, local copra dealers were loud in \textsuperscript{79}Sugar rose 654 per cent, hemp 279 per cent, and leaf tobacco 150 per cent over prewar levels. Since sugar had practically disappeared from the export market, copra by 1950 represented the greatest export item in quantity as well as value. \textsuperscript{80}\textit{Manila Chronicle}, Year 3, No. 207, (December 12, 1947), p. 11. their complaints that it had failed to make deliveries on contracts for September and October; this had "adversely affected commitments of large amounts of the product to buyers abroad and disrupted dealings in this Philippine export product."81 Five large dealers were hurt, and one began court proceedings for damages suffered as a result of NACOCO's failure. One dealer who was reported to have made a commitment of some $345,000 had been waiting since October for NACOCO to make the delivery. Dr. Kalaw, general manager of the NACOCO, admitted the failure in deliveries, which he said was caused by typhoon damage to NACOCO stocks. While there was some truth in this statement, it did not wholly cover the situation. A Canadian representative reported that confidence in the corporation had suffered severely in the Dominion as a result of the "ragged business methods" of the NACOCO. He mentioned that, in one transaction with the corporation, involving 2,000 tons of copra at $165 a ton, the delivery was delayed over a month, and it was discovered that a shrinkage of 13.84 per cent had occurred. In addition, he said, the company experienced difficulty in obtaining a rebate on the freight charges. The whole transaction was considered highly unbusinesslike, with the result that his company would have no further dealings with the NACOCO. Later in 1948, the trading activities of the NACOCO were suspended because of the heavy losses incurred in buying, selling and research. Its personnel was reduced 70 per cent and its agencies in the provinces were closed. The corporation, in the three years following its resumption of the business in 1946, in the words of an outstanding Manila reporter, had squandered more than three and a half million pesos of the people's hard earned money."82 It was charged that Kalaw had been dealing in futures, entering into contracts with foreign buyers during a period of constantly rising prices. He failed in deliveries because he could not secure stocks from producers, who refused to sell at the low early price. To remedy the situation, it was charged, Kalaw bought stocks at the higher price and sold at a loss of one and a half million pesos "to save his face...."83 It was further charged that large amounts were paid as commissions. One broker was to receive nearly P20,000 when payment was stopped by the Government Enterprises Council as being "anomalous."84 The director replied that he could have made a profit for the corporation had the Philippine National Bank granted him a requested loan. The Bank replied that it had refused to grant the loan because the corporation was "mismanaged." Since the function of the NACOCO was not to act as a dealer and exporter, Mr. Kalaw's defense of his position was assailed by many critics; among them former Representative I. Vameta, one of the authors of the bill creating the corporation.85 While this sad story was little different in nature from others concerning the Government and its agencies throughout 1948 and 1949, it differed from the others in the fact that it concerned a very important government corporation charged with the direction of one of the most important Philippine industries. It strengthened the opponents of government participation in business, and deepened the disillusion of many --- 81Manila Bulletin, Vol. 133, No. 6 (January 7, 1948), p. 1. 82Leon O. Ty: "Another Kalaw Fiasco," The Philippines Free Press, Vol. 39, No. 26 (June 26, 1948), p. 4. 83Ibid. 84Ibid. 85"No Fiasco Says Maximo Kalaw," Philippines Free Press, Vol. 39, No. 28 (July 10, 1948), pp. 26-27. Cf. "Kalaw Stresses NACOCO Services," Manila Times, Vol. 4, No. 32 (September 18, 1948), pp. 1, 12. observers with the post-war policies of a government which could not afford to waste even the most insignificant funds. The cause of the Hukbalahap and the Communist was thereby aided, and the cynicism of the citizen made more profound. By contrast, the activities of one of the few successful private cooperative business enterprises in the country made cheerful reading for those interested in the healthy growth of Philippine economy. The CAPCA (Camarines Norte Copra and Abaca Producers Cooperative Association), of Daet, Camarines Norte, in the words of its supporters, belied "the charges that no Filipino business enterprise can succeed in a field dominated by aliens." Its activities were successful in bringing a good life to the people it served. It aided in the stabilization of copra and rice prices and broke the alien monopoly of rice. It ended the hoarding of gasoline, and attacking the age-old problem of usury, opened credit facilities to tenants and landowners alike. Its activities in the copra field have been noteworthy. Small as it was, according to one writer, it could have aided NACOCO in its difficulties had "Dr. Maximo Kalaw... been farsighted." The CAPCA had repeatedly asked financial assistance from the NACOCO but in each instance such assistance was denied and instead the activities of NACOCO were "entrusted... to personnel of doubtful ability in Tabaco and Legaspi." The writer pointedly referred to the fact that NACOCO spent "millions for the improvement of the quality of copra," whereas the CAPCA, with a fraction of the capital of the corporation, achieved the utmost in such improvement, according to statements of various Bureau of Commerce inspectors in the region. Summary:— Coconut industries were generally profitable during the early years of the Commonwealth but remained at the mercy of disease and natural disasters which, with important foreign developments, slowly ate away the high returns hoped for by planters. In general, the major profits were gained by processors and shippers. In an effort to aid planters, the Government began a move to provide warehouses for cheap storage. It had little effect, however, on the situation since the ills of the industry centered around poor production habits and improper drying and processing methods. Since so much of the prosperity of the Philippines depended upon a prosperous coconut industry, the many adverse factors — uncertain legislative practices in America, shipping strikes in America, natural disasters and disease, lack of producers in remedying the manifold problems, excessive dependence upon the American market, confusion within the industry, confusion in the Government as to programs of solution — which were characteristics during this period led to a general retrogression within the industry with a resultant decay throughout the nation's economy. Associations were formed and reformed in feeble attempts to meet an increasingly uncertain future but the members had a general lack of understanding as to proper procedures for action and, indeed, disagreement as to what was actually needed. As a result, they became more and more concerned with securing assistance from the United States and the Commonwealth Government. Their meetings soon fell into the usual pat- --- 86Julio Q. Liwag: "CAPCA — A Great Success," The Philippines Free Press, Vol. 39, No. 45 (November 6, 1948), p. 18. 87Ibid. tern of forums for complaint and passing of meaningless resolutions although the expenditure of more funds and energy in research was needed. The enthusiasm for ending, by the Government, uncertainty that existed for the industry, was soon dissipated. This occurred when individuals in the Government began to appreciate what the oil excise tax would provide in cash for the Government. Although the representatives of the Government had loudly protested the imposition of this tax, they soon realized that the treasury would be increased each year by P20,000,000, a sum whose magnitude eventually obscured the vision of Philippine legislators. Many proposals were made as to expenditure of these funds but no concrete steps were taken to utilize them in solving the basic economic problems of the nation. The industry, therefore, had to rely upon the uncertain generosity of the American Congress. Members of the industry, realizing this if nothing more, appealed to the United States, through the Joint Preparatory Committee, for succor by the extension of free-trade relations for an indefinite period or a longer Commonwealth period of transition. They failed to secure this objective, through the opposition of politicos at home and "anti-imperialists" in America, and were urged to bear up under adverse conditions since their sacrifice was made, they were told, in the interest of the whole country. Actually, only a few benefitted and these only temporarily. The President suggested that the difficulties were caused by the activities of middlemen and told the members of the coconut industry that he had directed a study be made with regard to the development of marketing and cooperatives. However, the difficulties of forming such cooperatives were either not appreciated or were ignored. In any event, it was the exception to the rule which succeeded. An observer was sent abroad to study production and marketing methods in competing countries but his report, published after his return, was largely ignored. The only result was the establishment of the National Coconut Corporation which was to prove to be one of the most dismal of failures. The industry, then, was hardly in any position to face the disruption of a major war and its pre-war weaknesses foreshadowed the dark development of the post-war period. One cannot escape the conclusion that the industry continued to wait passively for some foreign miracle to occur to bring the industry back to prosperous days and preferred domination to determination. The drop in copra prices early in the Republic caused some apprehension among producers and sellers. Many observers believed, however, that the situation would be of material assistance in the development of better methods of production and the eventual stabilization of the industry. Local businessmen repeatedly pointed to the fact that the industry could not be aided as long as its raw materials were shipped out of the country for processing. They hoped that the 1949-1950 slump in prices would encourage producers and capitalists to improve methods of production and the quality of the products, leading to a standardization of product that would bring continuing better prices in the world market. One writer accurately stated, in 1949, that "if we are really set upon industrialization, we should first learn to discard antiquated agrarian concepts, accept transitory dislocations and prepare to fight for world markets against established industrial nations."88 88Hernando G. Cosio: "Cushioning Copra Prices," Manila Times Midweek Review, Vol. 2, No. 27 (February 2, 1949), p. 5. 89 businessmen were not prepared to do that, he added, "let us stop kidding ourselves and chasing the rainbow of industrial competency." "Let us instead concentrate on our traditional agricultural economy and do a good job of producing raw materials from our bountiful natural resources for the endlessly hungry machines of the industrial world." No more accurate observation has been made with regard to any of the country's industries and it should have been taken as a motto by a Government seeking a pathway out of the tangled jungles of its economy. Since the hectarage of coconuts was constant since 1948, and since production steadily declined because of disease and other factors, and since no decisive steps had been taken, up to 1950, to meet the situation, most observers predicted that no change could be expected for at least five years, if then. --- ABBREVIATIONS: PJC — Philippine Journal of Commerce. M.O.P. — Messages of the President. PHILCUSA-FOA PUMP IRRIGATION PROGRAM IN THE PHILIPPINES DOMINADOR Z. ROSELL* The pump irrigation program under the Department of Agriculture and Natural Resources is one of the many projects of the PHILCUSA-FOA in the Philippines. The office undertaking the implementation of the program is the Irrigation Service Unit, which was formally organized on October 1, 1952, in accordance with the provisions of the Memorandum of Agreement jointly made in August, 1952, by the Secretary of the Department of Agriculture and Natural Resources, the chief of the Mutual Security Agency (MSA) here and the chairman of the Philippine Council for U. S. Aid (PHILCUSA). This program extends irrigation facilities through the installation of irrigation pumps, thus providing water to communities of small farmers whose fields depend solely upon rainfall for water supply during the rainy period of the year and are left uncultivated during the dry season. The immediate objective is to bring about increase in the yield of the regular season crop of rice in the irrigated areas, make possible the planting of a second crop during the dry season, and improve general farm practices to raise the level of the farm income and the standard of living of the farmers served. There are three important fundamental factors to be considered before a pump irrigation system is installed, namely: 1. The people 2. The land 3. The water (1) The people in the area where the pump irrigation system will be installed, such as the landowners and the farmers or tenants, must show and signify their interest in the irrigation project. They must indicate their desire by filing with the ISU, DANR, the necessary application for the purpose. When the application is approved, they must form an association duly registered with the Securities and Exchange Commission, in order to have a legal personality, thus enabling them to transact business and sign contracts with the Irrigation Service Unit. (2) The land must at least be 75 hectares contiguous and irrigable once the irrigation water is made available. The soil should be of such texture that it will retain a good part of the water brought up for irrigation. It must be at least level or flat so that the construction of canals and their maintenance will be at the minimum cost. The distance from centers of population and good roads must be such that transportation of fuel for the engine and other necessities will not be prohibitive, thus making the operation, maintenance and administration economical. (3) The water in the river should be able to supply a minimum of 2,000 gallons of water per minute, without being exhausted during the dry season. It must be free from harmful salts and substances toxic to the crop. The height from the water level to the discharge pipe should not be more than twenty-eight (28) feet. Administrator, Irrigation Service Unit, DANR; Secretary, Philippine Geographical Society. If one of these factors is not satisfied we simply disapprove the application and no irrigation pump will be installed. The ISU is at present in a position to investigate and study these factors every time an application is received from interested parties. We have competent soil technologists to inspect the land and soils; agricultural engineers to see the elements involved in the construction of canals, water duty, and other factors; civil engineers for both designing and specification; and construction engineers to determine the proper designing, installation, and construction of the pump and engine on the site. We guarantee the proper installation and working condition of the project before the system is turned over to the association of landowners and tenants, who will eventually own the whole system. We have set a certain standard for a finished project called PHILCUSA-FOA STANDARD which consists of the following important requirements: 1. Engine and pump foundation should be permanent. 2. Pump and engine should be of correct size and design for the pump site. 3. Protective housing for the engine must be constructed. 4. Irrigable area to be comparable to full pump capacity considering type of soil and water duty. 5. Canals should have hydraulic elements which will minimize initial cost, maintenance cost, erosion and silting. 6. Canal structure should be of permanent nature. Of the fifty 16-inch and the ten 42-inch pump units, bought under the dollar allocation, here is the score to date: (1) 39 units now in operation (1 came from the former IRPA); (2) 11 units under construction; (3) 11 units to be bidded for construction. The rest of the bids will be ready for construction before the end of 1953. The condition for the operation and maintenance of the project is that the irrigation system shall be operated by the association which shall employ mechanics and watertenders as may be necessary, with the approval of the Administration of the ISU until the unit has been fully paid for. A head mechanic will be assisted by a second mechanic whenever it becomes necessary. There must be several watertenders for a unit and it is advisable to have at least one watertender for every 100 hectares irrigated to insure adequate distribution of water. To date, farmers' associations operating these pump units report excellent performance of both the engines and pumps. The people can hardly believe that the 16-inch Johnston pumps and 95 HP Cummins engines brought by the American people could deliver so much water sufficient to irrigate 200 hectares. In the words of Mr. Flavio Vasquez, vice president, Vasquez Irrigation Association, Sto. Tomas, Jaen, Nueva Ecija, "During plowing and planting times I let the Cummins engines run 24 hours a day and I have no complaint. They are just perfect." The people's reaction to having the irrigation water is certainly optimistic. They believe that with the irrigation water during this coming harvest season, they will significantly increase their yield from 50 to 75 per cent or more. The effect of the pump is by now felt in places where farmers and tenants have moved and built their houses in areas covered by the irrigation. The community's morale has increased and the people are now busy preparing the seed beds for the coming rice season. With more dollar allocation to buy more pumps and engines, and more peso counterpart to install and manage them, the Philippines will within a few years experience a bountiful harvest of rice in the rice-producing areas. CONSTRUCTION OF PUMP IRRIGATION SYSTEM To be able to appreciate the task of the Irrigation Service Unit in the construction of pump irrigation systems throughout the Islands under the PHILCUSA-FOA program, the reader should know the answer to the question: How does the ISU construct a pump irrigation system? According to the guiding principles jointly approved by FOA, PHILCUSA, and the Department of Agriculture and Natural Resources, a group of landowners with at least 75 hectares of land to be irrigated and with a suitable source of water supply may apply to the ISU (using ISU Form No. 2) for one or more pump units, depending upon the size of the irrigable area. Upon receipt of the application, the site is investigated by ISU fieldmen and if found feasible, the project is surveyed and designed. The applicants are advised to organize themselves into an association duly registered with the Securities and Exchange Commission. Upon the approval of the plans, construction is undertaken either by the association which spends for the labor with the ISU furnishing all the materials, or by contract whereby the ISU shoulders all the expenses. Excerpt from article by Mr. Pacifico Pinili, Chief Engineer, ISU 98 A pump irrigation project, as constructed by the ISU, must satisfy the following specifications: 1. A pump well and foundation on which the pump is mounted. It is located ordinarily a few meters inside the bank of the river in order to protect it from drifts during floods. 2. Engine foundation on which the engine driving the pump is set. 3. Housing for the engine and the mechanic. 4. Stilling basin to check the high velocity of the water discharged by the pump. 5. Main canal and laterals with the necessary structure to convey the water to the different parts of the irrigable area. For every 200-hectares irrigable area, 16"-diameter pump and a 65-HP or 95-HP engine (depending upon the lift, that is, the approximate distance from the top of bank to the minimum water level in the river) are installed. For a project of 1,000 hectares, a 42"-diameter pump a 200-HP or 250-HP engine are allocated. The ISU in one year of existence has been able to complete the construction of 34 projects with 38 16"-diameter and one 42"-diameter pump units installed, and nine other projects (six 16"-diameter and three 42"-diameter pumps), more or less completed. Irrigation water delivered by a 16-inch pump to irrigate 200 hectares in Sta. Cruz, Laguna. Main canal and a check gate for a 16-inch pump in Bayaniwan, La Paz, Tarlac. The main problems or handicaps under which the engineering division of the ISU worked during the first year were: 1. Lack of the necessary technical personnel with sufficient experience in pump irrigation design and construction. Being new the office had to employ mostly inexperienced men in view of the unavailability of experienced ones. 2. Lack of engineering equipment and transportation facilities especially in the provinces. 3. Lack of hydrographic data for almost all sources of water supply. In this connection, it may be suggested that funds to finance the hydrographic survey by the corresponding office of the Bureau of Public Works of all streams in the Philippines for use in the consideration of the construction of pump and gravity irrigation systems be provided. 4. Difficulty in the acquisition of rights-of-way for canals and for pump and engine sites. In most cases small landowners would not part with their lots if traversed by a canal. Likewise some landowners who have not signed the necessary applications but whose lands are within the irrigable area, refuse to give rights-of-way, without certain privileges even if the guiding principles of the ISU require that an application may be approved if signed by the farmers representing at least 3/4 of the area to be irrigated. 5. Difficulty in securing signatures of illiterate small landowners. Many are reluctant to sign the regular application for the installation of a pump for their benefit and also the power of attorney authorizing the ISU to apply for water rights in their behalf for fear that their lands may be forfeited. The benefits from the construction of 39 pump units by the ISU during its one year of existence will be appreciated when as a result of the operation of these pumps increased yields are obtained by the farmers. By then the ISU shall have justified its establishment. PULP AND PAPER PROJECT ERNESTO P. VILLAREAL Problems that greatly retard the development of the pulp and paper industry in the Philippines, while sundry and manifold, are surmountable; but the major hindrance is still the immediate unavailability of the basic raw material, the right kind that is abundant locally yet cheap enough for commercial purpose as to place the industry on a competitive position. Nearly all pulp and paper-making countries use as basic raw material the pulp wood, and whatever common wood predominating in their natural forests such as the pine and the spruce and few other species of trees indigenous to temperate climate regions. In a tropical country like ours, on the other hand, the natural forests consist of more than 3,000 species of trees of which 75 per cent of the forest stand are of the dipterocarps commonly called "lauan" to which the "Philippine mahogany" belong. None of these trees possess the necessary qualities equal or comparable to the spruce or pine as a pulp and paper material, except for our only long fibered tree, the Benguet Pine, whose growth and natural habitat, however, is limited to the higher altitudes of the Mountain Province and the supply of which could hardly meet the present demand. Most of the production is used by gold mines in the Province as tunnel timber. Fortunately, of the locally-available dipterocarps, the "white lauan" and some light colored trees, mostly second growth, possess some qualities, though inferior to those of the pine and spruce, that may to a certain extent be employed for special kinds of paper. At present, however, there is no attempt to produce on a commercial scale either newsprint or cardboard from these woods. All efforts, although still on an experimental basis, indicate a fairly bright prospect for paper-making from local woods. Aside from wood, other fibrous materials may be considered for manufacturing into pulp and paper. Full utilization and conversion of sugarcane bagasse into paper has been long advocated as an alternative solution that may yet save the sugar industry whose future, like that of our coconut industry, is none too bright. While it is true that the Philippines is the first country in the world to have made and produced paper from bagasse on a commercial scale, yet the limited demand for the kind of paper produced in Bais may not warrant the further establishment of similar paper mills. (India is the second country to make high grade paper from bagasse.) Furthermore, the sugar mills were forced to resort to very expensive bunker oil as replacement of bagasse as fuel. Considering the high cost of fuel oil (a drum of bunker fuel, 200 liters, is quoted at ₱28.32 per drum in Manila) and the scarcity of local coal, substituting these for bagasse as fuel, in order to utilize bagasse as a material for producing cheaper grades of paper whose market value is not substantial enough to make the enterprise profitable evidently can not really be called a solution to the problem of sugar industry at all as it was originally conceived. For one thing, surplus bagasse from different sugar centrals can serve as a raw material for paper only if this surplus could be processed into semi-pulp form, then baled and transported to the nearby pulp and paper mill. Again, the uncertainty of the bagasse supply and the ab- * Paper Technologist, Cebu Portland Cement Company, Manila sence of long term contracts between the sugar mill owner on the one hand and the paper mill owner on the other, in addition to the complexity of the problem that may arise from such a scheme, are obstacles that can discourage prospective paper manufacturers from investing millions of pesos in an industry which will necessarily have to be dependent on another industry for its continued existence. Making paper from rice straw has been advanced by a group of local scientists. There is a move at present to establish a cottage industry for making handmade paper from rice straw. This proposed cottage industry has the objective of providing additional income to our farmers during the off-season. Laudable as the project is, rice straw, however, has not been found to be good material for making paper. In the first place, fibers from the straw are short and fine. The silica content is high. Moreover, the presence of nodes in the rice stalks produces dark spots and impurities which cannot be removed during the pulping and bleaching processes. Also, the resulting paper from rice straw is stiff and hard, lacking the absorbent qualities of paper from wood pulp. Lastly, pure rice straw pulp could not be used for paper. Unless mixed with a long fibered pulp, with the percentage of rice straw pulp relatively small, rice straw pulp cannot be made into paper and therefore, can not be regarded as a basic paper material. An industry of this kind may have future possibilities, no doubt, if more suitable material for making a special hand-made paper is used and there is a demand and ready market of it. Japan has been favored with cottage industries, aided and encouraged by several government research laboratories, which guide the people in the use of suitable raw materials, not to mention cheap labor and the existence of a ready home market. The raw material known as "kozo," which grows wild or is cultivated on the hill sides, and the bark of the mulberry tree, harvested every three years, offer very good basic materials for making hand-made paper. Practically all houses of the masses use hand-made paper for windows and walls and other decorative fixtures for which the Japanese craftsman is noted. Umbrella making, also a traditional household industry, uses handmade paper to a great extent. These factors, coupled with cheap skilled labor, justify the existence of a sound and solid cottage industry that has flourished for hundreds of years and has remained unhampered by the establishment of modern paper mills. In England, the making of special hand-made paper for bank notes, official documents, parchment papers, etc., is a flourishing industry and a highly developed art. Efforts by the machine paper manufacturers to duplicate such kind of papers have been to no avail. Rice straw would be more suited to the manufacture of strawboard than bleached paper. Making strawboard is much simpler and more economical. Rice straw will cost less to make into strawboard, being a farm waste aside from being easily available especially in the Central Luzon provinces. There is a great demand for pasteboards and cardboards either as insulation board or as boxes, as containers for haberdashery, toilet articles produced locally, canned foods, etc. Our importation of pasteboard and cardboard amounted to P10,127,594 in 1949 and P7,129,258 in 1950. For importing boxes alone, we spend around two and a half million pesos yearly. These materials could be made from rice straw. The government should, therefore, encourage the establishment of a strawboard industry, even by foreigners, should local capitalists feel hesitant in investing money in such a venture. Bamboo also has good paper-making possibilities. There are well-established paper mills utilizing bamboo as raw material in some countries, notably Siam and India. Local bamboo, especially "boho" (Schizostachyum limampao), are found in thick stands in the national parks and U.S. army and naval reservations in Bataan and in some parts of Zambales. "Boho" could also be grown and cultivated in other areas. A medium-sized paper factory utilizing "boho" as raw material can be started now, but hand in hand with its operation should be the future acquisition of sufficient areas of public land where "boho" grows. This area must be developed and maintained to assure a continuous supply of the material. "Boho" is at present used as building material and in the manufacture of "sawali" as a substitute for wallboard; sawali-making as a household industry is developed in barrios where "boho" is available. However, it is regrettable that, as in the case with other forest products, there is rampant and indiscriminate cutting of "boho." If this harmful practice is not abated or stopped, there will be a rapid depletion of this forest product in the near future. Abaca waste is one of the best local raw materials for making paper. Because of encouraging results in making strong paper from abaca waste, the Cebu Portland Cement Company has decided to erect a small pulp plant building this material. A mixture of pulp produced from abaca waste and kraft waste paper and other cheap materials has produced a satisfactory if not excellent paper for cement bags. The limited supply of abaca waste, difficulty of collection because of the lack of roads from small stripping sheds to production centers, and lack of facilities for drying and baling are problems which will have to be solved before this material can be readily available at cheaper cost. The pulp mill should be located with the paper mill in Cebu because the existing pulp machinery there will not warrant its establishment elsewhere.* Pulp and paper mill machinery, like other machinery of a highly specialized industry, are very expensive and difficult to obtain at present. In the United States, where there is at present a shortage of metals particularly copper and steel, priorities for the defense program will have to be met first; in Europe, the big shops are filled with commitments, their standing orders and future deliveries running from two to three years ahead. In Japan, only a few essential machines may be ordered on shorter delivery time. Considering these present difficulties of acquiring new pulp and paper machinery, a new mill will not only be costly but may also take years to erect and be ready for operation. The distant position of the Philippines from other pulp and paper manufacturing countries is another difficulty which will have to be overcome. Executives for this new industry will have to plan one year ahead for mill supplies and spare parts, with provision for more capital for imported supplies and shop equipment for mill maintenance. There is no reason why pulp mills should not be, in the future, established in abaca-producing regions.—Ed. The present high cost of fuel and power are other problems that need thorough study and investigation. Moreover, being a new industry there is the necessity for hiring foreign experts, experienced pulp and paper makers and mechanics, during the early stages of operation so as to give training to local technicians and personnel. One great factor that will favor the development of the pulp and paper industry in the country, aside from high prices and the increasing consumption of paper, is the world wide shortage of pulp wood and the present tendency of using more and more short fiber (broad leaf and hardwood) trees as raw materials. With the exception of Canada, which is still underdeveloped, the great pulp and paper producing countries are suffering from a shortage of forest resources and other fiber materials. Better methods of pulping and processing hardwood have been discovered; materials that were never used as pulp materials before are now being processed into paper on a commercial scale. The production of sulphate and semi-chemical pulps from hardwood or broad leaf trees is now fully developed and there is a justification for using any of the above processes for local woods in the production of certain kinds of paper and even fiberboards. Although most local trees do not possess all the necessary qualities for the production of paper, yet there are some which can be utilized to a certain extent when the pulp obtained from them is mixed in a large proportion with long-fibered pulp which we may have to import. These trees are abundant and if systematic methods of logging and lumbering are properly carried out so as to utilize every portion of the tree including the slabs, small branches, etc., it is possible that costs can be brought to a minimum and enable the industry to stand on its own feet. Utilizing our forests to the maximum advantage, and producing only such products for which a tree is best suited, we might yet find a way to solve this distressing problem of raw material supply. A good case in point is that of the Benguet Pine, perhaps the only material found locally that is ideally suited for making pulp and paper, which is being used as mine timber, even if it is not appropriate for such purpose because of its brittleness under prolonged load. If the Benguet Pine is earmarked for use by the pulp and paper makers only, then we will be utilizing our forest trees to the best advantage. For reinforcing mine tunnels we can urge the use of common lumber which is abundant and could be transported to the mines cheaply.* A similar scheme can, likewise, be followed regarding other fiber materials. Using bagasse for paper-making was decidedly a good start; utilizing rice straw for strawboard may be another. Also producing paper from abaca wastes, bamboo, etc., are possibilities which we can exploit further. Using the pine tree and a few other local trees exclusively for pulp manufacture is a certain way of beginning this industry. Erecting an ambitiously large pulp and paper factory may not be possible because of certain limitations but erecting modest ones in places where the supply can be guaranteed may after all prove to be more practical and logical in the long run. This point may be disputed, in the absence of supporting statistics.—Ed. CLIMATE OF MANILA SEVERINO L. KOH TOPOGRAPHY Located at about 14°35' N. lat., and 121°E. long, Manila lies in the western coast of Luzon along Manila Bay. The bay, which covers an approximate area of 750 square miles, lies to the western quadrant of the city. Across the bay, in the province of Bataan are mountains highest among which are the Mariveles Mountains, with peaks up to 4,660 feet. North of these are higher mountains, the Zambales Mountain Range, which rise to heights as high as 6,700 feet. To the east of the Zambales Mountain Range and from the northwest to directly north of Manila is the Central Plain of Luzon. The Sierra Madre extends in a north-south direction to the east of the Plains and Manila. Two of the high peaks in the Sierra Madre near Manila are, Mt. Irid, 4,750 feet high, some 25 miles ENE of Manila, and Mt. Angilo, 4,300 feet high, about 25 miles NE of the city. Ten miles to the SSE of Manila is the northwestern shore of Laguna de Bay. Farther to the SE of the city and across Laguna de Bay is Mt. Banahao, 7,150 feet above sea level. Several other peaks are scattered nearby: Mt. Maquiling, 3,640 feet high, 35 miles S of Manila, and Mt. Batulao, 2,600 feet high, 40 miles SSW of the city. A general view of the topography around Manila presents an interesting picture. The high Sierra Madre Mountain Range shields the city from the direct effects of elements coming from the eastern semicircle. Through the Central Plains and between this Sierra Madre and the Bataan-Zambales Mountains, Manila is exposed from the north. In effect, this arrangement of mountains and plains serves like a channel oriented in a north-south direction with the city of Manila located at the southern end. Another interesting feature is the location of Manila Bay between the Bataan Mountains and those in Cavite and Batangas. The bay empties into the China Sea at a point (about 12 miles wide) some 40 miles to the southwest of Manila. Thus, southwesterly air currents, especially during the prevalence of the Southwest Monsoon, are "funneled" through this point, over the bay, and into Manila. Doubtless, these geographical features of the area surrounding Manila affect various weather elements as observed in the city. Frequently, characteristic climate peculiarities are satisfactorily explained by topography. In the discussion that follows, this influence of topography on climate is mentioned several times to explain and emphasize certain points. WIND As anywhere else in the Philippines, the seasonal variation in weather is largely influenced by wind flow. The high pressure area that predominates over the continent of Asia during the cold months of November to February causes northeasterly winds (frequently referred to as Northers) to prevail over Manila. With the subsequent weakening * Meteorologist, Weather Bureau, Department of Commerce and Industry, Manila and northward displacement of this Asiatic High after the month of February, the effects of the trade winds become more pronounced. Coming under the influence of the Northern Pacific subtropical anti-cyclone, the southwestern sector of this anti-cyclone spreads over the northern portion of the Philippines. Luzon, in particular, experiences southeasterly trades. Therefore, during the months of March, April, and May southeasterly winds are observed in Manila most frequently. We may add further that from November to May the prevailing winds over the city have an easterly component. Plate I (b) shows the percentage distribution of winds at Manila from November to May. During the rest of the year, i.e., from June to October, the Southwest Monsoon prevails over the Philippines. Consequently southwesterly winds are observed in Manila during these months. Plate I (a) shows the percentage of wind direction at Manila for this period. It must be mentioned here that June and October are transition months during which the southeasterly trades are replaced by the southwesterly winds in June and the southwesterly winds by the northerly in October. Due to other factors, the wind observed at Manila is at times at variance with the normal prevailing wind. A typhoon or a tropical depression in the vicinity of the Philippines might induce over Manila a wind from a direction entirely different from normal. Furthermore, being a coastal city, Manila is affected by the unequal heating and cooling of land and sea. Coming from the bay, a sea breeze is observed in the city during hot, sunny days when the land areas become rather heated. Land breeze is observed at night when these land areas have radiated much of the solar heat collected during the day and have become cooler than the surrounding water areas. The wind speed varies in strength with the months. The wind is quite steady and strong during the prevalence of the Southwest Monsoon. Particularly, the wind is strongest in August — that is considering the average wind speeds as computed on monthly basis from wind records of 46 years. Based on these same records, the wind is lightest in December. It might be interesting to note that the greatest wind force ever recorded at Manila was 118.5 miles per hour. This was on October 20, 1882. RAINFALL It was mentioned previously that topography has its influence on climate and that wind flow contributes greatly to seasonal variation in weather. The geographical distribution of land and water, mountains and plains, about a certain locality and the direction of wind flow (which is indirectly an index to the characteristics of the air mass over the locality)—the effects of these on rainfall variation in that locality cannot be denied. This is so particularly here in the Philippines. In fact, the existence of several types of rainfall characteristics over the Archipelago has served as the basis for classification of different climatic types in the Philippines. Manila, in particular, is characterized by two pronounced seasons in a year. The city is dry during the months of January to April. (According to Rev. Coronas' classification, a dry month is considered such in the Philippines if the rainfall for that month is less than 50 millimeters.) During the rest of the year, i.e., from May to December, Manila is wet with a marked rain period from June to October. The accompanying rainfall chart, Plate II, shows the monthly variation of average rainfall in Manila covering a period of 80 years. It is noteworthy that the pronounced rainy season from June to October coincides with the prevalence of the Southwest Monsoon. During this period the extremely moist air which is unhampered by mountains as it comes from the southwest over Manila Bay and into the city brings with it considerable amount of rain. The strong convective instability of this air mass, which is conducive to the formation of thunderstorms, further causes a great amount of rain to fall; thus, rainfall during these months is excessive in Manila. July, which is the wettest month in the city, has an average rainfall of 16.84 inches. This is nearly 20% of the average yearly rainfall of 81.52 inches. On the average, rain is observed for 24 days of the month in July. August and September are nearly as wet as July, with monthly rainfall averages for these two months equal to 16.58 and 13.95 inches, respectively. August has an average number of 23 rainy days in a month; while that for September is 22 days. This excessive rain period coincident with the prevalence of the Southwest Monsoon is also observed at several other localities in the Philippines, especially those that are situated in Western Philippines and to the west of the high mountains. Worth mentioning among these localities are Baguio City, Iba, and Olongapo. Baguio, which holds the world's record for greatest 24-hour rainfall collected, has an average rainfall of 44.70 inches in August, and 41.90 inches in July. Iba and Olongapo are wettest in July with rainfall averages of 38.92 and 38.83 inches, respectively, for this month. On the other hand, the other months of the year, November to May, are characterized by drier weather than that experienced from June to October. In Manila, the pronounced dry season from January to April coincides with the prevalence of the Northeast Monsoon. Coming from the northeast over the Pacific Ocean, the air is nearly as moist as the southwesterlies during the Southwest Monsoon. Excessive rain occurs over the eastern part of the Philippines, especially over the eastern slopes of the mountains in this region. Therefore, the air which has given off much rain becomes comparatively drier as it crosses the mountains. Streaming down the western slopes of Sierra Madre, the air is further dried and heated. Thus, little rain is available in the air as it streams over the city of Manila. This explains the pronounced dry season during the months of January to April. In Manila the driest month is February. An average of only 0.44 inch of rainfall is collected for this month. This amount is less than one per cent of the average annual rainfall. For the month of February an average of only three days a month is characterized by rain; the remaining days of the month are totally dry. During the dry season in Manila, other localities in the Philippines which are similarly situated in the western part of the Archipelago and to the west of mountains, such as the previously mentioned Baguio City, Iba, and Olongapo, also experience the dry season. From a maximum monthly average rainfall of 44.70 inches in August, Baguio City rainfall drops to a minimum of only 0.82 inch averaged for February. Similarly, Iba and Olongapo are quite dry during the months January to April. Iba is driest in February with an average rainfall of only 0.24 inch; while Olongapo is driest in March with an average of only 0.12 inch. These figures cited above are normal values computed from many years' records. For Manila, our normal values are based on records for a period of 80 years of observation (1865 to 1939; 1946 to 1950). There are times when the amount collected for a given period is far below the normal value. Then we have what is called drought. Manila has experienced several such droughts, the most severe of which, since the establishment of the Weather Bureau in 1865, is that which occurred in the year 1885. This is the driest year on record, with a total rainfall of only 35.69 inches collected throughout the year. This amount is less than 45% of the average annual rainfall of 81.52 inches. On the other hand, the rainfall accumulated for a given period may be far above the normal amount, in which case floods result due to insufficient drainage. These inundations occur especially if the rains remain heavy and continuous for long periods. Manila has been flooded many times in the past. Among the most remarkable floods that have ever been observed in Manila is the flood of 1919. This flood was mainly caused by the continuous heavy rains brought by the several typhoons that passed over or near the Philippines from the last week of July to the first week of September. An enormous total rainfall of 78.07 inches was accumulated in Manila during the 30 days it rained in the month of August. This amount of rainfall is on record as the greatest collected in Manila in a single month. It might also be mentioned here that this same year, 1919, holds the record for the greatest annual rainfall with a total amount of 154.35 inches. Equally remarkable is the flood of August 11-13, 1931. Of this flood which was one of the worst ever experienced in the Philippines, Mr. Bernardino Perez, who made a study of Manila floods from 1900 to 1943, has this to say: "The very heavy rainfall in coincidence with an unusual hour-foot tide rise caused the worst flood ever experienced in Manila. Three-fourths of the city was under water." TEMPERATURE In the tropics, within which region the Philippines is included, the annual variation in temperature is small compared with the variation during the day. A twenty-degree Fahrenheit variation in one day is common in Manila; whereas, for a year, the difference between the means of the warmest and coolest months rarely exceeds ten degrees Fahrenheit. To cite figures: temperature in Manila varies in a day from an average minimum of 69.0°F to an average maximum of 86.4°F in January. This is a variation of 17.4°F. In May, the variation in a day's time is from an average minimum of 75.3°F to an average maximum of 92.8°F or a range of 17.5°F. On the other hand, the annual variation is not as great. The mean monthly temperature varies from 83.4°F for May, the warmest month, to 76.8°F for January, the coolest month, giving an average annual range of 6.6°F only. Plate III shows the fluctuation of the temperature in Manila based on 61-year monthly averages. The annual mean temperature for the city is 80.0°F, which is slightly cooler than that averaged for the entire Philippines. It would be interesting to compare this 80.0° mean for Manila with those for other localities in the Philippines. To cite some: Baguio City, which is located at an altitude nearly 5000 feet above sea level, has a mean annual temperature equal to 64.2°F; Cebu City, 81.3°F; Davao, 80.4°F; Dumaguete, 80.8°F; Iloilo, 80.5°F; Legaspi, 80.7°F; and Zamboanga, 79.8°F. This further proves the point that in the Philippines latitude is not an important factor insofar as temperature observations go; what is more important is the altitude of the station. Finally, one would like to know the extreme temperature conditions observed at Manila. For a period of 61 years (1885 to 1940; 1946 to 1950), the temperature of 101.5°F on May 17, 1915 stands as the highest observed in Manila. In contrast to this, the lowest temperature ever registered in Manila was 58.1°F, observed on January 11, 1914. HUMIDITY AND CLOUDINESS Weather elements are closely related with one another. In particular, it is interesting how humidity, cloudiness, rain, and temperature are interdependent on each other. Everything else being the same, humid air is more conducive to cloud formation than dry air. Clouds under appropriate conditions result in rain. Then rain increases the humidity of the air. Further, the relative humidity of the air is high when the temperature is low; conversely, the relative humidity is low when the temperature is high. This fact is observed everyday as temperature fluctuates. However, this inter-relationship is not that simple. Although rain occurs only in the presence of clouds, since it is these clouds that give off rain, still it must be understood that the mere presence of clouds do not presage the occurrence of rain. Suffice it to say that there are types of clouds which ultimately result in rain, while there are other types which do not give any rain at all. Also, humid air does not always result in more cloudy weather. So on with the other relationships: rain and humidity, temperature and humidity, etc. Other factors have to be considered. Despite these complicated relationships between any two of the elements just mentioned, the mean values of these elements clearly show good correlation. Of humidity and rain: during the dry season, i.e. from January to April, the relative humidity is low. Averaging the monthly means for this period of the year gives an average relative humidity of 72.6%. These months of the dry season have monthly relative humidity means below 80%. That for the month of April is lowest with a value of 69.4%. The average relative humidity for the marked rain period based on the monthly means for the period from June to October is 83.7%. This is more than 11% greater than that averaged for the dry season. The mean relative humidity for each of the months of the rainy season is above 80%. September has the highest relative humidity with a monthly mean of 85.1%. In cloudiness, April, which has the lowest mean relative humidity, is also least cloudy. During this month an average of only four-tenths of the sky is covered by clouds. The other dry-season months, i.e. January, February, and March, are also characterized by nearly as small an average cloud coverage as that for April. The three months of July, August, and September, which are the months of greatest rainfall, are also the months during which the sky is cloudiest. On the average, nearly eight-tenths of sky is cloud-covered throughout this rainy period. Plate IV presents in comparative form rainfall, humidity and cloudiness variations in Manila throughout the year. THUNDERSTORMS Thunderstorms are frequently observed in Manila. For a period of five years, 1946 to 1950, thunderstorms were reported in the vicinity of Manila for 229 days. Of this total, 49 cases or 21% occurred in August, making this the month during which thunderstorms most frequently occur. Thunderstorms are observed in May, June, September, and nearly as frequently as in August. In fact, on the average no less than 4 cases are observed per month from April to October. During this same five-year period, no thunderstorm was observed in January. Several other months are also characterized by rare occurrence of thunderstorms. For the five years considered, a total of only 5 cases was observed in November, 3 in December, 1 in February, and 3 in March. On the average, only a single case a month occurs in each of the months of November, December, and March; none occurs in January and February. It is noteworthy that this "thunderstorm-less" period coincides with the dry season more or less; while the months of frequent thunderstorms, particularly the months from June to September, coincide with the rainy season. It must be remembered that this study on thunderstorm frequency is based on the record of five years only. This short-period record is not sufficient to establish normal values. In fact, it would not be surprising if the observations of the coming years would give results different from those based on this five-year record only. TROPICAL CYCLONES A study was made of the remarkable typhoons that passed over or near the Philippines for a period of 32 years. In this study, tropical cyclones of great intensities only were considered. A total of 123 such strong typhoons affected the Philippines during these 32 years under consideration. Of this number, 16.3%, or a total of 20 cases, passed within 75 miles of Manila. Forty-three or a percentage of 35, passed within 150 miles of Manila. The typhoon table appended at the end of this study shows the frequency distribution of strong typhoons and depressions over a period of 32 years. It is worth mentioning that during the months of January to April throughout the period under study, no remarkable typhoon passed within 75 miles of Manila. Again, none passed within 150 miles of the city during the months of January, February and March. Furthermore, it can be noted that despite the fact that strong typhoons affect the Philippines rather frequently in July and August, only a total of three such typhoons passed within 150 miles of Manila during the 32-year period. The majority of these typhoons passed north of the Philippines. SUMMARY It would be desirable to summarize the more important points raised above. Thus, we conclude with a brief summary of the climate of Manila: Easterly winds prevail over Manila from November to May—northeastly in November becoming southeasterly by May. From June to October west to southwest winds prevail over the city. The dry season starts from the latter part of December and continues to the end of April. May ushers in the rainy season. In temperature, Manila is slightly cooler than most of the other low-altitude localities in the Philippines. Mean annual temperature is 80.0°F. January is the coolest month, while May is the warmest. Finally, Manila is affected by strong typhoons most frequently in October and November. Very rarely do these typhoons occur in January, February, March or April. ACKNOWLEDGEMENT The author gratefully acknowledges the invaluable help and advice unselfishly extended to him by Mr. Lorenzo R. Goli, Chief, Statistics Section, Weather Bureau. Mr. Goli prepared most of statistical data used in this study. REFERENCES 1. Coronas, J. (1920): Climate and Weather of the Philippines, 1903-1918. Bureau of Printing, Manila. 2. Deppermann, C. E. (1940): Upper Air Circulation (1-6 Km.) over the Philippines and Adjacent Regions. Bureau of Printing, Manila. 3. Estoque, M. A. (1951): Circulation and Convergence over the Tropical Northwestern Pacific from May to December. Unpublished. 4. Maso, M. S. (1914): Annual Amount and Distribution of Rainfall in the Philippines. Bureau of Printing, Manila. 5. Selga, M. (1935): Charts of Remarkable Typhoons in the Philippines, 1902-1934. Bureau of Printing, Manila. ### TEMPERATURE IN °F. \| Month \| Mean (1885-1940; 1946-1950) \| Mean maximum (1885-1940, 1946-1950) \| \|-------------\|-----------------------------\|-------------------------------------\| \| January \| 76.8 \| 86.4 \| \| February \| 77.8 \| 88.0 \| \| March \| 80.1 \| 90.8 \| \| April \| 82.8 \| 93.4 \| \| May \| 83.4 \| 92.8 \| \| June \| 82.2 \| 96.5 \| \| July \| 80.6 \| 87.8 \| \| August \| 80.7 \| 87.4 \| \| September \| 80.3 \| 87.6 \| \| October \| 79.9 \| 88.0 \| \| November \| 78.5 \| 87.1 \| \| December \| 77.2 \| 86.4 \| Annual \| 80.0 \| 88.8 \| 72.8 \| Month \| Highest (1885-1939; 1946-1950) \| Lowest (1885-1939; 1946-1950) \| \|-------------\|---------------------------------\|-------------------------------\| \| \| Day and Year \| Day and Year \| \| January \| 95.4 6, 1947 \| 58.1 11, 1914 \| \| February \| 96.1 25, 1906 \| 60.1 18, 1920 \| \| March \| 98.1 (25, 1933) (17, 1946) \| 61.2 9, 10, 1911 \| \| April \| 100.4 30, 1915 \| 63.0 (2, 1913) (2, 1923) \| \| May \| 101.5 17, 1915 \| 68.0 (22, 1913) (1, 1921) \| \| June \| 99.7 (4, 1912) (5, 1946) \| 70.9 7, 1896 \| \| July \| 97.3 (7, 1915) (7, 1946) \| 69.4 2, 1919 \| \| August \| 95.4 6, 1900 \| 69.1 27, 1896 \| \| September \| 95.5 18, 1903 \| 69.4 24, 1919 \| \| October \| 95.2 10, 1903 \| 67.1 26, 1913 \| \| November \| 93.2 (5, 1923) (9, 1946) \| 62.2 30, 1911 \| \| December \| 94.3 14, 1947 \| 60.3 31, 1892 \| Annual \| 101.5 May 17, 1915 \| 58.1 Jan. 11, 1914 \| ## RAINFALL IN INCHES \| Month \| Average amount (1865-1939; 1946-1950) \| Average number of rainy days (1865-1939; 1946-1950) \| Greatest fall in 24 hours (1865-1939; 1946-1950) \| Day and Year \| \|-----------\|--------------------------------------\|-----------------------------------------------------\|--------------------------------------------------\|--------------\| \| January \| 0.91 \| 6 \| 7.33 \| 1, 1883 \| \| February \| 0.44 \| 3 \| 1.72 \| 3, 1921 \| \| March \| 0.69 \| 4 \| 2.36 \| 18, 1866 \| \| April \| 1.31 \| 4 \| 5.63 \| 29, 1905 \| \| May \| 5.06 \| 12 \| 8.32 \| 7, 1923 \| \| June \| 9.93 \| 17 \| 9.95 \| 15, 1891 \| \| July \| 16.84 \| 24 \| 11.56 \| 29, 1919 \| \| August \| 16.58 \| 23 \| 12.75 \| 10, 1947 \| \| September \| 13.95 \| 22 \| 13.23 \| 24, 1867 \| \| October \| 7.61 \| 19 \| 7.65 \| 15, 1918 \| \| November \| 5.53 \| 14 \| 10.96 \| 18, 1923 \| \| December \| 2.67 \| 11 \| 3.90 \| 19, 1889 \| \| Annual \| 81.62 \| 159 \| 13.23 \| Sept. 24, 1867 \| ### Highest monthly rainfall \| Month \| Year \| Year \| \|-----------\|------\|------\| \| January \| 1883 \| 0.00 \| \| February \| 1921 \| 0 \| \| March \| 1887 \| 0 \| \| April \| 1905 \| 0 \| \| May \| 1934 \| 0 \| \| June \| 1926 \| 0.98 \| \| July \| 1899 \| 5.28 \| \| August \| 1919 \| 2.80 \| \| September \| 1867 \| 2.00 \| \| October \| 1869 \| 0.38 \| \| November \| 1923 \| 0.24 \| \| December \| 1889 \| 0.01 \| \| Annual \| 1919 \| 35.69\| ### Lowest monthly rainfall \| Month \| Year \| Year \| \|-----------\|------\|------\| \| January \| 1905, 1946 \| Various \| \| February \| Various \| Various \| \| March \| Various \| Various \| \| April \| Various \| Various \| \| May \| 1875, 1889 \| 1893 \| \| June \| 1866 \| 1909 \| \| July \| 1885 \| 1911 \| \| August \| 1885 \| 1896 \| \| September \| 1911 \| 1885 \| \| October \| 1885 \| 1885 \| \| November \| 1885 \| 1885 \| \| December \| 1885 \| 1885 \| \| Month \| Prevailing wind direction \| Average wind velocity (1893-1918; 1946-1950) \| \|---------\|---------------------------\|---------------------------------------------\| \| January \| NE quadrant \| 4.1 \| \| February\| E quadrant \| 4.9 \| \| March \| SE \| 5.3 \| \| April \| SE \| 5.6 \| \| May \| SE \| 5.5 \| \| June \| W quadrant \| 5.6 \| \| July \| W quadrant \| 6.7 \| \| August \| W quadrant \| 7.7 \| \| September\| W quadrant \| 6.1 \| \| October \| W quadrant \| 4.1 \| \| November\| NE quadrant \| 4.1 \| \| December\| NE quadrant \| 3.8 \| \| Annual \| \| 5.3 \| \| Month \| Average Cloudiness (1885-1935; 1946-1950) \| Average relative humidity (1885-1910; 1946-1950) \| \|---------\|------------------------------------------\|--------------------------------------------------\| \| January \| 5.6 \| 77.2 \| \| February\| 5.0 \| 77.3 \| \| March \| 4.6 \| 70.7 \| \| April \| 4.1 \| 69.4 \| \| May \| 5.7 \| 75.2 \| \| June \| 7.0 \| 80.9 \| \| July \| 7.8 \| 84.4 \| \| August \| 7.8 \| 84.6 \| \| September\| 7.7 \| 85.1 \| \| October \| 6.8 \| 83.4 \| \| November\| 6.5 \| 82.2 \| \| December\| 6.2 \| 80.6 \| \| Annual \| 6.2 \| 78.9 \| ### FREQUENCY OF THUNDERSTORM DAYS * \| Month \| 1946 \| 1947 \| 1948 \| 1949 \| 1953 \| Total \| \|--------\|------\|------\|------\|------\|------\|-------\| \| January\| 0 \| 0 \| 0 \| 0 \| 0 \| 0 \| \| February\| 0 \| 0 \| 1 \| 0 \| 0 \| 1 \| \| March \| 1 \| 1 \| 0 \| 0 \| 1 \| 3 \| \| April \| 0 \| 8 \| 3 \| 1 \| 6 \| 18 \| \| May \| 9 \| 4 \| 2 \| 0 \| 16 \| 31 \| \| June \| 8 \| 5 \| 5 \| 11 \| 13 \| 42 \| \| July \| 2 \| 6 \| 5 \| 8 \| 3 \| 24 \| \| August \| 8 \| 6 \| 4 \| 17 \| 14 \| 49 \| \| September \| 6 \| 4 \| 3 \| 12 \| 9 \| 34 \| \| October\| 3 \| 0 \| 3 \| 10 \| 3 \| 19 \| \| November\| 1 \| 0 \| 0 \| 4 \| 0 \| 5 \| \| December\| 3 \| 0 \| 0 \| 0 \| 0 \| 3 \| \| Annual \| 41 \| 34 \| 26 \| 63 \| 65 \| 229 \| ### REMARKABLE TYPHOONS (1903-1934) \| Month \| Passed over the Philippines \| Passed within 75 miles of Manila \| Passed within 150 miles of Manila \| \|--------\|-----------------------------\|----------------------------------\|-----------------------------------\| \| January\| 3 \| 0 \| 0 \| \| February\| 0 \| 0 \| 0 \| \| March \| 0 \| 0 \| 0 \| \| April \| 3 \| 0 \| 1 \| \| May \| 4 \| 1 \| 3 \| \| June \| 11 \| 3 \| 5 \| \| July \| 15 \| 1 \| 1 \| \| August \| 17 \| 2 \| 2 \| \| September \| 22 \| 2 \| 6 \| \| October\| 22 \| 4 \| 8 \| \| November\| 17 \| 4 \| 13 \| \| December\| 9 \| 3 \| 4 \| \| Annual \| 123 \| 20 \| 43 \| ANNUAL AND SEASONAL PERCENTAGES OF WIND DIRECTIONS AT MANILA AFTER REV. CORONAS, 1920 PLATE II AVERAGE RAINFALL IN MANILA AMOUNT OF RAINFALL IN INCHES NUMBER OF RAINY DAYS (RAINFALL OF 0.01" OR MORE IS COLLECTED DURING A "RAINY DAY.") PLATE III TEMPERATURE IN MANILA (°F) 68 72 76 80 84 88 92 96 J F M A M J J A S O N D MEAN MAXIMUM MEAN MEAN MINIMUM PLATE IV ANNUAL VARIATION OF RAINFALL, CLOUDINESS AND RELATIVE HUMIDITY AT MANILA - AVERAGE RELATIVE HUMIDITY IN % - AVERAGE CLOUDINESS IN TENTHS OF SKY - AVERAGE RAINFALL IN INCHES 115 MARBLE DEVELOPMENT OF THE CEBU PORTLAND CEMENT COMPANY FILOMENO DUTERTE Jr.* In plain geological language, marble is merely a metamorphosed limestone. To the layman, however, any rock that takes a polish is, in general, marble. The layman would, therefore, take even polished granite for marble. As far as is known, there has been no attempt made by anyone to develop our granite deposits for building or ornamental purposes. This article covers only true marble rock, true as far as sense and form is concerned and as studied, explored and developed by the Cebu Portland Cement Co. At present there are three areas where marble is known definitely to exist in commercial quantity and is either being intensively explored or in its early stages of development. The first of these marble claims is in the Antipolo-Tercsa area on the island of Luzon, covering an area of approximately 450 hectares. Prior to the war, they were developed by the A. C. G. Marble Co., Inc. but for some cause this company sold its rights to the Cebu Portland Cement Co. Immediately after the war, the Elizalde interests through their energetic representative, Mr. P. Umberti, pioneered a post-war marble industry. Through negotiations, the Cebu Portland Cement Co., who has always wanted to encourage private capital, transferred its rights to the Elizalde interests, and this arrangement gave rise to the present "Marble Corporation of the Philippines." Because of the difficulty of securing locally, during early liberation, the essential machinery and equipment, Mr. Umberti endeavored to obtain from war surplus and other sources, enough parts to assemble a make-shift unit able to produce his requirements. There was no market then for finished marble slabs. Production of these, therefore had to be temporarily stopped and other products were produced. Chips, gravel for road paving, marble dust that could be used as filler in the rubber shoe industry, in glass-making, for agricultural purposes, etc. were produced. There is a market resistance to these products that must be overcome, since the present income derived is just enough to keep the present make-shift factory going. There could be no allowance for return of capital investment and purchase of new and modern machinery and equipment to be able to produce the desired products at low cost. This industry can only survive the competition presented by imported marble by-products, if and when additional, adequate capital is secured. Should this problem be impossible to overcome the government must come to the rescue. The government can be assured of a profitable return on investment in this industry and at the same time benefit the country. There is only one color of marble of any commercial value in this area: beige with white and golden streaks. It takes polish well. The finishing plant is composed of two gang saws, one big automatic cutting machine, two made-up polishing machines, a horizontal grind- * Technical Assistant, Cebu Portland Cement Company, Manila. ing table and several other equipment and tools adequate only for a fairly small scale production of finished marble slabs. Since there is no variety of colors for the customer to select from it would appear that production of marble slabs would only become a sideline in the production of marble chips and dust. Very recently the Marble Corporation of the Philippines expanded their quarrying operation to produce whitish marble blocks in Romblon for trans-shipment to their plant in Teresa, Rizal. The second marble claim is in the Naga-Toledo area of Cebu province. Before World War II, in the course of extensive operations of the Uling Coal Mines by the Cebu Portland Cement Co., a sizeable deposit of marble was discovered in the outlying vicinity of Uling, Naga, Cebu. Several samples were taken for study and determination of its economic value. It is not "true" marble but simply a limestone that takes a very high polish. There are several varieties of color ranging from grey to beige, golden and pink. The pink variety is slightly "terraceous" made up of alternate layers of beautiful pink and whitish veins in irregular formation. Several blocks of this variety were shaped from big boulders before the war, for slabbing at the Cement Plant by a gang saw that was transferred from Romblon. Due to the outbreak of the war, the plan did not materialize. Recently, these blocks were shipped to the newly established CEPOC marble plant in Manila and were sawed and finished into slabs. This is by far the most beautiful color variety ever produced locally. In the Naga-Toledo area there are seven placer claims of 64 hectares each, leased by the Cebu Portland Cement Co. The presence of other color varieties aside from the pink (Golden Cebu) are not attractive because of their existence in Romblon where the present development of marble industry is concentrated. However, the company has included in its program of future operation the production of "Golden Cebu" blocks which are to be finished in the Manila Plant. The third claim is in the Romblon area on the island of Romblon. The Island of Romblon has become a synonym for marble so that whenever this island is mentioned it is always linked to this "rock gold." Early Spanish "conquistadores" utilized Romblon marble in churches for the altar, fonts, flooring, etc. Up to the present, the natives of Romblon have been making almeris, sign bars, lapidas and many other typical products that are marketed throughout the archipelago. The production of marble in commercial form and quantity was started by the Philippine Marble Company in 1939. Gang saws, frasing and polishing machines, lathes and other finishing equipment were installed. Quarrying requirements were sufficient. Technical and practical know-how for the operations were furnished by Italians, who were brought to the Philippines. The company had reached a reasonable scale of operation as evidenced by the existence of present quarry faces and finished products. The Philippine Marble Company folded up just before the war and sold out to the Cebu Portland Cement Co. During the war nothing was done on this property. After liberation, the Cebu Portland Cement Co., because of the enormous amount of work involved in the rehabilitation of its cement plant and the expenditure it had to make in the expansion necessary to double its cement production, had neither the needed time, personnel nor capital to develop their other properties. Only recently has the company finally decided to launch a program to develop the marble industry into a large, full-producing enterprise to meet not only all local requirements but also for exportation. The old machineries in Romblon that were installed by the defunct Philippine Marble Co. were salvaged, in spite of having been looted during the war. Some of these are installed and are now in actual operation at the North Harbor Compound of the Cebu Portland Co. A complement personnel under the supervision of an experienced plant foreman, Mr. Claudio Basina, is being maintained. Finished typical marble products are produced daily. The present purpose of the plant is not for commercial production but to serve as a pilot plant to make studies of the different varieties of marble best suited for the market, to determine what proper machineries are needed to meet the requirements of producing the best product at a minimum cost and to train the men who will undertake the work. In this way, the company will be ready to meet the intricate problems inherent in full-scale production and keep to a minimum the risks involved thus insuring a reasonable dividend on the big capital investment. At the same time, the company is now maintaining a force at Romblon to locate and explore commercial quarry sites. Preparation of at least two quarries is well under way supervised by an experienced quarry foreman, Mr. Abraham Langara. By nature, quarrying operation is slow and expensive at the start because of the quantity of overburden and boulders to move out before the rock in place is reached. Several marble blocks have already been shaped and some were shipped to the Manila finishing plant for study. There is no argument that Romblon marble is a "true" marble. It is comparable to if not better than any foreign marble as to its hardness, color, texture and grain design. Marble from Romblon occurs in several color varieties ranging from whitish, golden pink, greenish, beige to gray. Different beautiful grain designs in the finished slabs are produced depending upon the angle the slab is sawn in relation to its definite line of color laminations. Undoubtedly, marble deposits in commercial quality and quantity exists in the Philippines. The set-back of its early development and expansion is primarily due to the coolness of capitalists who are averse to the investment of capital without being able to expect an immediate return on the investment in the first to third year of operation. There are groups of persons or associations very anxious to develop a marble industry in the Islands but because of either their lack of capital or the inability to raise this capital from subscription, their plans have never gone beyond the exploration stage. There are many ready local markets for finished marble for ornamental purposes in the construction of homes, churches, private and public edifices, tombs, etc. The import control imposed on foreign marble as an ornamental and building material is such as to make marble again a rarity and to command an exaggerated price. This can redound to the benefit of the industry in this country if marble production were immediately geared to present demand. Public edifices especially, which are now in the process of building, would serve as an outlet for finished products if available. Loading marble rock in Romblon. Agricultural Lime is also prepared as by-product of marble production. Ground Floor of the Philippine National Bank is furnished purely with Philippine Marble Unfortunately, private capital, although given encouragement along this line, is either too timid or too incompetent. Nothing remains therefore but for the government to immediately and on full scale undertake the operation and exploitation of an industry which not only means the conserving but also the making of dollars. Impetus should be give to the immediate development of the quarries and its product with the announced intention of the Battle Monuments Commission to use only Philippine Marble for the cemetery in Fort McKinley. It is hoped that the government will see in this enough encouragement to order the training of skilled technicians, to purchase the necessary machineries and to provide the required financial outlay to foster the marble industry in the Philippines. TYPHOONS AS A RETARDING INFLUENCE ON THE EAST COAST OF SAMAR MIKE McINTYRE* The recent typhoons of October, 1952, which were so exceedingly destructive of life and property, have reemphasized for the Filipinos, and brought to the attention of the world at large, the innate hazards involved in attempting sustained occupation along the east coasts of the northern Philippines. While this is not to intimate that other parts of the Islands are not subject to occasional typhoon disaster, the great bulk of the north Pacific's violent tropical disturbances is spawned along the intertropic front south of the Marianas, and they take a westward course toward the exposed eastern flanks of the archipelagos off the Asiatic mainland. There is evident a certain seasonal nomadism of these storms, associated with the advance and retreat of the high sun during the year, and likewise an annual frequency variation may be observed, with the greatest number appearing in the late summer and early fall. Thus, the season of maximum frequency coincides with the period when the major storm tracks are athwart the northern Philippines, and the east coasts, then, of these islands are particularly susceptible to frequent and serious devastation. Samar, situated toward the southern end of this danger zone, is relatively less afflicted than Luzon and the Batanes; nevertheless, fully 20% of the some 13 typhoons which cross the Philippines each year strike somewhere along the east coast of this land. Such storm frequency profoundly modifies the overall climate of Samar, although in some respects the effect is distinctly beneficial, especially in terms of reliable precipitation. But any region, where three to five storms each year with winds upward of 100 miles per hour, torrential rains, and dangerously high seas must be accepted as normal, is forced to adjust its economy drastically to cope with these unusual environmental stringencies. Wholly successful adjustment is a virtual impossibility. Where the situation is somewhat more critical, as in northeastern Luzon, the coast has simply been abandoned except by wandering hill tribes. Eastern Samar, however, does support a moderate population, but regional development has lagged perceptibly, retarded in large part by these periodic typhoon visitations. There are other factors, of course, aside from climatic considerations, which contribute to a lack of economic development of Samar's east coast. One of these is the peripheral location of the island within the archipelago and particularly the east coast, which faces out onto the Pacific instead of the busy inland seas. For although Samar was the site of Magellan's first Philippine landfall in 1521, it has since that time fallen more and more into the backwash of Philippine commerce and national affairs which tend to center strongly on distant Manila. The coastal district also exhibits a distinct paucity of low, flat land which in this region of subsistence rice culture constitutes a serious liability; for the number of people who can support themselves in a given area * Professor of Geography, Department of Geography, Wayne University, Detroit 1, Michigan, U.S.A. 1 U.S. Coast Pilot, Philippine Islands, Pt. I. Luzon, Mindanao, and the Visayas. 3rd Ed., Coast and Geodetic Survey, Washington, D.C., 1940, p. 32. is likely to be in rather direct proportion to the amount of land which can be converted into productive rice paddy. And the east coast plains are limited to a series of deltaic lowlands, none extensive and all non-contiguous. In addition, the rivers to which these deltas owe their origin, although carrying a considerable volume of water throughout the year, are seldom of gentle enough gradient to be classed as navigable for any distance; and all have their mouths obstructed by bars which allow nothing but small boat traffic. Nor do these isolated plains have access to one another by anything resembling a reasonable system of roads. The sole coastal road, a one lane affair from Guiuan to Oras, boasts of only three short sections which are passable in all seasons, and it is constantly interrupted by broad river estuaries which are crossed by non-vehicular ferries. There is a difficult mountain road leading across the narrow waist of the island from Taft to Wright, but fully a third of the east coast, from Oras to Palapag, is not served by any form of transportation artery other than forest trail. Seaborne coast-wise movement, which in many parts of the Philippines is an adequate substitute for undeveloped land facilities, is hampered here during the winter months by strong northeast monsoon winds. And potential anchorages in the strongly indented coastline are not only exposed to these winds, but are more often than not mangrove fringed and reef encumbered as well. These are admittedly serious difficulties, and to overcome them would require an imaginative and sustained planning project coupled with a considerable amount of working capital. But the problems are not insuperable, and it is eminently possible that a determined improvement program might achieve a great deal. For instance, centrally located Port Borongan might be transformed into a fine deep-water outlet for the entire region by the addition of a breakwater, a channel and turning basin cleared of coral snags, and additional wharfage construction. Such a course, however, assumes, for efficient use of the port, an adequate land transportation system which could funnel goods from the outlying areas and distribute imports. And this in turn assumes the production of a commodity surplus requiring markets outside of the immediate east coast district, and also a population with a standard of living well above the subsistence level which can create a demand for goods not locally available. All of these assumptions are far from valid at the moment, but improvements can be made. The prefabricated Bailey bridge has proved successful elsewhere in the Philippines as a cheap, yet reasonably permanent means of bridging rivers the width of those encountered along the east coast of Samar, and it might be introduced here to help tie together the poorly integrated road system. New roads can be built, existing ones overhauled, and primitive maintenance methods modernized. In the realm of agriculture, the basic industry of the region, a great deal could be done to increase the production and quality of both subsistence and commercial crops. The limited lowlands do not now produce sufficient rice to satisfy local requirements, but this is primarily due to backward farming methods rather than a shortage of land suited to rice. For even though the potential paddy land is in short supply, it could readily be made to produce triple its present yield. On the basis of annual rainfall and temperature, two crops a year is entirely feasible although seldom attempted at present. And, proper preparation of the field for planting by the use of plow and harrow (instead of merely the muddy trodding of carabao), fertilization, seed selection, and insecticides, could do much toward increasing yields. Concomitantly, vigorous attacks on an entrenched system of tenancy and a long tradition of privately financed rural credit at exorbitant rates of interest might well be instituted. There are on the east coast of Samar, two important commercial crops that for many years have entered the channels of international trade in small quantities. They are coconut and abaca. But through a combination of haphazard production methods and a needlessly involved system of marketing, the farmer has seldom realized a reasonable profit on the sale of his product, and has been the victim of a chronic fluctuating price structure. However, if the quality of Samar copra and fiber were improved so that it might enter the world market in a highly competitive position, price fluctuations could be held at a minimum. This upgrading of quality might be brought about, in the case of copra, by the adequate spacing of palms in the groves, by picking only the mature nuts, and by the application of modern drying techniques; and in the case of abaca by selective cutting of plants at the proper state of maturation, careful hand stripping of the fibers, and adequate drying and storage. But, above all, an incentive must be supplied the farmer by the elimination of the myriad middlemen in the marketing of his product and the establishment and enforcement of a workable grading system. These commercial crops are important in the east coast economy. Not only are they potential sources of capital, but they are complementary to rice culture rather than competing for land, for they occupy the sandy beaches and steep slopes which are useless for paddies. A few kilometers inland from barrio General MacArthur on the southeastern coast, is the largest post-war iron mining development in the Philippines. An open pit operation on what appears to be rather extensive deposits of 60% ore, this mine is really of post-war origin since production had just gotten underway at the outbreak of hostilities. The ores are readily salable to Japan; and although the annual output of something over 100,000 tons may seem small, the income from this mine in terms of working capital to be applied to the development of the east coast is of very major local significance. Thus it seems that the east coast of Samar has distinct possibilities of considerable regional development, given the requisite capital and planning. But this brings us back to our original proposition, namely, can any region which must accept several destructive typhoons each year as the normal order of things progress beyond the low economic and cultural level that has already been achieved in eastern Samar? It appears doubtful that a costly improvement program can be defended under these circumstances. There is little question but that some progress might be anticipated through the years in such things as seed selection, rural sanitation, simplifying of the marketing system, and the like; however, most public works are virtually ruled out by excessive maintenance costs in the face of serious and continuing typhoon devastation. Neither should the effects of these storms on the psychology of the people be underestimated. The prevalent "what's the use" attitude among the coast populace is easily understandable when every year coconut palms are damaged by high winds, weak abaca plants flattened, paddy fields flooded, grain shattered, PHYSIOGRAPHIC SAMAR SCALE 125 and houses and personal belongings carried away downwind. Even the massive loading dock of the Samar Iron Mining Company has sustained sufficient damage on several occasions to necessitate the closing down of mining operations for considerable periods of time. No one who has experienced a Samar typhoon doubts for a moment but that nature has the upper hand, and puny humans who at times are prone to boast of their mastery over nature find it expedient to simply hang on and pray. Science knows of no way to control typhoons. Actually, science is hard put to even explain their origin and the fickleness of their habits. At the present state of our knowledge the best that can be done is to attempt to track them once they have formed and to warn of their approach; but these are scarcely adequate measures to emancipate the east coast of Samar from its age old scourge. Perhaps the Sama-reño should be admired for his temerity in wresting a living from such a harsh environment rather than censored for his lack of progress. PAPERS ON PHILIPPINE GEOLOGY AND GEOGRAPHY COMPILED BY WARREN D. SMITH AND LISTED CHARLES O. HOUSTON, JR. 1. ( 1) Abella y Casariego, Enrique. Descripción Física, Geológica y Mínera En Bosquejo de la Isla de Panay. (Shorter title: Descripción de la Isla de Panay.) Publicación oficial. Manila: Tipo-Litografía de Chofré y Ca., 1890, 203 pages. Illus., engravings and maps. Author was Inspector-General of Mines. 2. ( 2) Centeno, D. Jose, D. Anacleto del Rosario y Sales, and D. Jose de Vera y Gomez. Memoria Descriptiva de los Manantiales Minero-Medicinales de la Isla De Luzon. Publicada de Real Orden. Madrid: Imprenta y Funcición De Manuel Tello, 1890, 120 pages. Tables. 3. ( 3) Abella and Casariego, Enrique José de Vera y Gomez and Anacleto del Rosario y Sales. Estudio Descriptivo de Algunos Manantiales Minerales De Filipinas. (Vol. 2 of above). Publicación Oficial. Manila: Tipo-Litografía De Chofré y Ca., 1893, 150 pages. Tables. 4. ( 4) Abella y Casariego, Enrique D. Terremotos Experimentados en la Isla de Luzon Durante Los Meses De Marzo y Abril de 1892, Especialmente Desastrosos en Pangasinan, Union y Benguet. Publicación Oficial. Manila; Tipo-Litografía de Chofré y Ca., 1893, 110 pages. Maps, drawings. All three of above bound in one volume. 5. ( 5) Burritt, Charles H. (Compiler). The Coal Measures of the Philippines. A Rapid History of the Discovery of Coal in the Archipelago and Subsequent Developments, with the Full Text of the Record of the MacLeod Coal Concession in Cebu, or The Uling-Sutac Coal and Railway Concession. (Report to the U.S. Military Government in the Philippines.) War Department, Division of Insular Affairs, August, 1901. Washington: U.S. Government Printing Office, 1901, 289 pages. Illus., maps, tables. Was Officer in Charge of the Mining Bureau (1st Lt., 11 Cavalry, U.S.V.). 6. ( 6) Eveland, A. J. A Preliminary Reconnaissance [sic] of the Mancayan-Suyoc Mineral Region, Lepanto, P.I. Department of the Interior, The Mining Bureau, Bulletin No. 4. Manila: Bureau of Printing, 1905, 58 pages. 43 illustrations, 6 maps and sections. Author was geologist, Mining Bureau. 7. ( 7) Hidalgo, J. G. Catálogo de Los Moluscos Testáceos de Las Islas Filipinas, Jolo y Marianas. I. Moluscos Marinos. Madrid: Imprenta de La "Gaceta de Madrid," 1904-1905, 408 pages. Author was member of the Royal Academy of Science. 8. ( 8) Leith, Andrew. The Geology of the Baguio Gold District. Commonwealth of the Philippines, Department of Agriculture and Commerce, Technical Bulletin 9. Manila: Bureau of Printing, 1938, 91 pages, paper; illus., maps. Author, Member of Geological Survey Division, Bureau of Mines. 9. ( 9) McCaskey, H. D. Report on A Geological Reconnaissance [sic] of the Iron Region of Angat, Bulacan. The Mining Bureau, Bulletin No. 3. Manila: Bureau of Public Printing, 1903, 62 pages. 41 illustrations, 15 maps; sketches and tables. Author was Mining Engineer for the Mining Bureau. 10. (10) Osias, Camilo. The Filipino Way of Life. The Pluralized Philosophy. (Boston: Ginn & Co., 1940), 321 pages. Cloth. 11. (11) Preliminary Report to the Board of Directors: Metropolitan Water District. Extension of Water Supply. (A. Gideon, Manager). August, 1922. Contains Report of Warren D. Smith: "Preliminary Report on the Angat River Project..." 79 typewritten pages; appendix, variously paged. 12. Smith, Warren D. The Coal Deposits of Batan Island. With notes on the General and Economic Geology of the Adjacent Region. Department of the Interior, The Mining Bureau, Bulletin No. 5. Manila: Bureau of Printing, 1905, 56 pages. Illustrated, 20 plates, 8 figures. Maps and Sections. 13. Worcester, Dean C. The Philippine Islands and Their People. A Record of Personal Observation and Experience, with a Short Summary of the More Important Facts on the History of the Archipelago. (New York: The Macmillan Co., 1898), 529 pages. Cloth. 12. Handbuch der Regionalen Geologie. 24 Volumes. (Figures in parentheses are editors' classificatory system.) (14) I. Niederlande. By G. A. F. Molengraaff and W. A. J. M. Van Waterschoot. 98 pages (I.3.). (15) II. Danemark. By N. V. Ussing. Heidelberg: Carl Winter's Universitäts-buchhandlung (printer of series), 1910, 38 pages. (I.2.). (16) III. A. Die Österreichischen Und Deutschen Alpen Bis Zur Alpino—Dinarischen Grenze (Ostalpen). By Franz Heritsch. Heidelberg: 1915, 153 pages. (II.6.a.). (17) IV. British Isles. By G. A. J. Cole, A. M. Davies, C. Davison, J. W. Evans, J. W. Gregory, A. Harker, O. T. Jones, P. F. Kendall, J. Parkinson, L. Richardson, W. W. Watts, and H. J. O. White. 356 pages. (III.1.). (18) V. La Peninsule Ibérique. A.—Espagne. Robert Douvillé, 175 pages. (III.3.). (19) VI. Island. H. Pjettruss. 22 pages. (IV.1.). (Iceland). (20) VII. Grönland. O. B. Boggild. 37 pages. (IV.3.). Heidelberg, 1917. (21) VIII. Die Nordatlantischen Polarinseln. Otto Nordenskjöld. 30 pages. (IV.2b.). Heidelberg, 1921. (22) IX. Finnoskandia (Skandinavien und Finnland). A. G. Hoghom. 197 pages. (IV.3.). (23) X. Armenien. Felix Oswald (übersetzt von Otto Wilckens [translator]). 40 pages. (V.3.). (24) XI. Kaukasus. A. F. V. Stahl. 79 pages. (V.5.). Heidelberg, 1923. (25) XII. Kleinasien. A. Philppsson. 178 pages. (V.2.). Heidelberg, 1918. (26) XIII. Persien. A. F. Stahl. 46 pages. (V.6.). (27) XIV. Zentralasien. Kurt Leuchs. 138 pages. (V.7.). Heidelberg, 1916. (28) XV. Balkenhalbinsel. A. Die Kustenländer Österreich—Ungarns. Richard Schubert. 51 pages. (V.1.a.). Heidelberg; 1914. (29) XVI. Syrien, Arabien und Mesopotamien. Max Blanckenhorn. 157 pages. (V.4.). (30) XVII. The Philippine Islands. Warren D. Smith. With a chapter on the Lithology by J. P. Iddings. 24 pages. (VI.5.). (31) XVIII. Madagascar. Paul Lemoine. 44 pages. (VII.4.). Heidelberg, 1911. (32) XIX. New Zealand and Adjacent Islands. P. Marshall. 78 pages. (VIII.1.). (33) XX. Oceania. P. Marshall. 36 pages. (VII.2.). (34) XXII. Afrique Occidentale. Paul Lemoine and: English Colonies on West Coast of Africa and Liberia. John Parkinson. 88 pages. (VII.6a.). Heidelberg, 1913. (35) XXIII. Aegypten. Max Blanckenhorn. 244 pages. (VII.9.). Heidelberg, 1921. (36) XXIV. Die Mittelatlantischen Vulkaninseln. C. Gagel. 32 pages. (VII.10.). Heidelberg, 1910. (37) XXV. United States of North America. Eliot Blackwelder. 258 pages. (VIII.2.). (38) XXVI. Antarktis. Otto Nordenskjöld. 29 pages. (VIII.6.). Heidelberg, 1913. 13. (39) Becker, George F. Report on the Geology of the Philippine Islands. Followed by a Version of Ueber Tertiäre Fossilen Von Den Philippinen (1895) by K. Martin. (Translation by G. F. Becker). Author was member of U.S. Geological Survey (1901). Accompanied Gen. E. S. Otis to Manila (app. in 1898), spent 14 months in Philippines; From: U.S. Geological Survey, 21st Annual Report, Part. III, 1901 (?) pp. 487-644. 14. (40) Eddingfield, F. T. Ore Deposits of the Philippine Islands. Reprinted from: The Philippine Journal of Science, Vol. VIII, No. 2, Sec. A., (April, 1913), pp. 81-104. Three plates and 4 text figures. Printed by: Bureau of Printing, 1913. (Damaged). 15. (41) Smith, Warren D. Contributions to the Stratigraphy and Fossil Invertebrate Fauna of the Philippine Islands. Reprinted from: The Philippine Journal of Science, Vol. VIII, No. 4, Sec. A., (Chemical and Geological Sciences and the Industries), (August, 1913), pp. 235-300. 20 plates. Printed by: The Bureau of Printing, 1913. (Damaged). 16. (42) Smith, Warren D. Geologic Reconnaissance of the Pidatan Oil Field, Cotabato Province, Mindanao. Separate from The Philippine Journal of Science, Vol. 20, No. 1 (January, 1922), pp. 23-42. Two plates and 3 text figures. Printed by: Bureau of Printing, 1922. 17. (43) Smith, Warren D. Brief Statement on Post-Cretaceous Formations in the Philippines. Reprinted from Special Publications of Bernice P. Bishop Museum, No. 7, pp. 770-774, 1921. 18. (44) Smith, Warren D. Areal Geological Mapping in the Philippine Islands. Reprinted from Special Publications of Bernice P. Bishop Museum, No. 7, pp. 673-676, 1921. 19. (45) Smith, Warren D. Pre-Tertiary Correlation in the Pacific Basin With Special Reference to Oregon and the Philippines. Reprinted from Proceedings of the Sixth Pacific Science Congress, 1939, pp. 429-435. 20. (46) Smith, Warren D. The Geological Structure of the Philippines. Reprinted from Special Publications of Bernice P. Bishop Museum, No. 7, pp. 890-894, 1921. 21. (47) Philippine Geological Papers, Bureau of Science, Volumes I-VI. Collected articles bound in 6 volumes. Contents of Volume I: (48) Bacon, Raymond Foss. "The Crater Lakes of Taal Volcano." Philippine Journal of Science, Vol. II, No. 2, pp. 115-126. (1907). (49) Cox, Alvin J. "Philippine Coals and Their Gas-Producing Power." Ibid., Vol. I, No. 8, pp. 877-902, 1906. (50) Cox, Alvin J. "The Proximate Analysis of Philippine Coals." Ibid., (V.2 (?), No. 1 (?) (1907?)), pp. 41-65. (51) Cox, Alvin J. "The Occurrence, Composition and Radio-activity of the Clays from Luzon, Philippine Islands." Ibid., Vol. II, No. 6 (December, 1907), pp. 413-437. (52) Eveland, A. J. "Notes on the Geology and Geography of the Baguio Mineral District." Ibid., Vol. II, No. 4 (August, 1907), pp. 207-233. (53) Ferguson, Henry G. "Notes on the Occurrence of Rhyolite in Cebu." Ibid., Vol. II, No. 6 (December, 1907), pp. 407-411. (54) Ferguson, Henry G. "Contributions to the Physiography of the Philippine Islands: II. The Batanes Islands." Ibid., Vol. III, No. 1 (February, 1908), pp. 1-25. (55) Smith, Warren D. "Orbitoides From the Binangonan Limestone (With Some Notes on Early Connections Between Formosa, the Philippines and Java.)" Ibid., Vol. I, No. 2, pp. 203-209, 1906. (56) Smith, Warren D. "Preliminary Geological Reconnaissance of the Loboo Mountains, Batangas Province." Ibid., Vol. I, No. 6, pp. 617-635, 1906. (57) Smith, Warren D. "Contributions to the Physiography of the Philippine Islands." Ibid., Vol. I, No. 10, pp. 1043-1059, 1906. (58) Smith, Warren D. "The Asbestos and Manganese Deposits of Ilocos Norte, with Notes on the Geology of the Region." Ibid., Vol. II, No. 3 (June, 1907), pp. 145-177. (59) Smith, Warren D. "Petrography of Some Rocks From Benguet Province, Luzon, P.I." Ibid., Vol. II, No. 4 (August, 1907), pp. 235-253. (60) Smith, Warren D. "The Geology of the Compostela-Davao Coal Field." Ibid., Vol. II, No. 6 (December, 1907), pp. 377-405. 22. Contents of Volume II: (Philippine Geological Papers.) (61) Adams, George I. "Geological Reconnaissance of the Island of Leyte—With Notes and Observations on the Adjacent Smaller Islands and Southwestern Samar." The Philippine Journal of Science, Vol. IV, No. 5 (September, 1909), pp. 339-357. (62) Adams, George I. "Sand, Gravel and Crushed Stone Available For Concrete Construction in Manila." Ibid., Vol. IV, No. 5 (September, 1909), pp. 463-481. (63) Adams, George I. "Editorial: The Marble and Schist Formations of Romblon Island." Reprinted from The Philippine Journal of Science, Vol. IV, No. 1 (January, 1909), pp. 87-89. (64) An Extract, possibly from the Bulletin Géologique, for November or December 1909, pp. 129-132. Primary reference is from Section, "Communications Orales" of meeting of 8 November 1909, by Henry Douvillé, "Sur Le Tertiaire des Philippines," pp. 130-131. Smith has added to chart printed on p. 131, the classification Miocene for "Sec. II, a, b, c," and Upper Oligocene to "Sec. I." (65) Cox, Alvin J. "Philippine Raw Cement Materials." Philippine Journal of Science, possibly Vol. IV, No. 2 (1909), pp. 211-229. (66) Cox, Alvin J. "Calorimetry, and the Determination of the Calorific Value of Philippine and Other Coals From the Results of Proximate Analysis." Ibid., Vol. IV, No. 3 (May, 1909), pp. 171-203. (67) Cox, Alvin J. "Volcanic Tuff as a Construction and A Cement Material." Ibid., n.d., pp. 391-407. (68) Cox, Alvin J. "Philippine Coals as Fuel." Ibid., Vol. III, No. 5 (November, 1908), pp. 301-355. (69) Cox, Alvin J. "Editorial: The Relationship Between the External Appearance and the Ash Content of Philippine Coal." Reprinted from Philippine Journal of Science, Vol. III, No. 2 (1908), pp. 91-93. (70) Dilworth, J.B. "Philippine Coal Fields." From: Transactions of the American Institute of Mining Engineers. (Delivered at the Chattanooga Meeting, October, 1908.), pp. 39-50. (71) Ferguson, H.G. "Editorial: Coal in the Cagayan Valley." Philippina Journal of Science, Vol. II, No. 6 (n.d.), pp. 595-537. (72) Ferguson, H.G. "Physiography of the Philippine Islands: III. Western Masbate." Ibid., Vol. IV, No. 1 (January, 1909), pp. 1-16. (73) Goodman, Maurice. "A Reconnaissance From Davao, Mindanao, Over the Divide of the Sahun River to Butuan, Including a Survey from Davao to Mati.—Narrative of the Expedition." Ibid., Vol. III, No. 6 (1908), pp. 501-511. (74) Smith, Warren DuPré. Geographical Work in the Philippines. "From 'The Geographical Journal' For November, 1909, pp. 529-544. (Printed by William Clowes and Sons, Ltd., London and Beecles). (75) Seven pages, interleaved, from the Philippine Journal of Science, notices of publications. Two pages contain list of publications to date of printing. (76) Short Article on "Starch Production in the Philippine Islands." No author, pp. 93-94, Philippine Journal of Science. Vol. III, No. 2 (1908). (77) Smith, Warren D. "The Coal Resources of the Philippine Islands." Reprinted from Economic Geology, Vol. IV, No. 3 (April, 1909), pp. 224-228. (78) Smith, Warren D. "Contributions to the Physiography of the Philippine Islands: IV. The Country Between Subig and Mount Pinatubo." Philippine Journal Science. Vol. IV, No. 1 (January, 1909), pp. 19-25. (79) Smith, Warren D. "A Geologic Reconnaissance of the Island of Mindanao and the Sulu Archipelago. I. Narrative of the Expedition." Ibid., Vol. III, No. 6 (1909?), pp. 473-499. (80) The Mineral Resources of the Philippine Islands, With a Statement of the Production of Commercial Products During the Year 1907. Issued by Warren D. Smith, Chief of the Division of Geology and Mines, Bureau of Science. Manila: Bureau of Printing, 1908, 39 pages. (81) The Mineral Resources of the Philippine Islands, With a Statement of the Production of Commercial Mineral Products During the Year 1908. Issued by Warren D. Smith, Chief of the Division of Geology and Mines, Bureau of Science, Manila: Bureau of Printing, 1909, 49 pages. 23. Contents of Volume III: (Philippine Geological Papers.) (82) Adams, George I. "Medical Survey of the Town of Taytay. II. Geology and Water Supply." Philippine Journal Science, Vol. IV, No. 2 (March, 1909?), pp. 211-218. (83) Adams, George I. "Geologie Reconnaissance of Southwestern Luzon." Ibid., Vol. V, No. 2 (March, 1910), pp. 57-116. (84) Adams, George I. and Wallace E. Pratt. "Philippine Pottery." Ibid., Vol. V, No. 2 (March, 1910), pp. 143-153. (85) Douvillé, Henri. "Les Foraminifères Dans Le Tertiare Des Philippines." Ibid., Vol. VI, No. 2 (April, 1911), pp. 53-80. (86) Ferguson, H.G. "The Gold Deposits of the Philippine Islands." Reprinted from Economic Geology, Vol. VI, No. 2 (March, 1911), pp. 109-137. (87) Iddings, Joseph P. "The Petrography of Some Igneous Rocks of the Philippines." Philippine Journal Science, Vol. V, No. 2 (March, 1910), pp. 155-170. (88) Pratt, Wallace E. "The Eruption of Taal Volcano, January 30, 1911." Ibid., Vol. VI, No. 2 (April, 1911), pp. 63-85. (89) Smith, Warren D. The Philippine Islands, With a Chapter on the Lithology by J. P. Iddings. Same as Handbuch der regionalen Geologie. (VI. 5). No. XVII, in Smith's numbering. (90) Smith, Warren D. "The Essential Features of the Geology of the Philippine Islands." Philippine Journal of Science, Vol. V, No. 5 (November, 1910), pp. 307-343. Contains a bibliography of important works. (91) Smith, Warren D. "Geologic Reconnaissance of Mindanao and Sulu: II. Physiography." Ibid., Vol. V, No. 5 (November, 1910), pp. 345-363. (92) The Mineral Resources of the Philippine Islands, With a Statement of the Production of Commercial Mineral Products During the Year 1909. Issued by the Division of Geology and Mines, Bureau of Science, Warren D. Smith, Chief, Manila: Bureau of Printing, 1910, 81 pages. Title page has errata sheet. 24. Contents of Volume IV: (Philippine Geological Papers.) (93) Cox, Alvin J. "Philippine Guano." Philippine Journal Science, Vol. ? No. ? (n.d.), pp. 195-199. (94) Cox, Alvin J. "The Oxidation and Deterioration of Coal." Ibid., Vol. VII, No. 5 (October, 1912), pp. 297-315. (Read before the section on Fuels and Asphalts, Eighth International Congress of Applied Chemistry, September 4-13, 1912, Washington and New York.). (95) Cox, Alvin J., W.C. Reibling, and F.D. Reyes. "Sand-Lime Brick and Artificial Sand Stones in the Philippines." Ibid., Vol. VII, No. 5 (October, 1912), pp. 317-335. (96) Eddingfield, F.T. "Ore Deposits of the Philippine Islands." Ibid., Vol. VIII, No. 2 (April, 1913), pp. 81-105. (97) Eddingfield, F.T. "Alteration and Enrichment in Calcite-Quartz-Manganese Gold Deposits in the Philippine Islands." Ibid., Vol. VIII, No. 2 (April, 1913), pp. 125-134. (98) Eddingfield, F.T. "Gogo, Entada Scandens Bentham, and Its Effect on Gold and Gold Solutions." Ibid., Vol. VIII, No. 2 (April, 1913), pp. 135-139. (99) Fanning, Paul R. "Geologic Reconnaissance of Northwestern Pangasinan." Ibid., Vol. VII, No. 4 (August, 1912), pp. 255-281. (100) Fanning, Paul R. "Editorial: A Philippine Natural Bridge." Ibid., Vol. VII, No. 4 (August, 1912), pp. 291-295. (101) Fanning, Paul R. and F.T. Eddingfield. "The Black Sands of Paracale." Ibid., Vol. VII, No. 4 (August, 1912), pp. 213-253. (102) Smith, Warren D. "Geological Reconnaissance of Mindanao and Sulu: III. General and Economic Geology." Ibid., Vol. VI, No. 5 (October, 1911), pp. 359-395. (103) Smith, Warren D. "The Geology of Luzon, P.I." Reprinted from the Journal of Geology, Vol. XXI, No. 1 (January-February, 1913), pp. 29-61. (104) Smith, Warren D. and Frank T. Eddingfield. "Additional Notes on the Economic Geology of the Baguio Mineral District." Philippine Journal Science, Vol. VI, No. 6 (December, 1911), pp. 429-447. (105) The Mineral Resources of the Philippine Islands for the Year 1911. Issued by the Division of Mines, Bureau of Science, Warren D. Smith, Chief, Manila: Bureau of Printing, 1912, 99 pages. 25. Contents of Volume V: (Philippine Geological Papers.) (106) Adams, George L. "Tests of Philippine Road Materials." Probably Philippine Journal Science, September, 1909, pp. 455-461. (107) Cox, Alvin J. and A.S. Arguelles. "The Soils of the Island of Luzon." Philippine Journal Science, Vol. IX, No. 1 (February, 1914), pp. 1-50. (108) Dalburg, F.A. and Wallace E. Pratt. "The Iron Ores of Bulacan Province, P.I." Ibid., Vol. IX, No. 3 (June, 1914), pp. 201-262. (109) Eddingfield, F.T. "Microscopic Study of the Bulacan Iron Ores." Ibid., Vol. IX, No. 3 (June, 1914), pp. 263-267. (110) Ferguson, H.G. "The Geology and Mineral Resources of the Aroroy District, Masbate." Ibid., Vol. VI, No. 5 (October, 1911), pp. 397-425. (111) Ickis, H.M. "A Geological Reconnaissance from Infanta, Tayabas, to Tanay, Rizal." Probably Philippine Journal Science, September, 1909, pp. 483-489. (Posthumous Publication). (112) Manó, Miguel Saderra and Warren D. Smith. "The Relation of Seismic Disturbances in the Philippines to Geologic Structure." Philippine Journal Science, Vol. VIII, No. 4 (August, 1913), pp. 199-233. (113) Pratt, Wallace E. "Geology and Field Relations of Portland Cement Raw Materials at Naga, Cebu." Ibid., Vol. IX, No. 2 (April, 1914), pp. 151-161. (114) Pratt, Wallace E. "Petroleum on Bondoc Peninsula, Tayabas Province, Philippines." Étude faite à la XIIe Session du Congrès géologique international, reproduite du Compte-Rendu, (Synopsis of article for Philippine Journal Science, by Pratt and Smith.), pp. 901-907. (1913)? (115) Pratt, Wallace E. and Warren D. Smith. "The Geology and Petroleum Resources of the Southern Part of Bondoc Peninsula, Tayabas Province, P.I." Philippine Journal Science, Vol. VIII, No. 5 (October, 1913), pp. 301-376. (116) Smith, Warren D. "Contributions to the Stratigraphy and Fossil Invertebrate Fauna of the Philippine Islands." Ibid., Vol. VIII, No. 4 (August, 1913), pp. 235-300. (117) Smith, Warren D. "Editorial: Notes on the Geology of Port Arthur and Vicinity." Ibid., Vol. IX No. 3 (June, 1914), pp. 269-270. (118) West, Augustus P. and Alvin J. Cox. "Burning Tests of Portland Cement Raw Materials." Ibid., Vol. IX, No. 1 (February, 1914), pp. 79-103. (119) Wright, J.R. and O.F. Smith. "A Quantitative Determination of the Radium Emanation in the Atmosphere and Its Variation With Altitude and Meteorological Conditions. Part I. A Quantitative Determination of the Radium Emanation in the Atmosphere at Manila." Ibid., Vol. IX, No. 1 (February, 1914), pp. 51-77. 26. Contents of Volume VI: (Philippine Geological Papers.) (120) Cox, Alvin J. and T. Dar Juan. "Salt Industry and Resources of the Philippine Islands." Philippine Journal Science, Vol. X, No. 6 (Dec., 1915), pp. 375-401. 17 plates. (121) Dickerson, Roy E. "A Fauna of the Vigo Group: Its Bearing on the Evolution of Marine Molluscan Faunas." Ibid., Vol. 18, No. 1 (January, 1921), pp. 1-23. (122) Goldsberry, J. P. "Eruption of Bulusan Volcano." Ibid., Vol. XI, No. 5 (September, 1916), pp. 251-256. (123) Heise, George W. "Notes on the Water Supply of the City of Manila." Ibid., Vol. XI, No. 1 (January, 1916), pp. 1-13. (124) Heise, George W. and R. H. Aguilar. "The Oxygen-Consuming Power of Natural Waters." Ibid., Vol. XI, No. 1 (January, 1916), pp. 37-47. (125) Juan, T. Dar and V. Eliciano. "Philippine Raw Materials for Glass Making." Ibid., Vol. XIV, No. 5 (September, 1917), pp. 465-479. (126) King, Albert E.W. "The Pozzuolian Properties of Meycauayan Volcanic Tuff." Ibid., Vol. XI, No. 1 (January, 1916), pp. 19-35. (127) Lednicky, V.E. "The Palidan-Slide Mine." Ibid., Vol. XI, No. 5 (September, 1916), pp. 241-249. (128) Pratt, Wallace E. "The Location of Artesian Wells in the Philippine Islands From a Geologic Viewpoint." Ibid., Vol. X, No. 4 (July, 1915), pp. 231-239. (129) Pratt, Wallace E. "Petroleum and Residual Bitumens in Leyte." Ibid., Vol. X, No. 4 (July, 1915), pp. 241-279. (130) Pratt, Wallace E. "On the Occurrence of Petroleum in the Province of Cebu." Ibid., Vol. X, No. 4 (July, 1915), pp. 281-287. (131) Pratt, Wallace E. "Geologic Reconnaissance in Caramoan Peninsula, Camarines Province." Ibid., Vol. X, No. 5 (September, 1915), pp. 303-321. (132) Pratt, Wallace E. "The Persistence of Philippine Coal Beds." Ibid., Vol. X, No. 5 (September, 1915), pp. 289-301. (133) Pratt, Wallace E. "Iron Ore on Calambayanga Island, Mambulao, Camarines." Ibid., Vol. X, No. 5 (September, 1915), pp. 323-333. (134) Pratt, Wallace E. "Philippine Lakes." Ibid., Vol. XI, No. 5 (September, 1916), pp. 223-239. (135) Pratt, Wallace E. and Victor Lednicky. "Iron Ore in Surigao Province." Ibid., Vol. X, No. 5 (September, 1915), pp. 335-347. (136) Smith, Warren D. "Notes on a Geologic Reconnaissance of Mountain Province, Luzon, P.I." Ibid., Vol. X, No. 3 (May, 1915), pp. 177-209. (Received for publication, November 27, 1914). (137) Smith, Warren D. "Notes on the Geology of Panay." Ibid., Vol. X, No. 3 (May, 1915), pp. 211-229. (138) Smith, Warren D. "Tropical Geology and Engineering." Ibid., Vol. 18, No. 3 (March, 1921), pp. 221-241. (139) The Mineral Resources of the Philippine Islands for the Year 1914. Issued by the Division of Mines, Bureau of Science. Manila: Bureau of Printing, 1915, 41 pages. (140) The Mineral Resources of the Philippine Islands for the Year 1915. Issued by the Division of Mines, Bureau of Science. Manila: Bureau of Printing, 1916, 39 pages. (141) The Mineral Resources of the Philippine Islands For the Year 1916. Issued by the Division of Mines, Bureau of Science. Manila: Bureau of Printing, 1917, 31 pages. (142) The Mineral Resources of the Philippine Islands for the Years 1917 and 1918. Issued by the Division of Mines, Bureau of Science. Manila: Bureau of Printing, 1920, 75 pages. (143) Witt, J. C. "Philippine Paving-Brick Materials: A Preliminary Report." Philippine Journal of Science, Vol. XI, No. 5 (September, 1916), pp. 203-221. GEOGRAPHICAL NOTES RICE PRODUCTION CONTEST The Soil Science Society of the Philippines is sponsoring the first Nationwide 100-cavan-per-hectare contest from April 15, 1953 to April 14, 1954, offering great opportunities to Filipino farmers to win prizes such as tractors, threshers, propeller pumps, diesel engines, rice and corn mills and other equipment for the farm. This was made possible through the cooperation of the various agencies of the Philippine Government, commercial houses and institutions of learning, engaged in promoting modern and progressive agriculture in the islands. The main purpose of the contest is to arouse the interest of the farmers in increasing rice production through proper and modern agricultural practices to attain self-sufficiency in rice and reducing its cost of production. Also to help establish model rice units in rural communities which will serve as examples to farmers in their respective areas, and to demonstrate the need for crop diversification and feasibility of livestock farming in the Philippines. Any farmer engaged in rice production and is interested in winning modern farm equipment and machineries can enter the contest provided he is willing to abide by the rules and regulations required. The said rules and regulations can be furnished by the Provincial and Municipal Agriculturist in charge of the area. Submit your entry form as soon as possible. The contest is classified into three divisions as follows: (1.) Upland, (2.) lowland, and (3.) lowland with irrigation. The number of entries needed to qualify in each type of culture should exceed or equal at least the number of registered municipalities or districts in the province or chartered city, respectively. The winner in each type of culture in each province automatically becomes candidate for the regional contest and the winners in the regional contest automatically become candidates for the National contest. So each entry has a chance to win three major prizes. Misinterpretation of the rules and regulations in some provinces hindered the progress and expansion of the contest. Many of the farmers do not want to enter because they thought they had to pay five pesos for soil analysis even if they are not interested in the analysis. Most of those complaining about the fees are owners of more or less half a hectare and the said amount means a great deal to them. So to avoid further misunderstanding the Soil Science Society amended the rules and regulations about soil analysis in such a way that soil sample analysis will be free to all contestants and those who gave and are willing to give five pesos will automatically become members of the Society. Now the only requirement that makes the farmers reluctant to enter is the procedure required in harvesting which calls for a specific length of time to finish the harvest. The contestant would provide plenty of "Bayani" to complete the harvesting and threshing within the specified time, and this will be a problem to most of the farmers for they plant and harvest almost at the same time. The end of the contest is fast approaching and still there are but 24 provinces that are qualified out of the 53 provinces in the Philippines as of October 15. Listed below are the qualified provinces and the types of culture entered: 1. Agusan — lowland and lowland with irrigation 2. Basilan — lowland 3. Bataan — lowland with irrigation 4. Batangas — upland 5. Bohol — upland and lowland 6. Bukidnon — lowland 7. Camarines Sur — upland, lowland with irrigation 8. Cavite — upland, lowland and lowland with irrigation 9. Cotabato — lowland 10. Ilocos Sur — lowland with irrigation 11. Isabela — lowland 12. Marinduque — lowland 13. Masbate — lowland 14. Mindoro Oriental — upland and lowland with irrigation 15. Misamis Occidental — upland and lowland with irrigation 16. Negros Occidental — lowland 17. Nueva Vizcaya — lowland with irrigation 18. Pampanga — upland, lowland and lowland with irrigation 19. Quezon — lowland 20. Romblon — lowland 21. Surigao — lowland 22. Tarlac — lowland and lowland with irrigation 23. Zambales — lowland 24. Zamboanga del Norte — lowland It is surprising to note that among the ten leading provinces in rice production, who are expected to join the contest, the following are not yet qualified: Nueva Ecija, Pangasinan, Iloilo, Lanao, Bulacan, and Samar. MARIA CRISTINA FALLS HYDROELECTRIC PROGRAM The Maria Cristina Falls Hydroelectric Plant and its sister project the Chemical Fertilizer Plant both located in Lanao are now realities. Their materialization is the culmination of many years of planning by the National Power Corporation. The harnessing of the Maria Cristina Falls is prompted by the need to abort the tendency of concentrating industrial development in Manila and its suburbs. The hydroelectric plant is capable of supplying power of 25,000 kilowatts while the Chemical Fertilizer Plant is expected to give a yearly output of 50,000 metric tons of fertilizer. The estimated yearly income of P13,000,000 from the chemical fertilizer product alone and the rice yield from fields using the locally produced ammonium sulphate which is about P12,000,000 can justify the P19,000,000 invested in these two projects. These will not only benefit the nation as a whole which is manifested in the acceleration of our national economic program but it will also affect the people individually by making possible the employment of many due to the expected increase in the industrialization of that part of the Philippines. All in all, the Maria Cristina Falls Project is a boon to our program toward the economic stabilization of the country. THE EIGHTH PACIFIC SCIENCE CONGRESS The Eighth Pacific Science Congress which will be held in the Philippines from November 16 to November 28, 1953 is under the auspices of the Republic of the Philippines and the National Research Council. The latter is a recognized scientific body of the country. This conference will convene delegates representing the constituent countries, those appointed by the Representative Institutions of the Pacific Science Association, scientists personally invited by the Society and officers of the Congress for the purpose of helping improve the welfare of all peoples of the Pacific area. The Pacific Science Council consists of fourteen member-countries: Australia, Canada, China, France, Hawaii, Indonesia, United States, Japan, Netherlands, New Zealand, Philippines, U.S.S.R., United Kingdom and Vietnam, although there are many other countries eligible for membership in the Association that have participated in previous congresses with their representative institutions. During the conference, various activities are scheduled. Representatives will read their papers on original researches. There will be meetings to consider matters presented before the body and symposia to discuss scientific findings for the welfare of the people not only of the Pacific but of the whole human race. Tours of interest to geologists, botanists, ethnologists, agriculturists and others are organized to be held immediately before and after the Congress and also during the week end between the two weekly sessions. This Congress is of great importance to peoples of the vast Pacific area — home of more than one-half of the human race. Findings of these group are the basis of most of human progress. They delve deeply into the great unknown and make startling discoveries that usually make world history. It is of significance to the Philippines because it is the first time, since the organization of the Pacific Science Society in 1920, that their conference is held here. ONE YEAR OF THE IRRIGATION SERVICE UNIT The Irrigation Service Unit, a PHILCUSA-FOA project for the implementation of pump irrigation program in the Philippines, aims to provide water thru the installation of irrigation pumps to communities of small farmers whose fields depend solely upon rainfall for water supply during the rainy season and remain idle and uncultivated during the dry season. This will increase the yield of regular season crops and make possible the planting of a second crop during the dry season. As a result the income of farmers will increase and ultimately raise their standard of living. Since October 1, 1952 when the ISU came into being, it has gained a foothold in our agricultural development program and its activities have been geared to the demands of our fast becoming science-minded farmers. As of today, thirty-nine pump units have been installed in thirty-four projects. Eleven are under construction, eight are to be bidded and four are under study. This amounts to 62 irrigation pump units which will be completely functioning at the end of this year as per schedule. The thirty-four pump irrigation projects now in operation cover an area of 7,800 hectares principally planted with rice. Thru pump irrigation, the potential rice production to be added yearly to the rice supply of the country will be 468,000 cavanes. With irrigation system, almost any crop can be planted the year round provided the soil and the climate will permit and if the farmer is interested in working. COLOMBO PLAN TRAINEES Colombo Plan trainees shown at Woodstock Farm, England, Shell's experimental agricultural station. Mr. Justo C. Gonzales, Instructor at the Araneta Institute of Agriculture and Mr. Ambrosio Ramos, Assistant Entomologist at the Bureau of Plant Industry, may be seen sitting in the front row. Among a party of Colombo Plan Students who recently underwent a period of training with the Shell Company in the United Kingdom were Ambrosio Ramos, Assistant Entomologist, Bureau of Plant Industry, Manila, and Justo C. Gonzales, Instructor, Araneta Institute of Agriculture, Malabon. During their period of training with Shell, the students were given an insight into the important role played by that Company in the never-ending fight against the many plant pests and diseases existing today throughout the world. They were given up-to-date information concerning Shell's research centers situated in the United Kingdom, Holland and the U.S.A. They were also told of the many facilities existing for extensive field trials in almost every country in the world. A visit was made to one of Shell's experimental farms situated near Sittingborne in Kent, England, where a ceaseless research is carried on in order to find improved agricultural chemicals and where improvements in the technique for the application of existing chemicals are made. This included Shell Company's own engineering workshop where experimental farm machineries, designed by Shell's engineers, are experimented with in connection with the application of new crop protection material. Particular interest was shown by the visitors in the outstanding new insecticides—Aldrin and Dieldrin—which are today contributing greatly to increased crops and improved public health throughout the world. FEATI INSTITUTE OF TECHNOLOGY Member, Philippine Association of Colleges & Universities COURSES OFFERED Four-year courses leading to the degree of Bachelor of Science in Engineering Two-year courses leading to the title of Associate in Engineering Two-year courses leading to the diploma in vocational courses Flying Courses (Enrollment open any time) One-year vocational courses Four-year Technical High School course Summer classes start May 2, 1953 For particulars, write or see the Registrar STA. CRUZ, MANILA TEL. 3-26-25 ARANETA INSTITUTE OF AGRICULTURE Villanueva Park, Malabon, Rizal The only private college in agriculture offering courses in: Four-Year Courses leading to the following Degrees: ★ Bachelor of Science in Agricultural Administration (B.S.A.A.) ★ Bachelor of Science in Agriculture (B.S.Ag.) ★ Bachelor of Science in Horticulture (B.S. Agro.) ★ Bachelor of Science in Animal Husbandry (B.S.A.H.) ★ Bachelor of Science in Agricultural Engineering (B.S.A.E.) Four-Year Course with the Degree of B.S.A., B.S. Agro. or B.S.A.E. with Major in Soils Two-Year Course leading to the Title of Associate in Agriculture (A.Agr.) One-Year Special Courses leading to the following: ★ Bachelor of Science in Agricultural Engineering (B.S.A.E.) ★ Bachelor of Science in Agricultural Education (B.S.A.Ed.) Four-Year Course in Vocational Agriculture One-Year Special Course in Farm Mechanics Short Courses in Poultry, Piggery, Dairy, Tractor Operation and Nursery Management. Subject to Special Arrangement Correspondence Course in Supervised Poultry Management OSMEÑA COLLEGES Masbate, Masbate Courses offered: Education, Junior Normal, Academic Secondary, Secondary Commercial, Complete Elementary. Vocational: Stenography & Typewriting. ANACLETO R. PELEÑO — Director SWORN STATEMENT (Required by Act 2580) The undersigned, Dominador Z. Rosell, business manager of Philippine Geographical Journal, published Quarterly in English at Soil Conservation Bldg., Manila, after having been duly sworn in accordance with law, hereby submits the following statement of ownership, management, circulation, etc., which is required by Act 2580, as amended by Commonwealth Act No. 201: Name: Post-Office Address Editor: Charles O. Houston, Jr. ........... University of Manila Business Manager: D. Z. Rosell .................. Soil Conservation Bldg., Manila Owner: Philippine Geographical Society ..... Soil Conservation Bldg., Manila Publisher: Philippine Geographical Society ..... Soil Conservation Bldg., Manila Printer: Carmelo & Bauermann, Inc. ...... 2057 Azcarraga, Manila Office of publication: .. Soil Conservation Bldg., Manila If publication is owned by a corporation, stockholders owning one percent or more of the total amount of stocks: None Bondholders, mortgagees, or other security holders owning one percent or more of total amount of security: None In case of publication other than daily, total number of copies printed and circulated of the last issue dated March, 1953: 1. Sent to paid subscribers ............... 372 2. Sent to others than paid subscribers .. 557 Total ........................................ 929 (Sgd.) DOMINADOR Z. ROSELL Business Manager Subscribed and sworn to before me this 7th day of October, 1953 at Manila, the affiant exhibiting his Residence Certificate No. A-0125005 issued at Manila, on January 15, 1953. (Sgd.) D. P. TUCAY Notary Public until Dec. 31, 1954 Doc. No. 588; Page 32 Book No. 8; Series of 1953. (Note):—This form is exempt from the payment of documentary stamp tax. MAGAZINES ANNUALS BROCHURES DIRECT MAIL PAPER BOXES POSTERS BOOKLETS OFFICE FORMS CHECKS TAGS LABELS CALENDARS COLOR REPRODUCTIONS ETC. what you want... how you want it! for quality printing work geared to you— sales and institutional requirements. We invite you to make C & B's* ex- tensive facilities, talents and services, yours TELEPHONE 3-72-51 CONNECTING ALL DEPARTMENTS BOX 665 ■ 2057 AZCARRAGA MANILA ■ PHILIPPINES Carmelo & Bauermann INCORPORATED Devoted to the production of fine printing since 1887 OFFSET LITHOGRAPHERS ■ PRINTERS
INSTRUCTION MANUAL AC-Q41SP VANDAL RESISTANT WATERPROOF STAND-ALONE ACCESS CONTROL UNIT InteliDoor Smart Access Control www.rosslaresecurity.com ## Contents INTRODUCTION ................................................................. 4 Technical Specifications .................................................. 5 Key Features ........................................................................... 7 INSTALLATION ................................................................. 9 Mounting the AC-Q41SP Controller ........................................ 9 Wiring Diagrams ..................................................................... 10 FEATURES AND CONCEPTS .............................................. 16 Normal, Secure, & Master Users ............................................ 16 Modes Of Operation ............................................................... 17 Changing the Modes of Operation .......................................... 18 Auxiliary Output and Input .................................................... 19 Request to Exit (REX) Button .................................................. 20 Cass and Bells Tones .............................................................. 20 BL-D40 External Sounder ....................................................... 21 PROGRAMMING THE AC-Q41SP ........................................ 22 Entering Programming Mode ................................................ 23 Exiting Programming Mode .................................................. 23 1 Changing the Open Code ..................................................... 24 2 Changing the Auxiliary Code ............................................... 24 3 Changing the Programming Code ......................................... 25 4 Changing the Normal / Secure Code ..................................... 26 5 Changing the Normal / Bypass Code ..................................... 26 Door Chime Settings ............................................................. 26 6 Setting Fail Safe / Secure Operation ..................................... 28 Setting Tamper Switch Time ................................................... 28 Setting the Lock Strike Release Time ....................................... 28 Setting the Auxiliary Mode .................................................... 29 Auxiliary Mode Quick Reference Table .................................... 30 Auxiliary Mode Reference ..................................................... 31 Enabling or Disabling the built-in Keypad Heater ..................... 34 7 Enrolling Primary and Secondary Codes ............................... 35 8 Deleting Primary and Secondary Codes ................................. 38 9 Lock Strike Relay and Auxiliary Relay ................................... 40 Code Assignment ................................................................... 41 0 Return to Factory Default Settings ......................................... 42 Replacing a lost Programming Code ........................................ 43 Replacing a lost Normal / Secure Code .................................... 43 APPENDIX ........................................................................... 44 Glossary ................................................................................ 44 WARRANTY ........................................................................... 47 TECHNICAL SUPPORT ....................................................... 49 Introduction The AC-Q41SP is a waterproof and vandal resistant keypad access control unit suitable for internal or external applications. The unit accepts up to 500 users and provides entry via the use of PIN codes. Equipment provided The following is provided as part of every AC-Q41SP package: - AC-Q41SP Access Control Unit. - Installation Kit - Installation and Operating Instructions Additional Equipment Required 1) Electric Lock Strike Mechanism Fail Safe (Power to Lock) or Fail Secure (Power to Open) 2) Power Supply with Backup Battery 12 to 24V DC (From a Regulated Power Supply) or 12 to 24V AC (From a Transformer) 3) Request To Exit (REX) Button Normally Open Type - Switch is closed when pressed. 4) BL-D40 External Sounder (Optional) Provides Siren, Bell, and Chime functions for AC-Q41SP Other Rosslare accessories can be found at Rosslare’s Web Site: www.rosslaresecurity.com Technical Specification Electrical Characteristics Operating Voltage Range: - 12 to 24V DC From a Regulated Power Supply or - 12 to 24V AC From a Transformer Maximum Input Current (when heating is inactive): - At 12V DC - Standby: 40mA Not including attached devices - Max: 100mA Not including attached devices Maximum Input Current (when heating is active): - At 12V DC Max. 630mA Not including attached devices - At 24V AC Max: 350mA Not including attached devices Relay Outputs: - Lock Strike Relay Form C, 2A - Auxiliary Relay Form C, 2A Inputs: - REX N.O., Dry Contact - Auxiliary Input (In / Monitor) N.O., Dry Contact LEDs Two Tri-colored LEDs Environmental Characteristics Operating Temperature: - Keypad Heater Enabled: -4°F to 145°F (-20°C to 63°C) - Keypad Heater Disabled: -32°F to 145°F (0°C to 63°C) Operating Humidity: 0 to 95% (Non-Condensing) Suitable for outdoor use. (IP 65) Mechanical Characteristics Dimensions: 4.72” (120mm) L x 3” (76mm) W x 0.85” (22mm) D Weight: 1.2 lbs (521g) --- Key Features Here are some of the AC-Q41SP’s key features: - Waterproof and Vandal Resistant - Built in keypad for PIN code entry - Auxiliary Input & Auxiliary Output - Ten Auxiliary Modes including: - Door Ajar - Forced Door - Shunt - Door Monitor - Normal / Secure - LED Control - Internal Buzzer - Comes with security screw and security screw tool - Two Tri Color LED’s for Status / Programming Interface - Three User Levels: - Normal User - Secure User - Master User - Three Modes of Operation - Normal Mode - Bypass Mode - Secure Mode - Code Search feature that helps make maintaining user codes easier. - Input for Request to Exit (REX) button. - Comes with mounting template for easier installation. - Built in Case and Back Tamper - Bell, Chime, Siren, and Strobe features available with BL-D40. - Programmable Siren Time (for BL-D40) - Programmable Lock Strike Release Time Installation Mounting the AC-Q41SP Controller 1) Before starting, select the location to mount the AC-Q41SP controller. 2) Drill holes into the back of the metal according to how you want to mount the AC-Q41SP. (See explanation and diagram below). US Gang Box There are two hole indicators on the back of the metal cover specifically for the US Gang Box. (Shown marked as A) 4 Screw Custom There are four indicators on the back. (Shown marked as B) 3) Screw the AC-Q41SP back cover to its mounting location. 4) Wire the controller according to the wiring instructions on the following page. 5) Mount the AC-Q41SP to the back cover. 6) Secure the AC-Q41SP by using the supplied security screw in the controllers accessories kit. An L-Shaped tool is provided for use when tightening the security screw. Wiring the AC-Q41SP The controller is supplied with a 40” (1 m) pigtail, having a 10 conductor cable. To wire the AC-Q41SP, perform the following steps. 1) Prepare the controller cable to the required length. 2) Splice the controller pigtail wires to the corresponding devices and cover each connection. Refer to the wire color guide provided below and to the wiring diagrams provided on the following pages. ### Wire Color Guide \| COLOR \| DESCRIPTION \| \|---------\|-------------\| \| RED \| V/INPUT \| \| BLACK \| GROUND \| \| GREEN \| REX / BL \| \| WHITE \| IN / MONITOR\| \| PURPLE \| LOCK: COM \| \| GRAY \| N.O. \| \| BROWN \| N.C. \| \| BLUE \| AUX: COM \| \| YELLOW \| N.O. \| \| ORANGE \| N.C. \| 3) Trim and cover all conductors that are not used. A few of the typical wiring diagrams are shown on the next three pages; for other wiring diagram examples refer to the support section of the Rosslare Web Site. www.rosslaresecurity.com Wiring Diagrams Wiring the Lock Strike Relay and REX AC-Q41SP - GROUND: BLACK - V IN: RED - REX / SL: GREEN - AUX: BLUE - LOCK: N.O. - RED/GREEN - COM: PURPLE ELECTRIC LOCK STRIKE FAIL SAFE FAIL SECURE COMMON POWER SUPPLY 12V - 24V DC 12V - 24V AC FROM TRANSFORMER Wiring the Auxiliary Input and Output AC-Q41SP - IN / MONITOR: WHITE - AUX: YELLOW - N.O.: RED - N.C.: BLUE - GROUND: BLACK - V IN: RED COMMON POWER SUPPLY 12V - 24V DC 12V - 24V AC FROM TRANSFORMER Auxiliary Relay - COM: RED - N.C.: BLUE - N.O.: RED Wiring the BL-D40 External Sounder AC-Q41SP GROUND BLACK TIN RED REX / SL GREEN COMMON POWER SUPPLY 12V - 24V DC or 12V - 24V AC (TRANSFORMER) BL-D40 Normal, Secure, & Master Users The AC-Q41SP accepts up to 500 users and provides entry via the use of PIN codes. Each user is provided with two code memory slots, Memory Slot 1 (Primary Code) and Memory Slot 2 (Secondary Code). The way in which the two memory slots are programmed determines a user’s access level and also determines the way in which the AC-Q41SP grants access in its three Modes of Operation. There are three user levels: Normal User A Normal User only has a Primary Code, and is only granted access when the AC-Q41SP is in Normal or Bypass Mode. Secure User A Secure User must have a Primary and Secondary Code programmed; the two codes must not be the same. The Secure User can gain access when the AC-Q41SP is in any of its three Modes of Operation. In Normal Mode the Secure User must use their Primary Code to gain entry. In Secure Mode the Secure User must first present their Primary and then their Secondary Code in order to gain entry. Master User A Master User must have both Primary and Secondary Codes programmed with the same Proximity Card or PIN code. The Master User can gain access during any Mode of Operation by entering their PIN code to the controller. Modes of Operation The AC-Q41SP has 3 Modes of Operation: 1) Normal Mode. - Mode LED is green Normal Mode is the default mode. In Normal Mode the door is locked until a Primary Code is presented to the controller. The controller can only be programmed in Normal Mode. 2) Bypass Mode. - Mode LED is orange In Bypass Mode, access to the premises is dependent on whether the controller’s Lock Strike Relay is programmed for Fail Safe Operation or Fail Secure Operation. When the Lock Strike Relay is programmed for Fail Secure Operation, the door is locked until the star button is pressed. When the Lock Strike Relay is programmed for Fail Safe Operation, the door is constantly unlocked. In case of power failure, for security reasons when power is restored the controller will be in Normal Mode. 3) Secure Mode. - Mode LED is red Only Secure and Master Users can access the premises during the Secured Mode. A Secure User must enter their Primary and Secondary Codes to gain entry. After entering their Primary code the Door LED will flash green for 10 seconds, during which the Secondary Code must be entered. A Master User only needs to enter their PIN code once to gain entry. Changing the Modes of Operation Changing from Normal Mode to Secure Mode: The default factory setting for the Normal / Secure Code is 3838. 1) Enter the 4-digit Normal / Secure Code. - Mode LED will flash red. 2) Press the "" key to confirm the Mode change. - Mode LED is red. Changing from Secure Mode to Normal Mode: The default factory setting for the Normal / Secure Code 3838. 1) Enter the 4-digit Normal / Secure Code. - Mode LED will flash green. 2) Press the "" key to confirm the Mode change. - Mode LED will turn green. The Auxiliary Input of the AC-Q41SP can also be used to switch the mode of operation from Secure to Normal Mode and vice versa. Refer to “Setting the Auxiliary Mode” on Page 29. Changing from Normal Mode to Bypass Mode: By default there is no Normal / Bypass code. The Normal / Bypass code must first be programmed to use this function. Refer to page 26 to create / modify the Normal / Bypass Code. 1) Enter the 4 digit Normal / Bypass Code. - Mode LED will flash orange. 2) Press the "" key to confirm the Mode change. - Mode LED will turn orange. Changing from Bypass Mode to Normal Mode: 1) Enter the 4 digit Normal / Bypass Code. - Mode LED will flash green. 2) Press the "" key to confirm the Mode change. - Mode LED will turn green. Auxiliary Input and Output The AC-Q41SP auxiliary input and output can be configured in ten different modes of operation, for optimum usability in different applications. For more information, refer to “Setting the Auxiliary Mode” on Page 29. Request to Exit (REX) Button The REX button must be located inside the premises to be secured and is used to open the door without the use of a proximity card or PIN code. It is usually located in a convenient location, e.g., beside the door or near the security guard’s station. The function of the REX button depends on whether the Lock Strike Relay is programmed for Fail Safe Operation or Fail Secure Operation. The door chime in the BL-D40 does not sound when the REX button is used to open the door. 1) Fail Secure Operation: From the moment the REX button is pressed, the door will be unlocked until the “Lock Strike Release Time” has passed. After this time, the door will be locked even if the REX button has not been released. 2) Fail Safe Operation: From the moment the REX button is pressed, the door will be unlocked until the REX button is released, plus the “Lock Strike Release Time”. In this case the “Lock Strike Relay” only begins its count down once the REX button has been released. This feature is designed to keep the door open when used in conjunction with fire systems. Case and Back Tamper If the case of the controller is opened or the controller is removed from the wall, a tamper event is triggered. A coded tamper signal is then sent to the BL-D40 or other compatible device. If the BL-D40 External Sounder receives a Tamper Event Signal, it will activate a Siren and a Strobe Light. The Siren time can be easily programmed in the AC-Q41SP from 0 to 9 minutes. The tamper event can activate the Auxiliary Output if the controller is in Auxiliary Mode 3. Refer to “Setting the Auxiliary Mode” on page 29. Clearing a tamper event is done by entering a valid User Code. BL-D40 External Sounder The BL-D40 External Sounder is designed to operate indoors and installed within the premises to be secured. The Sounder can be powered by 12 to 24V DC power supply or by 12 to 24V AC from a transformer. The BL-D40 is capable of emitting four different types of alerts both audible and visual; Bell, Door Chime, Siren, and Strobe Light. 1) The Bell always sounds when the controller’s door bell button is pressed. 2) The Door Chime can be programmed to sound whenever a valid code is entered. 3) The Siren can be programmed to sound when the case of the controller is tempered i.e. opened or when the controller is removed from the wall. The controller can also program the length of the Siren time in the BL-D40. The Controller communicates with the BL-D40 using a coded proprietary Rossilare communications protocol. This provides a secure link between the Controller and the BL-D40. If the BL-D40 receives any unrecognized codes on its communication line or communication between the controller and the BL-D40 are severed, the Strobe will flash repeatedly until the communication problem has been resolved. Programming the AC-Q41SP Programming the AC-Q41SP is done solely via the unit's keypad driven Programming Menu System. To reach the Programming Menu System the AC-Q41SP must first be placed into Programming Mode. See "Entering Programming Mode" on Page 23 for more information. During the AC-Q41SP's manufacturing process certain codes and settings are pre-programmed. These settings are the called the "Default Factory Settings". The table below shows the names of all the AC-Q41SP menus. It also shows of all the AC-Q41SP's default factory codes and settings. ### Programming Menu \| Factory Settings \| Menu Description \| Menu Number \| \|------------------\|-------------------------------------------------------\|-------------\| \| 2580 \| Change Open Code \| 1 \| \| 0852 \| Change Auxiliary Code \| 2 \| \| 1234 \| Change Program Code \| 3 \| \| 3838 \| Change Normal / Bypass Code \| 4 \| \| \| Change Normal / Bypass Code \| 5 \| \| 0004 \| Change Door Release Time \| 6 \| \| 2004 \| Define Auxiliary Inputs / Outputs \| 6 \| \| \| Enroll Proximity Cards, PIN Code or both. \| 7 \| \| \| Delete Proximity Cards or PIN Code \| 8 \| \| \| Code Assignment with Strike/Auxiliary \| 9 \| \| \| Return to Default Factory Setting \| 0 \| You will find a complete description and instructions for each of the above menu items on the following pages. ### Entering Programming Mode 1) Press the "#" key two times within 2 seconds. - Mode LED will turn off - Door LED will turn red 2) Enter your 4-digit Programming Code. If the Programming Code is valid the door LED will turn green and the AC-Q41SP will be in Programming Mode. Note: - The AC-Q41SP must be in Normal Mode to enter the Programming Mode. - The factory default Programming Code is 1234 - If a Programming Code is not entered within 5 seconds, The AC-Q41SP will return to Normal Mode. ### Exiting Programming Mode 1) To exit the Programming Mode at any time: - Press the "#" key. - You will hear 1 long beep - The Door LED will be off - The Mode LED will turn green This indicates that the AC-Q41SP has returned to Normal Mode. 2) Wrong entries may reset the controller back to Normal Mode. 3) While in Programming Mode if no key is pressed for 1 minute the AC-Q41SP will exit programming mode and return to Normal Mode. 4) While in enrolling users, deleting users, or code assignment modes, to exit Programming Mode press the "#" key three times. Changing the Open Code The Open Code is mainly used as a method to quickly test the Lock Strike Relay during installation. The Default Factory Setting for the Open Code is 2580. When the first user is added to the controller, the default Open Code will automatically be deleted, ready for a new Open Code to be re-entered. 1) Enter Programming Mode 2) Press “1” to enter Menu 1 - The Mode LED will turn red 3) Enter the new 4-digit code you wish to set as Open Code. 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off - The Mode LED will turn green Note: - Open Code does not function in Secure Mode. - Wrong entries; you will hear a long beep and the controller will return to Normal Mode. - Code 0000 will erase and deactivate the Open Code. Changing the Auxiliary Code The Auxiliary Code is mainly used as a method to quickly test the Auxiliary Relay during installation. The Default Factory Setting for the Auxiliary Code is 0682. For security reasons, when the first user is added to the controller or the open code is changed, the default Auxiliary Code will automatically be deleted, ready for an Auxiliary Code to be assigned. 1) Enter Programming Mode 2) Press “2” to enter Menu 2 - The Mode LED will turn orange 3) Enter the new 4-digit code you wish to set as Auxiliary Code. 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off - The Mode LED will turn green Note: - Auxiliary Code does not work in Secure Mode. - Auxiliary Code only works when the Auxiliary Mode is 0, 1, 8 or 9. - Code 0000 will erase and deactivate the Auxiliary Code. Changing the Programming Code 1) Enter Programming Mode 2) Press “3” to enter Menu 3 - The Mode LED will turn green. 3) Enter the new 4-digit code you wish to set as Programming Code 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off - The Mode LED will turn green Note: - Programming Code can not be erased, i.e. the code 0000 is not valid and will not erase the Programming Code. Changing the Normal / Secure Code 1) Enter Programming Mode 2) Press "4" to enter Menu 4 - The Mode LED will flash red 3) Enter the new 4-digit code you wish to set as Normal / Secure Code 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off - The Mode LED will turn green Note: - When the Auxiliary Mode is 1, 2, 3, or 4 the Auxiliary Input takes priority over the Normal / Secure Code. Changing the Normal / Bypass Code and Door Chime Settings The Normal / Bypass Code is also used to turn the Door Chime off and on. Chime only functions with BL-D40 External Sounder. 1) Enter Programming Mode 2) Press "5" to enter Menu 5 - The Mode LED will flash orange. 3) Below is a list of the four different ways that the Normal / Bypass Code and Door Chime can be programmed. a) Disable Bypass Mode - Disable Door Chime b) Disable Bypass Mode - Enable Door Chime c) Enable Bypass Mode - Disable Door Chime d) Enable Bypass Mode - Enable Door Chime a) Disable Bypass Code - Disable Door Chime Enter the 4-digit code 0000 b) Disable Bypass Code - Enable Door Chime Enter the 4-digit code 0001 c) Enable Bypass Code - Disable Door Chime Enter any 4-digit code ending with 0 d) Enable Bypass Code - Enable Door Chime Enter any 4-digit code not ending with 0 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off - The Mode LED will turn green Note: - The chime is only generated when the Lock Strike Relay is activated due to a valid code entry. Setting Fail Safe/Secure Operation Setting Tamper Siren Time Setting the Lock Strike Release Time The Tamper Siren feature requires the BL-D40 External Sounder. 1) Enter Programming Mode 2) Press "6" to enter Menu 6 - The Mode LED will flash green 3) Construct the 4-digit code using the instructions below: First Digit For Fail Secure Operation the first digit should be "0" For Fail Safe Operation the first digit should be "1" Second Digit Tamper Siren Time, enter any number from 0 to 9 minutes. Third and Fourth Digit Enter the number of seconds from (1 to 99 seconds) that you want the Lock Strike to be released. For example 0 5 1 2 means Fail Secure Operation, with a 5 minute Tamper Siren Time, and a 12 second Lock Strike release time. 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off. - The Mode LED will turn green Setting the Auxiliary Mode The default auxiliary setting is 2004. 1) Enter Programming Mode 2) Press "6" to enter Menu 6 - The Mode LED will flash green 3) Construct the 4-digit code using the instructions below: Auxiliary Mode In addition to the Lock Strike Relay and Lock Strike REX, the AC-Q41SP features an Auxiliary Output Relay and an Auxiliary Input. The Auxiliary Mode defines the function of the Auxiliary Input and Output. The Auxiliary Mode also determines if the Auxiliary Output Relay is set for Fail Safe or Fail Secure Operation. Auxiliary Settings Each of the Auxiliary Modes has a two digit setting that affects how the Auxiliary Mode functions. 4) System returns to Normal Mode - You will hear three beeps - The Door LED will turn off. - The Mode LED will turn green The Auxiliary Mode Quick Reference Table can be found on the next page. For a more detailed explanation on each auxiliary mode refer to the 'Auxiliary Mode Reference Guide' on page 31. ## Auxiliary Mode Quick Reference Table \| Auxiliary Mode \| Auxiliary Input Function \| Auxiliary Output Activated On \| Aux. Relay \| Aux. Relay Release Time \| \|----------------\|--------------------------\|-------------------------------\|------------\|-------------------------\| \| 0 \| REX-2 \| Valid Code or REX-2 \| N.O. \| 01 to 99 Aux. Relay Toggles \| \| 1 \| Normal / Secure \| Valid Code \| N.O. \| 01 to 99 Aux. Relay Release Time \| \| 2 \| Normal / Secure \| Star Button \| N.O. \| 01 to 99 Aux. Relay Release Time \| \| 3 \| Normal / Secure \| Tamper Event \| N.C. \| 01 to 99 Aux. Relay Release Time \| \| 4 \| Normal / Secure \| Direct Shunt \| N.O. \| 00 to 99 Shunt Time \| \| 5 \| Door Monitor \| Shunt \| N.C. \| 00 to 99 Maximum Shunt Time \| \| 6 \| Door Monitor \| Forced Door \| N.C. \| 00 to 99 Forced Delay \| \| 7 \| Door Monitor \| Door Ajar \| N.C. \| 00 to 99 Ajar Delay \| \| 8 \| LED Ctrl - Red \| Valid Code \| N.O. \| 01 to 99 Aux. Relay Release Time \| \| 9 \| LED Ctrl - Green \| Valid Code \| N.O. \| 01 to 99 Aux. Relay Release Time \| --- ## Auxiliary Mode Reference Guide The following are brief descriptions of each of the AC-Q41SP’s auxiliary modes. To use these features refer to “Setting the Auxiliary Mode” on page 29. ### AUXILIARY MODE 0 In auxiliary mode 0 the AC-Q41SP can function as a two door controller. The auxiliary relay should be attached to the lock on the second door. The auxiliary setting defines the door open time for the second door. The auxiliary input should be attached to the REX button for the second door. ### AUXILIARY MODE 1 In auxiliary mode 1 the AC-Q41SP can function as a two door controller. The auxiliary relay should be attached to the lock on the second door. The auxiliary setting defines the door open time for the second door. The auxiliary input can switch the mode of operation of the controller between Normal and Secure Mode. By connecting a switch timer to the auxiliary input you can for example automatically switch the AC-Q41SP from Normal Mode during office hours to Secure mode after office hours. ### AUXILIARY MODE 2 In auxiliary mode 2 the auxiliary relay can function as a general purpose timed switch that can be activated when the star button on the AC-Q41SP is pressed. The auxiliary setting defines how long the auxiliary relay should remain active. The auxiliary input can switch the mode of operation of the controller between Normal and Secure Mode. By connecting a switch timer to the auxiliary input you can for example automatically switch the AC-Q41SP from Normal Mode during office hours to Secure mode after office hours. ### AUXILIARY MODE 3 In auxiliary mode 3 the auxiliary output is activated if the AC-Q41SP is tampered, i.e. the case tamper or back tamper is triggered. The auxiliary input can switch the mode of operation of the controller between Normal and Secure Mode. By connecting a switch timer to the auxiliary input you can for example automatically switch the AC-Q41SP from Normal Mode during office hours to secure mode after office hours. AUXILIARY MODE 4 In auxiliary mode 4 the AC-Q41SP is capable of shunting an alarm system’s door sensor. The auxiliary output should be wired in parallel to the door sensor. When in use the auxiliary output is normally open and the door sensor functions normally. When a valid code is entered the auxiliary relay will shut the door sensor for the duration of the shunt time as defined by the auxiliary setting. If the door is left open longer than the shunt time the alarm will be triggered. AUXILIARY MODE 5 In auxiliary mode 5 the AC-Q41SP is capable of shunting an alarm system. In this mode the auxiliary input should be wired to the magnetic contact switch on the door. The auxiliary relay should be wired to the alarm system. Without a valid code entered the auxiliary relay will match the condition of the magnetic contact switch, if the door opens the auxiliary relay will open, if the door closes the auxiliary relay will close. When a valid code is entered a count down timer with a shunt time as defined by the auxiliary setting begins, if the door is not closed before the maximum shunt time, the alarm will be triggered. AUXILIARY MODE 6 In auxiliary mode 6 the AC-Q41SP can trigger the auxiliary relay if it detects that the door has been forced. In this mode the auxiliary input should be wired to the magnetic contact switch on the door. The auxiliary relay should be wired to the alarm system. If the door is forced open the controller will wait for the forced door delay time and then activate the auxiliary relay. The auxiliary setting defines the forced door delay. AUXILIARY MODE 7 In auxiliary mode 7 the AC-Q41SP can trigger the auxiliary relay if it detects that the door has been ajar for long. In this mode the auxiliary input should be wired to the magnetic contact switch on the door. The auxiliary relay should be wired to the alarm system. If the door is opened the controller will wait for the door ajar delay time, if the door does not close before the ajar delay time the controller will activate the auxiliary relay. The auxiliary setting defines the door ajar time. If the BL-D40 is connected and an ajar event occurs the BL-D40 will chime every few seconds for 1 minute or till the door is closed. AUXILIARY MODE 8 In auxiliary mode 8 the AC-Q41SP can function as a two door controller and also provide LED Control functionality. The auxiliary relay should be attached to the lock on the second door. The auxiliary setting defines the door open time for the second door. The auxiliary input is used to control the LED, if the auxiliary input is open the Door LED will flash red, if the auxiliary input is closed the Door LED will flash green. Note: This mode takes control of the Door LED. The Door LED will no longer activate when a valid code is entered or when in Secure Mode waiting for a Secondary Code to be entered. AUXILIARY MODE 9 In auxiliary mode 9 the AC-Q41SP can function as a two door controller and also provide LED Control functionality. The auxiliary relay should be attached to the lock on the second door. The auxiliary setting defines the door open time for the second door. The auxiliary input is used to control the LED, if the auxiliary input is open the Door LED will flash green, if the auxiliary input is closed the Door LED will flash red. Note: This mode takes control of the Door LED. The Door LED will no longer activate when a valid code is entered or when in Secure Mode waiting for a Secondary Code to be entered. Enabling or Disabling the built in keypad heater The AC-Q41SP contains a built-in keypad heater. When the keypad heater is enabled the keypad heater will activate when the ambient temperature drops to approx. 37°F to 41°F (3°C to 5°C) and will remain active until the keypad temperate rises to approx 43°F to 48°F (6°C to 9°C). When the heater is enabled the AC-Q42H's lowest operating temperature is -4°F (-20°C). When the heater is disabled the AC-Q42H’s lowest operating temperature is 32°F (0°C). The factory default setting for the keypad heater is disabled. 1) Enter Programming Mode 2) Press "6" to enter Menu 6 • The Mode LED will flash green 3) Construct the 4-digit code using the instructions below: Fourth Digit To DISABLE the heater the fourth digit should be “0” To ENABLE the heater the fourth digit should be “1” Enrolling Primary & Secondary Codes Primary Codes - Primary Codes can only be enrolled to an empty User Slot, i.e a slot where there is no existing Primary Code. - Primary Codes must be unique, i.e. one users Primary Code may not be the same as another users Primary Code. - Primary Codes cannot be the same as any system codes, such as the Normal / Secure Code or Open Code. - Users who hold a Primary Code can gain entry during Normal Mode and Bypass Mode. Secondary Codes - Secondary Codes can only be enrolled to User Slot that already has a Primary Code - Secondary Codes do not have to be unique, i.e. multiple users can all hold the same Secondary Code. - Secondary Codes cannot be the same as any system codes, such as the Normal / Secure Code or Open Code. - Users who hold Secondary Codes can gain entry in any Mode of Operation. Enrolling Primary and Secondary Codes There are two methods to enroll Primary and Secondary codes, the Standard Method and the Code Search Method. A. The Standard Method is mainly used when the User Slot number for the user you wish to program is known. You can program both Primary and Secondary Codes using the Standard method. (See Enrolling Users with the Standard Method on Page 36) B. The Code Search Method is mainly used when enrolling a users Secondary Code and the User Slot Code is unknown. The Code Search method only works if a users Primary Code is already enrolled but the Secondary Code is not. (See Enrolling Users with the Code Search Method on Page 37) Enrolling Primary and Secondary Codes using the Standard Method 1) Enter Programming Mode 2) Press "7" to enter Menu 7 - The Door LED will turn orange 3) Enter the 3-digit User Slot number between 001 to 500 that you wish to enroll a Primary or Secondary code to. For example, the User Slot 003 represents User #3. 4) a. If the selected slot has no Primary Code, the Mode LED will flash green, indicating that the controller is ready to accept a Primary Code. b. If the selected slot already has a Primary Code but no Secondary Code, the Mode LED will flash red, indicating that the controller is ready to accept a Secondary Code. c. If the selected slot already has a Primary and Secondary Code, you will hear a long beep and the controller will return to Normal Mode. 5) Enter the 4-digit PIN that you want to assign as the Primary or Secondary Code for this slot number. If the PIN that is entered is valid the Mode LED will stop flashing and then the controller is ready for you to enter the next 3 Digit slot number (refer to step 3) that you want to assign a code to, or press the "#" key to move to the next slot number (refer to step 4). If you do not wish to continue enrolling codes, press the "" key twice and the controller will return to Normal Mode. Enrolling Secondary Codes using the Code Search Method The Code Search feature enables you to quickly enroll a Secondary Code to a user who's slot number is unknown but who's primary code is known. 1) Enter Programming Mode 2) Press "7" to enter Menu 7 - The Door LED will turn orange 3) Enter the 3-digit User Slot number 000 - The Door LED will flash orange The controller is now waiting for the Primary Code of the User you want to add a Secondary Code to. 4) Enter the 4 Digit PIN Code of the Primary Code belonging to the user you want to add a Secondary Code to. - The Mode LED will flash red If the Primary Code entered is not valid, you will hear a long beep and the AC-Q41SP will continue to wait for a valid Primary Code. 5) Enter the 4-digit PIN Code to be used as the Secondary Code. If the Secondary Code is valid the controller will beep three times and return to Normal Mode. If the Secondary Code is invalid the controller will make a long beep and then the AC-Q41SP will continue to wait for a valid Secondary code to be entered. Deleting Primary & Secondary Codes There are two methods to delete Primary and Secondary codes, the Standard Method and the Code Search Method. When deleting a User Slot, both the Primary Code and the Secondary code are erased. Deleting Primary and Secondary Codes using the Standard Method 1) Enter Programming Mode 2) Press “8” to enter Menu 8 - The Mode LED will turn red 3) Enter the 3-digit User Slot codes you wish to delete. - The Mode LED will flash red Indicating the controller is waiting for the Programming Code to confirm the deletion. If the User Slot is empty you will hear a long beep and the AC-Q41SP will return to Normal Mode 4) Enter your Programming Code to confirm the deletion. If the Programming Code is valid, you will hear three beeps and the AC-Q41SP will return to Normal Mode. If the Programming Code is invalid, you will hear a long beep and the AC-Q41SP will return to Normal Mode. Note: - It is recommended that a record be kept of added and deleted users so that it will be easier to keep track of which user slots are empty and which user slots are not. Deleting Primary and Secondary Codes using the Code Search Method 1) Enter Programming Mode 2) Press “8” to enter Menu 8 - The Mode LED will turn red 3) Enter the 3-digit User Slot 000 - The Door LED will flash orange The controller is now waiting for the Primary Code of the User you want to delete. 4) Enter the 4-digit PIN Code of the Primary Code belonging to the user you want to delete. - The Mode LED will flash red 5) Enter your Programming Code to confirm the deletion. If the Programming Code is valid, you will hear three beeps and the AC-Q41SP will return to Normal Mode. If the Programming Code is invalid, you will hear a long beep and the AC-Q41SP will return to Normal Mode. Note: - It is recommended that a record be kept of added and deleted users so that it will be easier to keep track of which user slots are empty and which user slots are not. Lock Strike Relay and Auxiliary Relay Code Assignment When a Primary Code is enrolled for any user, that user is assigned rights to activate the Lock Strike Relay when they present a valid code to the controller. The Code Assignment Menu allows you to assign whether the Lock Strike Relay and/or the Auxiliary Relay is activated when a user enters a valid code. There are two methods to Assign Codes, Standard Method and the Code Search Method. Lock Strike Relay and Auxiliary Relay Code Assignment using the Standard Method* 1) Enter Programming Mode 2) Press "9" to enter Menu 9 - The Mode LED will turn green 3) Enter the 3-digit User Slot that you want to assign a code to. - The Door LED will flash green 4) Enter the assignment digit for the current User Slot: - "1" assigns the Lock Strike Relay only - "2" assigns the Auxiliary Relay only - "3" assigns the Lock Strike and Auxiliary Relay - If the assignment code is valid the Mode LED will stop flashing. The controller is now waiting for another slot number. Press the "#" key to go to the next slot or enter a new slot number, or if you do not wish to continue press the "#" key twice and the controller will return to Normal Mode. Lock Strike and Auxiliary Relay Code Assignment using the Code Search Method 1) Enter Programming Mode 2) Press "9" to enter Menu 9 - The Mode LED will turn red 3) Enter the 3-digit User Slot 000 - The Door LED will flash orange The controller is now waiting for the Primary Code of the user you want to Code Assign 4) Enter the 4-digit PIN Code of the Primary Code belonging to the user you want to assign a code to. - The Mode LED will flash green 5) Enter the assignment digit for the current User Slot: - "1" assigns the Lock Strike Relay only - "2" assigns the Auxiliary Relay only - "3" assigns the Lock Strike and Auxiliary Relay - If the assignment digit is valid, you will hear three beeps and then the controller will return to Normal Mode. - If the assignment digit is invalid, you will hear a long beep and the controller will wait for another assignment digit to be entered. Return To Factory Default Settings Warning: You must be very careful before using this command! Doing so will erase the entire memory which includes all User and Special Codes, and return all codes to their factory default settings. 1) Enter Programming Mode 2) Press “0” to enter Menu 0 - The Mode LED will flash red - The Door LED will flash red 3) Enter your 4-digit Programming Code. - If the Programming Code is valid, all memory will be erased, you will hear three beeps and the controller will return to Normal Mode - If the Programming Code is invalid you will hear a long beep and the controller will return to Normal Mode without erasing the memory of the controller. --- Replacing a lost Programming Code Note: The AC-Q41SP must be in Normal Mode otherwise this will not work. Make sure that the Mode LED is green before proceeding. 1) Remove power from the AC-Q41SP 2) Press the REX button 3) Apply power to the unit with REX button pressed 4) Release the REX button 5) You now have 15 seconds to program a new Programming Code into the unit using the initial default code 1234, before the controller reverts to the existing code. --- Replacing a lost Normal / Secure Code Note: The AC-Q41SP must be in Secure Mode otherwise this will not work. Make sure that the Mode LED is red before proceeding. 1) Remove power from the AC-Q41SP 2) Press the REX Button 3) Apply power to the unit with REX button pressed. 4) Release the REX Button 5) You now have 15 seconds to use the default Normal / Secure code 3838 in order to return to normal mode, where you may enter programming mode to program a new Normal / Secure Code. Glossary A Access Control: Primarily refers to a device or set of devices controlling the entry of people travelling through a door or set of doors. Ajar Delay: The time allowed for a door to be left open before sounding an alarm and / or activating the Auxiliary Relay. Amplitude Shift Keying (ASK): The type of data communication between the Proximity Card and the Proximity Reader. ASK: An abbreviation of Amplitude Shift Keying Auxiliary Input: The term used for the programmable input electrical signal from an external device such as a Door Monitor switch or Primary REX button. Auxiliary Code: The four digit code used to activate the Auxiliary Output for testing purposes during installation. Auxiliary Output: The term used for the Relay Output in the AC-Q41SP that may be programmed to activate upon different system events such as Tamper, Forced Door Event, Door Ajar etc. B Back Tamper: The electronic tamper signal advising the controller that the contact has been removed from the wall. Bypass Code: The four digit code used to change the Mode of Operation of the AC-Q41SP from Normal to Bypass Mode or vice versa. Bypass Mode: A Mode of Operation where door access is not restricted to valid users. In this mode the door may be released by anyone pressing the star button. C Case Tamper: The electronic tamper signal advising the controller that the case has been opened. Code Assignment: The process of assigning which Output(s) (Lock Strike Relay and / or Auxiliary Relay) are to be activated when a valid code is entered. D Direct Shunt: The arrangement in which an external input (such as a door monitor) is connected directly to the Auxiliary Relay allowing the Auxiliary Output to be activated after the direct shunt delay period. This leaves the Auxiliary Input available for Normal / Secure mode toggle. Default Factory Setting: The settings of the controller is preprogrammed with when the controller is manufactured. Direct/Count Delay: The delay time (user programmable) used in Direct Shunt (See Direct Shunt). Door Bell: The alert sound activated when the doorbell button on the AC-Q41SP is pressed. (Requires the BL-D40 External Sounder) Door Chime: The alert sound activated when the lock strike unlocks the door after a valid code has been presented. (Requires the BL-D40 External Sounder) F Fail Safe: The system setting in which a total power loss leaves the connected door unlocked. Fail Secure: The system setting in which a total power loss leaves the connected door locked. Forced Door: A door which has been physically opened without the access control device having released the lock. Forced Door Time: The amount of time (user programmed) the controller waits in the event of a Forced Door before the Auxiliary Output is activated. L Lock Strike: Term used for the electric or electromagnetic door lock used for locking or unlocking the door. Lock Strike Release Time: The amount of time (user programmed) that the Lock Strike remains unlocked when a valid code is entered. M Master User: A user which has a Primary and Secondary Code which are the same and can gain access in any Mode of Operation. Mode of Operation: The state of operation of the controller. There are three Modes: Normal Mode, Bypass Mode, and Secure Mode. N Normal Mode: The system setting (Mode of Operation) in which all valid users have access after entering a valid PIN Code (Primary Code). Normal / Bypass Code: The four digit code used to change the controllers Modes of Operation from Normal to Bypass Mode or vice versa. Normal / Secure Code: The four digit code used to change the controllers Mode of Operation from Normal to Secure Mode or vice versa. Normal User: A user who only has a Primary Code and can only gain access in Normal Mode. Normally Closed: A relay output from the controller that is activated (closed circuit) under normal conditions. Normally Open: A relay output from the controller that is de-activated (open circuit) under normal conditions. O Open Code: The four digit code used to activate the Lock Strike Relay for testing purposes during installation. P Primary Code: The unique code issued to enable access in Normal Mode. Users with only primary codes are limited to this mode. Programming Code: The four digit code required when entering programming mode, deleting users, and resetting the AC-Q41SP to its factory default settings. Programming Mode: The mode used for programming the AC-Q41SP’s system settings. Proximity Cards: A contactless (RFID) identification card each with unique code. R Relay: An electronically controlled switch used for providing an Open Circuit or Closed Circuit output to external devices. REX: An abbreviation of “Request To Exit.” Request To Exit (REX): Refers to a button which can release the door from inside. Commonly located at the reception desk, or near a door as an emergency door release. S Secondary Code: An additional code issued to enable access in Secured Mode. Users with non-identical Primary and Secondary Codes are Secure Users. Users with identical Primary and Secondary Codes are Master Users. Secure Mode: The system setting (Mode) in which only valid Secure and Master Users have access upon presenting a valid code. Secure User: A user which has a Primary and Secondary Code that are non-identical, and can gain access in any Mode of Operation. Shunt: The condition when an external input (such as a door monitor) is connected to the Auxiliary Input, allowing the auxiliary output to be activated after the Shunt Delay elapses. Shunt Delay: Is the delay time (user programmed) used in Shunt (See Shunt) Strike: See Lock Strike T Tamper Siren: The alert sound activated when a Back Tamper or Case Tamper event occurs. (Requires the BL-D40 External Sounder) Tamper Siren Time: The time (user programmed) that the Tamper Siren will sound when activated. --- Limited Lifetime Warranty ROSSLARE ENTERPRISES LIMITED’S (Rosslare) LIMITED LIFETIME WARRANTY is applicable worldwide. This warranty supersedes any other warranty. Rosslare’s LIMITED LIFETIME WARRANTY is subject to the following conditions: WARRANTY Warranty of Rosslare’s products extends to the original purchaser of the Rosslare product and is not transferable. WARRANTY DURATION Rosslare warrants this product against defects in material and/or workmanship for the life of the product from the date of original purchase to the original purchaser. WARRANTY COVERAGE Rosslare will repair or replace, at its option, any product which under normal conditions of use and service proves to be defective in material or workmanship. No charge will be made for labor or parts with respect to defects covered by this warranty, provided that the work is done by Rosslare or a Rosslare authorized service center. This warranty does not cover expenses incurred in the transportation, removal or reinstallation of the product, whether or not proven defective. Replacement or repairs furnished under this warranty are subject to the same terms and conditions of the original warranty. EXCLUSIONS AND LIMITATIONS Specifically excluded from this warranty are failures caused by abuse, neglect, misuse, improper operation, normal wear, accident, improper maintenance or modification. This warranty does not cover repair or replacement where normal use has exhausted the life of a part or instrument. Service life of the product is dependent upon the care it receives and the conditions under which it has to operate. In no event shall Rosslare be liable for incidental or consequential damages. LIMITED LIFETIME WARRANTY TERMS: The terms of this warranty may not be varied by any person, whether or not purportedly represented on behalf of Rossiare. This warranty represents the full extent of Rossiare’s responsibility. Repair, replacement, or refund of the original purchase price, of the product is the exclusive remedy. This limited lifetime warranty is provided in lieu of all other warranties. All other warranties expressed or implied, including without limitation, all express warranties of merchantability and fitness for a particular purpose, are specifically excluded. In no event shall Rossiare be liable for damages in excess of the purchase price of the product, or for any other incidental, consequential or special damages, including but not limited to loss of use, loss of time, commercial loss, loss of income, and loss of profits, arising out of the installation, use, or inability to use such product, to the fullest extent that any such loss or damage may be disclaimed by law. This warranty shall become null and void in the event of a violation of the provisions of this limited warranty. Technical Support International: Rossiare Enterprises Ltd. 805-912 Wing Fat Industrial Bldg., 12 Wang Tai Road, Kowloon Bay, Hong Kong. www.rossiare.com.hk/support Tel: (852) 2795 5630 Fax: (852) 2795 1508 E-mail: email@example.com United States and Canada: Rossiare NAPDC Suite 238, 200 East Howard Street, Des Plaines, IL 60018 USA www.rossiarsecurity.com Tel: (866) 262 8633 (Toll-Free) Tel: (847) 827 6330 (Direct) Fax: (847) 827 6433 E-mail: firstname.lastname@example.org
2017 October Cisco Official New Released 400-051 Dumps in Lead2pass.com! 100% Free Download! 100% Pass Guaranteed! 2017 timesaving comprehensive guides for Cisco 400-051 exam: Using latest released Lead2pass 400-051 exam questions, quickly pass 400-051 exam 100%! Following questions and answers are all new published by Cisco Official Exam Center! Following questions and answers are all new published by Cisco Official Exam Center: https://www.lead2pass.com/400-051.html QUESTION 181 You have implemented 5-digit forced authorization codes to all international route patterns on Cisco Unified Communications Manager. Your users report that after entering the FAC codes, they must wait for more than 10 seconds before the call is routed. Which procedure eliminates the wait time? A. Check and eliminate any existing route patterns that overlap with the international route pattern. B. Go to the Cisco Unified Communications Manager Service Parameters and reduce the T-304 number to 5000 milliseconds. C. Request your long distance telephone service provider to reduce the call setup time to 5 seconds. D. Configure a # (hash) sign to the end of the forced authorization codes to signal the end of dialing. E. Educate the users to press # (hash) after entering the forced authorization codes. Answer: E QUESTION 182 Refer to the exhibit. A user is going through a series of dialing steps on a SIP Type A IP phone to call a SCCP IP phone. Both phones are registered to the same Cisco Unified Communications Manager cluster. Assume that the calling SIP phone is not associated with any SIP dial rules. Which statement about how digits are forwarded to the Cisco Unified Communications Manager for further call processing is true? A. As each digit is pressed on the SIP IP phone, it is sent to the Cisco Unified Communications Manager in a SIP NOTIFY message as a KPML event. B. The SIP IP phone waits for the inter-digit timer expiry and then sends each digit to the Cisco Unified Communications Manager as a separate KPML event in a SIP NOTIFY message. C. The SIP IP phone waits for the inter-digit timer expiry or for the Dial soft key to be selected before it sends all digits to the Cisco Unified Communications Manager in a SIP INVITE message. D. The SIP IP phone waits for the inter-digit timer expiry, or for the Dial soft key to be selected before it sends the first digit in a SIP INVITE and the subsequent digits in SIP NOTIFY messages. E. The SIP IP phone sends all digits to the Cisco Unified Communications Manager in a SIP INVITE message as soon as the fourth digit is pressed. Answer: C QUESTION 183 Refer to the exhibit. A user is going through a series of dialing steps on a SIP Type B IP phone to call a SCCP IP phone. Both phones are registered to the same Cisco Unified Communications Manager cluster. Assume that the calling SIP phone is associated with a SIP dial rule with a pattern value of “2001”. Which statement about how digits are forwarded to the Cisco Unified Communications Manager for further call processing is true? A. As each digit is pressed on the SIP IP phone, it is sent to the Cisco Unified Communications Manager in a SIP NOTIFY message as a KPML event. B. The SIP IP phone waits for the inter-digit timer expiry and then sends each digit to the Cisco Unified Communications Manager as a separate KPML event in a SIP NOTIFY message. C. As soon as the user selects the Dial soft key, the SIP IP phone forwards all digits to the Cisco Unified Communications Manager in a SIP INVITE message. D. As soon as the Dial soft key is selected, the SIP IP phone forwards the first digit in a SIP INVITE and the subsequent digits in SIP INFORMATION messages. E. The SIP IP phone waits for the inter-digit timer expiry, and then sends all digits to the Cisco Unified Communications Manager in a SIP INVITE message. Answer: C QUESTION 184 Refer to the exhibit. All displayed devices are registered to the same Cisco Unified Communications Manager server and the phones are engaged in an active call. Assume that the provided configurations exist at the phone line level and multicast MOH is disabled cluster wide. Which description of what happens when the user of IP phone B presses the Transfer soft key is true? A. IP phone A user hears audio source 3 from MOH server A. B. IP phone A user hears audio source 4 from MOH server B. C. IP phone A user hears audio source 3 from MOH server B. D. IP phone A user hears tone on-hold beep tones. E. IP phone A user hears no on-hold music or beep tones. Answer: E QUESTION 185 Which option is a characteristic of the Enhanced Location Call Admission Control mechanism on Cisco Unified Communications Manager? A. It accounts for network protocol rerouting. B. It accounts for network downtime and failures. C. It supports dynamic bandwidth adjustments based on WAN topology changes. D. It supports asymmetric media flows such that different bit rates in each direction are deducted accordingly. E. Unidirectional media flows are deducted as if they were bidirectional. Answer: E QUESTION 186 Which two host portion format conditions are true for directory URI on Cisco Unified Communications Manager? (Choose two.) A. It is case sensitive. B. It cannot start with a hyphen. C. It must have at least one character. D. It supports IPv4 or IPv6 addresses, or fully qualified domain names. E. It cannot end with a hyphen. F. It supports the & character. Answer: BE QUESTION 187 Which two user portion format conditions are true for directory URI on Cisco Unified Communications Manager 9.1 or later? (Choose two.) A. It supports the $ character. B. It support space between characters. C. It has a maximum length of 50 characters. D. It has a maximum length of 25-4 characters. E. It is always case-sensitive. F. It cannot be a directory number. Answer: AB QUESTION 188 Which configuration parameter defines whether or not the user portion of a directory URI is case sensitive on Cisco Unified Communications Manager 9.1 or later? A. URI Dialing Display Preference in Cisco CallManager Service Parameter B. URI Lookup Policy in Cisco CallManager Service Parameter C. URI Dialing Display Preference in Enterprise Parameters D. URI Lookup Policy in Enterprise Parameters E. The user portion of a directory URI is always case sensitive and cannot be changed. Answer: D QUESTION 189 When IP phone A was provisioned in a Cisco Unified Communications Manager, 2001 was configured as the directory number for its first line. Also, firstname.lastname@example.org was defined as the only directory URI on the Directory Number configuration page for this line. A few days later, an end user was created in the same Cisco Unified Communications Manager and was associated with the same phone with the primary extension set to 2001. Also, email@example.com was defined as a directory URI for that end user. Which option about the primary directory URI for IP phone A is true? A. firstname.lastname@example.org B. email@example.com C. It depends on which radio button was selected next to the Directory URI entries on the Directory Configuration page. D. Both are primary directory URIs in a manner like a shared line for DNs. E. Neither are primary directory URIs for IP phone A. Answer: B QUESTION 190 The number of calls waiting in a Cisco Unified Communications Manager native call queue has reached its maximum limit. Which statement about what happens to additional incoming calls is true? A. Calls are handled according to the Forward Hunt Busy settings on the Hunt Pilot configuration page. B. Calls are handled according to the Forward Hunt No Answer settings on the Hunt Pilot configuration page. C. Calls are handled according to the Forward Hunt Busy settings on the Line Group members. D. Calls are handled according to the Hunt Options settings on the Line Group Configuration page. E. Calls are handled according to the When Queue Is Full settings on the Hunt Pilot Configuration page. Answer: E QUESTION 191 A queued call has reached the maximum wait time configured for a Cisco Unified Communications Manager native call queue. Which statement about what happens to this queued call is true? A. Calls are handled according to the Forward Hunt No Answer settings on the Hunt Pilot configuration page. B. Calls are handled according to the When Maximum Wait Time Is Met settings on the Hunt Pilot Configuration page. C. Calls are handled according to the When Maximum Wait Time Is Met settings in Cisco Unified Communications Manager Service Parameters. D. Calls are handled according to the Not Available Hunt Option settings on the Line Group Configuration page. E. Calls are handled according to the When Queue Is Full settings on the Hunt Pilot Configuration page. Answer: B QUESTION 192 Which statement about what happens to incoming calls to a Cisco Unified Communications Manager native call queue when no hunt members are logged in or registered is true? A. Calls are handled according to the Forward Hunt No Answer settings on the Hunt Pilot configuration page. B. Calls are handled according to the Not Available Hunt Option settings on the Line Group Configuration page. C. Calls are handled according to the Forward Hunt Busy settings on the Hunt Pilot configuration page. D. Calls are forward to the Forward Busy Calls To destination if configured; otherwise the calls are disconnected. E. Calls are handled according to the correspondent parameters under the Queuing section on the Hunt Pilot Configuration page. Answer: E QUESTION 193 Which statement about what happens to a hunt member who does not answer queuing-enabled hunt-list call in Cisco Unified Communications Manager 9.1 is true? A. The hunt member is logged off automatically and must press HLOG to log back in. B. The hunt member remains logged in if Automatically Logout Hunt Member on No Answer is not selected in Cisco Unified Communications Manager Service Parameters. C. The hunt member is logged off automatically and must manually reset the phone to log back in. D. The hunt member is logged off if Automatically Logout Hunt Member on No Answer is selected on the Line Group configuration page. E. The hunt member remains logged in if Automatically Logout Hunt Member on No Answer is not selected in Hunt Pilot configuration page. Answer: D QUESTION 194 Which SIP request is used by a Cisco 9971 IP Phone to signal DND status changes to Cisco Unified Communications Manager? A. REGISTER B. NOTIFY C. INFO D. PUBLISH E. UPDATE Answer: D QUESTION 195 Which SIP request is used by Cisco Unified Communications Manager to signal DND status changes to a Cisco 9971 IP Phone? A. OPTIONS B. NOTIFY C. INFO D. REFER E. UPDATE Answer: D QUESTION 196 Which two call processing features have a lower priority than the Do Not Disturb settings on a Cisco IP phone? (Choose two.) A. park reversion for a locally parked call B. hold reversion C. intercom D. pickup notification E. terminating side of a call back F. originating side of a call back Answer: DE QUESTION 197 Which call processing feature overrides the Do Not Disturb settings on a Cisco IP phone? A. park reversion for remotely parked calls by a shared line B. hold reversion C. remotely placed pickup request by a shared line D. pickup notification E. terminating side of a call back Answer: B QUESTION 198 You are assisting a customer to troubleshoot a SIP early-offer problem with a SIP service provider. You have enabled Cisco CallManager trace and set the debug trace level to Detailed for SIP Call Processing trace on their standalone Cisco Unified Communications Manager 9.1 system. Using the RTMT tool, your customer has remote browsed to the Cisco UCM and asked you which trace file to download. What is the trace file name syntax in which detailed SIP messages are logged? A. SDL B. SDI C. CCM D. Call logs E. Traces Answer: A QUESTION 199 Which tag in the SIP header is used by Cisco Unified Communications Manager to deliver a blended identity of alpha URI and number? A. x-cisco-callinfo B. x-cisco-service-control C. x-cisco-serviceuri D. x-cisco-number E. x-cisco-uri Answer: D QUESTION 200 Which SIP header is used by Cisco Unified Communication Manager to support the Redirected Number ID Service? A. replaces B. RPID C. diversion D. join E. P-charging-vector Answer: C Lead2pass is confident that our NEW UPDATED 400-051 exam questions and answers are changed with Cisco Official Exam Center. If you cannot pass 400-051 exam, never mind, we will return your full money back! Visit Lead2pass exam dumps collection website now and download 400-051 exam dumps instantly today! 400-051 new questions on Google Drive: https://drive.google.com/open?id=0B3Syig5i8gplDeVpYVlZNcjVzaW8 2017 Cisco 400-051 exam dumps (All 542 Q&As) from Lead2pass: http://www.lead2pass.com/400-051.html [100% Exam Pass Guaranteed]
WEB DESIGN AND I.T SOLUTIONS WEB DESIGN AND RESOURCES PROPOSAL OPEN PROPOSAL © Copyright 2009. MEDIA EMPIRE LIMITED. ALL RIGHTS RESERVED. By accepting a copy of this document and signing it, the recipient is bound by the terms laid out as a legal agreement governed by the company codes of Ghana. The recipient agrees not to reproduce it in whole or in part, not to use it for any other purpose than keeping as a legal reference to terms and conditions of the contract between the recipient and MEDIA EMPIRE LIMITED, using as reference in business dealings with MEDIA EMPIRE LIMITED and its divisions, and not to disclose any of its contents to third parties without written permission of MEDIA EMPIRE LIMITED. MEDIA EMPIRE LIMITED accepts full responsibility of the accuracy and completeness of the information, and everything contained in the document is, or shall be, relied on as a promise or representation of all dealings regarding the purpose of this contract. The MEDIA EMPIRE LIMITED logo, MEDIA EMPIRE LIMITED name, logos and all other brand logos are registered trademarks of MEDIA EMPIRE LIMITED. All Rights Reserved. ABOUT US Media Empire Ghana Limited is a multi-media limited company, registered under the Ghana Companies registration code of the Registrar General’s Department of the Republic of Ghana on December 17th 2007. WHO WE ARE We are a huge Empire of Business Solutions made up of efficient service delivery, creative innovations, advance technology utilization and above all optimum customer relation and satisfaction in website designing and I.T solutions. As a Multi-Media ICT and Web Design Company, our business is to make information accessibility and dissemination easy and affordable through the application of the most advanced technology by our team of I.T experts, Web Designers, experienced writers and content developers. With our technical know-how, professional experience and advantage in Information Communication Technologies Intelligence, (ICTI) we offer opportunities that put our clients ahead of their competitors through the design of comprehensive standardized websites that sells their products and services to the outside world. The acquisition of your own domain name will prove enough incentives to bring returns on your investment. OUR SERVICES Our services include a basket full of products and services that have been developed by IT personnel of many years of practicing experience in web design and IT solutions. We render one of the most reliable, cost effective and professionally customized web design services in the world. No matter what your needs are, call it web site, high impact graphics or e-mail marketing, etc, you just name it and we will work up to your expectation, giving you a high quality product on time and at an affordable price without compromise. With our intensive knowledge and background experience in conceptualization, dynamic multimedia systems, developments and application, we combine a high quality graphic design techniques with stylish interactive customer friendly website delivery services in the following web / ICT solutions categories. - E-mail Marketing - Custom web design - Flash Animation - Mobile Advertising: - Sign Up To - Graphic Design Suite - Web Affiliations - 3D / 2D Animations - Search Engine Optimisation - Web Marketing Tools - Mobile Content Integration. (WAP) Our services are geared towards projecting the positive image of the global business community especially the small to medium-sized businesses and individuals. We are poised to give our clients a focal point of identity on the World Wide Web, (www) especially their online presence. WEB DESIGN With our core of professional web designers, we offer broad based web design services with comprehensive content presentations, that stimulates interest by conveying the most important message about your business, products and services for the benefit of your clients, customers and potential ones who are yet to know about your services. Our Web Design products range from affordable web site packages (for small businesses) to robust customized e-commerce solutions. You are always assured of quality and professional web design techniques that deliver front-graphic design and clean back-end programming. We deliver up to date technological experience in web technologies such as Flash, Action script, Java Script, HTML, XHTML/CSS, PHP, ASP, NET, Cold Fusion, mySQL and SQL databases. With combination of creativity, innovation and good business ethics, our clients are ensured of high performance websites every time they call for our services. Our custom websites starts around US $ 1,000.00 and takes three to five (3 – 5) weeks to finish. We encourage you to contact us now for solutions to all your website and IT problems or expectations. PACKAGES At Media Empire Ghana Ltd, the client is our most cherished assets; hence our services are tailored to suit their specific needs through varieties of packages such as: - The SME package - The Corporate Organizations package - The Schools and Educational Institutions incentives package - Churches and Non Governmental Organizations discounts package. Small and Medium Enterprise (SMEs) Our special package for the growth and opportunity of up and coming small and Medium Size Enterprises allows them to enjoy a 30% discount on all web – designs contracts. Contact us now for this special offer for your company to own its website in [5 weeks] NGO, CHURCHES AND SCHOOLS Non Governmental Organisations, churches and schools also enjoy special packages and discounts under the Media Empire’s “Schools, NGOs and Churches Support Programme”. This package includes design of a comprehensive website for basic educational institutions, such as International or preparatory schools, Second Cycle educational institutions, like senior High Schools and Tertiary schools at the Polytechnic or University levels. It is our desire and corporate responsibility to help these important institutions to take advantage of the magic of the internet to have their own website at a minimum cost. The good news is that, Media Empire Ghana Ltd, by our self will bear 30% of the cost for all NGOs and schools wishing to own their websites. INDIVIDUALS WHO WANT TO HAVE THEIR OWN WEBSITE At Media Empire Ghana Ltd, we encourage individuals to explore the vast opportunities derived in the new information communication age. This they can maximize by owing their customized website by contacting us to build to suit their choice and taste. CORPORATE ORGANISATIONS (COs) Media Empire Ghana Ltd’s corporate organisations’ special package includes a 10% discount on all ICT solutions, including new website designs, hosting, management, upgrading, search engine optimization, mobile advertising; sign up – to, e – mail marketing, flash animation, graphic design suits, 3D / 2D Animations, web affiliation services etc. OTHER PRODUCTS & SERVICES THAT WE OFFER FLASH: This product is customized to the personification of the client’s corporate or individual specification or character. It combines robust animation with multimedia content interactivity. Our flash presentation is an appealing business communication tool that allows business and individuals to present themselves to their audiences and prospective customers in a more effective way. Our technique allows us to capture the attraction of our audience and pay very much attention to interactivity. This is vividly captured in two forms; namely Interactive and Linear Presentations. WHAT OUR LINEAR PRESENTATION CAN OFFER This product has the capacity to encapsulate the essence of the corporate organization by putting forth its corporate ideals, core objectives and competences. It really showcase the crux of the organization with the presentation of captivating animations, interspersed with expertise from our 2D animators and flash designers, viewers interest does not fade off easily, even for a flash of just a second. INTERACTIVE PRESENTATION Our interactive presentation gives you an experience of a mesmerizing mélange of audio, visual and interactive elements. It provides the best possible way for one-on-one communication with the target audience, thereby, allowing audience to navigate the course of the presentation. This is guaranteed to keep them enthralled and entertained. 2D & 3D ANIMATION Our 2D and 3D Animation professionals enable us to provide reliable services in the areas of modeling, character design, motion graphics and compositing. With our expertise in animation techniques, we work to bring life into every aspect of product delivery. WEB ON MOBILE / WAP With the application of our Web on Mobile technique, we deliver the website in (wap) enable manner with WAP meaning [wireless application protocol], we design the website that allow people who uses mobile phone which have WML version 1.1 to see your website in a packed manner without the animations and colours but see the website in black and white pages on their mobile phone. CD AND DVD MEDIA Optical media discs including CDs, CD business cards, shaped CDs, mini CDs and DVDs can be used to deliver audio and Video Contents. Media Empire Ghana Ltd utilizes the state of art experiences to author, replicate, and duplicate optical media to our clients’ satisfaction. We create powerful multimedia and interactive presentations for CDs using Adobe Director and Adobe Flash and also use Final Cut Pro to edit and work on videos. Other CD projects include Power Point Presentations, website to CD conversions, resource CDs (including software, PDFs, MP3 and other media files. SKILLS & EXPERIENCE At Media Empire Ghana Ltd, we blend innovation with skills and experience. This allows us to be a step ahead of our competitors in the industry. We depend largely on talented individuals who have the passion for perfection and precision no matter how challenging the task is. Our strength stem from the fact that we have a culture of competent team that works and ensures that ideas are translated into viable business propositions for the benefit of our clients. Our store of experienced men and women is ably supported by efficient personnel made of - Content Systems Management - Graphic Designers - 2D & 3D Animations - Web Designers - Php, Script, Databases and MySQL - Software Programmers - Content Editors - Client Servicing Executives - Internet Marketing Gurus INTERLECTUAL PROPERTY CAUTION Media Empire Ghana Limited, web/IT solutions reserved the right to take a legal action against an individual or organisation or group of people who duplicate this document in its entire form or part of it or in any other form without prior authority. All information contained in this proposal remained the sole property of Media Empire Ghana Limited web/IT solution. ADDITIONAL FEATURES & SERVICES Additional features and services that are tailored to project your corporate image to outside world includes: - EMAIL MARKETING SERVICES – This is one feature of the internet marketing, there are over 2 billion websites on the internet whiles the internet welcomes it 1 billionth user in 2006. This makes it impossible to just design a website and not do any web advert online, and one of the most powerful web marketing tool is email marketing where you send bulk mails by a click to all your customers. - BULK MAIL SENDER – Send bulk mail messages with just a click and design your newsletter using the same programme. - EMAIL RETRIEVER – Retrieve emails as many emails as you want by putting in your parameters and what you want the retriever to retrieve for you. E.g. I want to retrieve emails of Ghanaians who live in London between the ages of 24 to 70 years and the retriever will do just that. - Bonded Sender Program: - This is a constant contact reseller tool of Bonded Sender Certification suitable for small and medium size organizations. - Custom E-mail Template Design Service: - Is a service that leaves a lasting impression with custom e-mail templates that uncompromisingly reflects your company’s unique brand and style. INTERLECTUAL PROPERTY CAUTION Media Empire Ghana Limited, web/IT solutions reserved the right to take a legal action against an individual or organisation or group of people who duplicate this document in its entire form or part of it or in any other form with out prior authority. All information contained in this proposal remained the sole property of Media Empire Ghana Limited web/IT solutions. MEDIA EMPIRE GHANA LIMITED WEB PACKAGES COMES IN TWO (2 CATEGORIES) 1. STANDARD WEBSITE PACKAGES 2. CUSTOMISED PACKAGES STATIC WEBSITE PACKAGES CORPORATE PROLITE: - $2,000.00 - Free Domain - Free Disk Space - Up to 40 comprehensively designed web pages - Up to 200 e-mails SEMI CORPORATE PROLITE: - $1,500.00 - Free Domain - Free Disk Space - Up to 20 comprehensively designed web pages - Up to 100 e-mails SME’s PROLITE: - $1,200.00 - Free Domain - Free Disk Space - Up to 60 comprehensively designed web pages - Up to 20 e-mails SEMI SME’s PROLITE: - $1,000.00 - Free Domain - Free Disk Space - Up to 8 comprehensively designed web pages - Up to 30 e-mails EDUCATIONAL INSTITUTIONS: - $800.00 - Free Domain - Free Disk Space - Up to 10 comprehensively designed web pages - Up to 100 e-mails CONTENT MANAGEMENT FACILITY $4,000.00 - Free Domain - Free Disk Space - Administration Panel - Add, Edit, Delete Content etc - CMS manual NOTE: - Pricing on complex CMS website development will be based on an agreed project description. Domain hosting shall be subject to yearly renewal as follows: a) Corporate Prolite = $350 b) Semi Corporate Prolite = $300 c) SME’s = $250 d) Semi SME Prolite = $200 All prices are VAT & NHL inclusive. WHAT YOU NEED TO KNOW DETAILS OF INPUT FOR CUSTOMISED WEBSITE CHARGES Layout and Prototype your customised web pages. Domain Hosting and Annual Renewals DOMAIN REGISTRATION: - this is the registration of your web address e.g. www.mycompany.com. You are allowed to choose a name for your domain. PROFESSIONALLY DESIGNED WEBSITE Having a web site is very essential for any organization that wants to be taken seriously in today’s environment. This will serve as a portal for presenting your vision both local and the world over. Websites are a cheaper, more effective way to present your organization or project to the whole wide world. It is time to take advantage of this opportunity. It is important for a website to be attractive, intuitive and yet functional. Benefit from our professionally designed websites that are tailored to reflect your organization’s message and image. Importance of a website for your organization - It serves as a portal where people can be informed about your organization. - Your information can be accessed by anyone anywhere in the world - Your information is available 24 hours a day, 7 days a week - It provides a portal to present your organization. - It gives your organization better recognition and appreciation - An opportunity to explain your vision or service in detail - Reach a global market segment - A place where information on your location and your activities can be accessed. This is what is included in this package Included in this amazing package is - A captivating and professionally designed Website - Your own unique Domain name e.g. www.yourdomain.com etc - Detailed traffic reports on number of visits to your site daily, monthly etc. - Create over 5,000 email addresses; enough for all managers, staff and departments - Directory listing - Email system that is compatible with Outlook E*press, Incredimail etc. UNIQUE EMAIL ADDRESSES Providing you with a quantity of e-mail address using your own domain name forms part of all packages. Create your own unique email address that reflects your departments or staff of your organization. email@example.com REGISTERED DOMAIN NAME Choose your own domain name that reflects your business, organization or church. This is the registration of your web address e.g. www.mycompany.com. You are allowed to choose a name for your domain. We register domains for .com .org .edu.net and so on. EMAIL TRANSFER TO YOUR PC Our POP3 & SMTP service delivers your emails directly to your laptop or PC at no extra charge. Finally! You get to use Outlook and other mail client software to send, receive and manage your emails. WEB MAIL ENABLES SERVICE With this service, you can also check, send and receive e-mails from anywhere internet is available. WEBSITE TRAFFIC REPORT Receive regular report on the number of visitors to your site free of charge. SEARCH SYSTEMS & E-MAILS The search system also helps you to locate any information that otherwise would have difficult to access ordinarily. THIS IS WHAT IS INCLUDED IN THIS PACKAGE Included in this amazing package is - A captivating and professionally designed Website - Your own unique Domain name e.g. www.yourdomain.com etc - Detailed traffic reports on number of visits to your site daily, monthly etc. - Create over 5,000 email addresses; enough for all managers, staff and departments - Directory listing - Email system that is compatible with Outlook Express, Incredimail etc. CONTENT MANAGEMENT SYSTEM This facility also known as CMS forms an important part of your web management system. It allows the website to be regularly updated, such as news and pictures to suit current trends. All security apparatus and features necessary for this facility will be provided while somebody from your company would be trained as ‘Web Administrator’ to use the CMS to edit and manage the website. Media Empire Ghana Ltd expertise would also be available to help should the need arise. PAGE DEVELOPMENT This includes number of pages of the website, pictures and all other materials on demand for professional web page. INTERFACE OPTIMISATION This is the linking of your company’s website to major search engines like google.com, yahoo.com etc; thereby making it easy for prospecting company’s website in any part of the world. THE PHOTO GALLERY A gallery of pictures “known as online photo album” is provided to showcase products and services the company is engaged in. ANIMATIONS These are moving images and their other information such as logos, banners etc, that goes into the designing of a web page. DATA BASE It enables accessibility to database utilisation on members of staff, management associations etc. NEWS COLUMNS This is the column for news about the company, its activities as well as the related industry. HIT COUNTERS This advanced high tech – allows the company to retrieve data on the exact number of visitors to their website on the day, time, week and moth basis. ONLINE PRESENTATION Unique feature of Power Point Presentation of products / services, upcoming events etc. FORUM / CHART This is a platform for instant communication online. It provides opportunity for instant feedback between company employees, management. REGISTRATIONS ONLINE This allows institutions to conduct online registration of customers or applicants to minimise cost. ELECTRONIC ORDER FORMS Electronically generated forms imbedded for customers to place orders for products from your website on the internet. FEED BACK FORM This allows clients to answer questionnaires or make enquiries, post comments or make complaints about the company’s products /services. LIBRARY / ARCHIVES A facility that allows access to detail information to stored materials for the references and research activities. DOWNLOAD PROTOCOL This facility is similar to the library menu. It facilitates easy downloads of general information. AUDIO / VIDEO STREAMING A live coverage online or pre-recorded coverage of corporate events such as workshops, seminars, end of year dinner, product launching, dinner dance, etc is adequately taken care of by this facility. MEMBER LOG IN FEATURE This allows limited access of information from your website by people who are enthused only through the use of ID and Password application. PLEASE NOTE Your selection of any of this customised website packages do not have specific price structure, due to variation of customer interest, taste and demand. Subscription to any of the above added features to client’s website shall be subject to price negotiation. INTERLECTUAL PROPERTY CAUTION Media Empire Ghana Limited, web/IT solutions reserved the right to take a legal action against an individual or organisation or group of people who duplicate this document in its entire form or part of it or in any other form without prior authority. All information contained in this proposal remained the sole property of Media Empire Ghana Limited. OUR CORE VALUES 1. WE WORK HARD FOR FAIR REWARDS 2. WE ARE FAIR AND THOUGHTFUL 3. WE GIVE BEFORE WE GET IN LIFE 4. WE SEEK TO UNDERSTAND THE OTHER PERSON BEFORE NEEDING TO BE UNDERSTOOD OURSELVES 5. WE ARE PROACTIVE NOT REACTIVE 6. WE STRIVE TO LEARN SOMETHING NEW EVERYDAY 7. WE OFFER A CONSTRUCTIVE SOLUTION WHEN WE COMPLAIN 8. WE ARE NOT AFRAID TO MAKE MISTAKES AND LEARN FROM THEM 9. WE CORRECT OUR MISTAKES AND LEARN FROM THEM 10. WE MAKE A COMMITMENT TO OUR WORK AND OURSELVES 11. WE TAKE RESPONSIBILITY FOR POOR RESULTS AND OWN FUTURE 12. WE SET CLEAR GOALS BEFORE BEGINNING A TASK 13. WE CHOOSE TO OVER COMMUNICATE RATHER THAN UNDER COMMUNICATE 14. WE ASK GOOD QUESTIONS OF OURSELVES AND OTHERS TO GET GOOD ANSWERS 15. WE KNOW OUR FUTURE REWARDS AND SUCCESSES ARE DIRECTLY LINK TO ACTION AND COMMITMENT WE MAKE TODAY PORTFOLIO www.mediaempireghana.com www.mychoicefmghana.com www.entertainmentghana.com www.northernwingtoday.com www.ghanaleagueclubs.com www.tackieyarboi.com www.abundantgracechristiancentre.com www.timewareghana.net www.ghanaloop.com www.aliveghana.com www.vbcint.com www.allafricanyouth.org www.vbcint.net www.womeninthegap.com www.victoryonlinestore.com www.cablegoldtv.com www.atwimakwanwoma.com www.travelmatters.com www.od4gh.com www.empireentertainmentgh.com www.pinkpantherhotel.net www.faceofwisconsin2009.com www.blaydinacademy.com www.royalpriesthoodschool.com www.rogerscoconstruction.com www.erikanstours.com www.deradectours.com www.montecarloghana.com www.empireentertainmentgh.com www.kingsfurnituregh.com www.empireradioschool.com www.spinalclinicghana.com www.hisupperroom.com www.actionchapel.com www.duncanwilliamsministry.com www.citypublicityghana.com www.wisebrainmagazine.com www.kenteradio.com
Deep Triplet-Driven Semi-supervised Embedding Clustering This is the author's manuscript Original Citation: Availability: This version is available http://hdl.handle.net/2318/1714021 since 2020-04-26T14:26:13Z Publisher: Springer Published version: DOI:10.1007/978-3-030-33778-0_18 Terms of use: Open Access Anyone can freely access the full text of works made available as "Open Access". Works made available under a Creative Commons license can be used according to the terms and conditions of said license. Use of all other works requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law. (Article begins on next page) 02 December 2021 Deep Triplet-Driven Semi-Supervised Embedding Clustering Dino Ienco\textsuperscript{1} and Ruggero G. Pensa\textsuperscript{2} \textsuperscript{1} IRSTEA, UMR TETIS, LJRM, Univ. Montpellier, Montpellier, France \email@example.com} \textsuperscript{2} University of Turin, Dept. of Computer Science, Turin, Italy \firstname.lastname@example.org} \textbf{Abstract.} In most real world scenarios, experts dispose of limited background knowledge that they can exploit for guiding the analysis process. In this context, semi-supervised clustering can be employed to leverage such knowledge and enable the discovery of clusters that meet the analysts’ expectations. To this end, we propose a semi-supervised deep embedding clustering algorithm that exploits triplet constraints as background knowledge within the whole learning process. The latter consists in a two-stage approach where, initially, a low-dimensional data embedding is computed and, successively, cluster assignment is refined via the introduction of an auxiliary target distribution. Our algorithm is evaluated on real-world benchmarks in comparison with state-of-the-art unsupervised and semi-supervised clustering methods. Experimental results highlight the quality of the proposed framework as well as the added value of the new learnt data representation. \section{Introduction} Clustering is by far one of the most popular machine learning task among computer scientists, machine learning specialists and statisticians. Although it is conceived to work in fully unsupervised scenarios, very often, its application in real-world domains is supported by the availability of some, scarce, background knowledge. Unfortunately, producing or extracting such background knowledge (in terms of available class labels or constraints) is a time consuming and expensive task. Hence, the amount of available background knowledge is not sufficient for driving a supervised task. Still, it can be helpful in guiding a semi-supervised learning process. The aim of semi-supervised clustering is to take advantage of the few available side information to guide the clustering process towards a partitioning that takes into account both the natural distribution of the data and the expectations of the domain experts. One of the most popular class of semi-supervised clustering algorithms exploit the so-called pairwise constraints: the clustering process is driven by a set of \textit{must-link} (or similarity) and \textit{cannot-link} (or dissimilarity) pairs modeling the fact that two data examples involved in any of these constraints should belong to the same cluster (must-link) or not (cannot-link). Such constraints are successively exploited to either learning a distance metric [14, 7, 12] or forcing constraints during the clustering process [23], although the most effective methods usually combine both strategies [4, 3, 19]. However, all these strategies suffer from the same two problems: i) two examples involved in a cannot-link constraint may actually be assigned to the wrong clusters and still satisfy the constraint; ii) when constraints are generated from the labeled portion of the training set (a common practice in semi-supervised learning), and the class is rather loose (e.g., multiple clusters co-exist within the same class), the must-link constraints would mislead the clustering algorithm resulting in poor partitioning results. To address this issue, an alternative form of supervision has been proposed: given three data examples $x_a$, $x_p$ and $x_n$, one may impose that $x_a$ (called reference or anchor example) is closer to $x_p$ (called positive example) than to $x_n$ (called negative example). Such relative comparisons form the so-called triplet constraints [15]. In this paper, we propose $T^2DEC$ (Triplet Semi-Supervised Deep Embedding Clustering). $T^2DEC$ is a deep embedding-based clustering framework that leverages triplet constraints to inject supervision in the learning process. The framework consists of a two-stage approach: i) an autoencoder extracts a low-dimensional representation (embedding) of the original data and ii) an initial cluster assignment is refined via the introduction of an auxiliary target distribution [25]. Both stages are guided by the knowledge supplied by triplet constraints. By means of an extensive experimental study conducted on several real-world datasets, we show that our approach outperforms state-of-the-art semi-supervised clustering methods no matter how much supervision is considered. ## 2 Related Work Early semi-supervised approaches used pairwise (e.g., must-link and cannot-link) constraints to learn a metric space before applying standard clustering [14], or to drive the clustering process directly [23]. In [23], a simple adaptation of k-means that enforces must-link and cannot-link constraints during the clustering process is described. [2] proposes a constrained clustering approach that leverages labeled data during the initialization and clustering steps. Instead, [4] integrates both constraint-based and metric-based approaches in a k-means-like algorithm. Davis et al. propose an information-theoretic approach to learning a Mahalanobis distance function [7]. They leverage a Bregman optimization algorithm [1] to minimize the differential relative entropy between two multivariate Gaussians under constraints on the distance function. This approach has been recently extended by Nogueira et al., who combine distance metric learning and cluster-level constraints [19]. Zhu et al. present a pairwise similarity framework to perform an effective constraint diffusion handling noisy constraints as well [28]. In recent years, the advances in the deep learning field have also fostered new research in semi-supervised clustering. For instance, in [11], the author propose a semi-supervised clustering algorithm that directly exploits labels, instead of pairwise constraints. Their algorithm generates an ensemble of multiresolution semi-supervised autoencoders. The final partitioning is obtained by applying k-means on the new data representation obtained by stacking together all the different low-dimensional embeddings. In a very recent work [21], a semi-supervised extension to Deep Embedded Clustering (DEC [25]) is proposed. DEC learns a low-dimensional representation via autoencoder and, successively, it gradually refines clusters with an auxiliary target distribution derived from the current softcluster assignment. Its semi-supervised extension [21], called Semi-supervised Deep Embedded Clustering (SDEC) makes use of pairwise constraints in the cluster refinement stage. Therefore, the learned feature space is such that examples involved in a must-link (resp. cannot-link) constraint are forced to be close (resp. far away) from each other. Our approach is also based on DEC, but, contrary to [21], it exploits triplet constraints introducing the background knowledge at the different stages of the process: during the embeddings generation and during the clustering refinement. We remind that, the expressiveness of triplet constraints has already demonstrated to be effective in the constrained clustering task [15]. 3 Triplet Semi-Supervised Deep Embedding Clustering In this section, we introduce our semi-supervised clustering approach, called $T^2$2DEC (Triplet semi-supervised Deep Embedding Clustering). The goal is to group together a set of examples $X = \{x_i\}_{i=1}^{N}$ into $C$ clusters given some background knowledge in terms of constraints. To this purpose, we model our problem using neural networks. In a nutshell, given $ML = \{(x_j, x_i)\}$ (resp. $CL = \{(x_j, x_i)\}$) the set of must-link (resp. cannot-link) constraints, first we derive triplet constraints from these two sets. A triplet constraint is defined as a tuple $(x_a, x_p, x_n)$ where $x_a$ is the anchor example and $x_p$ (resp. $x_n$) is the positive (resp. negative) example with the associated semantic that $x_a$ and $x_p$ (resp. $x_a$ and $x_n$) belong (resp. do not belong) to the same cluster. Furthermore, due to transitivity, we also have that $x_p$ and $x_n$ do not belong to the same cluster. Successively, due to the exponential number of triplets we can generate, we adopt a simple and practical strategy to sample a subset of such triplets. We remind that triplet selection is an hard task and some research works are investigating how to smartly sample useful and informative subsets of triplet constraints [26]. It is out of the scope of this work supplying a method that competes with such strategies. On the other hand, we set up an easy and ready to use approach that well fits our scenario. Once the set of triplet constraints are chosen, we inject such background information into a deep-learning based clustering algorithm [26, 9, 27]. More in detail, we integrate the semi-supervision during: i) the data embedding generation, by alternating unsupervised and semi-supervised optimization of the network parameters and ii) the clustering refinement stage when cluster assignment hardening loss [18] is employed. Figure 1(a) and Figure 1(b) provide a general overview of the embedding generation and clustering refinement stage, respectively. For each stage, we depict with the rose color and the dotted line the components related to the semi-supervised optimization (working on triplet constraints) while we depict with the blue color and the solid line the fully unsupervised components (working on the whole set of data $X$). In the following, we provide the details of all the algorithmic steps of our approach. Triplets generation strategy The first preliminary step of $Ts2DEC$ is the generation of a set $T$ of triplet constraints from the set $ML$ and $CL$ of must-link and cannot-link constraints. To achieve this goal, first, we compute the transitive closure from both sets [6], then, we leverage it to generate all possible triplet constraints $(x_a, x_p, x_n)$. Generating triplets in such a way can produce a huge number of constraints. Consequently, we limit the number of triplet constraints by adopting the following strategy: for each pair $(x_a, x_p)$, we randomly sample a subset of possible examples that can play the role of negative examples $(x_n)$. Here, we give more importance to background knowledge that groups together similar examples (positive information) than information that forces examples to be clustered apart (negative information). We adopt this strategy because, during our experimental evaluations, we have empirically observed that positive information seems more effective in stretching the representation manifold thus respecting the given background knowledge. In our experiments, we sample 30% of all possible negative examples for each pair $(x_a, x_p)$ in order to obtain a reasonable trade off between performances and computational cost. The obtained set of triplet constraints is denoted by $T$. Embedding generation with background knowledge The core of $Ts2DEC$ involves a first stage in which semi-supervised embedding representations are generated by means of autoencoder neural networks. This stage is depicted in Figure 1(a). Autoencoders [16] are a particular kind of feed-forward neural network commonly employed to generate low-dimensional representation of the original data by setting up a reconstruction task. The autoencoder network is composed by two parts: i) an encoder network that transforms the original data $X$ into an embedding representation (EMB) and ii) a decoder network that reconstructs the original data from the embedding representation. Furthermore, the autoencoder network is layered and symmetric and the last layer of the encoder part is generally referred as bottleneck layer. The commonly adopted loss function optimized by an autoencoder network is the mean squared error between the original data and the reconstructed one: \[ L_{ac} = \frac{1}{\|X\|} \sum_{x_i \in X} \|\|x_i - dec(enc(x_i, \Theta_1), \Theta_2)\|\|^2_2 \] (1) where $enc(z, \Theta_1)$ is the encoder network with parameters $\Theta_1$, while $dec(\cdot, \Theta_2)$ is the decoder network that reconstructs the data, with parameters $\Theta_2$. For the encoder network, similarly to what proposed in [25], we adopt a feed-forward neural network with four layers (resp. 500, 500, 2000, 10 neurons per layer). The activation function associated to the first three (hidden) layers is the Rectifier Linear Unit (ReLU) while, for the last (bottleneck) layer, a simple linear activation function is employed [25]. The decoder is symmetrically derived from the encoder reversing the hidden layers. A semi-supervised autoencoder [8, 20] (denoted as SSAE), instead, is a multi-task network that, in addition to the reconstruction task via its autoencoder structure, also deals with a discrimination task (mainly classification) leveraging the embedded representation. Conversely to most previous works on semi-supervised autoencoders [8, 20, 11] where the SSAE exploits labeled data to perform classification as supervised task, here, we design a SSAE that, associated to the reconstruction task, exploits the set $T$ of triplet constraints to generate the low-dimensional data embeddings EMB. The set $T$ of triplets is used to learn a triplet network which consists of three different encoders/decoders with shared weights (highlighted in rose color and dotted line in Figure 1(a)). In addition to the standard reconstruction loss, the specific loss function (triplet loss) optimized by the model is defined as follows: \[ L'_{triplet} = \sum_{(x_a, x_p, x_n) \in T} [d(x_a, x_p) - d(x_a, x_n) + \alpha]_+ \] (2) with \[ d(b, c) = \|\|norm_{L_2}(enc(b, \Theta_1)) - norm_{L_2}(enc(c, \Theta_1))\|\|^2_2 \] (3) where $ T $ is the set of triplet constraints, $[x]_+ = max(0, x)$ is the hinge loss, $\|\|x\|\|^2_2$ is the squared $L_2$ norm of $x$, $enc(x, \Theta_1)$ is the encoder network, with weights parameters $\Theta_1$, applied on an example $x$, $norm_{L_2}$ is a function that performs the $L_2$ normalization of the output of the encoder and $\alpha$ is the margin hyperparameter usually involved in distance-based loss function to stretch the representation space [26]. We consider $\alpha$ equal to 1.0 since distances are derived by $L_2$ normalization. Additionally, we can observe that, due to the transitivity relation among the examples in the triplet tuple, we can also define a second triplet loss function: \[ L''_{triplet} = \sum_{(x_a, x_p, x_n) \in T} [d(x_a, x_p) - d(x_p, x_n) + \alpha]_+ \] (4) where the second term of the hinge loss, this time, consider the relationship between the $x_p$ and $x_n$ examples. In the rest of the paper, $L'_{triplet}$ and $L''_{triplet}$ are exploited to introduce semi-supervision in the clustering process and we use the notation $L_{triplet}$ to indicate the sum of the two triplet loss functions: $L_{triplet} = (L'_{triplet} + L''_{triplet})$. The overall architecture of our semi-supervised autoencoder involves the optimization of $L_{triplet}$ loss as well as the simultaneous reconstruction of the examples concerned by the constraints. Given $T$, the set of triplet constraints, the loss function optimized by the $SSAE$ is as follows: \[ L_{ssae} = \frac{1}{\|T\|} \left( \left[ \sum_{t \in T} \sum_{x_i \in t} \|\|x_i - dec(enc(x_i, \Theta_1), \Theta_2)\|\|^2_2 \right] + \lambda L_{triplet} \right) \] (5) where $t = (x_a, x_p, x_n)$ is a generic triplet, $\lambda$ is a hyperparameter that controls the importance of the triplet loss term. Such loss function optimizes the parameters $\Theta_1$ and $\Theta_2$ so as to optimize the data reconstruction as well as to meet the constraint relationships expressed by the background knowledge. In $L_{ssae}$, the reconstruction term is considered with the aim of regularizing the action of the $L_{triplet}$ loss. Therefore, we obtain embeddings that meet the requirements expressed by the constraints as well as with the main reconstruction task. We underline that, in our context, the embedding generation process involves two different stages: the first one implies the optimization of the autoencoder loss on the full set of data $X$ while, the second one regards the optimization of the semi-supervised autoencoder loss considering only the set of examples $X_t$ ($X_t = \{x_i \in t \| t \in T\}$) covered by the triplet constraint set. Algorithm 1 reports the joint optimization procedure we employ to learn the weight parameters $\Theta_1, \Theta_2$. In a generic epoch, the procedure optimizes: i) the unsupervised loss associated to data reconstruction on the set of data $X$ (line 3-4) and, ii) both reconstruction and triplet losses ($L_{ssae}$) considering the set of data involved in the set $T$ (line 5-6). The learning of parameters is achieved via a gradient descent based approach using mini-batches. Finally, the data embeddings are generated considering the $\Theta_1$ parameters associated to the encoder network $enc(\cdot, \Theta_1)$. Algorithm 1 Semi-supervised autoencoder optimization Require: $X$, $T$, N\_EPOCHS Ensure: $\Theta_1, \Theta_2$. 1: $i = 0$ 2: while $i < N\_EPOCHS$ do 3: Update $\Theta_1$ and $\Theta_2$ by descending the gradient: 4: $\nabla_{\Theta_1, \Theta_2} \frac{1}{\|X\|} \sum_{x_i \in X} \|\|x_i - dec(enc(x_i, \Theta_1), \Theta_2)\|\|^2_2$ 5: Update $\Theta_1$, $\Theta_2$ by descending the gradient: 6: $\nabla_{\Theta_1, \Theta_2} \frac{1}{\|T\|} \left( \left[ \sum_{i \in T} \sum_{x_i \in I} \|\|x_i - dec(enc(x_i, \Theta_1), \Theta_2)\|\|^2_2 \right] + \lambda L_{triplet}(T) \right)$ 7: $i = i + 1$ 8: end while 9: return $\Theta_1, \Theta_2$ Clustering refinement with background knowledge Once the embedding representation produced by the SSAE is obtained, the final stage consists in a clustering refinement step via cluster assignment hardening [18, 25] as depicted in Figure 1(b). Here, we iterate between computing an auxiliary target distribution and minimizing the Kullback-Leibler (KL) divergence with respect to it. More in detail, as depicted in Figure 1(b), we discard the decoder part of the previous model ($\Theta_2$ parameters) but we still allow modifications of encoder parameters $\Theta_1$. Given the initial cluster centroids $\{c_j\}_{j=1}^{C}$, the cluster assignment hardening technique tries to improve the partitioning using an unsupervised algorithm that alternates between two steps: i) compute a soft assignment between the embeddings and the cluster centroids and ii) update the embedded data representation and refine the cluster centroids by learning from current high confidence assignments leveraging an auxiliary target distribution. The process is repeated until convergence is achieved or a certain number of iterations is executed. To generate the clustering centroids we use K-Means on the embeddings produced by the encoder network. To compute the soft assignment, as commonly done in deep embedding clustering approaches, we exploit the Student’s t-distribution as a kernel to measure the similarity [17]: $$q_{ij} = \frac{(1 + \|\|EMB_i - c_j\|\|^2)^{-1}}{\sum_{l=1}^{C}(1 + \|\|EMB_i - c_l\|\|^2)^{-1}}$$ \hspace{1cm} (6) where $EMB_i$ is the embedded representation of the $i-th$ example obtained via $enc(x_i, \Theta_1)$, $c_j$ (resp. $c_l$) is the cluster centroid of the $j-th$ (resp. $l-th$) cluster, and $q_{ij}$ is the soft assignment between example $x_i$ and cluster $c_j$. Once the soft assignments are computed, they are iteratively refined by learning from their high-confidence assignments with the help of an auxiliary target distribution. The target distribution is defined as: $$p_{ij} = \frac{q_{ij}^2 / \sum_i q_{ij}}{\sum_j (q_{ij}^2 / \sum_i q_{ij})}$$ \hspace{1cm} (7) Such distribution forces the assignment to have stricter probabilities (closer to 0 or 1) by squaring the original distribution and then normalizing it [18]. To match the soft-assignment $q$ with the auxiliary target distribution $p$, we employ the Kullback-Leibler (KL) divergence as loss function to evaluate the distance between the two probability distributions. The KL divergence is computed between the soft assignment $q_i$ and the auxiliary distribution $p_i$: $KL(P\|\|Q) = \sum_i p_i \cdot \log \frac{p_i}{q_i}$. Furthermore, we integrate the semi-supervision supplied by the background knowledge in this step as well, by adding the information carried out by the triplet constraints to the overall loss function: $$L_{sscr} = KL(P\|\|Q) + \lambda L_{triplet}$$ (8) The resulting loss function considers the auxiliary target distribution together with the triplet constraints when upgrading the parameters of the encoder ($\Theta_j$). Hence, this last step has also an influence on the way embeddings are computed. As before, $\lambda$ is an hyperparameter controlling the importance of the triplet loss term and it is the same in the two steps of our framework. To optimize such semi-supervised loss $L_{sscr}$ we adopt a similar strategy to what proposed in Algorithm 1. Finally, once convergence is reached, each example is assigned to the cluster that maximizes its assignment score: $cluster(x_i) = argmax_j q_{ij}$. 4 Experiments In this section, we assess the effectiveness of $Ts2DEC$ on several real world datasets comparing its behavior w.r.t. competitors. Then, we consider the impact of the different components of $Ts2DEC$ by means of an ablation study. Finally, we provide a visual inspection of the representation learnt by our strategy. Competitors For the quantitative evaluation, we compare the performances of $Ts2DEC$ with those obtained by different unsupervised and semi-supervised competing algorithms. The former are employed as baselines to understand the gain related to the introduction of weak supervision; the latter consist of fair state-of-the-art competitors that are more closely related to the task at hand. As regards the unsupervised approaches, we consider $K$-Means and DEC [25], a recent deep learning unsupervised clustering approach (the unsupervised clustering algorithm $Ts2DEC$ is built upon). As semi-supervised clustering algorithms, we consider the following competitors: a semi-supervised variant of $K$-Means, named $MPCKmeans$ [4]; a recent constrained spectral clustering method [5], called Spectral; two very recent deep learning based methods named $MSAEClust$ [11] and $SDEC$ [21]. $MPCKmeans$ combines metric-learning and pairwise constraint processing to exploit the supplied supervision as much as possible. Spectral captures constrained clustering as a generalized eigenvalue problem via graph Laplacians. [11] employs an ensemble of semi-supervised autoencoders to learn embedding representations that fit the data as well as the background knowledge and that are finally used to perform clustering. Finally, $SDEC$ is a direct extension of $DEC$ that integrates the pairwise constraints in the clustering refinement stage, by adding an extra term to the cluster assignment hardening loss. Experimental settings and datasets To measure the clustering performances of all the methods, we use the Normalized Mutual Information (NMI) [22] as well the Adjusted Rand Index (ARI) [10]. Both NMI and ARI take their maximum value when the clustering partition completely matches the original one, i.e., the partition induced by the available class labels. The NMI measure ranges between $[0, 1]$ while the ARI index ranges between $[-1, 1]$. Both evaluation metrics can be considered as an indicator of the purity of the clustering result. For each dataset, both measures are computed considering the whole set of examples, including the ones on which the constraints are defined. We analyze the behavior of the different methods according to increasing levels of supervision. More in detail, we simulate the supervision in term of constraints, by selecting a number of labeled examples per class and, successively, inducing the correcorresponding full set of constraints. We vary such amount of labeled examples per class between 5 and 25 with a step of 5. Due to the randomness of the sample selection process and the non deterministic nature of the clustering algorithms, we repeat the sample selection step 5 times for each number of per-class labels and, successively, we repeat the clustering process 10 times. For $Ts2DEC$ we derive the corresponding triplet constraints as explained in Section 3. Finally, for each level of supervision, we report the average values of NMI and ARI. For all the methods, the number of clusters is equal to the number of classes. $Ts2DEC$ is implemented via the Tensorflow python library and the implementation is available online\footnote{\url{https://gitlab.irstea.fr/dino.ienco/ts2dec}}. Model parameters are learnt using the Adam optimizer [13] with a learning rate equal to $1 \times 10^{-3}$ for the autoencoder (reconstruction and triplet loss functions) and we use Stochastic Gradient Descent with learning rate equal to $1 \times 10^{-4}$ for the Clustering Refinement stage (KL loss function) as done in $DEC$ [25] and $SDEC$ [21]. We set the value of $\lambda$ equal to $1 \times 10^{-3}$, a batch size of 256 and a number of epochs equal to 50 for the semi-supervised autoencoder. For the refinement clustering stage, we iterate the procedure for 20000 batch iterations as done in $DEC$ [25] and $SDEC$ [21]. For all the competitors, we use publicly available implementations. For $SDEC$, the source code was kindly provided by the authors of the related paper. Experiments are carried out on a workstation equipped with an Intel(R) Xeon(R) W-2133, 3.6Ghz CPU, with 64Gb of RAM and one GTX1080 Ti GPU. To evaluate the behavior of all the competing approaches the experiments are performed on four publicly available datasets: (1) $USPS$ is a handwritten digit recognition benchmark (10 classes) containing 9298 grayscale images with size 16 x 16 pixels and provided by the United States Postal Service. (2) $MNIST$ is a dataset of Zalando’s article images (shirt, sneakers, bags, etc..) consisting of 70000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes. It serves as a more complex drop-in replacement for the original MNIST benchmark [24]. (3)$Reuters10k$ is an archive of English news stories labeled with a category tree that contains 810000 textual documents. Following [25], we used 4 root categories: corporate/industrial, government/social, markets and economics as labels and excluded all documents with multiple labels. We randomly sampled a subset of 10000 examples and computed TF-IDF features on the 2000 most frequent words. (4)$Optdigits$ is a dataset of the UCI repository involving optical recognition of handwritten digits. It contains 5620 examples described by 64 feature each. Quantitative evaluation Figure 2 and 3 report the performances of the different approaches on the four benchmarks in terms of NMI and ARI, respectively. Notice that $Spectral$ was not able to process the $MNIST$ benchmark due to the fact that the original implementation cannot handle a dataset with 70000 examples. We observe that both NMI and ARI depict a similar situation. At first look, we note that $Ts2DEC$ outperforms all the competing approaches regarding any amount of supervision for all the four benchmarks. In addition, the graphs generally show that the margin gained by $Ts2DEC$ increases with the amount of available supervision. This behavior is particularly evident in $USPS$, $fMNIST$ and $Reuters10k$. Considering $Optdigits$, we observe an improvement between the supervision value 5 and 10 while, later on, $Ts2DEC$ remains stable according to NMI and it slightly increases according to ARI. This is not the case for all the other semi-supervised competitors. For instance, considering the $fMNIST$ benchmark, we note that all competitors remain almost stable while varying the amount of supervision, underlying the fact that they are unable to exploit increasing amount of background knowledge properly. Unexpectedly, we observe that one of the best competitor is DEC, which is completely unsupervised. More surprisingly, $SDEC$ performs similarly to its unsupervised counterpart. Ablation and parameter analysis In this section, we study the impact of the different components of $Ts2DEC$ that involve supervision, as well as the sensitivity of our method to hyperparameter $\lambda$. To do this, we fix the amount of supervision by considering 10 labeled examples from each class. For the first study, we derive two variants of our method: i) $Ts2DEC_{v1}$ which considers backTable 1. Impact of the different components of $Ts2DEC$ considering the NMI measure. \| Dataset \| $Ts2DEC$ \| $Ts2DEC_{v1}$ \| $Ts2DEC_{v2}$ \| \|-------------\|------------\|---------------\|---------------\| \| USPS \| 0.86 ± 0.02 \| 0.86 ± 0.02 \| 0.82 ± 0.03 \| \| fMNIST \| 0.65 ± 0.01 \| 0.64 ± 0.01 \| 0.64 ± 0.02 \| \| Reuters10k \| 0.64 ± 0.03 \| 0.64 ± 0.03 \| 0.61 ± 0.03 \| \| Optdigits \| 0.92 ± 0.01 \| 0.91 ± 0.01 \| 0.91 ± 0.02 \| Fig. 4. Sensitivity analysis of the $\lambda$ hyperparameter: NMI (a) and ARI (b) are reported for increasing weight of semisupervision. ground knowledge only to generate embeddings via semi-supervised autoencoder, and ii) $Ts2DEC_{v2}$ which considers background knowledge only during the clustering refinement stage. Table 1 reports the results of this study in terms of NMI. We note that the best performances are obtained when semi-supervision is injected at both stages of our process. Furthermore, we observe that $Ts2DEC_{v1}$ consistently achieves slightly better results than $Ts2DEC_{v2}$ in terms of NMI. The results of the sensitivity analysis are given in Figure 4. In details, we let the hyperparameter $\lambda$ varies in the range $(10^{-4}, 10^{-3}, 10^{-2}, 10^{-1}, 10^0)$. At first look, USPS, fMNIST and Optdigits exhibit a similar behavior. When $\lambda$ is too small ($10^{-4}$), supervision is not that effective while, starting from $\lambda$ equal to $10^{-3}$, we observe that $Ts2DEC$ achieves stable performances and becomes insensitive to such parameter. On the other hand, for the Reuters10k dataset, we note that the performances slightly decrease when $\lambda$ increases. At a deeper inspection, we observe that raising the value of $\lambda$ results in an increase of the standard deviation associated to the average value plotted in Figure 4. We remind that this benchmark is characterized by a high-dimensional feature space (2000 features) and, the encoder/decoder architecture (inherited from the DEC method) is unable to compress the original data properly and, simultaneously, incorporate the supervision. This may explain the increasing instability and reduced performances when augmenting the importance of the semi-supervision. Visual inspection Here, we visually analyze the embedding generated by our approach on the Optdigits benchmark. To this end, we visually compare the embeddings derived by Ts2DEC with the embeddings generated by the other deep learning competitors considering an amount of labeled examples per class equal to 10 (i), and by increasing the amount of background knowledge from 5 to 20 labels per class (ii). To obtain the two dimensional representations, we apply the t-distributed stochastic neighbor embedding (TSNE) approach [17]. For this evaluation we consider 300 instances per class. In the former evaluation (Figure 5), we clearly note that the visual representation induced by Ts2DEC provides a better separation among examples belonging to different classes and, simultaneously, locates examples belonging to the same class close to each other. The latter experiment (Figure 5) shows the ability of Ts2DEC to modify the data manifold exploiting the increasing amount of background knowledge. We observe that clear differences exist between the embeddings learnt when 5 (Figure 6(a)) and 15 labeled examples (Figure 6(c)) per class are considered, the latter providing significant better class separation than the former. 5 Conclusion We have presented $TS2DEC$, a new semi-supervised deep embedding clustering technique that integrates background knowledge as triplet constraints. More precisely, $TS2DEC$ integrates the background knowledge at two stages: i) during the data embedding generation and ii) during the clustering refinement. Extensive evaluations on real-world benchmarks have shown that $TS2DEC$ outperforms state-of-the-art competitors w.r.t different amount of background knowledge. References 1. Banerjee, A., Merugu, S., Dhillon, I.S., Ghosh, J.: Clustering with bregman divergences. Journal of Machine Learning Research 6, 1705–1749 (2005) 2. Basu, S., Banerjee, A., Mooney, R.J.: Semi-supervised clustering by seeding. In: ICML, pp. 27–34 (2002) 3. Basu, S., Bilenko, M., Mooney, R.J.: A probabilistic framework for semi-supervised clustering. In: KDD, pp. 59–68 (2004) 4. Bilenko, M., Basu, S., Mooney, R.J.: Integrating constraints and metric learning in semi-supervised clustering. In: ICML, pp. 81–88 (2004) 5. Cucuringu, M., Koutis, I., Chawla, S., Miller, G.L., Peng, R.: Simple and scalable constrained clustering: a generalized spectral method. In: AISTATS, pp. 445–454 (2016) 6. Davidson, I., Ravi, S.S.: Intractability and clustering with constraints. In: ICML. pp. 201–208 (2007) 7. Davis, J.V., Kulis, B., Jain, P., Sra, S., Dhillon, I.S.: Information-theoretic metric learning. In: ICML. pp. 209–216 (2007) 8. Haiyan, W., Haomin, Y., Xueming, L., Haijun, R.: Semi-Supervised Autoencoder: A Joint Approach of Representation and Classification. In: CICN. pp. 1424–1430 (2015) 9. Harwood, B., G, V.K.B., Carneiro, G., Reid, I.D., Drummond, T.: Smart mining for deep metric learning. In: ICCV. pp. 2840–2848 (2017) 10. Hubert, L., Arabie, P.: Comparing partitions. Journal of Classification 2(1), 193–218 (1985) 11. Ienco, D., Pensa, R.G.: Semi-supervised clustering with multiresolution autoencoders. In: IJCNN. pp. 1–8 (2018) 12. Kainthla, W., Ono, S., Nurnao, M., Fukui, K.: Kernelized evolutionary distance metric learning for semi-supervised clustering. In: AAAI. pp. 4945–4946 (2017) 13. Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. CoRR abs/1412.6980 (2014). http://arxiv.org/abs/1412.6980 14. Klein, D., Kanvar, S.D., Manning, C.D.: From instance-level constraints to space-level constraints: Making the most of prior knowledge in data clustering. In: ICML. pp. 307–314 (2002) 15. Kumar, N., Kunnamamuru, K.: Semisupervised clustering with metric learning using relative comparisons. IEEE Trans. Knowl. Data Eng. 20(4), 496–503 (2008) 16. LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521, 436 (may 2015), https://doi.org/10.1038/nature14539 17. van der Maaten, L., Hinton, G.E.: Visualizing high-dimensional data using t-sne. JMLR 9, 2579–2605 (2008) 18. Min, E., Guo, X., Liu, Q., Zhang, G., Cui, J., Long, J.: A survey of clustering with deep learning: From the perspective of network architecture. IEEE Access 6, 39501–39514 (2018) 19. Nogueira, B.M., Tomas, Y.K.B., Marcaccini, R.M.: Integrating distance metric learning and cluster-level constraints in semi-supervised clustering. In: IJCNN. pp. 4118–4125 (2017) 20. Rasmus, A., Berglund, M., Honkala, M., Valpola, H., Raiko, T.: Semi-supervised learning with ladder networks. In: NIPS. pp. 3546–3554 (2015) 21. Ren, Y., Hu, K., Dai, X., Pan, L., Hoi, S.C.H., Xu, Z.: Semi-supervised deep embedded clustering. Neurocomputing 325, 121–130 (2019) 22. Strehl, A., Ghosh, J.: Cluster ensembles — A knowledge reuse framework for combining multiple partitions. Journal of Machine Learning Research 3, 583–617 (2002) 23. Wagstaff, K., Cardie, C., Rogers, S., Schrödl, S.: Constrained k-means clustering with background knowledge. In: ICML. pp. 577–584 (2001) 24. Xiao, H., Rasul, K., Vollgraf, R.: Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. CoRR abs/1708.07747 (2017) 25. Xie, J., Girshick, R.B., Farhadi, A.: Unsupervised deep embedding for clustering analysis. In: ICML. pp. 478–487 (2016) 26. Yu, B., Liu, T., Gong, M., Ding, C., Tao, D.: Correcting the triplet selection bias for triplet loss. In: ECCV. pp. 71–86 (2018) 27. Zhao, Y., Jin, Z., Qi, G., Lu, H., Hua, X.: An adversarial approach to hard triplet generation. In: ECCV. pp. 508–524 (2018) 28. Zhu, X., Loy, C.C., Gong, S.: Constrained clustering with imperfect oracles. IEEE Trans. Neural Netw. Learning Syst. 27(6), 1345–1357 (2016)
Nonlinear energy operators for defibrillation Shock Outcome Prediction Beatriz Chicote\textsuperscript{1}, Unai Irusta\textsuperscript{1}, Elisabete Aramendi\textsuperscript{1}, Iraia Isasi\textsuperscript{1}, Daniel Alonso\textsuperscript{2}, Fernando Vicente\textsuperscript{2}, María Sanchez\textsuperscript{2} \textsuperscript{1}University of the Basque Country (UPV-EHU), Bilbao, Spain \textsuperscript{2}Emergentziak Osakidetza (Basque Health Service), Bilbao, Spain Abstract Accurate prediction of shock success would avoid futile defibrillation attempts that may damage the myocardium, and would help optimizing treatment decisions for out-of-hospital cardiac arrest (OHCA) patients. This work applies the Smoothed Nonlinear Energy Operator (SNEO) to analyze the energy content of the pre-shock ventricular fibrillation (VF) waveform acquired by automated external defibrillators (AED). A database of 419 shocks was analyzed and shock outcome predictors were calculated in the a 5-s pre-shock ECG segment. The SNEO was compared to some classical VF features. For each feature a detector of successful shocks was designed minimizing the Balanced Error Rate (BER). Finally, using SNEO as shock outcome predictor the minimum pre-shock segment duration was determined. The SNEO has proven to be a good shock outcome predictor even for 2-s segments and it could be used to optimize treatment decisions for OHCA patients. 1. Introduction In sudden cardiac arrest (SCA) two early interventions are key for the survival of the patients, early defibrillation and early cardiopulmonary resuscitation (CPR). Useless interruptions or ineffective chest compressions (CC) during CPR, in addition to delays in CPR or in the access to the automated external defibrillator (AED) may adversely affect patients’ survival. The survival rate also decreases with every futile defibrillation attempt due to damage to the myocardium produced by shocks. Consequently an accurate prediction of optimal therapy, defibrillation or continuation of CPR, is of major relevance. Predicting defibrillation success, i.e the development of accurate shock outcome predictors, would help optimizing timing of defibrillaion. A noninvasive approach to shock outcome prediction is ECG analysis of the VF waveform, through which many predictors/features have been developed over the years [1–3]. In this work, we present the Smoothed Nonlinear Energy Operator (SNEO) [4] as a shock outcome predictor. SNEO is based on the analysis of the local energy content of the VF-waveform and in this work we also compare it to other classical shock outcome prediction features. After that, we determine the minimum ECG pre-shock segment duration for an accurate shock outcome prediction using SNEO. 2. Methods 2.1. Data collection and annotation A dataset of 1009 out of hospital cardiac arrest (OHCA) cases was used for this study. The OHCA patients were treated by the basic life Support (BLS) services of the Basque Health Service (Osakidetza) between January 2013 and June 2015. The Emergency service of the Basque Autonomous Community is a two-tier system, where BLS is the first at scene and the patients were treated with automated external defibrillators (AED). Data from the following AEDs were collected: LifePack 1000 (Physio-Control, Redmond, WA, USA), ZOLL AED PRO (ZOLL Medical, Chelmsford, MA, US) and Philips Medical Systems Heartstart FR2 (Philips Medical Systems, Andover, MA). The ECG resolution and sampling frequencies of the devices were: 4.8/4.8/2.5 µV and 125/250/200 Hz, respectively. The ECG data was recorded in the three AEDs, but only using LifePack 1000 the thoracic impedance (TI) was recorded. Using the manufacturer’s custom tools all data and the messages from the devices were exported to a common MATLAB format. The signals were resampled to a common sampling rate of 250 Hz. Shocks were identified using the messages from the AEDs, and a 30-s pre-shock ECG interval for analysis, and a 70-s post-shock interval to annotate the outcome were extracted. Shocks were considered successful if sustained QRS complexes (rate > 30 min$^{-1}$) appeared within one minute. Cases where there was no ECG signal, it was corrupted by CC-artefacts. or any other noise, or the rhythm annotation was not possible were removed. The 5-s preshock VF segments 1-s prior the defibrillation were extracted. The final dataset contained a total of 419 shocks from 163 patients, 107 of which (65 patients) were successful and 312 (125 patients) unsuccessful. ### 2.2. Shock outcome predictors The segments of the database were preprocessed using a 4th-order bandpass elliptic filter with 1 dB of passband ripple, 30 dB of stopband attenuation and a typical AED passband of 0.5-30 Hz. The filter suppressed high frequency noise and baseline oscillations. The non-linear Teager-Kaiser Energy Operator (TKEO) [5,6] was applied to the ECG segments. TKEO $(\psi_k[x(n)])$ is expressed in the discrete domain as the Equation (1), where the constant $k$ is the lag parameter and $x(n)$ is the VF-segment: \[ \psi_k[x(n)] = x^2(n) - x[n-k]x[n+k] \] (1) Then TKEO was convolved with a Kaiser window, to obtain the Smoothed Nonlinear Energy Operator (SNEO): \[ \psi_{S,L}[x(n)] = \psi_k[x(n)] \otimes w_L(n) \] (2) were $\otimes$ denotes convolution and $w_L(n)$ represents the smoothing Kaiser window of length $L + 1$. The Kaiser window is defined by: \[ w[n] = \frac{I_0(\beta \cdot \sqrt{1 - (2n/L - 1)^2})}{I_0(\beta)} \] (3) where $I_0$ is the zero order modified Bessel function. In SNEO, the window length is associated to the lag parameter by $L = 4k + 1$. The parameter $\beta$ of the kaiser window can be adjusted to approximate the most common windowing functions [7], as shown in Table 1. SNEO was computed also for different sampling frequencies: 250, 125, 85, 62 and 50 Hz. Therefore SNEO depends on the $k$, $\beta$ and $f_s$. Finally the shock outcome predictor based on SNEO was obtained by computing SNEO’s median value, as it is customarily done for the amplitude or slope [2]. To benchmark the performance of the new predictor several classical predictors were computed: average Peak-to-Peak amplitude (PPA) in the time domain [2, 8], Median Slope (MdS) in the slope domain, Amplitude Spectrum Analysis (AMSA) [1, 9] and Power Spectrum Analysis (PSA) in the spectral domain. To compute the spectral features, a hamming window and a 2048-point FFT were applied. \| Type of window \| $\beta$ \| \|----------------\|-----------\| \| Rectangular \| 0 \| \| Bartlett \| 1.33 \| \| Hanning \| 3.86 \| \| Hamming \| 4.86 \| \| Blackman \| 7.04 \| Table 1. Values of $\beta$ and window equivalences. ### 2.3. Data Analysis The optimization of the SNEO predictor was performed minimizing the Balanced Error Rate (BER) defined as: \[ \text{BER} = 1 - \frac{1}{2} \cdot (\text{Se} + \text{Sp}) \] (4) where sensitivity, Se, was defined as the percentage of successful shocks correctly classified, and specificity, Sp, the unsuccessful shocks correctly classified. The optimal working point was determined using a Leave one patient out cross validation (LOPCV) scheme for each feature. Results for every predictor were given in terms of Se, Sp, BER, Positive Predictive Value (PPV), Negative Predictive Value (NPV), and the Area under the Curve (AUC), from the Receiving Operating Curve (ROC), analysis. ### 3. Results #### 3.1. Optimization of SNEO To find the optimum working point the SNEO was computed for $k = 1$ up to 15, for $f_s = 250, 125, 85, 62$ and 50 Hz, and for $\beta = 0$ up to 12, increasing $\beta$ in 0.1. Figure 1 summarizes the results obtained for this simulation for a value of $\beta = 1.33$. As shown in the figure reducing $f_s$ and $L$ (or $k$) yields similar results, i.e. a lag of 8 for $f_s = 250$ Hz is equivalent to a lag of 4 for $f_s = 125$ Hz. Figure 2 shows the effect of the shape of the window ($\beta$). For each $\beta$ the BER plotted corresponds to the optimal value of $k$. As shown in the figure differences in BER are small so the choice of window does not affect the accuracy of the feature, and sampling frequencies should be kept above 80 Hz. Figure 3 shows the effect of $k$ in terms of the mean BER (averaged over all values of $\beta$), for $f_s = 250$ Hz. The optimal lag was $k = 8$ for this sampling frequency, but reducing $f_s$ should be accompanied by a proportional reduction of $k$. To analyze the effect of the window shape the analysis of Figure 3 was replicated for different shapes of the window, as shown in Figure 4. Optimal values of $\beta$ were around 9, which confirm the results shown in Figure 2. Finally we explored the possibility of reducing the analysis segment, as shown in Figure 5. The values were obtained for $f_s = 250 \text{ Hz}$, $\beta = 9$ and $k = 8$, and they show that segment lengths as short as 2-s can be used without significant losses in accuracy. ### 3.2. Comparison with classical predictors We compared the AUC for our new classifier (optimal working point) with the classical predictors as described in literature [3]. Results are shown in Table 2 and in Table 3. The SNEO showed the best results with a BER of 0.22, Se of 81.3% and Sp of 74.7% as shown in Table 3. These results correspond to the best results obtained with $k = 8$, $\beta = 9$ and $f_s = 250 \text{ Hz}$, as shown in Figure 4. \| Feature \| Se (Sp=90) \| Sp (Se=90) \| AUC \| \|---------\|------------\|------------\|------\| \| PPA \| 42.1 \| 53.5 \| 0.814\| \| MdS \| 49.5 \| 55.4 \| 0.826\| \| AMSA \| 53.3 \| 53.2 \| 0.814\| \| PSA \| 43.9 \| 52.6 \| 0.816\| \| SNEO \| 37.4 \| 55.1 \| 0.808\| Table 2. ROC analysis of the shock outcome prediction features, using 5-s segments, in terms of Se, Sp and AUC \| Feature \| Se \| Sp \| PPV \| NPV \| BER \| \|---------\|------\|------\|------\|------\|------\| \| PPA \| 78.5 \| 72.8 \| 49.7 \| 90.8 \| 0.244\| \| MdS \| 75.7 \| 75.3 \| 51.3 \| 90.0 \| 0.245\| \| AMSA \| 71.0 \| 76.9 \| 51.4 \| 88.6 \| 0.260\| \| PSA \| 73.8 \| 74.4 \| 49.7 \| 89.2 \| 0.259\| \| SNEO \| 81.3 \| 74.7 \| 52.4 \| 92.1 \| 0.220\| Table 3. Analysis of optimal working point of the shock outcome prediction features using 5-s segments in terms of Se, Sp, PPV, NPV and BER 4. Conclusions We introduced the SNEO as a new shock outcome predictor and we have compared its results with four classical VF-waveform features. We have used a database of 419 shocks extracted from OHCA cases treated by the BLS services. The SNEO with $f_s = 250Hz$, $k = 8$ and $\beta = 9$ showed the lowest BER. In addition, for segments as short as 2-s SNEO showed similar Se and Sp values. We conclude that the SNEO could be useful to optimize treatment decisions for OHCA patients. Acknowledgements This work received financial support from the Spanish Ministerio de Economía y Competitividad and FEDER through the projects TEC2012-31928 and TEC2015-64678-R, and from the UPV/EHU through the grant PIF15/190 and through its research unit UFI11/16. References [1] Ristagno G, Li Y, Fumagalli F, Finzi A, Quan W. Amplitude spectrum area to guide resuscitation-a retrospective analysis during out-of-hospital cardiopulmonary resuscitation in 609 patients with ventricular fibrillation cardiac arrest. Resuscitation Dec 2013;84(12):1697–1703. [2] Firoozabadi R, Nakagawa M, Helfenbein ED, Babaeizadeh S. Predicting defibrillation success in sudden cardiac arrest patients. J Electrocardiol 2013;46(6):473–479. [3] Chicote B, Irusta U, Aramendi E, Alonso D, Jover C, Corcuera C. Sample entropy as a shock outcome predictor during basic life support. Shock 2015;1:1. [4] Alam MJ, Kenny P, O'Shaughnessy D. Smoothed nonlinear energy operator-based amplitude modulation features for robust speech recognition. In Advances in Nonlinear Speech Processing. Springer, 2013. pp. 161–175. [5] Kaiser JF. Some useful properties of teager's energy operators. In Acoustics, Speech, and Signal Processing, 1993. ICASSP-93., 1993 IEEE International Conference on, volume 3. IEEE, 1993. 149–152. [6] Maragos P, Potamianos A. Higher order differential energy operators. Signal Processing Letters IEEE 1995; 2(8):152–154. [7] Oppenheim AV, Schafer RW, Buck JR, et al. Discrete-time signal processing, volume 2. Prentice hall Englewood Cliffs, NJ, 1989. [8] Neurauter A, Effstotl T, Kramer-Johansen J, Abella BS, Sunde K, Wenzel V, Lindner KH, Eilevstjnn J, Myklebust H, Steen PA, Strohmenger HU. Prediction of countershock success using single features from multiple ventricular fibrillation frequency bands and feature combinations using neural networks. Resuscitation May 2007;73(2):253–263. [9] Povoa HP, Weil MH, Tang W, Bisera J, Klouche K, Barbatsis A. Predicting the success of defibrillation by electrocardiographic analysis. Resuscitation Apr 2002; 53(1):77–82. Address for correspondence: Name: Beatriz Chicote Full postal address: Alda Urquijo s/n, 48013,Bilbao, Spain E-mail address: firstname.lastname@example.org
Patients Crossing Our Borders: An Ethical or Economic Conundrum? Scott R. Petersen, MD I would like to thank the Members of the Western Trauma Association for the opportunity to serve as your President for the past year. There are many reasons why I enjoy being part of this organization. The cultivation of lasting friendships, the camaraderie and collegiality of the scientific meetings and the family atmosphere that has been created make the Western Trauma Association one of the best kept secrets in American surgery. The memories that I have of our annual meetings often make me reflect on the priorities that we set for our careers. Skiing with my children throughout the West and Canada, having my sons be on a first name basis with leaders in the field of trauma and watching them take off with the “pack” not caring whether I was present or not was sometimes painful, but always gratifying. The title of my presidential address came from a communication I had with one of my former hospital administrators. He approached me about a critically ill trauma patient that I was caring for in the intensive care unit, saying, “Dr. Petersen, what are we going to do about the ethical conundrum concerning Case JC?” He did not say, “Mr. JC” or “your patient JC,” just “Case JC.” CASE PRESENTATION The patient was a 37-year-old Guatemalan National, from the town of San Marcos, an area in the Northern part of Guatemala that borders Mexico. To support his family, JC made the decision to leave Guatemala and come to El Otro Lado (the other side). He paid a smuggler, or “Coyote,” as they are termed in the Southwest, approximately one thousand dollars to transport him through Mexico and across the Rio Grande River into America. He described crossing the Rio Grande into Texas as a harrowing experience because he did not know how to swim and his “guides” abandoned him and his other migrant associates at the water’s edge. Three days later, after crossing the desert from Texas into New Mexico, they were picked up and taken to Phoenix. When he arrived, he did not even speak Spanish. He spoke a Guatemalan Indian dialect that had been handed down from his Mayan ancestors centuries earlier. JC contacted a cousin in Phoenix and was soon incorporated into the local economy of the day worker. JC would get picked up every day from a street corner, work hard, six or seven days a week for minimum wage (approximately $5 per hour), and was paid in cash. His employers paid no withholding or FICA taxes on his wages and no health insurance was provided. Most of the money he made was sent back to his wife and daughter in Guatemala. One evening, JC was struck by a pickup truck while riding his bicycle; a victim of a “hit and run accident.” According to witnesses, he was dragged underneath the vehicle for approximately one-half mile before being thrown free. He was transported to our trauma center “in extremis” with a serious closed head injury, fourth-degree abrasions to the skull, right shoulder, right hip and flank (approximately 15% total body surface area), a pelvic fracture, and bilateral ligamentous knee injuries. In addition, JC had eviscerated through his flank wound, and his entire small bowel and ascending colon were lying on the gurney beside him. He was intubated, resuscitated, and taken to the operating room where he underwent resection of all but 75 cm of jejunum with an end jejunostomy and a right hemicolecotomy with a mid-transverse colon mucous fistula. Over the course of the next six weeks, he underwent multiple debridements, skin grafting procedures, free flaps, and survived an episode of fungal sepsis. JC had no health care insurance and did not meet eligibility requirements for state Medicaid. Our social workers and case managers began working on arrangements for his eventual transfer to Guatemala for long-term care. This was the situation when our Chief Medical Officer approached me with the “ethical conundrum” question. He proceeded to tell me about MY patient—ventilator dependent, six inches of small intestine with the duodenum attached to his rectum. It was clear that he had not seen the patient, reviewed the chart or talked to any of the other physicians caring for him. I informed him that his facts were incorrect, and explained to him the patient’s current condition, the care that he had received, and the fact that it was a miracle that the patient was alive. During the conversation, the CMO, a physician, did not suggest that we should not have been as aggressive with the patient as we had been, but he implied it. He specifically asked me what I would do if the patient became septic, hypotensive and required vasopressors to maintain his blood pressure. I replied that we would proceed with a “full court press,” as I would treat any salvageable patient with a similar condition. The conversation then proceeded as to our treatment plans and our long-term goals for the patient’s care. I informed him, in some detail, of the plans for the eventual transfer of the patient to Roosevelt Hospital in Guatemala City, of the superb job our social work team, nursing staff, operating room personnel and my colleagues (plastic surgery, and infectious disease) had done to get the patient to his present state of health. “With proper care,” I told him, “JC will survive, regain complete gastrointestinal functionality and return to his prior life. To suggest any less of an effort, at our hospital, was unconscionable.” The problem, or conundrum, was not an ethical question, but purely an economic one. ILLEGAL IMMIGRANTS AND HEALTH CARE The United States is in the midst of the largest wave of immigration ever experienced.$^1$ Recent data from the U.S. Immigrations and Customs Enforcement Agency indicate that each day more than 8,200 immigrants enter the country. According to 2000 Census Bureau statistics, there are 30.7 million immigrants residing in the United States (11% of the population). Approximately 70% of these individuals are non-citizens (legal or illegal).$^2$ It is estimated that between 7–8 million immigrants are in the United States illegally; 30–40% of these people entered the country legally but overstayed their visas.$^3$ Figure 1 demonstrates the countries of origin of illegal migrants. Mexico and Central America account for over 72.8% of this migration.$^4$ The states with the largest population of illegal immigrants are shown in Figure 2.$^4$ California has the largest proportion of undocumented individuals with approximately 2.2 million or 31.5% of the total. The number of illegal immigrants in the U.S. has nearly doubled in the last 10 years from 3.6 million in 1990 to 7.0 million in 2000.$^3$ There is conflicting data concerning the cost of immigration to the economy. The Social Security Administration estimates that illegal workers paid over $20 billion in Social Security Taxes between 1990–1998.$^4$ A report from the National Research Council concluded that low-income immigrants contribute over $10 billion per year to the economy, but also use more government services.$^5$ Nonetheless, most immigrants and their descendents will pay $80,000 more in taxes than they will use in government services over their lifetime.$^6$ Nationwide, 31.6% of all immigrants have no health insurance compared with 11.4% of the U.S. born population. Low-income immigrants are twice as likely to lack insurance as low-income U.S. citizens. For 9.8 million low-income non-citizens in 1999, almost 59% had no health insurance compared with 30% of low-income citizens. In addition, citizens had nearly double the Medicaid coverage (Fig. 3). The likelihood of non-citizen immigrants obtaining health insurance depends upon a number of factors. Lack of education, specifically the failure to finish high school, nearly doubles the risk of being uninsured in non-citizens. Salary is also an important predictor of having employer-based health insurance. Fifty-three percent of immigrants who earn less than $25,000 per year are uninsured as opposed to 24.4% among those earning over $60,000 per year. Moreover, the uninsured rate for immigrants who have resided in the U.S. for less than five years is 48%, compared with 29% for those in the country for more than 15 years. Finally, 44.6% of non-citizen immigrant children under 18 are uninsured versus 17.6% of naturalized citizens and 14.4% of U.S. born. The type of insurance varies among immigrants. In 1998, 39% of non-citizens were uninsured, only 13% were covered under Medicaid; 42% had employer based coverage compared with 81% of U.S. citizens and 75.6% of naturalized citizens. Immigrants who lack health insurance often have no usual source of health care. They depend on public clinics, community health centers, and hospital outpatient departments—the so-called, health care “safety net.” Thirty-seven percent of non-citizens with incomes below the federal poverty level have no usual source of care, as opposed to 19% for U.S. born individuals. Interestingly, non-citizens are more likely to use clinics and less likely to go to physicians offices or emergency rooms. In addition to lack of insurance coverage, many immigrants experience additional barriers to obtaining care. The high cost of health care causes immigrants to forego or postpone care. Cultural differences, specifically language barriers, hinder accurate communication and create uncertainty concerning diagnosis and treatment. Lack of understanding of public programs or fear that seeking help from a government agency may lead to deportation, are other factors that limit health care access in non-citizens. GOVERNMENT FINANCED HEALTH CARE FOR IMMIGRANTS Immigration status is the most important reason whether or not low-income workers are eligible for government health benefits. Because of the 1996 Welfare Reform Act, even new legal immigrants have to wait 5 years before they are eligible for health care coverage. On the other hand, naturalized immigrants are eligible for the same benefits as U.S. born citizens. The impact of this policy has been to shift the fiscal burden of health care for legal and illegal immigrants from the federal government to the states and local communities. Fortunately, or unfortunately, depending on your point of view, the Emergency Medical Treatment and Labor Act of 1987 (EMTALA) mandates that hospitals must provide emergency care to all patients regardless of their immigration status or ability to pay. This further burdens states, and subsequently, hospitals to provide uncompensated care. Illegal and qualified aliens who are injured are always eligible for emergency services under the Federal Emergency Services Program (FES). It is unlikely that harsh enforcement of immigration laws will stop the influx of illegal migration. Thus, the population of undocumented individuals from Mexico and other countries will continue to grow and further economically burden our health care system. There are two major ideological differences concerning health care delivery to undocumented aliens. First, there are those who believe that illegal immigrants should not be eligible for any benefits except “safety net” services and those who acknowledge the economic contribution of immigrants and want to provide health care through state and federal funding. Caught in the middle of this ideological struggle between the patients’ health care needs and the economy of payment are the providers (physicians) and hospitals that assume, not only the physical load of providing care, but also the fiscal burden of lack of compensation. TRAUMA AND ILLEGAL IMMIGRANTS There is a paucity of information in the literature with respect to the impact of immigration status and its affect on trauma centers. Several reasons for this lack of data include: reluctance of patients to divulge their immigration status, unwillingness of hospital clerks and staff to question patients regarding citizenship to avoid violating an individual’s civil rights, and, on occasion, presentation of false information or forged documents from patients or their families. St. Joseph’s Hospital and Medical Center in Phoenix is an American College of Surgeons verified Level I Trauma Center. Thirty-six percent of our patients are of Hispanic origin (Table 1). Of those, 31.4% are uninsured (70% of which are undocumented). Although, this group of patients does not account for non-citizens who may have insurance coverage, this group represents a substantial proportion of patients at our trauma center (7.8% overall; 599 patients in FY 2003-2004) (Fig. 4). In addition, the number of uninsured illegal immigrants who present to our trauma center is increasing significantly (2003 = 207 patients, 5.7%; 2004 = 391 patients, 9.8%; $\chi^2 = 46.196; p < 0.001$). The overall uninsured trauma population at our institution is 22.6%. In 2004, our hospital’s charity committee spent $309,000 to facilitate safe discharges for undocumented patients who were not eligible for Medicaid or any follow up services. These expenditures were for transporting patients to their country of origin (primarily Mexico), purchasing mechanical ventilators for long-term care, paying for local home care, arranging for short-term stays in extended care facilities, and buying durable medical equipment or prescription drugs. This is the first year that this data has been consistently collected and no data are available for earlier years. Inpatient costs that were not reimbursed in FY 2004 equaled $11.7 million dollars, an increase of $2.6 million from FY 2003 (28.6%). Of the uncompensated and charity care provided by our hospital in FY 2004, $22 million (68.7%) was associated with the trauma service. The majority of this uncompensated care ($77% - $12.5 million) was provided to undocumented trauma patients. ILLEGAL IMMIGRANTS AND SOCIAL RESPONSIBILITY The professional obligation to assure proper medical care for all patients regardless of immigration status is a tenet that we as physicians must uphold. Many secular faiths espouse an ethic of reciprocity. For example, traditional Judeo-Christian doctrine implores us that “…if a stranger sojourn with thee in your land, ye shall not vex him. But the stranger that dwelleth with you shall be unto you as one born among you, and thou shalt love him as thyself” (Leviticus 19:34). In the New Testament, Christians are taught to treat their neighbors as themselves. The Hippocratic Oath we took as graduating medical students reminds us of our responsibilities to humanity, “I will remember that I remain a member of society, with special obligations to all my fellow human beings, those sound of mind and body as well as the infirm.” Finally, the Fellowship Pledge of the American College of Surgeons states: “…I pledge myself to pursue the practice of surgery with honesty and to place the welfare and rights of my patient above all else. I promise to deal with each patient as I would wish to be dealt with if I were in the patient’s position…” These moral duties recognize the vulnerability of the sick and acknowledge the ongoing responsibility that doctors have to act with beneficence toward patients under their care. Moreover, the ethical duty of physicians not to abandon patients who have no real opportunity to secure another source of care has been upheld by courts as a legal duty, regardless of the patient’s ability to pay for continued necessary treatment. The ethical question of whether societies have a responsibility to provide health care for illegal immigrants sometimes becomes a political issue. In 1996, Congress passed the Illegal Immigration Reform and Immigrant Responsibility --- Table 1 Trauma Ethnicity and Payor Mix (FY 2003-2004) \| Payor \| White \| Hispanic \| Black \| Native American \| Asian, Pacific \| Other \| Total \| \|----------------\|-------\|----------\|-------\|-----------------\|----------------\|-------\|-------\| \| Medicare \| 315 \| 43 \| 23 \| 10 \| 0 \| 15 \| 406 \| \| Medicaid \| 895 \| 1160 \| 206 \| 176 \| 11 \| 227 \| 2675 \| \| Workers Comp \| 138 \| 105 \| 11 \| 4 \| 0 \| 11 \| 269 \| \| Uninsured \| 561 \| 855 \| 82 \| 69 \| 12 \| 104 \| 1683 \| \| Managed Care \| 723 \| 220 \| 40 \| 20 \| 8 \| 54 \| 1063 \| \| Commercial \| 909 \| 236 \| 39 \| 32 \| 9 \| 71 \| 1296 \| \| Military/Champus \| 81 \| 17 \| 12 \| 2 \| 2 \| 4 \| 118 \| \| Other \| 17 \| 90 \| 1 \| 5 \| 0 \| 10 \| 123 \| \| Total (%) \| 3639 (47.7) \| 2726 (35.7) \| 414 (5.4) \| 318 (4.2) \| 42 (0.5) \| 496 (6.5) \| 7635 (100) \| --- Fig. 4. Payor Source with White and Hispanic Ethnicity at St. Joseph’s Hospital and Medical Center, Phoenix, FY 2003–2004. Act. This law made all immigrants ineligible for Medicaid, although it did allow the federal government to reimburse states for emergency treatment of these individuals. In 1994, the citizens of California debated Proposition 187, an even more restrictive measure. This initiative proposed to deny publicly funded health care, social services, and education to illegal immigrants. The Proposition was approved by 59% of the voters. It has never been implemented because the courts found problems with its constitutionality, but the debate for and against its provisions remains active.\textsuperscript{11} Despite the limited public budget for health care, the argument that U.S. citizens and legal residents are more deserving of benefits than are illegal aliens is wrong. It frames the issue as a choice between competing “goods.”\textsuperscript{11} It is true that a society cannot have everything; universal health care, public schools, public parks, public services, and low taxes. What is false is that we have to make a choice between basic health care for illegal immigrants and basic health care for our citizens. Many tradeoffs are possible, including an increase in public funding, changes in other entitlements or decreases in defense spending. We have a societal responsibility to provide all members with basic care. Ziv and Lo, in discussing the physician’s responsibility to care for patients in medical need, regardless of nationality, residency, immigration status or ability to pay, state that it is immoral to turn patients away when society has no other provision or institution to provide them with care.\textsuperscript{12} This notion does go beyond professional ethics and has more to do with social justice than the practice of medicine. However, it makes the ethical argument, based on a belief of social responsibility, that illegal immigrants are contributing members to the economic and social life of our nation.\textsuperscript{13} They are workers, parents, students, and part of our community life. During the industrial revolution, children, women, and men without property were also not treated as full citizens. They were vulnerable people, doing undesirable work for which society needed to take responsibility. Such is the case with undocumented workers today. I believe we have a social responsibility, as well as a professional one, to care for them. OUTCOME JC survived his injuries. Our discharge challenge was to meet his long-term needs of nutrition and rehabilitation. Beyond his emergency care, he was not eligible for Medicaid, and other than friends and cousins, he had little social support in the United States. The best concentration of resources (clinical, financial, and social) was in Guatemala, where JC had the social support of his family and the medical treatment necessary for his recovery. We closed his jejunostomy, weaned him from ventilator support, and subsequently flew him to Guatemala. He was admitted to the intensive care unit of Roosevelt Hospital in Guatemala City, his care was assumed by the surgical staff, and he was reunited with his family. I am told that he was eventually weaned from total parenteral nutrition to oral alimentation and has returned to his home in San Marcos. JC’s bill for 83 days in the hospital was $965,354. FES reimbursement equaled $141,859. Our hospital provided resources for his eventual transfer, and established a fund through our charity committee to assure that his parenteral nutritional needs would be met as he transitioned to enteral feedings. His outcome was engineered by our staff based on moral and ethical principles, not purely on economic ones; namely, treat all patients with respect, and as people whose best interests matter. I believe it is our duty to recognize the vulnerability of all injured patients in our care and to continually advocate for the provision of excellent treatment. Although JC was an illegal worker in relationship to our state and federal governments, he should be viewed as a legal member of our community. Our moral responsibility is to assure that trauma care is available to all individuals and that a person’s immigration status, the color of his skin, or the foundation of his religious beliefs does not diminish that responsibility. Indeed, for society, what seems on the surface as an economic issue based on access and payment of health care truly is an ethical problem based on social justice and societal responsibility. REFERENCES 1. Weitzman M, DuPlessis HM, Fisch SI, et al. Health care for children of immigrant families. Pediatrics. 1997;109:153–156. 2. Immigrants’ Health Care: Coverage and Access. The Kaiser Commission on Medicaid and the Uninsured. Washington, D.C. March, 2001. 3. Burkholder B. Magnet Force: Immigration, Health and Social Policy in Arizona. In Arizona Health Futures. February, 2002. 4. U.S. Immigration and Naturalization Service. Illegal Alien Resident Population. Available at: http://www.ins.gov/govt/aboutins/statistics/illegalien/illegal.pdf, Accessed January 5, 2005. 5. The New Americans: Economic, Demographic and Fiscal Effects of Immigration. The National Research Council of the National Academy of Science. Washington, D.C.: National Academic Press; 1997. Available at: http://www.nap.edu/html/newamer. Accessed January 7, 2005. 6. Lillie-Blancoo M, Hudman J. Untangling the web: race / ethnicity, immigration and the nation’s health. Amér J Pub Health. 2001; 91:1736–1737. 7. Carasquillo O, Carasquillo AL, Shea S. Health insurance among immigrants. Amér J Pub Health. 2000;90:917–293. 8. Lasagna L. Hippocratic Oath—Modern Version. Tufts University. Boston. 1964. Available at: http://www.pbs.org/wgbh/nova/doctors/oath_modern.html. Accessed February 5, 2005. 9. American College of Surgeons. Fellowship Pledge. Available at: http://www.facs.org. Accessed February 1, 2005. 10. Hall MA, Ellinor IM, Strouse DH. Health Care Law and Ethics in a Nutshell. 2nd ed. St. Paul, MN: West Group 1999; pp 112–116. 11. Illegal Immigrants, Health Care and Social Responsibility. The Hastings Center Report, January, 2004. 12. Ziv TA, Lo B. Denial of care to illegal immigrants. New Engl J Med. 1995;332:1095–1098. 13. Farmer P. The major infectious disease in the world—To treat or not to treat? New Engl J Med. 2001;345:208–210.
MINISTRY OF EDUCATION AND TRAINING UNIVERSITY OF DANANG This thesis has been completed at the College of Foreign Languages – The University of Danang NGUYỄN UY DŨNG Supervisor: Ngũ Thiện Hùng, Ph.D AN INVESTIGATION INTO STYLISTIC DEVICES IN POLITICAL SPEECHES BY US PRESIDENTS Examiner 1: .................................................................. Examiner 2: .................................................................. Field Study : THE ENGLISH LANGUAGE Code : 60.22.15 M.A. THESIS (SUMMARY) DANANG - 2010 This thesis will be presented to the Examining Committee at the University of Danang on October 30th 2010 This thesis is available found at the library of - College of Foreign Languages, University of Danang. - Information Resources Center, University of Danang. Chapter 1 INTRODUCTION 1.1. RATIONALE The effectiveness of a speech depends on many factors, including the mood of the crowd, the ability of the orator, the situation the crowd confronts, the topic of the speech and the using of words - the using of stylistic devices. The goal of any political speech is persuasion - you want to bring the crowd around to your point of view, whether that means convincing them to vote for you. So the frequent and wide use of stylistic devices is an important characteristic of political speeches which is an effective way to make these speeches more attractive, lively and more persuasive. A stylistic device is an example of the figurative use of words, which produces a particularly rhetorical effect when people use the language creatively in a specified context so all politicians use stylistic devices in political speeches. In the realm of discourse analysis, the typical linguistic features of political speeches have so far been the focus of attention of many linguists and researchers. However, there are not many studies of the language used in political speeches, especially stylistic devices used in political speeches of US Presidents because the majority of people who listen to political speeches pay more attention to the content of what is being said rather than how it is being said. Thus, the audience, though being affected by the style of the speaker’s speech in general, may have little interest in the stylistic analysis of what is being said in particular. Accordingly, they, though are able to evaluate a speaker’s delivery style and mannerisms in general, may pay rare attention to the colorful meaning that each of the stylistic devices can bring in the speech. In order to create favorable conditions for those who want to understand more about political speeches and to identify some of the common characteristics of political speeches and highlight the differences between the stylistic features commonly present in them, I decided to carry out a detailed analysis of these features: An Investigation into Stylistic Devices in Political Speeches by US Presidents. 1.2. AIMS AND OBJECTIVES 1.2.1. Aims: The study is aimed to investigate into Stylistic Devices in political speeches by US Presidents. Its purpose is to provide learners of English with practical information and to raise their awareness of Stylistic Devices in political speeches in English. 1.2.2. Objectives The study attempts to fulfill the following objectives: 1. To investigate the prominent stylistic features of stylistic devices, and the frequencies of occurrence of stylistic devices in political speeches by US Presidents. 2. To highlight the significant contribution of stylistic devices to the success of the political speeches by US Presidents. 3. To suggest some implications for teachers and learners of English, especially for who wishes to write speeches effectively. 1.3. RESEARCH QUESTIONS To fulfill the objectives mentioned above, the study tries to seek the information for the following questions: 1. What are linguistic features of the stylistic devices in political speeches by US Presidents? 2. Which stylistic devices are most frequently and distinctively used in political speeches by US Presidents? 3. What are their artistic or aesthetic values and contributions to the success of the political speeches by US Presidents? 1.4. SCOPE OF THE STUDY This study focuses on the findings of stylistic features of stylistic devices namely alliteration, metaphor, metonymy, rhetorical questions, repetition and personification in one hundred famous political speeches by US Presidents, and the frequencies of occurrence of the stylistic devices in these speeches. 1.5 DEFINITION OF TERMS Stylistic Devices (Rhetoric) is a branch of general linguistic which is regarded as a language science. It deals with the result of the act of communication. [10, p.12] It means that Rhetoric takes into consideration the “output of the act of communication”. The most frequent definition of rhetoric is one defined as the ability to write clearly, correctly and in a manner calculated to interest the readers. [10, p.12] 1.6. SIGNIFICANCE OF THE STUDY To some extent, the research is expected to be a valuable contribution to the teaching and learning of English Stylistics in general and of stylistic devices in political speeches in particular. The findings of the research will help Vietnamese learners aware and achieve the beauty of using language in political speeches. Moreover, it will not only help them get special styles but also provide them with some necessary strategies and techniques in the creative design or in the use of words in writing speeches. 1.7. ORGANIZATION OF THE STUDY The study is organized into five chapters as follows: Chapter 1: Introduction Chapter 2: Review of Literature & Theoretical Background Chapter 3: Methods and Procedures Chapter 4: Findings and Discussion Chapter 5: Conclusions and Implication Chapter 2 LITERATURE REVIEW AND THEORETICAL BACKGROUND 2.1. REVIEW OF PRIOR STUDIES So far, there have been a lot of studies on the stylistic devices in political speeches. In 1985, Lecoq made an investigation of several speeches by Roosevelt, Nixon and Reagan with regard to the function of metaphor in political discourse, and Darmon did a research about thematic and stylistic analysis of some inaugural addresses of presidents of the United States in 1990. In 1998, Browning, Marina carried out an analysis of Ronald Reagan’s political epideictic rhetoric as a form of propaganda for the conservative ideology, then in 2005, Wenzlawski had a research about using words at war and the pragmatic rhetoric of war speeches. However, there are not many studies on Stylistic Devices and so far no reported research has been found on analysis of stylistic devices in political speeches. In Vietnam, there are some studies about stylistic devices and political speeches such as: Nguyen Thi Dieu Tram (2005), in “An investigation into lexical stylistic devices in “Vanity Fair” by William Makepeace Thackeray”, investigated into the Metaphor and Irony in the novel Vanity Fair and highlight the artistic merits of the novel as well as Thackeray’s unique style. Phan Thi Uyen Uyen (2006) investigated into some commonly used stylistic devices in advertising language in English and Vietnamese newspapers. This author highlighted the differences and the similarities of using stylistic devices in advertising language in English and Vietnamese newspapers. Especially, there are few studies on investigating stylistic devices in political speeches in the perspectives of Appraisal Theories by Martin, a new model to evaluate the text in terms of dialogistic positioning. I hope that this thesis “An investigation into stylistic devices in political speeches by US Presidents” will contribute a minor part to fulfill the overall picture of this field. 2.2. THEORETICAL BACKGROUND 2.2.1. Stylistic devices 188.8.131.52. Stylistic devices (Rhetoric) 184.108.40.206. Functions of stylistic devices 2.2.2. Repetition 2.2.3. Metaphor 2.2.4 Rhetorical Questions 2.2.6. Metonymy 2.2.7. Alliteration 2.2.8. Political Speeches 2.3. APPRAISAL THEORY 2.3.1. Dialogistic Positioning When it is seemed as “dialogic” in the case that the writer presents himself as imaging how his readers will be reacting at this very point of the text and presents himself as responding to what he believes would be their voiced objections or questions had they been there, in the room with him, engaged in a face-to-face conversation. 2.3.2. Affectation Evaluation by means of the writer/speaker indicates how they are emotionally disposed to the person, thing, happening or state of affairs. For example, “I love jazz”; “This new proposal by the government terrifies me”. Under Affect, we are concerned with emotions, with positive and negative emotional responses and dispositions. 220.127.116.11. Authorial (1st-person) versus non-Authorial (2nd & 3rd person) Affect 18.104.22.168. Non-authorial (2nd and 3rd person) Affect 2.4. SUMMARY Chapter 3 METHODS AND PROCEDURE 3.1 RESEARCH DESIGN 3.2 RESEARCH METHODS AND PROCEDURE 3.3 SAMPLING 100 well-known political speeches by US Presidents has been collected on the Internet. 3.4 INSTRUMENTATION 3.5 DATA COLLECTION 3.6 DATA ANALYSIS The method which was used in this thesis was mainly the qualitative approach. The analysis looked into the figurative meanings that SDs may bring into the interpretation of the message in the speeches. The interpretation will take into consideration the deviation of the intended meaning from the source or literal one. Especially the judgment of the novelty and uniqueness; the creativeness and the freshness of SD used in a certain speech by a certain politician will be considered against what have been used so far. As mentioned above, the analysis of the SDs was done using the model of Appraisal Theory by Martin with a hope to bring a better insight into the interpretation of the effect of SDs used in political speeches. The findings of stylistic devices in political speeches was qualitatively presented from the descriptive analysis of stylistic devices. The data were then quantitatively analyzed and presented in tables to show the frequencies of occurrence of those stylistic devices. As has been set up from the outset of the study, the data analysis has been done concerning the dimensions of the research questions stated in chapter 1. - The linguistic features of the stylistic devices in political speeches by US Presidents. - The stylistic devices are most frequently and distinctively used in political speeches by US Presidents. - Their artistic or aesthetic values and contributions to the success of the political speeches by US Presidents. 3.7. RELIABILITY AND VALIDITY Chapter 4 DISCUSSION OF FINDINGS 4.1 THE LINGUISTIC FEATURES OF THE TYPICAL STYLISTIC DEVICES 4.1.1 Metaphor in Political Speeches by the US Presidents Metaphor was one of the most potent means of creating images in political speeches and it was preferred by the speakers due to its special effects on the audience such as emphasizing, appealing to our imagination and creating a vivid picture in the listeners/readers’ mind. Here are several instances of metaphors: (1) America is a friend of each nation and every man, woman, and child who seeks a future of peace and dignity, and we are ready to lead once more. [100] “America” is the tenor that is compared with “a friend of each nation” – the vehicle. Here an implied comparison is made to convey the message that America is friendly and kind to every nation which seeks a future of peace and dignity and it also makes a positive evaluation of Obama with a friendly attitude which may be the intended effect that the speaker aimed at sending the message to the listener. (2) My fellow citizens, today we celebrate the mystery of American renewal. [54] Metaphorical noun “renewal” makes a picture that something new will be created again and by resorting this image, the president wished to stress that he would have new policies to change and develop the USA in his term, this first sentence in his inaugural address created a persuasive situation and a lively image that the hearers or the readers are sure to be impressed by his intention. Metaphors are not only used with nouns but also with verbs and adjectives as well. A large number of collected instances in the study were found carrying cases of metaphorical verbs and adjectives. Following are some examples of this type of metaphor: (3) We have drifted, and that drifting has eroded our resources, fractured our economy, and shaken our confidence. [54] (4) We will defend ourselves and our future against terror and lawless violence. [44] This metaphorical adjectives “lawless” in (4) has the same meaning with “illegal” but here the presidents wanted to emphasize “lawless” and used them as evident in labeling of actions perceived to be against the interests of the USA. This implication was done by defining the behaviour of those who are opposed to the USA and the world as illegal and evaluating the actions of the speakers’ government as restorative forms of punishment. Apart from the cases mentioned above, our corpus also yields a number of interesting metaphorical collocations which can be used to make the speeches more persuasive, emotional and attractive as follows: (5) …since the preservation of the sacred fire of liberty and the destiny of the republican model of government… [126] In conclusion, metaphor is a significant and common phenomenon of language all over the world. It is also a basic ingredient of successful interpersonal conveying of ideas. In political speeches, it is valuable in satisfying the discourse goals. 4.1.2. Metonymy in Political Speeches by US Presidents (6) And tonight, a few miles from the damaged Pentagon, I have a message for our military… [38] In (6) the word “Pentagon” stands for “the United States Department of Defense”. This word was not used to refer to a shape with five sides as its literal meaning counts, but it directs the reference to something more abstract, basing on the association of related characteristics between the source object (the shape of the Pentagon) and the target object (The United States Department of Defense). - A concrete thing used instead of an abstract notion. In this case the thing becomes a symbol of the notion, as in: (7) This must be more than a fresh start between the Kremlin and the White House. [101] In (7) “the Kremlin” and “the White House” are concrete substitutions for “the President and staff” of the Russia and the USA respectively. - The relations of correspondence of particular parts of the body with particular actions, as in: (8)… man holds in his mortal hands the power to abolish all forms of human poverty and all forms of human life. [83] - The instrument which the doer uses in performing the action instead of the action, as in: (9) We are Americans, determined to defend the frontiers of freedom, by an honorable peace if peace is possible, but by arms if arms are used against us. [84] - The relation of proximity, as in: (10) They are likely to be successful only if both sides reach an agreement which both regard as preferable to the status quo—an agreement in which each side can consider its own situation to be improved. [84] “Both sides, each side” are metonymically associated with the substitutions for the US and its adversary. - The material instead of the action, as in: (11) The men who sat around the table in Paris knew that the time had come when the people were no longer going to consent to live under masters, but were going to live their lives that they chose themselves, to live under such governments as they chose themselves to erect. [97] In conclusion, we found out that there is a mix of classic metonymies and innovative ones which were actually used in political speeches by the US presidents. Some basic metonymies like “White House”, “Washington” or “Wall Street”, though losing their freshness and are no longer unpredictable, still contributes to the attraction of the speeches in as much as they show a property or an essential quality of the concept. Newly-used metonymies mentioned above have brought with them an interesting novelty serving the listeners or the readers’ interest. ### 4.1.3 Personification in Political Speeches by the US Presidents Let’s take some following examples for the analysis: (12) America is never wholly herself unless she is engaged in high moral principle. [48] (13) …the United States can maintain her interests intact and can secure respect for her just demands. [122] Both presidents personify “America”, “The United States” as a woman because women are generally seen as morally superior and as being innately gentle and good. And certainly a woman who characteristically sacrifices herself to others is expected to need supporting and protecting in return. By aiming at this personified image, the presidents aim to get the listeners’ support the actions and the policies of America. In this study, we found many cases like the instances above about NATION AS A PERSON AND NATION ACTING HUMAN as in: (14) The economic ills we suffer have come upon us over several decades. [107] In addition, personification is also a way of making the abstract ideological issues meaningful and is there a major leadership strategy during times of national crisis in the UAS, as in: (15) All nations should know: America will do what is necessary to ensure our nation’s security. We will be deliberate, yet time is not outside. I will not wait on events, while dangers gather. I will not stand by, as peril draws closer to closer. The United States of America will not permit the world’s most dangerous regimes to threaten us with the world’s most destructive weapons. [40] There is a shift from “American” to “we” to “I” and back to the USA. The speaker emphasized the equivalence that was established between nation, government and leader, which was supposedly designed to create an impression of national unity and a shared sense of the common purpose that is necessary to overcome a national crisis. The boundary between leader and nation was removed so that the views of the leaders was assumed to become the voice of the nation. In conclusion, the persuasive power of language in political speeches can be achieved through personification and the leaders proved that they know how to apply it when they wanted to give the message to the listeners and wished them to share their ideas and to join them to solve the national problems. 4.1.4. Alliteration in Political Speeches by the US Presidents In my corpus, there were a number of instances of alliteration with the repetition of similar sounds, in particular consonant sounds, in close succession, especially at the beginning of successive words. For example, (16)… governments that protect these rights are ultimately more stable, successful and secure. [99] In this example, there is a repetition of the initial consonant “s” which surely imposes sound effects and deep impression on the listeners or the readers. Maybe, the three successive fricatives at the initial position of each word could create an impression of something perfect and stable in its similarity. The audiences who are listening this speech are likely to have something left in their mind or remember the ideas that the speaker want to give deeply. The repeated consonant sounds can be completely successive. For example: (17) America has a clear goal: to disrupt, dismantle and defeat al-Qaida and its allies in Afghanistan and Pakistan. [101] In addition, we also found most of the alliterations that the initial consonant sounds in words were not next to each other. They could draw attention to the phrase and were often used for emphasis, as in: (18) Make no mistake: we do not want to keep our troops in Afghanistan. [99] It is concluded that alliteration is one of the important tools in producing a successful speech. 4.1.5. Rhetorical Questions in Political Speeches by the US Presidents 22.214.171.124 Yes- no question forms (19) Will we be one nation, one people, with one common destiny, or not? Will we all come together, or come apart? [55] Here, Clinton used the first plural personal pronoun “we” in these questions to link the listeners and him and share sense of the common purpose that is necessary to decide. Though the question was put in the form of an alternative one with two options, it is obvious that he wanted the listeners to agree with his ideas and his policies that he planed. The personal pronoun “we” suggested an idea of solidarity and somewhat reflected his positive strategy in interaction. 126.96.36.199 Wh- Question form In Wh- question form, we also found that the rhetorical questions were often used to produce an effect on listener’s attitude, emotion and psychology with the aim to get the listeners’ approval and support, as in: (20) Why, then, should we think that collectively, as a nation, we are not bound by that same limitation? (21) why shouldn’t we believe that? [107] Besides, we found some rhetorical questions might be encouraging and motivating. They also had the effect of urging and challenging the prospects to make decision or take part in the action, as in: (22) Who shall live up to the great trust? Who dares fail to try? [128] In addition, some rhetorical questions were found to be used to create attention by giving the compared images so that they might provoke the audience, as in: (23) How many white children have gone uneducated? How many white families have lived in stark poverty? How many white lives have been scarred by fear, because we’ve wasted our energy and our substance to maintain the barriers of hatred and terror? [78] Moreover, the speakers used rhetorical question in political speeches to give a persuasive way that can make the audience join them to act and find the solution for the policies or the problems as in: (24) What are we going to do now? [60] (25) What will we do in 1970 when elementary school enrollment will be 5 million greater than 1960? [79] In conclusion, rhetorical question used in political speeches is the way of the effective persuasion and also creates attention as well as provide the important affairs of the nation in order to get the listeners’ approval and support. 4.1.6. Repetition in Political Speeches by US Presidents For example: (26) That is in Israel’s interest, Palestine’s interest, America’s interest, and the world’s interest. [101] Here, the words “interest” was intentionally repeated. In this speech, the president was discussing about the conflict between Israel and Palestine and he wanted them to live in peace and security and that was the hope of the people all over the world so the words “interest” was repeated four times to highlight its importance and to made a good effect on the hearers/ the readers. 188.8.131.52. Repetition of words When the repeated word (or phrase) comes at the beginning of two or more consecutive sentences, clauses or phrases, we call anaphora as in the following examples: (27) We seek peace. We seek freedom. We seek to enrich the life of man. [78] In addition, some other kinds of repetition of words have also been found in this corpus, for example: (28) It does require, however, our best effort, and our willingness to believe in ourselves and to believe in our capacity to perform great deeds; to believe that together, with God’s help, we can and will resolve the problems which now confront us. [107] 184.108.40.206 Repetition of structures S BEpassive Ved2 CONJ S BEpassive Ved2 (29) The rights of every man are diminished when the rights of one man are threatened. [85] THE MORE S V THE MORE S V (30) The more the plans fail, the more the planners plan. [85] S V ADV+ER; S V ADV+ER (31) Each year the need grows greater; the program grows greater. [108] EVERY N S V, EVERY N S V, EVERY N S V (32) Every decision I have made, every executive action I have taken, every bill I have proposed and signed. [53] S WILL (NOT) V, WILL (NOT) V, WILL (NOT) V (33) They know this brutal dictator will do anything, will use any weapon, will commit any outrage. [49] S BE ADJ, S BE ADJ AND S BE ADJ (34) Our faith is sure, our resolve is firm, and our union is strong. [49] S SEE NO EVEL AND HEAR NO EVIL (35) Thy see no evil and hear no evil. [116] S WHO V, WHO V, AND WHO V (36) I promised you a President who is not isolated from the people, who feels your pain, and who shares your dreams, and who draws his strength and his wisdom from you. [51] NOT HOW W, BUT HOW X; NOT ONLY HOW Y BUT HOW Z (37) The Great Society asks not how much, but how good; not only how to create wealth but how to use it. [81] In brief, repetition is one of the stylistic devices employed most in political speeches in order to help the listeners or the readers to remember the idea easily. This is also the purpose of the speakers in creating distinctive features for his speeches. In addition, repetition is also a major rhetorical strategy for producing emphasis, amplification, or emotional effect so it is an effective way to attract readers. 4.2. THE DISTINCTIVE FEATURES OF STYLISTIC DEVICES IN POLITICAL SPEECHES IN VIEW OF APPRAISAL THEORIES 4.2.1 The Lexical Salient Features After analyzing the stylistic devices above, we found that metaphor was used most, and then repetition device was the second. Both stylistic devices are valuable in satisfying the discourse goals of political speeches and have their important contributions to the success of creating persuasive, inspirational and informative political speeches. In the total corpus size is 428,465 words with 1993 metaphors, we found that Bill Clinton used approximately one metaphor every 160 words while the other presidents used about one metaphor every 200 words average so we found that Bill Clinton was the president who used the most metaphors in his speeches and this could make him become one of the most famous leaders with many great speeches which are highlighted forever in history in the world. Let take an example of excellent using metaphor in his speech: (38) My fellow citizens, today we celebrate the mystery of American renewal. [54] In addiction, in this study we also found that George Bush was the president who used the most metaphor in his speeches next to Clinton, so this makes metaphor the most lexical salience device with a lot of distinctive features that have important contributions to creating many great speeches. Especially, Bush used a lot of metaphors in the speeches which had the contents about the attack on Iraq and called action against the terrorism and this provided an illustration of the famous events of the day during the era of George Bush. Here are some instances: (39) We will defend ourselves and our future against terror and lawless violence. [44] (40)...at President Shevardnadze’s request, the United States is planning to send up to 150 military trainers to prepare Georgian soldiers to reestablish control in this lawless region. [45] (41) This nation, in world war and in Cold war, has never permitted the brutal and lawless to set history’s course. [46] Moreover, we also found the lexical distinctive features of metaphor, which were the metaphorical personification cases that created rhetorical strategies to convince the audience and help the listeners to visualize what is meant by a phrase or expression, as in: (42) To those neighbours who share our freedom, we will strengthen our historic ties and assure them of our support and firm commitment. We will match loyalty with loyalty. We will strive for mutually beneficial relations. We will not use our friendship to impose on their sovereignty, for our own sovereignty is not for sale. [107] In conclusion, the analysis has showed that some lexical salient features of stylistic devices, especially metaphors are powerful, valuable in satisfying the discourse goals, highlighting the current political issues. Moreover, political speeches are not primarily politics itself, but are a way to present current issues in a way so that the people realize the problems and the visions of the politicians. 4.2.2 The Effect in Signaling Affection As we have discussed the Affect in Appraisal Theories in Chapter 2, the Affect is the evaluation by means of the writer/speaker indicating how they are emotionally disposed to the person, thing, happening or state of affairs. Under Affect, we are concerned with emotions, with positive and negative emotional responses and dispositions. Let’s take some instances with distinctive features of stylistic devices in political speeches in view of Appraisal Theories: (43) Let us put aside the personal advantage so that we can feel the pain and see the promise of America. [54] It is a negative case through the attributive relation of Affect “we feel the pain”. Clinton might not be pleased with what was happening and is happening in the America. So he seemed to call for and encourage the people of the nation to make the USA better and better. He might aim for the people to first and foremost acknowledge the needs and prospects of America. He seemed to seek support and appeals to the people to look beyond their own needs and to see the bigger pictures. He might want to appeal to the emotions of responsibility and support of the listeners. (44) The leadership of America would be in good hands. [97] In (44) it is a kind of metonymy, “hands” stand for persons. It is evaluated positively as he uses “good” to describe the persons who are the leaders. He wants to inspire the listeners with the new hope, belief and confidence in the new leaders of America. (45) America, at its best, is compassionate. In the quiet of the America conscience, we know that deep persistent poverty is unworthy of our nation’s promise. [36] It is a case of indetermination, we cannot decide it is positive or negative because Bush claims that America is a compassionate nation, however, he may suggest that America is not at its best and is therefore currently not compassionate. In times of need and desperation, the nation may not be at all compassionate and think of others. He implied that America is a humble nation that looks after others and that America is decent and respectable. That makes the effect of the emotional images in the listeners about a compassionate nation- America. In conclusion, using the Affect is the evaluation by means of the US presidents as writer/speaker indicating how they are emotionally disposed to the person, thing, happening or state of affairs through the stylistic devices in order to highlight the emotion of the speakers and they help us or the American people as listeners/readers understand and share the ideas with the speakers more easily. And we can clearly identify that the US presidents invite their audience to share that emotional response, or at least to see that response as appropriate and well motivated, or at least as understandable. When that invitation is accepted, then, solidarity or sympathy between speaker and listener will be enhanced. Chapter 5 CONCLUSIONS AND IMPLICATIONS 5.1. CONCLUSIONS By the combination of qualitative and quantitative methods, we draw some conclusions as followed: Metaphor, metonymy, personification, alliteration, rhetorical question, repetition are the most frequently used stylistic devices in political speeches by US presidents. Among them, metaphor ranks first, which is used the most in 100 speeches. Generally speaking, metaphor carries with it various rhetorical effects and strategies needed by the presidents for the purpose of persuasive and interesting speeches. Moreover we’ve found the distinctive features of metaphors in signaling affection and they help us identify the aesthetic in speeches and understand the speakers’ emotion. Each stylistic device has its own features and rhetorical effects so that it is the unique. Metaphor is preferred by the speakers thanks to its special effects on the audiences such as persuading emphasizing, appealing to our imagination and creating a vivid picture in the listeners/ readers’ mind. On the other hand, metonymy is actually able to create visual effects striking the eyes of readers. Personification and repetition are regarded the best way to express emotional attitude of the speakers or to emphasize his main messages. Rhetorical question is employed in speeches for the purpose of imposing a sense of support and approval on the audience while alliteration is aimed to produce sound and music effects. Another important in the research is the stylistic devices mentioned in the thesis closely linked with the view of Appraisal Theories to get the effects in signaling affection. They help us to identify the speakers’ emotion and his rhetorical strategies about attracting attention, arousing interest, stimulating desire, creating conviction and getting action in the audience. 5.2. IMPLICATIONS 5.2.1. For Learning of English Generally speaking, stylistic devices are one of the most difficult aspects of any languages. Therefore, for students of English, how to master the knowledge about stylistic devices and then to apply them in speaking or writing is indeed a big question. From the findings of this paper, it is hoped that students would grasp some necessary information helping them with the study of stylistic devices. Moreover, thanks to the rhetorical effects of stylistic devices identified in this study, students of English would find it more interesting to learn about stylistic devices. Besides, the findings of the study will help them appreciate the beauty of political language in famous speeches. It will not only help them get better understanding language in political speeches but also provide them with some necessary strategies and techniques in writing speeches. I also hope that this study will provide learners with precious experience of how to write and translate a political speech. 5.2.2. For Teaching of English How to find out the most effective method for teachers of English to help their students master stylistic devices is still a big concern at the moment. This study would give a suggestion on teaching stylistic devices through sentences in political speeches containing those stylistic devices. By doing in that way, teachers not only inspire their students with the help of interesting ideas in English political speeches but also bring the students clear examples of how good to use stylistic devices. 5.3. LIMITATIONS Due to the lack of time, space as well as relevant materials, the topic under study may not have been thoroughly discussed as it should be. Stylistic devices are complicated and abstract problems of stylistics. There are a lot of stylistic devices used in political speeches. However, this thesis only investigates some commonly used ones in political speeches. With those difficulties and limitation of personal ability, sources of materials relating to the problem under investigation and outside factors, some weaknesses are inevitable. 5.4. RECOMMENDATIONS Firstly, as we found out in the graduation paper, metaphor is the most frequently used stylistic device in English newspaper headlines. Therefore, it is really useful to carry out a research with the topic “An investigation into the use metaphor in political speeches by the US presidents” in which metaphor will be more thoroughly analyzed and categorized. Secondly, a comparison between stylistic devices in English political speeches and those in Vietnamese political speeches is also suggested as a topic for a graduation paper.
AMENDMENTS TO FEDERAL LAW ENFORCEMENT LIMITATIONS 2014 GENERAL SESSION STATE OF UTAH Chief Sponsor: Michael E. Noel Senate Sponsor: David P. Hinkins LONG TITLE General Description: This bill modifies the Public Safety Code regarding the authority of federal, state, and local law enforcement officers. Highlighted Provisions: This bill: - modifies current law regarding state and local law enforcement officers' recognition of law enforcement authority by federal agencies and employees; - defines the exercise of law enforcement authority, including on state land, private land, and federal land; - defines federal employee for the purposes of this bill; - defines proprietary jurisdiction of federally managed land; - describes when state and local law enforcement officers may recognize a federal employee's exercise of law enforcement authority; - describes the scope of law enforcement action as it relates to the federal Assimilative Crimes Act, and proprietary jurisdiction federally managed land; - provides that state and local law enforcement officers may not recognize a federal employee's exercise of law enforcement authority when the exercise is based on a state or local law or ordinance; - authorizes state and local law enforcement to assist a federal agency or employee under specified circumstances; ▶ addresses federal authority on federally managed land regarding violation of a state or local law in the case of an emergency; ▶ prohibits a federal agency's use of state or local law enforcement correctional or communication facilities without consent of the state or local law enforcement agency; ▶ provides procedures, requirements, and duration regarding entering into agreements with federal employees to exercise law enforcement powers regarding state and federal law; ᴴ→ [and] ▶ allows [local law enforcement agencies] county sheriffs to enter into agreements with federal agencies requiring fair compensation for assisting the federal agency; and ᴴ ▶ requires that county sheriffs regularly review the duties and activities of federal agencies that have law enforcement responsibilities and are acting within the jurisdictional area of a county. Money Appropriated in this Bill: None Other Special Clauses: None Utah Code Sections Affected: ᴴ→ [AMENDS: 53-13-106, as last amended by Laws of Utah 2013, First Special Session, Chapter 4] ᴴ ENACTS: 53-13-106.1, Utah Code Annotated 1953 53-13-106.2, Utah Code Annotated 1953 53-13-106.3, Utah Code Annotated 1953 53-13-106.4, Utah Code Annotated 1953 53-13-106.6, Utah Code Annotated 1953 53-13-106.7, Utah Code Annotated 1953 53-13-106.8, Utah Code Annotated 1953 53-13-106.9, Utah Code Annotated 1953 53-13-106.10, Utah Code Annotated 1953 Be it enacted by the Legislature of the state of Utah: Section 1. Section 53-13-106 is amended to read: 53-13-106. Federal officers -- State law enforcement authority. (1) (a) "Federal officer" includes: (i) a special agent of the Federal Bureau of Investigation; (ii) a special agent of the United States Secret Service; (iii) a special agent of the United States Department of Homeland Security, excluding a customs inspector or detention removal officer; (iv) a special agent of the Bureau of Alcohol, Tobacco and Firearms; (v) a special agent of the Drug Enforcement Administration; (vi) a United States marshal, deputy marshal, and special deputy United States marshal; and (vii) a U.S. postal inspector of the United States Postal Inspection Service. (b) (i) Federal officers listed in Subsection (1)(a) have statewide law enforcement authority relating to felony offenses under the laws of this state. This Subsection (1)(b)(i) takes precedence over Subsection (2): (ii) [Federal] Subject to Sections 53-13-106.1 through 53-13-106.10, state and local law enforcement officers are authorized to recognize the exercise of law enforcement authority by federal agencies and federal employees [may exercise law enforcement authority related to misdemeanor and felony offenses under Utah law only as established by an agreement] as defined in Section 53-13-106.1. This Subsection (1)(b)(ii) takes precedence over Subsection (2): (c) The council may designate other federal peace officers, as necessary, if the officers: (i) are persons employed full-time by the United States government as federally recognized law enforcement officers primarily responsible for the investigation and enforcement of the federal laws; (ii) have successfully completed formal law enforcement training offered by an agency of the federal government consisting of not less than 400 hours; and (iii) maintain in-service training in accordance with the standards set forth in Section 53-13-103; (2) Except as otherwise provided under Title 63L, Chapter 1, Federal Jurisdiction, and Title 77, Chapter 9, Uniform Act on Fresh Pursuit, a federal officer may exercise state law. H.B. 149 enforcement authority only if: (a) the state law enforcement agencies and county sheriffs with jurisdiction enter into an agreement with the federal agency to be given authority; and (b) except as provided in Subsection (3), each federal officer employed by the federal agency meets the waiver requirements set forth in Section 53-6-206. (3) A federal officer working as such in the state on or before July 1, 1995, may exercise state law enforcement authority without meeting the waiver requirement. (4) At any time, consistent with any contract with a federal agency, a state or local law enforcement authority may withdraw state law enforcement authority from any individual federal officer by sending written notice to the federal agency and to the division. (5) The authority of a federal officer under this section is limited to the jurisdiction of the authorizing state or local agency, and may be further limited by the state or local agency to enforcing specific statutes, codes, or ordinances.] Section H-[2] 1 H. Section 53-13-106.1 is enacted to read: 53-13-106.1. State and local law enforcement officers and federal employees -- Definitions. As used in this section and in Sections 53-13-106.2 through 53-13-106.10: (1) "Exercise law enforcement authority" and "exercise of law enforcement authority" means: (a) to take any action on private land, state-owned land, or federally managed land, to investigate, stop, serve process, search, arrest, cite, book, or incarcerate a person for a federal, state, or local criminal violation when the action is based on: (i) a federal statute, regulation, or rule; (ii) a state or local statute, ordinance, regulation, or rule; or (iii) a state or local statute, ordinance, regulation, or rule that is being enforced by a federal agency pursuant to the Assimilative Crimes Act, 18 U.S.C. Sec. 13; or (b) to gain access to or use the correctional or communication facilities and equipment of any state or local law enforcement agency. (2) "Federal agency" means a federal agency that manages federally managed land or regulates activities on that land, including: (a) the United States Bureau of Land Management; (b) the United States Forest Service; (c) the National Park Service; (d) the United States Fish and Wildlife Service; (e) the United States Bureau of Reclamation; (f) the United States Environmental Protection Agency; and (g) the United States Army Corps of Engineers. (3) "Federal employee" means an employee or other agent of a federal agency, but does not include: (a) a special agent of the Federal Bureau of Investigation; (b) a special agent of the United States Secret Service; (c) a special agent of the United States Department of Homeland Security, unless the employee is a customs inspector or detention removal officer; (d) a special agent of the Bureau of Alcohol, Tobacco, Firearms, and Explosives; (e) a special agent of the United States Drug Enforcement Administration; (f) a United States marshal, deputy marshal, or special deputy United States marshal; or (g) a United States postal inspector of the United States Postal Inspection Service. (4) "Federally managed land" means land managed by the following federal agencies: (a) the United States Bureau of Land Management; (b) the United States Forest Service; (c) the National Park Service; (d) the United States Fish and Wildlife Service; and (e) the United States Bureau of Reclamation. (5) "Proprietary jurisdiction federally managed land" means all federally managed land as defined in this section except: (a) buildings, installations, and other structures under the exclusive jurisdiction of the Congress of the United States pursuant to the United States Constitution, Article I, Section 8, Clause 17; and (b) parcels that constitute federal enclaves subject to the concurrent jurisdiction of the United States and the state of Utah. Section H→[3]2←H. Section 53-13-106.2 is enacted to read: 53-13-106.2. State and local law enforcement officers and federal employees -- Exercise of federal law enforcement authority when based on a federal enactment. Subject to Sections 53-13-106.6 and 53-13-106.7, and Subsection 53-13-106.9 (1): (1) State and local law enforcement officers are authorized to may recognize a federal employee's exercise of law enforcement authority, either on or off federally managed land, only when the exercise is consistent with the Constitution of the United States and based on: (a) a federal statute other than the Assimilative Crimes Act, 18 U.S.C. Sec. 13; or (b) a federal regulation that is authorized by a federal statute other than the Assimilative Crimes Act, 18 U.S.C. Sec. 13. (2) Notwithstanding Subsection 53-13-106.2(1), state and local law enforcement officers are authorized to may recognize a federal employee's exercise of law enforcement authority, on federally managed land other than proprietary jurisdiction federally managed land, only when the exercise is consistent with the Constitution of the United States and based on: (a) a federal statute, including the Assimilative Crimes Act, 18 U.S.C. Sec. 13; or (b) a federal regulation that is authorized by a federal statute including the Assimilative Crimes Act, 18 U.S.C. Sec. 13. Section 4(3). Section 53-13-106.3 is enacted to read: 53-13-106.3. State and local law enforcement officers and federal employees -- Exercise of federal law enforcement authority when based on a state or local enactment. Subject to Section 53-13-106.7 and Subsection 53-13-106.9 (1), state and local law enforcement officers are not authorized to recognize a federal employee's exercise of law enforcement authority, either on or off federally managed land, when the exercise is based on a state or local statute, ordinance, regulation, or rule. Section 5(4). Section 53-13-106.4 is enacted to read: 53-13-106.4. State and local county sheriff law enforcement officers and federal employees -- Enforcement of federal laws and regulations by state and local county sheriff officers. A state or local law enforcement agency or a county sheriff is authorized to may assist a federal agency or federal employee to enforce federal statutes and regulations on lands managed pursuant to 43 U.S.C. Secs. 1701-1736 and Secs. 1737-1782, Federal Land Policy Management Act, (only) after the United States secretary of the interior has achieved maximum feasible reliance upon the state or local law enforcement officials in enforcing federal laws and regulations, as (2) the United States secretary of the interior has otherwise complied with the requirement of 43 U.S.C. Sec. 1733(c)(1); and (3) the state or local law enforcement agency or a county sheriff has complied with an agreement authorized by Section Subsection 53-13-106.9 (3). Section 6. Section 53-13-106.6 is enacted to read: 53-13-106.6. State and local law enforcement officers and federal employees -- Exercise of federal law enforcement authority to enforce the Federal Land Policy Management Act. Notwithstanding Section 53-13-106.2, state and local law enforcement officers are authorized to recognize a federal employee's exercise of law enforcement authority to enforce the provisions of the Federal Land Policy Management Act on proprietary jurisdiction federally managed land, only if such the exercise is consistent with the Constitution of the United States and based on: (1) a federal statute other than the Assimilative Crimes Act, 18 U.S.C. Sec. 13; or (2) a federal regulation that is: (a) authorized by a federal statute other than the Assimilative Crimes Act, 18 U.S.C. Sec. 13; and (b) necessary to implement the provisions of the Federal Land Policy Management Act with respect to the management, use, and protection of the public lands, including the property located on those lands, as provided in 43 U.S.C. Sec. 1733(a). Section 7. Section 53-13-106.7 is enacted to read: 53-13-106.7. State and local law enforcement officers and federal employees -- Exercise of federal law enforcement authority based on state law during emergency. Notwithstanding Section 53-13-106.3, state and local law enforcement officers are authorized to recognize a federal employee's limited exercise of law enforcement authority on federally managed land in cases of a violation of a state or local statute, ordinance, regulation, or rule when: (1) the offense is an emergency and poses an immediate risk of bodily injury or damage to property; (2) a state, county, or municipal law enforcement officer is not reasonably available to take action; (3) the action is within the scope of the employee's or official's law enforcement power; and (4) the federal employee turns the matter, as well as the custody of any detained citizen, over to the state, county, or municipal law enforcement officer for further action as soon as the officer becomes available. Section H→ [8] 7 ←H . Section 53-13-106.8 is enacted to read: 53-13-106.8. State and local law enforcement officers and federal employees -- Use of correctional and communication facilities. State and local government agencies may not allow any federal agency access to or use of the correctional and communication facilities and equipment of any state or local law enforcement agency without the express written consent of the appropriate responsible official of the state or local law enforcement agency. Section H→ [9] 8 ←H . Section 53-13-106.9 is enacted to read: 53-13-106.9. State and H→ [local-law] county sheriff law ←H enforcement officers and federal employees -- Interagency agreements. Notwithstanding Section 53-13-106.3: (1) H→ [Local-law enforcement agencies] County sheriffs ←H may enter into agreements with federal agencies granting limited authority to specific federal employees to exercise law enforcement powers to enforce federal state and local laws, provided the agreements are limited to a term not to exceed two years and the officers granted authority have completed a 20-hour course focusing on Utah law and process approved by the director of the Peace Officer Standards and Training Division. (2) State law enforcement agencies may, with the consent of the local county sheriff, enter into agreements as described in Subsection (1), provided that the agreements may not exceed a duration of two years. (3) H→ [Local-law-enforcement agencies] Local county sheriffs ←H may enter into agreements with federal agencies requiring fair compensation for assisting a federal agency or federal employee to enforce federal statutes and regulations managed pursuant to 43 U.S.C. Secs. 1701-1736 and 43 U.S.C. Secs. 1737-1782, Federal Land Policy Management Act. Section H→ [10] 9 ←H . Section 53-13-106.10 is enacted to read: 53-13-106.10. State and local law enforcement officers and federal employees -- Review by county sheriffs. County sheriffs shall regularly review the duties and activities of federal agencies that have law enforcement responsibilities and that are acting within the jurisdictional area of the county to determine if the federal agencies are acting consistently with this section. Legislative Review Note as of 2-18-14 5:04 PM Office of Legislative Research and General Counsel
The First 50 Years A Social Experiment Sol Tax, Paul Fejos, and the Origins of Current Anthropology Sydel Silverman (E-mail: firstname.lastname@example.org) 1 IX 09 Current Anthropology was born out of a meeting of minds of two remarkable individuals. Sol Tax and Paul Fejos were very different personalities, almost polar opposites in their biographies and temperaments, yet they had in common unusual daring, an expansiveness of vision that acknowledged few limits, and a belief in anthropology that went beyond its scholarly value. Paul Fejos, born in Hungary to once-landed gentry, was cosmopolitan, multilingual, multitalented. During World War I he rode with the hussars and flew as a pilot with the air cavalry. He completed medical school, then became a theatrical and film director, first in Europe and later, to considerable acclaim, in Hollywood (a review of his first movie was headlined “Introducing You to Mr. Paul Fejos, Genius”). He came to anthropology by chance, through ventures as an explorer, ethnographic filmmaker, and self-taught archaeologist and ethnologist. After saving the life of Axel Wenner-Gren on a tiger hunt, Fejos persuaded him to establish the Viking Fund in 1941, which 10 years later was renamed for its benefactor. Fejos was a charismatic figure, master of the extravagant gesture, five-times married, whose influence on anthropology through the Wenner-Gren Foundation (WGF) was guided as much by his own instinct as by the advice of the professionals who counseled him. Sol Tax was born in Milwaukee, Wisconsin, to Russian Jewish immigrants, secular despite their descent from rabbis, socialist in their political leanings. Tax was small and shy but also entrepreneurial as a boy (Stocking 2000, 173).¹ He discovered anthropology through Ralph Linton at the University of Wisconsin, then completed graduate studies at the University of Chicago, later his long-time academic home. He did extensive fieldwork in Middle America, but his forte was more organizational than theoretical. Tax was earnest, hardworking, personally unprepossessing, a family man. His approach to the world, as summarized by Stocking (2000, 172), was “a mix of liberal democratic enthusiasm, relativist humility, and universalist hubris.” Fejos and Tax met up as members of the National Research Council’s committee on Latin America, and they worked together on a comprehensive state-of-the-art conference on Middle American ethnology in 1949. Tax edited the conference volume as well as three volumes of the concurrent International Congress of Americanists. At this time of postwar expansion of anthropology, Fejos was envisioning a more ambitious role for the WGF, beginning with a massive international symposium in 1952 that he thought of as “a concrete basis on which to erect its future policies” (Fejos 1953, v). Under the presidency of A. L. Kroeber, and with a planning group that brought in eight other major figures, its goal was to inventory the totality of what was known to anthropology at the time. Fifty papers presented at the symposium were published as Anthropology Today: An Encyclopedic Inventory (Kroeber and International Symposium on Anthropology, 1953). After the symposium Tax was brought in to oversee the task of condensing and editing the voluminous transcriptions of the discussions (Tax [and others] 1953). During the 1950s Tax was also developing an “action anthropology” program, serving (with vigor and innovation) a term as editor of the American Anthropologist (1952–1955), and organizing an elaborate celebration of the Darwinian centennial (which included a musical reenactment of Darwin’s life) as well as an equally ambitious event that brought together diverse North American tribes and factions in the American Indian Conference in Chicago in 1961. But it was with the project that became Current Anthropology (CA) that Tax moved firmly into the role of master impresario of world anthropology. His contemporary Margaret Mead was better known to the public for her advocacy of anthropology on a world stage, but she did so mainly as an individual enterprise. Tax, in contrast, was an institution builder, and his efforts were directed primarily at the anthropologists themselves. CA became his vehicle of choice. He did not, however, set aside other activities, and toward the end of his editorship, he also took on the task of hosting the 1973 International Congress of Anthropological and Ethnological Sciences (ICAES) in Chicago. True to form, for him this was not merely a matter of coordinating some meetings; he planned for the Congress to yield some 300 volumes (of which 92 were actually published), and he may be the only conference organizer in the history of anthropology to stage a full-scale opera in honor of the event, commissioning no less a figure than Gian Carlo Menotti for the assignment (Stocking 2000, 211). ¹ In addition to the publications cited, sources for this article were the files on CA and Tax in the Wenner-Gren archives and in-house publications such as annual reports of the WGF. The most valuable source on Paul Fejos was Doidge (1973), itself based largely on Fejos’s own account in the Columbia University Oral History Archives of his life. Many of the events and anecdotes recounted there stretch credulity, but when I once asked Liv Osmundsen which were true and which were exaggerations, she said that the more unbelievable ones are true. What Fejos and Tax shared was a vision of a world of seamless intercultural understanding to which anthropology would lead the way. Fejos, in fact, had persuaded Axel Wenner-Gren to devote his Fund to anthropology by appealing to his ambition to be a mediator of international relations (which had impelled an uninvited visit to Hermann Goering to try to persuade him of the folly of the Nazi military buildup). For both Fejos and Tax, anthropology embodied the most comprehensive knowledge of humankind, and it was a connected enterprise of scholars worldwide. They believed that that connectedness was also a means of addressing the large problems of the day. (In Wenner-Gren’s words, written for him by Fejos, anthropology had the potential to touch on universal issues.) Thus, the venture that became CA was always more than a publishing project; it was part and parcel of a worldview and a mission shared by the dashing Hungarian Renaissance man and the Jewish anthropologist from Milwaukee who had always been eager to do good in the world (Stocking 2000, 173). The two of them were undertaking what they thought of as a social experiment that could change the face of international anthropology and have an impact beyond the discipline. The background to the origin of CA was a shift in priorities within the WGF. In the immediate postwar years it had been among the very few sources of support for research in anthropology, but the creation in 1950 of the U.S. National Science Foundation and National Institutes of Health, both with far more lavish funding, began to leave the WGF behind in the competition for good projects. In response, Fejos decided to focus on “communication” functions. In the first instance, that meant conferences, building on the supper conferences that had been held at the townhouse in New York since it was acquired in 1945; to these were added major gatherings held at universities and other sites and, from 1958 on, the extremely productive summertime program of international symposia at the “castle,” Burg Wartenstein in Austria. The budget for grants was reduced to about a third of the total, and conferences were assigned an equal portion. (The remaining third was divided among several other activities, the most important of which was the casting program begun in 1958; once CA was established, it absorbed only about 5% of the budget, much less than the amount committed for its subsidy, because of the income it generated.) However, another aspect of the WGF’s communications role was to be to develop publications and other mechanisms that would enable anthropologists of all specialties, worldwide, to be in contact with one another and continually to be informed about the state of the field. The “Anthropology Today” symposium had proven a great success, and Fejos hoped to sustain and update the “encyclopedic inventory” it had yielded. He experimented with the idea of a Yearbook of Anthropology and published one volume, in a limited edition, in 1955; parts of it were reproduced for public sale under the title Current Anthropology: A Supplement to “Anthropology Today” (1956, edited by William L. Thomas, Jr., who was on the WGF staff). However, Fejos felt it was impractical for the Foundation to produce further editions in-house, and he was disappointed that some of the authors did not actually write inventories. Moreover, he thought that what was needed was a more rapid and more flexible means of communication. Fejos asked Tax to develop a plan that would more effectively meet these goals, and in April 1957 Tax agreed to take on the project, seeing in it an opportunity “to develop new patterns of worldwide intercommunication in the profession” (Stocking 2000, 204). Tax submitted an initial plan that envisioned a series of five books (to be called Current Anthropology) to be published every second year, which, along with revisions of the original Anthropology Today (AT) articles, would yield a definitive encyclopedia of anthropology. But before he was ready to put a plan into effect, he felt that he needed to discover what the scholars of the world themselves thought would be most useful. With full WGF support, he then embarked on a hectic round of meetings—44 of them between 1957 and 1960, covering more than 30 countries and a dozen venues in the United States, as well as several meetings at Burg Wartenstein and in New York, and involving some 650 individuals. Tax’s journeys took him everywhere except to Australia and the Pacific Islands. He estimated that there were about 3,000 anthropologists all over the world at the time, and he was determined to talk to as many of them as possible. The idea that Tax took with him was that there should be a “participatory democracy of scholars in exchanging information.” True to his conviction, he was bent on soliciting a wide range of suggestions. What he got was a dizzying array of reactions and ideas: participatory democracy meant everyone had his (sometimes her) own notions. It was Tax’s genius that he was able to sort these into coherent plans while seeming to accommodate most opinions, even contradictory ones. (He also seems to have been quite willing to have his own opinions overruled.) In 1957, Tax held seven supper conferences throughout the United States involving more than a hundred anthropologists. In these meetings, and continuing into the early months of 1958, the basic assumption was that CA would be not a journal but a series of yearbooks. As Tax summarized it at this time: “This new editorial venture . . . is intended to keep Anthropology Today up to date, and perform some of the functions in continuation of the Yearbook of Anthropology.” Its articles would review large fields of knowledge, inventory subjects not included in AT, and summarize current knowledge on major anthropological topics and problems. At the first meeting, in Michigan in July 1957, the specific proposal of the group was for a series of four yearbooks to be published every two years, which after eight years might lead to a publication much like AT but broader in scope (essentially the plan that Tax began with). The proposal was accompanied by an “Outline of the Encyclopedia Inventory” (prepared by Tax and Betty Potash), elaborating on the specific articles that might be included. That list of suggested articles grew over the course of the next several meetings until it amounted to more than 80 items. An additional section was to cover “Peoples and Cultures of the World.” Finally, the proposal envisioned a “Field Manual,” a handbook of items that a field-worker might use in data collection, along the lines of Notes and Queries. It was suggested that the manual be printed on microcards for portability. In November of that year, J. S. Slotkin elaborated on this idea and produced a full outline of a handbook, a document that seems to have been consigned to a file cabinet. By the end of 1957, Tax assembled the comments and suggestions of the several meetings into a report that set forth the central aim of CA as that of the updating and completion of AT. Other notions were included under “subsidiary purposes” and “possible side projects,” to be kept to one side but possibly incorporated in some way (these included a range of ideas, such as abstracts, reprintings, and translations of material published elsewhere; book reviews; an inventory of training aids; and opinion pieces). “Peoples of the World” and the “Field Manual” were left as “not conclusively resolved.” That the notion of a journal was still not adopted is clear from an exchange during a February 1958 meeting in Chicago. Melville Herskovits expressed concern that the publication might be “unstructured like a journal.” He continued, “If it’s a journal that’s all right but CA is not a journal.” Tax agreed: “A journal is Kroeker’s idea. This will be a book . . . .”, but, he added, it might continue, and there is “no reason why we need limit the publication to four issues.” In these early meetings, the unanimous point of agreement was that the publication would be international, interdisciplinary, and geared to professionals and graduate students (rather than a wider readership). There was also general support for Tax’s emphasis on review articles, but some added the proviso that these should reflect current trends and “be forward looking in nature.” A third point of wide agreement was that the articles should be contributed by individuals, although there could be room for coverage of conferences. But there was debate over whether these should come from volunteered submissions or solicited (with some proposing that committees representing each field decide whom to invite). As Tax’s year-end summary reveals, these gatherings yielded a plethora of suggestions, each with points of dispute. The idea of a “Peoples of the World” section came with differences of opinion as to whether it should be oriented toward data, theory, methodology, or problems. Some ideas, such as the proposal for a microcard library, went beyond the journal format. As Tax collected specific suggestions for articles, he organized the long list into categories: articles dealing with the particular field as a whole; methods, concepts, and theories; reviews of major blocks of knowledge; substantive results in various areas; interdisciplinary approaches. He sent the list to all fellows of the American Anthropological Association, asking for volunteered submissions and suggested authors, and he received several hundred replies. In the summer of 1958 Tax held seven regional conferences throughout Europe, involving more than 150 participants. He encountered conflicting reactions to the CA idea, both among different countries and within each one. Writing to Fejos in July, he said that in England there was “resentment” that CA might take resources away from WGF’s support of research and of students. The English were in favor of communications within specialties but “positively and loudly opposed to any sort of marriage of these specialties.” The Cambridge group wanted only reference materials—inventory or review articles “left them cold”—while the opposite was true in London, where the interest was only in “middle-range review articles.” In Paris, only Alfred Metraux favored inventory articles; others stressed “classifiable pieces of new knowledge.” According to Levi-Strauss, “the French were tired of philosophy and wanted information for empirical research.” The variety of opinions multiplied as Tax continued his travels. In August of that year, Burg Wartenstein was inaugurated. Its first event (after a day of ceremonies to which all Austrian anthropologists were invited) was a conference on CA, which included 12 participants along with Tax and Fejos. That conference resulted in a specific proposal that established the future contours of CA. At the conference “the idea of an open ended journal rather than a closed encyclopedia crystallized” (Tax 1965, 242) because it was felt that this would provide more rapid dissemination and would allow material to appear in units small enough to be easily handled. The proposal declared three firm desiderata for the journal: (1) that it be as broad and open as the changing sciences of man require, with the widest variety of relevant ideas and data, facilitating communication throughout the world; (2) that it be unitary (rather than divided by fields or specialities), with a single set of cross-cutting materials to be available to all; and (3) that it provide communication that is both fast and convenient. Specifically, the journal would be addressed to the world audience of all the anthropological sciences (“physical anthropology, ethnology, prehistory, folklore, linguistics, social anthropology, and all of the subsidiary and related sciences by whatever names”). It would publish two classes of materials. One would be major reviews of broad scope, comprising several subdisciplines and/or different scholarly traditions; these could be new considerations of traditional subjects, but preferably they would be subjects of new, current, and growing interest. The second kind of content would be current news and reference materials: news of research activities and discoveries, guides to materials (bibliographies of bibliographies, surveys of research and museum collections, etc.), and “Want Ads” (requests for information and research suggestions). The idea of Associates was also formalized, as Tax articulated his vision of a “community of scholars” that would develop through the journal and in turn determine the policy of the journal. Thus, Tax welcomed the shift to a journal model and immediately wedded it to his “strong and optimistic belief in humanity, community, and democracy” (Tax 1965, 242). He now saw CA as a new kind of scholarly institution that should be “permitted to evolve after the fashion of a ‘natural institution,’ like the family or a hunting party, with which anthropologists are so familiar” (1965, 248). With the basic parameters of the journal in place, Tax continued to solicit input and support through further travels in 1959. In January, he held regional conferences in Russia, India, Thailand, the Philippines, Hong Kong, and Japan and others in Hawaii and California. During the spring, he ranged across Latin America, visiting Mexico, Guatemala, Colombia, Peru, Chile, Argentina, Brazil, and Venezuela. The summer took him to Sweden, Poland, Czechoslovakia, Turkey, Egypt, Ethiopia, Uganda, the Sudan, Nigeria, Ghana, Senegal, and England. These meetings yielded numerous ideas and suggestions for changes. In particular, there was much interest in expanding the role of Associates. Throughout this time, Tax was preparing a Pre-Issue, which was mailed in September 1959, and the first full issue (dated January 1960 but mailed in November 1959). The pace of his work is hard to comprehend: by mid-January 1960, the March issue was completed, and by the end of January, the May issue was in press. He was also assembling his corps of Associates. Invitations were extended to some 3,000 anthropologists to become Associates and to participate in the network; Tax hoped to eventually recruit 4,000. To facilitate communications with the Associates, he inaugurated the Reply Letter device in the Pre-Issue, and by the end of January 1960 he had received about 1,000 replies. The philosophy of the journal was laid out in the Pre-Issue. The subtitle summarizes its basic goal, A World Journal of the Sciences of Man, and the cover design (showing the globe and the spectrum of fields) symbolizes the world community of anthropologists. “It is a journal, yes,” writes Tax, but it is more than that. It is the means by which individual scholars can communicate with each other all over the world: to exchange and pool ideas, information, research materials, and new knowledge; review the results of past research; and lay the basis for further intercommunication on current developments. (The language of “world” and “worldwide” rather than “international” was deliberate, reflecting that anthropologists participated as individuals, not through institutions or national bodies.) It intends to represent all of anthropology, to “go from the biological to the social and cultural, and also range from the purely historical to the analytical and to scientific generalizing,” but at the same time to recognize the differences among countries in how the sciences are organized. CA, Tax says, is a social experiment, and all Associates are part of it. The WGF has promised to pay the deficit for five years, after which “we citizens of the world community of scholars” might be able to pay for it. The Foundation was leaving to us (Associates) complete freedom to develop CA in any way that it can serve the sciences of man. The journal was to appear six times a year (soon reduced to five). Each issue would include review articles or other major articles, which would have “CA☆ treatment”; this innovation, a device for organizing the intercommunication among specialists that was a basic goal of CA, was in place from the first issue. There would also be sections for news, reference materials, and information exchange, serving its function as a clearinghouse. The editor’s Letter to Associates and a Reply Letter would be inserted in each issue, inviting Associates to comment on anything and everything from the design to the content to editorial policies and stylistic niceties. But none of this was immutable: as the first issue declares, “CA will eventually publish whatever it is that Associates wish to learn from and to teach one another.” The concept of Associates, who make up the “community,” was fundamental to CA. In return for a special subscription rate, Associates had obligations: (1) to keep the list of Associates updated and to nominate others; (2) to pay their fee (bills or reminders would not be sent); (3) to look through each issue upon receipt and respond to requests “meant for you”; (4) to give the Editor suggestions for material to be included, especially your own proposals; and (5) to respond to the Editor’s request for submissions, comments on articles, and other matters. The fee for Associates was set as the equivalent of the purchasing power of $2 in North America. Individual subscribers recommended by Associates (who were considered Potential Associates) were charged twice that amount, and the institutional rate was five times greater. The fee was payable in local currency. In some cases where currency restrictions prevented sending payment abroad, the accumulated fees were left in the country to be used for CA-related purposes. Informally, Tax described the fees as equivalent to the cost of a good meal in the country. Still, Associates and students in some countries simply could not pay fees, and no effort was ever made to collect them, the costs being absorbed as part of the deficit covered by the WGF. (To encourage paying subscribers, the Foundation added a bonus of two or three volumes in the Viking Fund Publications in Anthropology series each year.) In the Pre-Issue, Tax responded to the concern over whether CA would be competing with other journals. Not really, he said. The clearinghouse function was truly new, and while some of the major articles might be published elsewhere, the CA audience represented the broadest range of anthropologists geographically and in terms of interests. Tax was well aware that language was a thorny question. The 1958 Burg Wartenstein conference, after long discussion, had reluctantly concluded that publication in one language was the only practical alternative and that it had to be English, as the most widely read language. That decision was overwhelmingly endorsed by the Associates. Still, there was concontinuing debate over how other languages should be accommodated. It was decided that initial submissions needed to be made in English, however rough, and that accepted items would be fully translated and edited. The primary concession to the Associates’ linguistic diversity was the publication of the Editor’s statement (about CA principles and policies) in a different language in each issue, rotating among major languages (initially French, Japanese, Russian, Spanish, German, Chinese, and Italian). What the major articles were intended to be was somewhat ambiguous. While Tax had given up the yearbook notion, he continued to use the terms “major review” and “review article,” but the contributors were shifting the meaning of the terms. The first issue’s two major articles conformed to Tax’s original concept, providing synthesis and systematic coverage of topics (“Lexicostatistics So Far” by Dell Hymes and “Twenty Years of Peyote Studies” by Weston LaBarre). But later issues increasingly featured articles emphasizing new developments in the field. The question of what qualifies as a review article or how far the meaning should be stretched continued to be debated by the Associates, with the majority favoring a flexible approach. Tax attempted to accommodate these views with a revised policy statement in 1961. “A ‘Review Article’ [i.e., the long articles with CA treatment, usually two in an issue] is a major survey of knowledge on any topic relevant to the sciences of man. New material at the growing points of anthropology, and new evaluations, are preferred over syntheses of what has become well-known, but the new should be placed in the context of the known.” He also affirmed that there was no length limit “as long as it is consistent with the material it covers, provided that sensible canons of selection are observed.” His working principle was that “an article should be as long as it needs to be for what it says, and not one word longer” (Tax 1965, 255). In 1960, Tax summarized the status and future challenges of CA for the WGF board: (1) It is a “co-operative of scholars,” but that requires “institutionalization” so as not to depend on a particular editor (him). Its goal is “free and democratic participation and communication without bureaucracy,” but other mechanisms than those already set up may be needed. (2) It aspires to be an inventory of knowledge, but that will require a sufficiently large and representative flow of material to allow for selectivity and to ensure a backlog, given the time needed for CA treatment. (3) It is a medium for interchange, but clearer patterns need to be established for the kinds of items to be included, and the mailing needs to be speeded up. (4) Its function of information interchange could be expanded into a more active role in research cooperation. (Tax envisioned CA as coordinating simultaneous inquiry in different parts of the world—facilitating exchanges of ideas for related studies, discussions of research results, and the planning of subsequent stages of work.) Over the next few years, the Reply Letter became for Tax an important means of gathering opinions of the Associates, which to some extent led to changes in CA policy. There were constant suggestions for new departments and projects, many of which were implemented. But there was also complaint and criticism. From Tax’s 1965 reflection on the history of CA, it appears that there were three main areas that were proving problematic. One was the definition and role of Associates, that is, criteria of inclusion, fees, handling failures to fulfill obligations, and means of communication and cooperation among Associates. A second area had to do with the CA treatment. There were questions of how commentators were selected, how many rounds of comments and revisions should be circulated (Tax reluctantly settled on just one round), how criticisms and suggestions were handled by the authors and the editor, and how credit was assigned. Finally, there were the recurrent problems of language. Beyond the choice of English as the sole language of publication, there were issues of usage, idioms, and spelling; translation (or not) of place names and technical terms; politically charged terminology; use of the word “primitive”; ad hominem phrasings; the idea of printing abstracts in different languages (a compromise solution was ultimately reached); complaints that oversimplification was forced on writers not fluent in English; and more. But after experiencing a year of editing articles, Anna Pikelis concluded that the problem was not so much language as “communication”: “By far more potentially troublesome than language are the complex implications of inter-cultural and inter-disciplinary communication” (Tax 1965, 253). Underlying the language issues was the unavoidable American dominance, given a U.S.-based editor and sponsor. There was (and continued to be) a clear overrepresentation of U.S.-Americans among the authors of major articles, although more comments and reports came from outside the United States. U.S.-Americans were the majority of Associates at the beginning, but by the end of the first year they accounted for less than 40% (Tax 1965, 266). Both Tax and Fejos struggled heroically (in a preelectronic age of unreliable mails) to realize their commitment to making the journal truly international, and they achieved at least partial success. They managed to include in the CA community anthropologists from every part of the world, including countries that were politically closed off from the West. Probably the journal had its greatest importance in reaching the more isolated scholars, a role that never diminished until the advent of the Internet. (Fejos’s successor, Lita Osmundsen, made it a point to select the next two editors from outside the United States, albeit from English-speaking countries.) Fejos was enthusiastic about Tax’s efforts, and he reported to the WGF board that CA was providing a “means of constant interchange for the anthropologists of the world.” Both in its achievements and its relatively modest cost (although it never reached Tax’s optimistic goal of becoming self-supporting), CA was an important part of the Foundation’s vibrancy during the 1960s, a decade as yet untouched by the financial crisis that was to come. With the end of Tax’s editorship in 1974, and over the years that followed, several of CA’s innovations were solidified, some even copied by other journals, while others (including most of Tax’s as yet unfulfilled hopes for it) fell by the wayside. Inevitably, CA came to resemble more a conventional, albeit distinctive, journal than the linchpin of an anthropological world citizenry. Still, from our perspective today, we can appreciate that the social experiment engineered by Tax and Fejos succeeded in creating, and to a degree institutionalizing, new modes of communication among anthropologists of all specialties and nationalities. We continue to benefit from the vision of two extraordinary men. Acknowledgments I thank Barbara Metzger, the universally respected copy editor who served under five editors, for reminding me of the many ways in which CA came near to its goal of being truly international. It is worth remembering that Tax had no models to follow. In the mid-1950s he proposed to the ICAES the idea of a world journal that would bring together individual scholars; inquiring at the United Nations about precedents, he was told that all international associations were based on national groups, not individuals (Tax 1965, 241–242). References Cited Dodds, John W. 1973. The several lives of Paul Fejos: a Hungarian-American odyssey. New York: Wenner-Gren Foundation. Fejos, Paul. 1953. Preface. In Anthropology today: an encyclopedic inventory. A. L. Kroeber and International Symposium on Anthropology, eds. Pp. v–vii. Chicago: University of Chicago Press. Kroeber, A. L., and International Symposium on Anthropology, eds. 1953. Anthropology today: an encyclopedic inventory. Chicago: University of Chicago Press. Stocking, George W., Jr. 2000. “Do good, young man”: Sol Tax and the world mission of liberal democratic anthropology. In Excluded ancestors: inventible traditions: essays toward a more inclusive history of anthropology. Richard Handler, ed. Madison: University of Wisconsin Press. Tax, Sol [and others]. 1953. An appraisal of anthropology today. Chicago: University of Chicago Press. Tax, Sol. 1965. The history and philosophy of Current Anthropology. Current Anthropology 6:238, 242–269. Thomas, William L., Jr., ed. 1956. Current anthropology: a supplement to Anthropology Today. Chicago: University of Chicago Press.
ACID TEST Passivated micro-circuit resists hot dichromic acid, p 47 HOT CARRIER DIODES Faster switching than ever, p 53 TOMORROW'S ECLIPSE What will we learn from it, p 26 TRANSISTOR GAIN CONTROL New age circuit is a servo, p 60 NEW PIP SERIES ULTRAMINIATURE TRANSISTOR PULSE TRANSFORMERS UNITS SHOWN ACTUAL SIZE—IMMEDIATE DELIVERY FROM STOCK DEFINITIONS Amplitude: Intersection of leading pulse edge with smooth curve approaching top of pulse. Pulse width: Microseconds between 50% amplitude points on leading and trailing pulse edges. Rise Time: Microseconds required to increase from 10% to 90% amplitude. Overshoot: Percentage by which first excursion of pulse exceeds 100% amplitude. Droop: Percentage reduction from 100% amplitude a specified time after 100% amplitude point. Backswing: Negative swing after trailing edge as percentage of 100% amplitude. RUGGED—COMPLETELY METAL CASED Manufactured & Guaranteed to MIL-T-21038B 5/16" Dia. x 3/16" Ht.; Wt. 1/20 oz. Ratios—4:4:1 and 5:3:1 Anchored leads, withstands 10 lb. pull test Printed circuit use, plastic insulated leads Can be suspended by leads or clip mounted \| Type No. \| APPROX. DCR. OHMS \| BLOCKING OSCILLATOR PULSE \| COUPLING CIRCUIT CHARACTERISTICS \| \|----------\|-------------------\|---------------------------\|----------------------------------\| \| \| 1-Brn 2-Rd \| 3-Org 4-Yel \| Width µSec Rise Time % Over Shoot Droop % Back Swing P Width µSec Volt Unit Rise Time % Over Shoot Droop % Back Swing Imp. in/out* \| \| PIP-1 \| .18 \| .20 \| .07 .05 .02 0 0 37 .05 9 .018 0 0 12 50 \| \| PIP-2 \| .47 \| .56 \| .17 .1 .025 0 0 25 .1 8 .02 0 0 5 50 \| \| PIP-3 \| 1.01 \| 1.25 \| .37 .2 .03 0 0 15 .2 7 .035 0 0 5 100 \| \| PIP-4 \| 1.5 \| 1.85 \| .54 .5 .05 0 0 15 .5 7 .06 0 0 0 100 \| \| PIP-5 \| 2.45 \| 3.1 \| .9 1 .08 0 0 14 1 6.8 .15 0 0 5 100 \| \| PIP-6 \| 3.0 \| 3.7 \| 1.1 2 .10 0 0 15 2 6.6 .18 0 0 2 100 \| \| PIP-7 \| 4.9 \| 6.05 \| 1.8 3 .20 0 0 14 3 6.8 .20 0 0 2 100 \| \| PIP-8 \| 8.0 \| 9.7 \| 2.9 5 .30 0 0 3 5 7.9 .22 0 13 25.5 200 \| \| PIP-9 \| 13.1 \| 15.9 \| 4.7 10 .35 0 5 12 10 6.5 .4 0 15 20 200 \| \| PIP-100 \| Transistor pulse transformer kit, consisting of PIP-1 thru PIP-9 in plastic case. \| RATIO 5:3:1 MIL TYPE TPGRX531OCZ \| .55 .41 15 1 .01 0 20 1 .8 .01 0 0 5 140.50 \| \| PIP-10 \| \| \| 2.9 2.2 .82 1 .02 4 4 6 1 6.6 .05 0 6 12 280.100 \| \| PIP-11 \| \| \| 9.4 7.1 2.6 5 .05 0 12 12 5 8 .09 2 12 25 560.200 \| * Input winding leads Brn-Rd (1-2); output winding leads Org-Yel (3-4); leads Grn-Blu (5-6) open. AND CUSTOM BUILT UNITS TO YOUR SPECS. Write for catalog for full details on these and 1000 other stock items. UNITED TRANSFORMER CORP. 150 VARICK STREET, NEW YORK 13, N.Y. PACIFIC MFG. DIVISION: 3630 EASTHAM DRIVE, CULVER CITY, CALIF. EXPORT DIVISION: 13 EAST 40th STREET, NEW YORK 16, N.Y. CABLES "ARLAB" CIRCLE 900 ON READER SERVICE CARD THIN GOLD WIRE suspends a combined thin-film and semiconductor microcircuit in hot dichromic acid. Thus CBS Labs demonstrates successful surface passivation of thin-film components on the same silicon wafer that contains planar passivated transistors. Technique adds several degrees of freedom to microelectronics technology. See p 47 NEW LOOK in This Year's Comsats. Starting with Syncom II, to be launched next week, the second round of active and passive communications satellites will incorporate significant design improvements over their earlier counterparts. Designers of Relay II have taken steps to prevent any repeat mishaps and Echo II has a completely new balloon to improve r-f reflectivity LEM ELECTRONICS Faces New Design Challenges. High reliability in all flight model components and subsystems, as well as specially designed gear for the long lunar flight, are a prime requirement of designers of the lunar excursion module (LEM). But, say the experts, none of the electronics going aboard LEM is beyond the present state-of-the-art ECLIPSE TO BARE IONOSPHERE'S Secrets. Late tomorrow afternoon, astronomers and engineers from all over the world will take up observation posts between Maine and Alaska to make detailed solar and atmospheric measurements during the total solar eclipse. The long list of participants includes government agencies, scientific organizations, the military, universities, and private companies PROTOTYPE ELECTRON-BEAM welding system for in-space fabrication operations will be built for the Air Force by Hamilton Standard. System will weigh about 350 lb, use an 80-Kv gun. Main use of welding equipment in space will probably be to seal manned spacecraft structures in order to cut air supply losses INDUSTRIAL SPY—Aristocrat of Thieves. Is there a common element in situations where company secrets are stolen and then sold to other firms? Some experts blame a lot of these cases on lax management practices. Tightening security measures will undoubtedly help, but in the last analysis it is the moral atmosphere in which the employee works that matters. Ethical behavior must start with management MICROPOWER CIRCUITS: New Frontier in Solid State. When you start packing $\frac{1}{2}$ million components or more into a cubic inch of space, power consumption of individual circuits must be reduced to microwatts. But as you reduce you magnify problems. Now a pioneer in microelectronics helps iron out many of the kinks. By W. W. Gaertner and M. Schuller, CBS Labs CONTENTS continued HOT-CARRIER DIODES Switch in Picoseconds. New metal-semiconductor-interface diodes are removing speed limitations formerly imposed by minority-carrier storage in $p-n$ junction diodes. Making the new diodes requires very pure semiconductors and special methods for epitaxial growth and surface cleansing. By S. M. Krakauer and R. W. Soshea, Hp Associates 53 LEADLESS TRANSISTOR PROBE WORKS UNDERWATER. Probe contains 7-Mc tunnel-diode oscillator that is frequency modulated by signal to be measured. It can measure the near-field pattern of a dipole immersed in sea water. The same principles are useful in designing medical devices like the radio pill. By K. Iizuka, Harvard University 56 TRANSISTOR AGC: Novel Solution to a Dilemma. Automatic gain control of transistor i-f stages has never been easy because of changes in transistor parameters with operating point. Here a transistor amplifier is used as an attenuator in a servo system. Rectified output from feedback amplifier sets attenuator impedance. By F. Susi, Sylvania 60 DEPARTMENTS Crosstalk. The Fragmented Engineer 3 Comment. Negative Resistance VI. Metric System II 4 Electronics Newsletter. DOD Awards Zoom 7 Washington This Week. Project Forecast Gets Underway 12 Meetings Ahead. Aerospace Electro-Technology International Conference 40 Research and Development. What Happens to Dielectrics in Space 68 Components and Materials. How Ceramic Benders Control Light Rays 80 Production Techniques. Module Production-Quality Assured by Thermal Cycling and Life Testing 92 New Products. Synchronizer Accepts PCM at Rates to 1 Mc 104 Literature of the Week 120 People and Plants. Airpax Completes $1 Million Plant 122 Index to Advertisers 129 The Fragmented Engineer THE TITLE "fragmented engineer" does not mean some new engineering discipline; rather it means, in some cases, the absence of engineering discipline. It represents, unfortunately, what happens today to many of our most competent and creative engineers. Once an engineer is recognized as being knowledgeable and dependable, it is not long before more and more people begin calling upon his time and talent for any number of worthwhile causes. And he soon becomes not an electronics engineer but a fragmented engineer. Where he once spent most of his time planning and carrying forward his own research and development work with perhaps some time out to confer with colleagues about mutual problems or help a younger engineer over a rough spot, he now finds his professional life slowly being carved up into little bits and pieces. Within his own company he becomes a member of a papers review committee, design review committee, reliability review committee, and maybe more. Within his professional society he may be asked to serve on a section committee, professional group committee or technical committee. He may even, God forbid, become its chairman or secretary. He may represent his company on any number of trade association committees, or joint groups for this or that. Then, when he has really arrived, professionally speaking, he may become a member of one or more of the many committees and panels that are forming all the time, it seems, to advise the government about things technical and scientific. Meanwhile, he will also be delivering speeches to professional and trade association meetings, writing technical papers and articles, and possibly even testifying before various committees of the Congress. There is no doubt that all these activities are of great value to the engineer, his company, the country and the profession. But when does the engineer now find the time to do any engineering? It is a potentially dangerous trend that siphons off our best engineering talent and often leaves the actual research and development work either to immature or noncreative individuals. It well may be time to examine the cost of meetings in terms of lost hours of engineering creativity and question whether we are not in danger of cross fertilizing ourselves to death in the process of creating a giant, interdisciplinary nothing. If the engineer who feels himself becoming fragmented would look into the mirror every morning and as a minor rite repeat: "I am an engineer, I am an engineer. I AM AN ENGINEER, DAMMIT!" he might be better able to fend off the blandishments of those who would turn him into some kind of amateur statesman. Likewise, men in public life and directing the activities of professional groups or societies should consider carefully the cost in lost engineering time and talent of every new technical or scientific committee they propose. In this day of the race to the moon and other manifestations of the technological cold war, engineering creativity is indeed a national resource to be used only where it can do the most good. Let's ask ourselves whether we are out to create hot circuits and devices or just hot air. COMMENT Negative Resistance VI Concerning Mr. Lyon's letter in the May 24 issue (Comment, p 4), he points out correctly that negative resistance is a perplexing concept. Many of us, however, will object strenuously to using conductance to mean negative resistance. The definitions of these two terms are not the same. Conductance, in mhos, is of course the inverse of resistance, in ohms. One is equal to $I/E$; the other to $E/I$, providing we deal with pure resistance, and in either the d-c or a-c case. Negative resistance is a term used, generally, in the case of varying currents, rather than steady ones. Restricting ourselves to the case where we have a varying current and a varying voltage, we may define negative resistance as follows: Consider the voltage applied to, and the current through, an element. If the change in current is proportional to the change in voltage but in the opposite direction, that element is said to have negative resistance. To make the concept quite graphic, imagine a black box to which we apply a voltage, this time a direct steady voltage. If we apply various voltages and find to our surprise that each time we get a proportional current, but in exactly the opposite direction from that we expect with a resistive element, then we say we have a negative resistance, d-c variety. In either the case of the d-c or a-c negative resistance, there must be power generated within the element (d-c or a-c power). This power is furnished to the circuit. The element acts like a generator as far as direction of power flow is concerned. Perhaps the term power donor is helpful in contrast to the power absorber, or resistor. Formulae for negative resistance would use a minus sign to indicate that the current and power flow in the negative direction. For example, watts consumed equals $I^2$ times a negative $R$; power is produced, not consumed. OTIS N. MINOT Minot Informatic Devices Lexington, Massachusetts Metric System II When I read Mr. Poulsen's letter in the June 7 Comment column (p 4), I found myself in complete accord with one small exception. I would like to see you give all measurements and dimensions in the metric system (with the English equivalent following in parenthesis) immediately. In the past few years, I have been actively engaged in metrical photogrammetry and nucleonics mensuration. Both use the metric system and, through my associations with these fields, I have already started my reorientation in which systems to use. While I still think in the English system, I find it increasingly easy to do my engineering solely in metric units. My greatest trouble, now, is communicating with my colleagues. Any of your readers that are interested in helping with this vital change might do as I have just done: write The Metric Association, 502 Albee Building, 1426 G Street N.W., Washington 5, D.C. I haven't received their answer yet, but I feel better knowing that others feel as I do. As Mr. Poulsen mentions in his P.S., your range of articles and coverage make ELECTRONICS a most desirable journal, one that I look forward to each week. CLAUDE W. ASHBURN Albany, California Amplifier Gain Your very excellent article on consumer electronics in the May 17 issue had a small error that might mislead someone. You indicate (p 49) that the amplifier of the Acousticon hearing aid has an average gain of 93 db. As a matter of fact, we do not publish the gain of the amplifier itself. The HAIC gain (average of 500, 1,000 and 2,000 cps sound pressure gain as standardized by the Hearing Aid Industry Committee) is 39 db. Could it be possible that the typesetter transposed the numbers? HAROLD H. BEIZER Dictograph Products Inc. Danbury, Connecticut It was the author, not the typesetter. ONLY 7 INCHES OF RACK HEIGHT for this Tektronix Oscilloscope THAT ACCEPTS 12 PLUG-IN UNITS Type RM561A Oscilloscope—with Type 3A1 Dual-Trace Amplifier Unit and Type 3B3 Time-Base Unit—provides high sensitivity, wide-band, dual-trace operation combined with calibrated sweep delay in a compact, moderately priced package. You can use any of 10 other amplifier and time-base units for differential, multi-trace, sampling, other applications, including matched X-Y displays using the same type amplifier units in both channels. Also, you can observe no-parallax displays and sharp trace photography. For the new rectangular ceramic CRT has an internal graticule with controllable edge lighting. TYPE RM561A OSCILLOSCOPE (without plug-ins) ........................................ $525 TYPE 3A1 DUAL-TRACE AMPLIFIER UNIT ......................................................... $410 Passband from dc to 10 Mc for each channel • Sensitivity from 10 mv/cm to 10 v/cm in 10 calibrated steps, 1-2-5 sequence, with variable control • 6-cm linear scan • No signal delay TYPE 3B3 TIME-BASE UNIT ............................................................................. $525 Normal and Delayed Sweeps from 0.1 μsec/cm to 1 sec/cm in calibrated steps, 1-2-5 sequence, with variable control • 5X Magnifier • Single Sweep for Normal Sweep • Delay Interval from 0.5 μsec to 10 sec • Flexible Triggering Facilities with triggered operation to above 10 Mc. U. S. Sales Prices, f.o.b. Beaverton, Oregon FOR A DEMONSTRATION, PLEASE CALL YOUR TEKTRONIX FIELD ENGINEER Tektronix, Inc. / P. O. BOX 500 • BEAVERTON, OREGON I [Area Code 503] Mitchell 4-0161 • TWX: 503-291-6805 • Telex: 036-691 • Cable: TEKTRONIX • OVERSEAS DISTRIBUTORS IN 27 COUNTRIES Tektronix Field Offices are located in principal cities throughout the United States. Please consult your Telephone Directory. Tektronix Canada Ltd: Montreal, Quebec • Toronto (Willowdale) Ontario • Tektronix Ltd., Guernsey, Channel Islands Never be casual with a complex wave Be militant! Tackle it with a Sierra 125B, the brisk, businesslike Frequency Selective Voltmeter that never lets you stray more than a db away from the readings you ought to be getting. Poised on the brink of a wave, the 125B literally ripples with diverse capabilities. It's a tunable voltmeter with one continuous tuning range from 3 to 620 kc and dual selectivity bandwidths. Or, as the occasion demands, it can serve you as a broadband a-c voltmeter with essentially flat response from 1 to 620 kc. Its sensitivity varies from 24.5 µV to 30 V in eleven attenuator steps. The 125B costs a brisk, businesslike $895 — a touch more for models with carrier reinsertion or multiple input switching. Memo us in singlicate for a brochure that communicates on your wavelength. Or you might try waving at your Sierra sales representative for a response that'll be anything but flat. SIERRA ELECTRONIC DIV. OF PHILCO. A SUBSIDIARY OF Ford Motor Company. Sierra Electronic Division / 3885 Bohannon Drive / Menlo Park 1, California WASHINGTON—The Pentagon has had good news lately for the electronics industry. Although the volume of prime contracts is still behind the amount scheduled for the fiscal year which ended June 30 (p 7, June 28), a spurt in awards during the last days of the year has done much to correct the imbalance. The awards are just now being announced and include several in the $20-million-and-up category and many others that total less than $1 million. The Pentagon has no statistical data yet on the contracting zoom. Such figures aren't normally available for at least two months after the fact. Pentagon spokesmen say there is no great mystery about the rise. The last couple of weeks of a fiscal year are traditionally a period in which formal contract awards pour out at a spectacular rate. This is because the procurement agencies must commit the funds before June 30 or risk going through lots of red tape to get authority to contract out the money again in the next fiscal year. Ship's Engines Get Automatic Controls AN AUTOMATIC control system for a marine propulsion unit has been developed by Allgemeine Elektricitats-Gesellschaft (AEG). The system is now being demonstrated on a test stand in which the AEG remote-control system has been combined with a 5500-horsepower, five-cylinder diesel engine commonly used for shipboard installations. The system eliminates the human intermediary link between the bridge command and the engine room. The commands are still generated on an engine-room telegraph, but the new system automatically checks the acceptability of the command, and initiates whatever commands are necessary to control the diesel engine. The programming of the control system is designed to prevent damage caused by improper commands. Laser Gives Hope For Cancer Cure BOSTON — Researchers at Tufts-New England Medical Center here have cautiously reported that a method for destroying cancerous tissue may result from studies now being conducted with the laser. The scientists, using a device supplied by Raytheon, say they will also investigate the laser's effect on occlusive lesions of the arterial system. The experiments were begun in October, 1962, by Dr. Paul E. McGuff of the center and David Bushnell of Raytheon. "Since that time, an intensive series of experiments has been carried on using hamsters and human biopsy and autopsy specimens," the center announced. "The major emphasis to date has been in the study of 14 types of malignant tumors, with well over 200 tissue specimens or hamsters with transplanted human tumors having been exposed to laser energy." "It has been observed that the laser has a maximal effect on certain malignant tumor tissues and a minimal effect on normal tissues," center explained. "The extensive early destructive effect of the laser on certain types of transplanted human malignant tumors lends some encouragement to the belief that this energy may have an ultimate medical use. However, until the lasting effects can be determined and the restricting factors relating to the application of the laser to humans have been determined, it is premature to draw any firm conclusions at this time." Energy densities ranging from 17 to 12,000 joules per square centimeter were used in the experiments. 400-Kilogauss Magnet Predicted A MAGNET capable of producing fields in excess of 400 kilogauss will hopefully be built and operated within the coming year, it was reported at Oxford, England last week by staff members of the National Magnet Laboratory at MIT. Speaking at the second International Conference on High Magnetic Fields at Oxford University, MIT professor and magnet pioneer Francis Bitter reported that the plan is to make full use of the new magnet lab's four generators with flywheel energy storage, harnessing 32 megawatts for a few seconds of 400 Kg operation (p 56, May 17). This would be the world's most powerful magnet yet. The strength problem, though severe, appears to be soluble in terms of techniques already developed for the NML's 250 Kg magnet, Bitter told the international conference, adding that the main problem is to provide adequate cooling. The NML scientists are planning to use a pulsed cooling technique. timeter were applied to the surface of the tissue samples through a double convex lens. A 360-joule Raytheon laser of the same design as used in these tests has blasted holes in $\frac{1}{8}$-inch thick girders. Capacitors Would Store Solar Energy for Rockets HUNTSVILLE, ALA.—A NASA man at Marshall Space Flight Center here says he has invented a new electrical power system for space rockets. The system would gather solar energy and store it in capacitors, which would also serve as part of rocket's bulkhead structure. Robert J. Schwinghammer, the inventor, is section chief of one of the MSFC methods research and development branches. He says the stored energy could be used to energize on-board components, such as lasers, radar, and electric engines, and for vehicle coupling activities where short bursts of power are needed for fabrication-assembly operation. The inventor says the capacitors would have no moving parts, no hydraulics and no pneumatics and would not be subject to the lubrication, sealing and vapor-pressure problems present when conventional power systems are used in space. Lower-Cost Computer Offered for Typesetting NEW YORK—IBM moved into the automatic typesetting field in a big way last week, announcing that five daily newspapers, a book printer and a job printer have ordered its new system, built around the IBM 1620 computer. The IBM system sells for as little as $82,650, depending on attachments, but does not punch tape until all editorial corrections have been made. The system introduced last winter by RCA, priced about $300,000, punches tape directly from reporter's typewriter. Other computer companies are expected to introduce typesetting systems in the near future, ELECTRONICS learned. The potential market is huge. There are more than 1,700 daily newspapers in the U.S. Significantly, the IBM announcement mentions orders from newspapers, notably The New York Daily News and The Washington Evening Star, that have agreements with unionized typographers. Papers ordering the RCA system last winter employed non-union printers. Bertram Powers, president of the New York Typographical Union, Local 6 told ELECTRONICS his union has yet to come to an agreement with The Daily News on the new system but he "hopes this thing can be faced with reason." He said the system might result in fewer jobs but that his union would not oppose it so long as those printers currently employed are protected. Titan III First Stage Static Fired Tomorrow FIRST STAGE of the Titan III-C will be static fired for the first time tomorrow. The 85-foot high, 120-inch-in-diameter booster is the largest solid fuel motor to be developed and static fired in the U.S. It will be fired upside down, exerting a million pounds of thrust into a concrete thrust block, 30 feet square and 20 feet thick. Launch site will be United Aircraft's Development and Test Center, Coyote, Calif. In an operational mission, two of these boosters will be "strapped-on" to the Titan III-A (thus converting it into a Titan III-C) and fired simultaneously. The Titan III-A consists of a modified Titan II "core" with a new upper stage and control module mounted on top. Titan III will be capable of putting payloads of 5,000 to 25,000 pounds into relatively low orbits, accommodating payloads ranging from ten tons in a 100-nautical-mile orbit to 13,000 pounds in a 1,000-mile orbit. Developed specifically for space work, Titan III will be able to handle a variety of manned and unmanned missions. First manned mission will be Dynasoar. Associate contractors include: Martin-Marietta—airframe, assembly, test and systems integration; Aerojet General—liquid propulsion system for the first and second stages of the core in the transtage; United Technology Center—120-inch segmented solid rocket motors; AC Spark Plug—inertial guidance. In Brief . . . ATOMIC ENERGY Commission has extended its contract with Edgerton Germeshausen and Grier through 1968. The contract for fiscal year 1964 totals $32 million. COMMUNICATIONS Satellite Corp. is requesting proposals for study of multiple access systems for communications satellites, to replace the single-channel system in use today. Requests went to RCA, ITT, Bell Telephone Laboratories, General Telephone and Hughes. NASA has given Lear Siegler a $3.5-million contract to build data acquisition systems at Marshall Space Flight Center. Lear Siegler had previously received contracts totaling $6.5 million for this program. NATIONAL INSTITUTES of Health and NASA are supporting a major center for computer technology and biomedical research. The center will be managed by 12 New England universities and will be located in Cambridge, Mass. ITT PLANS to buy Cannon Electric Co. THE COMMON MARKET, in an attempt to close the five-year gap between Europe and the USA, is planning to expand computer installations tenfold by 1970. This will represent an investment of $600 to $700 million. RADIATION INC. was awarded a $1.2 million contract from AC Spark Plug for telemetry systems to be used in the Titan III. EMI ELECTRONICS in Great Britain is demonstrating a separate-luminance color TV technique using a four-tube camera operating in NTSC, SECAM and PAL systems but requiring only enough light for a monochrome studio. MARTIN received an Air Force contract for $100,000 to develop a new type of direct-energy conversion device. The device would be used in space missions where long-term reliability is important. New Bridge Design For Safe, Accurate, Easy Measurement of 'Lytic Capacitors The Sprague Model 1W2A Capacitance Bridge introduces new, improved technical refinements as well as restyling for added attractiveness and ease of operation. Built by capacitor engineers for capacitor users, it incorporates the best features of bridges used for many years in Sprague laboratories and production facilities. Precision Measurements over Entire Range from 0 to 120,000 μF The internal generator of the 1W2A Bridge is a line-driven frequency converter, and detection is obtained from an internal tuned transistor amplifier-null detector, whose sensitivity increases as the balance point is approached. It has provision for 2-terminal, 3-terminal, and 4-terminal capacitance measurements, which are essential for accurate measurement... ±1% of reading + 10μF... of medium, low, and high capacitance values, respectively. No Damage to Capacitors The model 1W2A Capacitance Bridge will not cause degradation or failure in electrolytic or low-voltage ceramic capacitors during test, as is the case in many conventional bridges and test circuits. The 120 cycle A-C voltage, applied to capacitors under test from a built-in source, never exceeds 0.5 volt! It is usually unnecessary to apply d-c polarizing voltage to electrolytic capacitors because of this safe, low voltage. Complete Specifications Available For complete technical data on this precision instrument, write for Engineering Bulletin 90.010A to Technical Literature Service, Sprague Electric Company, 35 Marshall Street, North Adams, Massachusetts. --- Did you know Sprague makes...? - Magnetic Logic Devices - Core-diode and core transistor magnetic shift registers and magnetic counters for switching and storage applications in computer and logic circuitry. - Molded Pulse Transformers - Miniature Pulse Transformers with tough molded cases for increased protection against physical damage and severe atmospheric conditions. - Nanosecond Pulse Transformers in TO-5 Transistor Cases - Special design offers distinct advantages: (1) Mini-fied size. (2) Welded hermetic seal. (3) Increased reliability. (4) Compatibility with transistor mounting techniques. - Something New in Counting Techniques - Simple yet versatile, low-cost yet reliable counters available for predetermined (2 to 11) or selectable (5 through 10) counting cycles. - Dynacor® Bobbin Cores - Series "300" Cores with logical flux values in popular physical sizes are stocked in production quantities for fast delivery. They're value engineered for quality with economy! - Hermetically-Sealed TO-5 Encased Switch Cores - Designed especially for high-speed, low-power switching up to 100 kc, adaptability with conventional transistor packaging techniques, and performance under MIL-S-21038 environmental conditions. - Electronic Modules to Customer Requirements - Custom packaging is no novelty at Sprague's Special Products Division, where "specials" are continually being developed and produced with countless variations in electrical characteristics and mechanical configurations. --- For application engineering assistance (without obligation, of course) on any of the above products, write or call the Special Products Division, Sprague Electric Company, 35 Union Street, North Adams, Massachusetts. --- Sprague and ® are registered trademarks of the Sprague Electric Co. How much do you know about today's high-frequency scopes? First aid? First aid? Internal medicine...Even if you're not the chief maintenance man on your oscilloscope, alignment, calibration and adjustment can be real problems for scope users, as well. Add to that the down time required by routine maintenance, and you see the true importance of having a scope that's both reliable and easy to keep on the air. It's pretty basic that the simpler the circuitry, the easier the alignment and maintenance, and when you can have simplified circuitry plus the best performance, it's well worth investigating. Our 175A Oscilloscope uses simple circuitry, as witness the total output amplifier of its main vertical amplifier: If you've ever calibrated and aligned a high-frequency scope — even if you've made the periodic routine adjustments — you know it's no easy matter. Even changing a tube can require quite a bit of readjustment of circuit components. Consider, then, the problem: Conventional high-frequency scopes employ complicated distributed amplifiers in the vertical deflection system. These amplifiers produce the large voltages needed to drive conventional high-frequency crt's. They require inductance coils and trimmer capacitors which need to be delicately tuned because they're interdependent. Next time you attempt this task, time yourself. For the 175A 50 MC Oscilloscope, we developed a new 12 kv cathode ray tube, a tube so sensitive that deflection circuitry can be simplified at no sacrifice in performance. Gone are distributed amplifiers and their tedious adjustments. Alignment time is vastly reduced from that required for conventional scopes. What's more, the 12 kv crt in the 175A provides a picture 6x10 cm, larger than the 4x10 cm picture on the conventional high-frequency scope. An internal graticule eliminates parallax error, and a specially etched glass produces a non-glare faceplate. There you've got unprecedented accuracy. Next, check your present scope instruction manual for adjustments required on your vertical delay circuitry... another time-consuming procedure. For the 175A we developed a high-frequency balanced cable delay line which has no adjustments — it's sealed. Altogether, the main vertical amplifier in the 175A Oscilloscope requires only five simple, independent adjustments. You're familiar with the number of components you have to keep on hand to maintain the conventional scope... and you're familiar with the term "selected" components. Most scopes use many tube types, for example, including some "selected" tubes... components that can only be replaced with tubes of identical characteristics or with considerable realignment. The 175A uses only 7 tube types and 5 transistor types... and none of them are "selected." Each can be replaced with any tube or transistor of the same type. See what that fact does to your parts inventory... and use your valuable shelf space for something else. Other virtues of the 175A: Horizontal and vertical plug-ins... such as dual- and four-trace and high sensitivity vertical viewing, sweep delay and x-y recording capabilities... when, but not unless you need them. They're all available with the 175A — as plug-ins. Positive preset syncing over the entire bandwidth, the easiest triggering and the most dependable triggering you can find. Fewer controls, easier to use. A novice can learn to use the 175A in half the time it'll take him to learn the intricacies of other scopes. Now consider the mechanical convenience (or inconvenience) of maintenance on your present high-frequency scope... inaccessible circuits and components, solder joints that are hard to get at. With some scopes maintenance is practically like factory assembly. The 175A is packaged in our modular cabinet, sides, top and bottom of which are easily removed for access to all components. Circuit boards with plated-through holes are connected to their circuits with "edge-on" connectors, greatly simplifying removal and replacement. We're not just "demonstrating" the 175A these days. We're actually letting you use it, stack it up alongside any high-frequency scope you can find, compare them feature by feature. And feature by feature you'll say to yourself, "This is great." Feature by feature our competitors will ask, "Why didn't we think of that?" hp 175A, $1325 (without plug-ins). Data subject to change without notice. Price f.o.b. factory. HEWLETT PACKARD COMPANY 1501 Page Mill Rd., Palo Alto, Calif., (415) 326-7000. Sales and service representatives in principal areas. Europe, Hewlett-Packard S.A., 54 Route des Acacias, Geneva, Switzerland; Canada, Hewlett-Packard (Canada) Ltd., 8270 Mayrand St., Montreal, Quebec. WASHINGTON THIS WEEK THE AIR FORCE has launched a major study on what its mission and weapons systems will be through the 1970's. Operating under the name of Project Forecast, more than 300 top level military, industry and scientific experts are working on the project under General Bernard A. Schriever, commander, Air Force Systems Command. The group is projecting the probable military missions for this time period, then appraising technology to determine what weapons can be developed to meet the mission requirements. It will be several months before the study is completed, with the recommendations certain to be highly classified. The Air Force will undoubtedly recommend development of military space weapons. Meanwhile, Air Force is optimistic that Defense Secretary McNamara will allow it to continue development of the manned Dynasoar spacecraft. A decision on Dynasoar's future is due soon. PENTAGON ESTIMATES savings of $40-million during first 10 months of fiscal year 1963 by shifting from noncompetitive to price competitive buying of electronic and communications equipment. Examples: price of the R-442 radio receiver was trimmed from a non-competitive quote of $1,519 to $1,034 in competition (total gross savings $908,890); RT-246 receiver/transmitter was cut from $3,976 to $2,692, with total savings of $927,048; price of the RT-524 receiver/transmitter was reduced from $3,074 to $2,036, resulting in $7.4-million total savings; and the AN/PRC-25 man pack radio (see p 36) was trimmed from $2,278 to $843, saving $10.5 million. The original sole-source supplier for the PRC-25 was RCA's Military Products division. In the follow-on competition, the firm's Commercial Products division came in with a bid price some 60 percent lower, McNamara said. McNAMARA REPORTED supply savings of $1 billion in the past year and set a target of almost 4-billion annual cost reductions over the next three years. The savings, McNamara said, result from increased use of excess inventories, "refining requirements calculations," eliminating "gold-plating" from specifications (one example involved use of an ordinary women's hair wave clip, cut to size, instead of a heat sink, to prevent soldering heat from damaging transistors: this reduced the unit cost from $1.35 to 2 cents); shifting from noncompetitive to competitive procurement; and shifting from cost-plus to fixed price and incentive contracts. THE GOVERNMENT has revised its official definition of electronics small business. Up to now, firms with no more than 500 employees were so classified, and thus qualified for whatever preference was granted to small firms in government procurement. The work force maximum has now been lifted to 750 for manufacturers of radio and tv receiving sets (except communications); radio and tv transmitting, signaling, and detection equipment; cathode-ray picture tubes; and transmitting, industrial, and special-purpose electron tubes. The limit is now raised to 1,000 workers for producers of radio and tv receiving-type electron tubes, except cathode ray. If a TV signal were beamed earthward from the surface of the moon tomorrow, a wideband phase-lock demodulator built by STL could pick it up and faithfully reproduce it. The new demodulator is already aboard Project Relay. Two important advances in FM design distinguish the STL demodulator. First, it allows reduction of spacecraft transmitter power by a factor of four. Second, its loop 3 db bandwidth of 7.5 megacycles is wide enough for real-time transmission of full bandwidth television or multichannel telephony. Scientists and engineers interested in advancing the art of space communications will find STL an active place. STL builds spacecraft for NASA and Air Force-ARPA, and continues Systems Management for the Air Force's Atlas, Titan and Minuteman programs. These activities create immediate openings for Space Physics, Radar Systems, Applied Mathematics, Space Communications, Antennas and Microwaves, Analog Computers, Computer Design, Digital Computers, Guidance and Navigation, Electromechanical Devices, Engineering Mechanics, Propulsion Systems, Materials Research. For Southern California or Cape Canaveral positions, write Dr. R. C. Potter, One Space Park, Redondo Beach, California, Department G-7, or Post Office Box 4277, Patrick Air Force Base, Florida. STL is an equal opportunity employer. SPACE TECHNOLOGY LABORATORIES, INC. a subsidiary of Thompson Ramo Wooldridge Inc. Los Angeles • Vandenberg AFB • Norton AFB, San Bernardino • Cape Canaveral • Washington, D.C. • Boston • Huntsville • Dayton • Houston Those attending the Aerospace Conference in Washington August 4-9 are cordially invited to visit STL Booth 93 3 new G-E photoconductive devices offer a host of new uses Photoconductive Arrays: Newly developed photoconductor arrays (illustrated) offer many promising applications, especially for data processing, industrial automation, telephony, photocopying, proportional control devices, and alarm systems of all sorts. The cadmium sulfide or cadmium selenide cells in the arrays can provide a wide range of electrical characteristics to meet specific needs: spectral response 5,500-7,300 angstroms; light resistance 100-125 K ohms @ 1-5 ft-c; dark resistance 15-100 megohms; power dissipation 50-400 mw. Tightly compact 20-, 40-, 90-cell and even more complex G-E photoconductor arrays are now under development. Pinpoint photosensitive areas can be tailored to almost any design configuration. Photosensitive "Siamese twins." Double-cell G-E photoconductors, in two sizes, also are now available. Each highly sensitive, hermetically sealed unit has three flexible leads, one interconnecting both photosensitive areas. Electrical characteristics can be varied considerably to meet special needs. New PC-L devices make noiseless controls and rugged low-level switches. G.E.'s new PC-L (photoconductor-lamp combination devices) are light-tight packages which enclose a photoconductive cell and a variable illumination source. Photoconduction is controlled by varying the voltage to the light source. Circuit isolation, noiseless potentiometry, and reliable low-level switching are just a few of the new product possibilities that G-E PC-L devices open up. Here are typical characteristics: \| Developmental Types: \| Y 1079 \| Y 1128 \| Y 1138 \| Y 1248 \| \|----------------------\|--------\|--------\|--------\|--------\| \| Lamp rating \| 5v 60ma\| 5v 60ma\| 5v 60ma\| 28v 40ma\| \| Max. photoconductor voltage \| 60v \| 30v \| 60v \| 60v \| \| Power dissipation \| 150 mw \| 75 mw \| 250 mw \| 100 mw \| \| Photoconductor resistance (ohms) for various lamp input voltages \| 45 @ 4v \| 150 @ 4v \| 45 @ 4v \| 50 @ 20v \| Free booklet gives application data and specifications on the complete line of "standard" G-E photocells—get yours today! Compactron IF amplifier gives 15% higher gain than comparably priced frame-grid types IF amplifier cost-and-performance figures were determined for TV sets of three manufacturers: Two manufacturers, Co. "A" and Co. "B," used the popular 6EH7 and 6EJ7 frame grid tubes plus a frame grid mixer. The third, Co. "C," used a 3-stage IF amplifier consisting of a 6AR11 compactron plus a non-frame grid 6JN8 pentode and a less expensive non-frame grid mixer. The results showed that all three amplifiers fell in the cost range of $1.40-$1.49.* However, the compactron version gave 90 db maximum gain as compared to only 78 db for the frame grid types. In all cases, gain was measured from mixer grid to detector; bandwidth was 3.2 mc. To increase gain, Co. "A" and Co. "B" have chosen to allow IF "pole shifting" to occur so that at maximum gain (fringe area signal) the bandwidth reduces to about 1.5 mc. This yields about 6 db more gain—still 6 db less than the compactron version. Includes price of IF tubes, plus component and socket cost differences and tuner cost difference for frame grid mixer required by Co. "A" and Co. "B." CIRCLE 200 ON READER SERVICE CARD July 19, 1963 • electronics More G-E compactrons in tomorrow's radio, TV, hi-fi, and industrial equipment Over 40 of today's major equipment manufacturers have joined the move to compactrons. Many of the 65 compactron types now available have been designed into equipment such as: portable and console TV sets, electronic organs, telemetering units, sweep generators, mobile and fixed communication equipment, stereo tuners, multiplex adapters, microwave amplifiers, halogen leak detectors, and precision power sources. Major reasons for this mushrooming growth of G-E compactrons are: (1) performance, (2) lower costs. Compactrons overcome the limitations of tubes and transistors and deliver more watts per cubic inch than any other component. They have a lower initial cost per function and offer savings in labor and materials. By combining several functions into one low-profile envelope requiring fewer pins, stems, sockets, welds, and handling, compactrons provide increased reliability and more compact circuitry, when compared to present-day components. They cost less than tubes or transistors and use up to 35% less power to perform the same function. Compactrons reduce hardware, wiring, and soldering connections and lower assembly time. Heat dissipation is up to 35% better than with conventional tubes, increasing life and reliability. Multifunction design provides more compact circuits, allowing use of a smaller chassis and cabinet with resultant savings in materials. New areas of design opened up by highly sensitive subminiature magnetron Two curves, illustrated above, emphasize the Z-2935's ability to produce large changes in frequency with slight changes in plate voltage and magnetic field strength. Typical sensitivities: for voltage, 1CPS/mv, or for magnetic field strength, 100 CPS/gauss. Specifications of the Z-2935 are: \| Parameter \| Value \| \|----------------------------------\|----------------\| \| Heater Voltage, AC or DC+ \| 6.3 ± 0.3 Volts\| \| Heater Current \| 0.15 Amperes \| \| Plate Voltage \| 25 Volts \| \| Inter-electrode Voltage \| \| \| Positive \| 5 Volts \| \| Negative \| 5 Volts \| \| Cathode Current \| 6.0 Milliamperes\| \| Heater-Cathode Voltage \| \| \| Heater Positive with Respect to Cathode \| 25 Volts \| \| Heater Negative with Respect to Cathode \| 25 Volts \| Versatile G-E reed switches outlive, outspeed, outswitch any other mechanical switching device G-E dry reed switches are designed to work perfectly, with milliwatt sensitivity, up to 100 million cycles or more. They're small, simple, rugged, and fast acting. All types are magnetically actuated. Contacts are pure gold, silver, or rhodium. Dry reed switches appeal especially to the designer who can't let well enough alone. Appliances, alarm systems, telephony and data processing gear, and virtually all other electromechanical equipment can be improved with economical dry reed switches. Some typical performance characteristics: \| SWITCH TYPE: \| 2DR15 \| 2DR50 \| Y1027 \| Y1136 \| \|--------------\|-------\|-------\|-------\|-------\| \| Life expectancy (at half load) \| 100 million operations \| 100 million operations \| 15 million operations \| 100 million operations \| \| Ampere turns Pull-in \| 90 ± 15 \| 90 ± 15 \| 43 ± 7 \| 90 ± 15 \| \| Release \| 35 ± 10 \| 40 ± 10 \| 25 ± 6 \| 65 ± 10 \| \| Max. contact rating (DC resistive) \| 15 watts 250 volts 1 amp \| 50 watts 250 volts 3 amps \| 4 watts 250 volts 250 mA \| 15 watts 250 volts 1 amp \| \| Max. contact resistance \| 50 milli-ohms \| 150 milli-ohms \| 150 milli-ohms \| 50 milli-ohms \| \| Length (excluding leads) \| 2.10" \| 2.10" \| 0.84" \| 2.10" \| Progress Is Our Most Important Product For more information: Write G-E Receiving Tube Dept., Technical Information and Product Service (TIPS), Room 7013-C, Owensboro, Ky. Please specify product(s). CIRCLE 201 ON READER SERVICE CARD CIRCLE 202 ON READER SERVICE CARD CIRCLE 15 ON READER SERVICE CARD The FM Stereo Modulator Type 219-A is designed to provide a multiplex output signal in accordance with FCC Docket 13506 when fed with Left (L) and Right (R) audio stereo channel inputs and/or subsidiary communications FM sub-carriers (SCA). The output of the modulator may be switched to provide either (L + R), (L - R), 19 KC pilot carrier, 38 KC residual carrier or the complete multiplex signal which can then be used to modulate a suitable FM Signal Generator. When used with the BRC Type 202-E, no external audio oscillator or other equipment is required. A peak reading metering system, calibrated in % of system deviation, is provided for setting and monitoring the levels of the individual sub-carriers. The internal matrix may be switched from the normal condition to provide either (L+R) or (L-R) null for checking the matrix in the receiver under test. The modulator is completely self-contained and housed in a single cabinet which may be adapted for standard rack mounting. BOONTON RADIO COMPANY A Division of Hewlett-Packard Company GREEN POND ROAD, ROCKAWAY, NEW JERSEY Tel. 627-6400 (Area Code 201) TWX: 201-627-3912 Cable Address: Boonraco Price: $975.00 F.O.B. Rockaway, N.J. INPUT CHARACTERISTICS LEFT (L) & RIGHT (R) INPUTS Frequency Range: 50 cps – 15 KC Level: 7.5 volts peak For approx. 10% peak multiplex output; simultaneous (L) and (R) inputs yield 90% peak multiplex output Preemphasis: 75 μsec preemphasis switchable in or out of circuit SUBSIDIARY COMMUNICATIONS (SCA) INPUT Frequency range: 20 – 75 KC Level: 1.0 volt rms For approx. 10% peak multiplex output MODULATING OSCILLATOR CHARACTERISTICS Osc Frequency: 1 KC Osc Output: Switchable into either (L) or (R) input OUTPUT CHARACTERISTICS Level: 0 – 7.5 volts peak Multiplex output Residual Hum & Noise: > 60 db below 100% output Crosstalk: > 40 db below 100% output (L-R) into (L + R) METERING Range: 0 – 10%* (19 KC and 38 KC only); 0 – 100%* Multiplex output; output adjustable 0 – 7.5 volts peak for 100% Output Modes: Switchable for (L + R), (L - R), 19 KC pilot carrier, 38 KC residual carrier, or multiplex signal PILOT CARRIER Frequency: 19 KC Accuracy: ± 0.01% MONAURAL (L + R) Fidelity: 50 cps – 15 KC ± 1 db ± 0.2 db and ± 1.5° relative to (L - R) DOUBLE SIDEBAND SUPPRESSED CARRIER (L - R) Frequency: 38 KC Fidelity: 50 cps – 15 KC ± 1 db ± 0.2 db and ± 1.5° relative to (L + R) SUBSIDIARY COMMUNICATIONS (SCA) Fidelity: 20 – 75 KC ± 0.5 db OSCILLOSCOPE SYNC SIGNAL Frequency: 19 KC Output Level: 0.5 volts rms Price: $975.00 F.O.B. Rockaway, N.J. New from Transitron The most recent achievement of Transitron's Total Reliability Program, these new planar versions of the highly popular 2N1722 and 2N1724 NPN silicon power transistors have the greater inherent reliability associated with advanced surface passivation techniques. They readily replace mesa versions in existing circuitry. Both offer lower leakage current, lower saturation voltage, better power dissipation and improved beta linearity. The 2N1722 is packaged in a flange type mounting, the 2N1724 is an 11/16" hex base, stud-mounted package. Both are available through your Transitron Distributor. For complete information, write to Transitron, Wakefield, Massachusetts. \| Type \| DC Current Gain @ Ic = 2 Amps (β) \| Typical Collector Saturation Voltage @ Ic = 2 Amps (Volts) \| Minimum Sustaining Voltage @ Ic = 200mA (Volts) \| Typical Cut-Off Frequency @ Ic = 100mA (Hz) \| Power Dissipation Rating @ 100°C Case (Watts) \| \|--------\|----------------------------------\|----------------------------------------------------------\|-------------------------------------------------\|---------------------------------------------\|-----------------------------------------------\| \| 2N1722 \| 20-90 \| 0.3 \| 80 \| 20 \| 50 \| \| 2N1724 \| \| \| \| \| \| Transitron electronic corporation wakefield, melrose, boston, mass. SALES OFFICES IN PRINCIPAL CITIES THROUGHOUT THE U.S.A. AND EUROPE • CABLE ADDRESS TRELCO CIRCLE 17 ON READER SERVICE CARD New Look in Next week it's Syncom; design changes set for new Relay and Echo, too By JOEL STRASSER Assistant Editor DESIGN IMPROVEMENTS will be incorporated into the second round of communications satellites to be launched next week and later this year using lessons learned from their predecessors. A beefed-up Syncom II will make another attempt into orbit no earlier than Tuesday, while both Relay II and Echo II will probably go up during the last few days in December. SYNCOM II—To prevent repetition of Syncom I's failure, the second Syncom, now being readied on the launch pad at Cape Canaveral, has been modified. Originally, it was thought that a defect in the power supply may have caused the failure. The prevailing theory now, however, is that one of the titanium tanks that store nitrogen used for vernier attitude control and period control may have ruptured in Syncom I, blowing the satellite apart. In tests conducted within the last several weeks, a tank was punctured causing an explosion that ejected a weight similar to that displaced by the apogee kick on Syncom I. In the second Syncom, clearance between the tanks was increased. As a precautionary measure against power failure, as indicated earlier (ELECTRONICS, p 25, May 17) all wires to power-consuming subsystems are being fully duplicated. Another set of batteries in the beacon will give 10 to 15 extra minutes of telemetry for location purposes, if the spacecraft gets lost again. While these improvements will add several pounds to the overall weight of Syncom II, a higher-powered apogee kick motor, built by Jet Propulsion Laboratory, will be used instead of the Thiokol motor used in Syncom I. Hughes Aircraft built the satellite. RELAY II—To avoid the problems that plagued Relay I with a faulty This Year's Comsats switching transistor in the transponder, changes are being made in the voltage regulator switching circuits. A different transistor, less sensitive to temperature variation, is being incorporated into this circuit. An electromagnetic relay is being placed in series with the voltage regulator switch to provide a second means of turning the satellite off. N-on-p solar cells, which offer much greater radiation resistance than Relay I's p-on-n units, are being incorporated into the satellite. They will have a 60-mil glass shielding. Radiation experiments conducted by Bell Telephone Laboratories and State University of Iowa will be the same, but a minor change will be made on the radiation damage experiment conducted by Goddard Space Flight Center. For redundancy, three gallium-arsenide cells will be used instead of one, and unlike Relay I's, these will have 6- to 10-mil glass shields. The tests, which were to be conducted with high-energy radiation, were voided by low-energy protons, which the glass shield is designed to protect against. More r-f shielding will be used in the wide-band system because of command anomalies. It was determined that Relay I's command circuit was particularly susceptible to rock and roll music. Command receiver characteristics are being changed—more shielding, the operating squelch level will be less sensitive to spurious transmissions, and a new filter will be placed in the antenna feed system. An extra cell will be added in each of the satellite's three batteries to achieve greater ampere-hour capacity. This will provide 23 v, rather than the former 22 v, at the end of the operating discharge cycle. Unpressurized traveling wave tubes may be used in the wideband system. Pressurized twt's were used in Relay I because there was no conductive or convective heat transfer. The redesigned twt would have the same electrical characteristics, but physical changes in the pitch angle of the helix and an outer glass coating would eliminate the heat problem. Pressurized tubes would then be unnecessary. Changes in the charge-controller circuit, which were affected by radiation damage in Relay I, will give longer periods of operation. Telemetry sensors are being relocated in some instances. A thermistor that could have picked up the failure of the 2N174 transistor immediately in Relay I is being moved to the satellite's base ring. When the satellite gets hot, a thermal controller activates a hydraulic system to let the heat radiate out through vanes in the spacecraft. A sensing device is being added to the vane system. Total added weight to the spacecraft for all these changes is about one pound. ECHO II—Biggest change in the Echo II balloon satellite will be a new inflation system that uses higher pressure acetamide as the inflating material in a specially designed compartmented balloon (Electronics, p 7, Feb. 15). Previous plans called for using lower pressure benzoic acid in a regular balloon. In a recent test, it was determined that the increased pressure is better for r-f. The balloon, which underwent its third static inflation test last week at Lakehurst, N. J., will be 135 feet in diameter compared to Echo I's 100 feet. Tv system to monitor the deployment of the balloon for five minutes following the launch is being incorporated into the second (Agena) stage of the Thor-Agena booster. Readout will be at Johannesburg, South Africa. The initial trajectory will be changed somewhat, but this will not affect orbit or period. Collins Radio is now using the old Echo I balloon for voice communications between Cedar Rapids, Iowa and Dallas, Texas. Under test is a new station at Dallas with a 60-foot parabolic dish and phase-lock loop receivers to aid in tracking. Collins reports that Echo I has shrunk from 100 feet to 60 feet in diameter but is still very much alive. MIT Lincoln Laboratory and the Naval Research Laboratory facility at Stump Neck, Md. are also using it successfully. You can take your work to the HONEYWELL 8100 TAPE RECORDER or you can take the 8100 to your work! Completely portable, completely self-contained, the new Honeywell 8100 recorder/reproducer offers instrumentation quality in a package you can carry from test to test with one hand. Or you can rack-mount the 8100, as shown at left with a Honeywell Model 1508 Visicorder Oscillograph. The 8100 contains all its own electronics, plus such convenience features as a built-in calibration panel and automatic switching of center frequencies. You can record analog data from DC to 10,000 cps on up to 8 data channels. And there's an optional built-in scope for monitoring data while you record. Available with either ¼ or ½ inch tape. Weight only 85 pounds. Price from $5900 to $11,300. Write for brochure #2038 to Honeywell, Denver Division, Denver 10, Colorado. Honeywell It's the reliability CULL line we can draw on the basis of current noise level for any lot of Corning metal-oxide film resistors. The good guys are on the left. Test them at 2½ times rated power and 25°C. We've been doing it for 23,000,000 part hours without one failure. If one had failed, you'd have a failure rate of 0.01001%/1000 hours at 90% confidence, 0.00398 at 60%, and 0.00022 at 5%. But none have! The CULL line falls at different db levels for different styles and sizes of Corning resistors. But we can draw one for any of them that is practical in cost and unsmirched to date for reliability. Resistors to the right of the line may be OK, but our findings indicate that all mavericks in TC and load life will be on the noisy side of the line. Read all about this new, non-destructive reliability screening tool, and how we'll put it to work for you at modest cost. Write for our new folder, "Current Noise Level: New Reliability Screening Technique for Corning Metal-Oxide Resistors," to Corning Glass Works, 3901 Electronics Drive, Raleigh, N. C. Corning Uniformity Limit Level CORNING Electronic Components HOUSTON—A minor roadblock in the procurement of electronic components has arisen in the National Aeronautics and Space Administration's Apollo project, ELECTRONICS learned this week. It is not serious enough, however, to cause any delay in NASA's overall goal of putting Americans on the moon this decade. Some component manufacturers are having difficulty meeting NASA's rigid quality control specifications. This is causing some delay in getting 'type A' components for the guidance and control systems, David Gilbert, manager of Apollo spacecraft systems office of guidance and control, reports. Gilbert says the problem is not serious, at least not at this stage. "We are going ahead and releasing initial systems designs," he said, "based on 'type B' components. These can be used in early test units. We will need 'type A' for the flight vehicles, however," Gilbert said. In the Apollo program, NASA requirements call for minute tracing of quality control steps by parts manufacturers. The degree of control, Gilbert reports, "has proved annoying to some manufacturers" leading to "negotiations and renegotiations. "The parts problem has been a surprise to us. We didn't expect it. We are working around it, however, and since we don't have to have all 'type A' components for the first few systems, we see no reason for any delay in the overall program," Gilbert said. SUBSYSTEMS — The electronic systems concept and basic designs of both the command and lunar excursion modules have been "pretty well completed" now, Gilbert said. LEM's electronic systems, which duplicate where possible those in the command module, will for the most part be handled by RCA under subcontract with Grumman Aircraft. Five areas that RCA, under a $50-million plus contract, will be responsible for include systems engineering, communications, radar, in-flight systems, and ground test out. RCA will also handle certain other undefined jobs. - Communication system includes: S-band transponder, S-band power amplifier (20 watts), S-band diplexer, S-band erectible antenna (lunar surface) 10-ft diameter, S-band high-gain antenna (spacecraft) 2-ft diameter, S-band omnidirectional antenna, vhf transceiver, vhf lunar stay antenna, vhf omnidirectional antenna, pcm telemetry equipment, tv equipment, intercom equipment, and miscellaneous microwave components. - Instrumentation system consists of data storage equipment, central timing equipment, signal conditioning equipment, sensor/transducer equipment, and in-flight test systems. - Navigation and guidance includes rendezvous radar, radar altimeter, computer, and signal conditioning equipment. Challenges CUSTOM-MADE GEAR—Of the communications and instrumentation systems, "almost nothing is off the shelf hardware for LEM except transducers and microwave switches," Owen Maynard, acting manager of spacecraft systems for LEM reports. However, he and Douglas Broone, Jr., of his staff, report that virtually all systems now called for are "within the state-of-the-art." For example, LEM will require some exacting antenna systems, but the main task facing designers is to match present systems to LEM's configuration. An exception may be the vhf lunar-stay antenna which will have to scan from 80 to 100 miles when the command module's orbit is directly above it, and up to 232 miles away as command module begins making its pass. Broone reports NASA is now looking at a turnstile-type antenna used on one of the Goddard earth orbital satellites. "Patterns to date indicate it is what we want. Basically, it will be a spiral antenna with two crosses in it." Although the adaptation of antennas to LEM's configuration is causing Frequency stable performance... 100 times faster than normal Frequency stable performance is achieved upon application of high voltage nearly 100 times faster with the ML-7855 than with conventional planar triodes. Even with an unregulated plate supply the ML-7855 will provide this same frequency stability within 2 seconds. High-voltage, arc-suppressed, operation is provided by the Phormat (matrix) cathode. . . . Plate-pulse to 3000 Mc, with 3000 v eb, 3.0 a ib, t 3 usec at 0.0025 Du. For complete write The Machlett Laboratories, Inc., Springdale, Conn. An affiliate of Raytheon Company. Now from Sprague! WIDE SELECTION OF Silicon Chopper Transistors NOW IN VOLUME PRODUCTION Specifically designed for use as low-level choppers, Sprague Silicon Precision Alloy Transistors are inherently stable. Every Sprague chopper transistor undergoes a rigid production conditioning of 40 temperature cycles from $-55^\circ C$ to $+140^\circ C$, a 200-hour bake at $+140^\circ C$, and a 2-hour 125mW operational burn-in! \| Model \| Description \| \|-------\|-------------\| \| 2N2162/2N2165 \| Guaranteed 30 volt rating. Typical $f_T$ of 20 Mc and low offset voltage make these transistors ideal where high voltage is required. \| \| 2N2163/2N2166 \| Have 15 volt rating and same high frequency performance and low offset voltage as 2N2162/2N2165. \| \| 2N2164/2N2167 \| Highest frequency P-N-P Silicon Choppers available as standard types. \| \| 2N2185 \| Extremely low leakage current of 1 nano-ampere at 100 volts with 30 volt rating. \| \| 2N2274 \| Similar to 2N2165 but has lower inverted dynamic saturation resistance. \| \| 2N2276 \| Low-cost version of 2N2274. \| \| 2N2278 \| Very low offset voltage of 1.75 mV at 1 mA. \| \| 2N2187 \| Matched pair of 2N2185 with $\Delta V_{OFF} = 50 \text{ mV max. from } +25^\circ C \text{ to } +85^\circ C$. \| \| 2N2275 \| Matched pair of 2N2274 with $\Delta V_{OFF} = 100 \text{ mV max. from } +25^\circ C \text{ to } +65^\circ C$. \| \| 2N2277 \| Matched pair of 2N2276 with $\Delta V_{OFF} = 100 \text{ mV max. from } +25^\circ C \text{ to } +85^\circ C$. \| \| 2N2279 \| Matched pair of 2N2278 with $\Delta V_{OFF} = 50 \text{ mV max. from } +25^\circ C \text{ to } +85^\circ C$. \| For additional information, write Technical Literature Service, Sprague Electric Co., 35 Marshall Street, North Adams, Mass., indicating the types in which you are interested. It will have a keyboard to allow the astronauts to insert information from the ground, and a digital display readout. Computers used in the command module and LEM will be original designs, with accuracy the primary factor. "Speed will not be a big factor" on the computers, says Gilbert. "Its critical portions will be during entry and landing. It has to solve problems in real time." One big question still under debate is whether to provide extensive back-up electronic systems on LEM. On the command module, NASA officials have generally favored in-flight maintenance rather than heavy redundancy of equipment. "It is quite possible we will favor this approach on LEM, which, compared to airplane missions, will be a long flight. It will have about six minutes of powered flight, although its overall mission may range from 10 to 48 hours. It makes sense, considering this, to consider in-flight maintenance." One exception may be the navigation-guidance systems. This will be one of the "very crucial" areas, as far as systems is concerned, and "you can bet it will be very refined and backed up where it has to be," Maynard says. PLANS—Although Project Gemini will be responsible for the nation's next manned space flight, Project Apollo is being pursued separately and "we can't wait for hardware" tested by Gemini, Gilbert says. NASA has officially said it hopes to land men on the moon by 1970. Nobody talks of "official schedules here, but there is hope that the big shot—sending the three-module Apollo to the moon for a two-man landing and return—can come as early as the end of 1967. To meet any kind of schedule such as this, Gilbert says, several Apollo earth orbital shots will be required within the next two to three years. Approximately 80 percent of the money spent for LEM and the command module will be for equipment common to both. Only 20 percent will be spent on specialized gear peculiar to the needs of one or the other. The design of the LEM computer, "now pretty well along," uses Fairchild's micro-logic circuits. Circuits contain three transducers and one resistor. It is a parallel machine packaged so it will have removable module trays. Some Stackpole resistors are not for you Stackpole COLDITE 70+® Fixed Composition Resistors roll down our line, and out to customers, by the tens of thousands every day. But does this mean sheer volume at the sacrifice of quality? On the contrary! It calls for one of the most rigid, comprehensive testing and quality control programs in the resistor industry... just to spot the one Stackpole resistor that is not for you (or any other Stackpole customer). Two of the tests are shown above. This is why you can specify COLDITE 70+ resistors without a second thought about their solderability, load life, humidity and temperature characteristics. And why they exceed every important MIL-R-11 requirement. If you're ever in our area, drop in and say, "Hello!" We'll be glad to chat about your design problem. Meantime, buzz your Stackpole distributor and prove our point: you get more out of Stackpole resistors because we put more in them. Or write: Electronic Components Division, Stackpole Carbon Company, St. Marys, Pennsylvania. ECLIPSE TO BARE IONOSPHERE'S SECRETS Experimenters converge on path of totality from Alaska to Maine By THOMAS MAGUIRE New England Editor ELECTRONIC INSTRUMENTATION on the ground, in jet aircraft and aboard rockets will be probing the ionosphere and "listening in" on the sun and moon tomorrow (July 20) during one of nature's rare spectacles—a total eclipse of the sun. Besides information on the structure and processes of the ionosphere, the eclipse is expected to yield data on the solar corona and chromosphere, airglow, sky brightness, and the exact size and shape of the moon. Some scientists will be using radiometric techniques in an attempt to get a millimeter-wave "picture" of the solar limb; and others will be checking once again Einstein's prediction that light waves passing near the sun are bent by its gravitational pull. In all parts of the North American continent, a partial eclipse will be observable. The total eclipse will be seen in a path from Alaska through Canada and central Maine. An estimated half-million professional scientists, amateurs and tourists will crowd into the most heavily populated section of the path of totality—a 55-mile-wide strip across Maine from Jackman to Bar Harbor. In this path, the totality will last about one minute. This eclipse has attracted unusual attention because of accessibility of the path of totality and its nearness to existing stations, approach of the International Quiet Sun Year, and relatively high angle of elevation of the sun and moon to stations on the North American continent. AIR FORCE — Cambridge Research Laboratories in Bedford, Mass., will concentrate heavily on this eclipse because of the proximity of the path to several of its highly instrumented field sites. Additional equipment has been set up in the Bangor, Me. area. At AFRL's Sagamore Hill Radio Astronomy Observatory in Hamilton, Mass., the 84-foot radio telescope will be used in a moon-bounce program in cooperation with the Army Signal Corps, the University of Illinois and Stanford University. Signals at 25, 50 and 150 Mc will be bounced off the moon before, during and after the eclipse. Receiving stations will measure Faraday rotation of electrons in the ionosphere and also total electron density changes. Measurements of ionospheric changes during a period of changing solar radiation constitute one of the most important observations to be made tomorrow. The new 150-foot dish at Sagamore Hill will be trained on the radio star Cassiopeia for scintillation studies as the ionosphere changes during the eclipse. CANADIAN ROCKET — From Fort Churchill, Canada, AFCRL will fire a Black Brant rocket carrying instrumentation for measuring variations in the D and E layers of the ionosphere during the eclipse. To measure changes in the D-layer, broad-band riometers located in Maine will receive Loran-C signals transmitted from a Coast Guard station on Nantucket Island. Effects on the D layer will also be measured by an instrumented KC-135 aircraft which will meet the eclipse over Fort Churchill, Canada. Also racing the moon's shadow will be Charles H. Smiley of Brown University, who will be observing from an F-104 jet plane which may be able to keep him in the shadow for five or six minutes as the dark image races eastward at about 3,300 mph. National Geographic Society and Douglas Aircraft are sponsoring a jet flight over the Canadian Northwest. NASA PARTICIPATION—Rocket firings will be made from Fort Churchill to probe the ionosphere as part of a NASA contract. The experiments, conducted by Geophysics Corp. of America, will simultaneously measure electron density and u-v and x-ray radiation under conditions of varying solar radiation. Hopefully, the experiments will provide a new insight into the precise effects of solar radiation on molecular particles and the temperature in the upper regions of the atmosphere. It is also hoped they will prove or disprove theories concerning the role of the sun's corona in generating the ionizing radiation that creates the atmosphere. NOW... Motorola Engineers Have Broken the Barrier Between Transistor Speed and Transistor Power! For the first time, the epitaxial concept has been applied to a power transistor to improve switching characteristics, to lower saturation voltage, and to increase gain. The result is the new Motorola high-frequency, high-current 2N2832-34 series germanium PNP device with collector-emitter voltage ratings of up to 160 volts, and 0.5 volt $V_{CE(\text{sat})}$ at 20 amps. This TO-3 packaged transistor will fill applications in television flyback circuits, inverters and converters, class C power amplifiers, ultrasonic generator circuits, and core drivers. Industrial and military designers of cathode ray tube deflection circuits will find that this new transistor excels in frequency, saturation voltage, transconductance, power, and switching time characteristics. For television flyback applications, a fall time of 0.7 $\mu$sec at 8 amperes of collector current is typical with this new device. The ring-emitter construction, very narrow base width, and conservative 20-ampere maximum current rating suggest applications in high-frequency inverter and converter circuits where there is a need to reduce transformer size and weight. \| Type Number \| $BV_{CEO}$ \| $BV_{EBO}$ \| $I_C$ (Typical) \| $F_{\text{Aa}} @ 1 \text{ Adc}$ \| $I_{CES}$ \| \|-------------\|------------\|------------\|-----------------\|-------------------------------\|----------\| \| 2N2832 \| 50V \| 2V \| 20A \| 100 \| @ $V_{ce} = 100V \ldots 20 \text{ mA}$ \| \| 2N2833 \| 75V \| 2V \| 20A \| 100 \| @ $V_{ce} = 140V \ldots 20 \text{ mA}$ \| \| 2N2834 \| 100V \| 2V \| 20A \| 100 \| @ $V_{ce} = 160V \ldots 20 \text{ mA}$ \| The 2N2832-34 power transistors are immediately available from your local Motorola Semiconductor Distributor or District Office. For complete technical information, write: Technical Information Center, Motorola Semiconductor Products Inc., P. O. Box 955, Phoenix 1, Arizona. "your most complete power transistor source" MOTOROLA Semiconductor Products Inc. BOX 955 • PHOENIX 1, ARIZONA A SUBSIDIARY OF MOTOROLA INC. D-63-043 One good electronic component... leads to another at Speer (Thanks to a Multi-Million Dollar Research and Development Program) At our R&D Center in Niagara Falls we are constantly developing new types of electronic components and improving the performance characteristics of our present types. We are also continually improving our production and quality control facilities throughout our various divisions. If you have a circuit design problem and want the cooperation of a modern, progressive company employing over 2000 people in the design and production of high quality components, call on a Speer representative. The broad line of electronic components manufactured by Jeffers Electronics and Speer Resistor Divisions includes: 1. Speer 1 Watt body tape-packaged resistors; 2. Jeffers phenolic coated PAC's; 3. Jeffers clear-coated PAC's; 4. Speer special leadless resistors; 5. Speer 2 Watt carbon composition resistors; 6. Jeffers Types 09, 15, 19, 22, 24, 28 and 30 molded inductors; 7. Speer ¼ Watt carbon composition resistors; 8. Jeffers special coils; 9. Jeffers JM 110 and 160 capacitors. Speer Carbon Co. Is A Division Of Air Reduction Company, Inc. Dept. 427, St. Marys, Pennsylvania PHASE II of a three-phase Air Force program to investigate joining materials in space by electron-beam welding techniques is now underway. Sponsored by the Aeronautical Systems Division Manufacturing Technology Division at Wright-Patterson AFB, the work is being carried out by Hamilton Standard div. of United Aircraft Corp. Objective of this phase of the program is to develop a prototype welding system—a project that will probably take a year to complete (Electronics, p 12, May 31). After that, in Phase III, the prototype system will undergo a series of operational tests in a simulated space vacuum environment. Initial phase of the program, which has just been completed, was aimed at defining the nature of the electron-beam system to be developed. The joining tasks that would be required in space were defined, the methods for using electron-beam welding in those tasks and the means of obtaining operating power were studied. APPLICATIONS—The Phase I study indicated that the main use of welding equipment in space will be to seal manned spacecraft structures to reduce losses of the vital on-board air supply (particularly on large, self-erecting space stations). Besides this, a fabrication and repair capability will also be needed in space for such things as meteoroid damages, docking and landing mishaps, station alterations, and rendezvous assembly operations. To handle such eventualities, the system was designed for mobility within the spacecraft while still being accessible to the power supply. Heart of the welder is a compact high-voltage welding gun column approximately 9 inches in diameter and 18 inches long. Power is fed to the welding gun through a flexible length of shielded high-voltage cable from a modular power supply and control package. Depending upon the nature of the power... Welder Begins source, the system is designed to operate in several modes, as shown in Fig. A. Complete independence from spacecraft power is possible for short durations by use of a rechargeable storage battery when operated in the "power pack" mode. SYSTEM SPECS—The welding system concept is shown schematically in Fig. B. In addition to the welding gun and its cable, the packaged power system (when operated in the power pack mode) will consist of a controls module, a high-voltage power supply module and a power pack which contains battery and inverter modules that can be removed to convert the system to other operational modes. The power system ultimately will fit into a package approximately $1 \times 2 \times 3$ ft, and the entire system will weigh about 350 lb. The electron-beam gun will normally be operated at 80 Kv and draw around 500 watts. But to provide adequate power for the variety of welding tasks that may be encountered in space, the system is being designed to operate at 80 Kv and 12.5 ma. The peak accelerating potential of 80 Kv results from a compromise among gun column size (and weight), weld penetration requirements, and the efficiency of the use of available power. Spectra-Physics LASERS within a millionth of an inch of perfection The difference between an ordinary monochromatic gas laser beam (left, above) and one with complete spatial coherence in a single phase wavefront (right) is very largely in the precision of the optics. The reflectors and Brewster's-angle windows in a continuous-wave gas laser, to qualify it as a precision laboratory instrument, require an optical finish of a very small fraction of a wavelength. You will find this precision in all Spectra-Physics CW gas lasers, a result not only of meticulous care in the preparation of the optics, but also of the practiced skill with which they are integrated into the instrument. The quality of a Spectra-Physics laser becomes immediately apparent when you turn on the power and observe the uniphase, truly coherent quality of its output. All its power is thus available for insertion into the diffraction limit of an optical system, with the important benefits of better collimation, greater intensity, and sharper focus. An impressive proportion of all significant laser work now underway is being conducted with equipment produced by Spectra-Physics. Among the many advantages now offered is complete assurance of reliability of operation, backed by a full year's warranty which includes even the plasma tube. Write for information on our Model 115, the higher-powered Model 112, and related accessories. We will also send you Laser Technical Bulletin #2, "Properties of laser resonators giving uniphase wave fronts." Address your inquiry to 1255 Terra Bella Avenue, Mountain View 3, California (or call collect (415) 968-4467). INDUSTRIAL SPY ARISTOCRAT OF THIEVES PART II He trades in ideas but is less clever than you would expect By DAN SMITH, Assistant Editor AS WE LEARNED from talking to experts at Norman Jaspan Associates, a management engineering firm that uses undercover agents to investigate the operations of client companies, the amount of industrial espionage in the electronics field should be a cause for alarm. In the past year, Norman Jaspan itself filed $1 million in claims to cover losses incurred by electronics firms from the theft of company secrets, and the total for the industry is undoubtedly much higher. How much, no one knows. Neither is it known how fast industrial spying is spreading—there simply aren't any reliable overall figures on the subject—but more general statistics indicate the activity is on the increase. So-called white-collar thievery was once a rarity; now it is the most common crime in the United States. "Compared to dishonest employees, the professional criminal is an amateur," the Jaspan firm says. "Recent Federal Bureau of Investigation figures indicate that the nation's armed robbers, burglars and auto thieves manage to steal about $500 million each year. White collar employees—rank and file, supervisory and executive—are stealing more than $4 million each working day in cash and property from their employers." The industrial spy is undoubtedly the aristocrat of white collar thieves, dealing as he does with the most esoteric material—ideas—and it is surprising to find how little ingenuity his modern brand of piracy requires. As with the research director who simply shipped out his company's secrets under the label of "Reprints," other wrongdoers described by Norman Jaspan seemingly needed little more to commit their larcenies than the desire to be dishonest. HIS OWN BOSS—One case involved a man who tried to destroy a company in revenge for what he considered a damaged ego. As added retaliation he walked off with the profits. Mr. X, principal of an electronics firm, sold the company to a larger organization with the stipulation that he would be appointed vice president in charge of electronics research. In time, however, he became disenchanted with the arrangement. He wanted to be his own boss, and he resented having to report to a professional manager who had no electronics background. To "get even" with his new associates for the situation he himself had arranged, he injected a critical modification into the prototype of an electromechanical item after it had proved out satisfactorily in the lab. The item, which had large market potential, was scuttled. Mr. X resigned, obtained financial backing, and on his own produced a commercial version of the item. It took the at your fingertips in Catalog 700... All These Relays and Hundreds More From Ohmite’s 12 Basic Models USED THROUGHOUT INDUSTRY: General purpose, industrial controls, mobile equipment, aircraft systems, commercial devices, and military equipment requiring 1 to 25-amp contacts. MANY CHOICES: High quality types, good economy units, unenclosed, plug-in, transparent enclosures, removable enclosures, dust-tight enclosures, hermetically sealed, miniature, sensitive, latching, indicator (light), U.L. approved, thyratron plate circuit models, and many special designs. FULL SELECTION OF TERMINALS: Quick-connect (push on), octal plug, screw type, solder, binding post, Jones plug, and banana plug. 338 TYPES CARRIED IN STOCK: Units stocked fall in the 5 to 25-amp contact range, and include latching types with transparent enclosures, U.L. approved models, hermetically sealed units, and thyratron plate circuit relays. Be sure to get a copy of relay Catalog 700 OHMITE MANUFACTURING COMPANY 3610 Howard Street, Skokie, Illinois Phone: (312) Orchard 5-2600 RHEOSTATS • POWER RESISTORS • PRECISION RESISTORS • VARIABLE TRANSFORMERS TANTALUM CAPACITORS • TAP SWITCHES • RELAYS • R.F. CHOKES • SEMICONDUCTOR DIODES electronics • July 19, 1963 CIRCLE 33 ON READER SERVICE CARD this is the new ZJ227 light activated controlled rectifier (it offers optional gate triggering—from either an isolated light source or direct electrical supply) this is a Controlled Avalanche Rectifier (it protects itself against destructive voltage transients... we have four different current ratings) this 200 volt C20B is the industry’s first $2.00 Silicon Controlled Rectifier for low-cost, high volume applications (press fit C22B sells for even less) What do these five rectifiers have in common? Each of these G-E developed devices can significantly improve existing circuit designs, or can open entirely new application areas. Our Application Engineering Center is always ready to show you how they can be used in your circuits. Just call your G-E Semiconductor Products District Sales Manager, or write Section 16G112, Rectifier Components Department, General Electric Company, Auburn, New York. In Canada: Canadian General Electric, 189 Dufferin Street, Toronto, Ont. Export: International General Electric, 159 Madison Ave., N.Y. 16, N.Y. AVAILABLE THROUGH YOUR G-E SEMICONDUCTOR DISTRIBUTOR GENERAL ELECTRIC ULTRA HIGH RELIABILITY tubular capacitors WITH A NEW DEGREE OF PRECISION SOUTHERN ELECTRONICS CORP. has long been a leader in the design and manufacture of high-precision tubular capacitors, and has pioneered in supplying them for critical applications in the classified military communications and other high-grade military and commercial equipment. They are made to the same standards as our high precision polystyrene capacitors so widely accepted for military applications. SEC tubular capacitors are manufactured under unusually critical quality control standards, resulting in tolerances as low as 0.5% in most values, and hermetic sealing guarantees accuracy over wide environmental changes. SEC tubulars are available in a wide range of tolerances to meet your needs, from 100 mmfd. to any higher value, in polystyrene, mylar, metallized mylar, teflon and dual-dielectrics. All SEC tubular capacitors meet or exceed the most rigid MIL-SPECS. In addition, we manufacture a complete line of tubular capacitors for commercial applications. Let us know your requirements. Write today for detailed technical data and general catalog. SOUTHERN ELECTRONICS Corporation 150 West Cypress Ave., Burbank, California A BOYHOOD FRIEND—A European scientist who had helped his firm develop an electronic measuring instrument bumped into a boyhood friend at a trade show where a rudimentary model of the instrument was on display. The friend, whom the scientist had not seen in years, was now an executive of a competitive firm. The two men reminisced about their youth and eventually the executive made an offer to the scientist. If the scientist would join forces with the executive, bringing along the secrets needed to manufacture the new instrument, he would be given a 50 percent interest in the executive's company. The scientist was receiving a substantial salary at his present company, in which he had previously been awarded a 10 percent interest. He, nevertheless, accepted his friend's offer. Within a short time, he and his new associates were giving his old firm considerable competition on the new instrument. MAINTENANCE COSTS — An East Coast electronics firm concentrated on turning out products that were the outgrowth of scientific breakthroughs, enjoying lead times of several months to a year over most of its rivals. There was one competitor, however, that consistently reduced these leads —so sharply that the firm became suspicious. It investigated and found that the head of its maintenance department was selling to the competitor the plans for production machinery, most of which had been specially designed and built by the firm's engineers. COMMON ELEMENT—Is there a common element in these cases? The Norman Jaspan organization thinks so. "This is basically a good management problem," we were told. "The best way for a company to prevent espionage is to run a good ship." Partly, this means laying down rules which will tell employees clearly what they can and can't do and adopting policies and techniques that will keep temptations to the minimum. But rules, locks, guards and classified files—which most firms have anyhow—don't go far. Norman Jaspan says: "In the final analysis, effective control hinges upon management's knowledge of what is actually taking place within the organization. It is a problem in communication, and TWO-WAY RADIOS FOR ARMY ARMY ELECTRONIC Material Agency is buying 10,944 portable two-way transceivers from RCA under a $9 million contract. Known as PRC-25, the unit weighs 17 lb. and has a five-mile range under combat conditions. It operates in 920 crystal-controlled channels of 50-Kc separation in the 80- to 76-Mc range. proven techniques are available to alert executives who understand the need for them." The Jaspan people also stress the importance of a company's attitude toward engineers and scientists, criticizing many firms for coddling them as "creative, highly emotional types" who shouldn't be subjected to the usual employee regulations. Jaspan also blasts firms on the opposite end of the spectrum: those that do not have the proper respect for the engineer and scientist and saddle him with make-work and other endeavors that rob him of his status as a professional. ULTIMATE PREVENTATIVE—Eventually even the shrewd, hard-headed men who run Norman Jaspan must talk more abstractly, and it is then, of course, that the trouble begins. No matter what safeguards are taken, no security program will succeed without the cooperation of the employees, and more than anything else, the moral atmosphere exuded by a corporation will influence the amount of cooperation received. Some people will be honest or dishonest no matter what, but the largest number will be affected by what they see going on around them. If management winks at executives who pad their expense accounts or juggle their inventories and efficiency ratings to make themselves look better, then it should not act surprised when workmen walk off with some of the tools and materials they handle or engineers and scientists sell to the highest bidder their knowhow. Ethical conduct will for the most part beget ethical conduct, although when it doesn't, the experience is likely to be a bitter one. It is up to individual managements to make sure companies are run on a high moral plane and, while help may be obtained from the outside, no outside conscience can ensure this. We would have to get even more airy if we were to pursue this subject much further. A discussion of the imperfectionality of the human race would probably come next—and that lies outside our scope here. Clean waveforms are characteristic of TI pulse generators. The Model 7505 provides coincident positive and negative 10 volt pulses into 50 ohms with unlimited duty cycle... and both pulses can be simultaneously controlled in width, rise time, fall time and delay. Amplitudes are independently variable and outputs are short-proof, with overload indicators and reset buttons. Specifications, briefly... rise/fall times from less than 20 to 500 nanoseconds; width 40 nanoseconds to 1 millisecond; delay 90 nanoseconds to 1 millisecond; repetition rates to 25 megacycles. Like all Texas Instruments pulse generators, the Model 7505 is compact, lightweight and portable, extremely convenient to use. Circuitry is all solid-state. Write for complete information. Your electronics BUYERS' GUIDE should be kept in your office at all times—as accessible as your telephone book. New silicone dielectrics New heat sink compound dissipates heat faster; improves performance, reliability Diodes, transistors, rectifiers... and light bulbs, too... operate cooler when Dow Corning 340 silicone heat sink compound is used. With triple the thermal conductivity of other materials, this new compound carries heat to the heat sink faster... devices operate longer, more efficiently, and with increased reliability. Formulated of heat stable silicones and thermally conductive fillers, this heat sink compound lowers equilibrium temperatures and assures reliable device performance. Dow Corning 340 silicone heat sink compound* is a grease-like material that fills all tiny air spaces, maintaining a positive seal between component and heat sink or chassis. This total contact means faster heat transfer, a uniform heat transfer path. Increased reliability is assured because this new silicone heat sink compound does not dry out, harden, gum or melt, even after long exposure to temperatures up to 200 C. Milk-white in color, it is chemically inert, won't attack or react with other materials, has low loss factor, low power factor, and excellent arc resistance. Optimum conductivity of the compound is demonstrated above. Dow Corning 340 silicone heat sink compound carries heat from base of bulb at right to heat sink faster... bulb base is 75 F cooler than bulb without heat sink compound. Typical Properties \| Dow Corning 340 silicone heat sink compound \| \|---------------------------------------------\| \| Color ........................................ White, opaque \| \| Consistency \| \| Penetration, worked and measured within one minute after working (ASTM D 217) \| 290 \| \| Bleed, percent after 24 hours at 200 C \| 0.6 \| \| Evaporation, percent after 24 hours at 200 C \| 1.0 \| \| Specific Gravity ......................... 2.85 \| \| Thermal Conductivity, K Factor 0.0015 gm/cal/sec/cm²/degrees C/cm \| CIRCLE 289 ON READER SERVICE CARD Dow Corning preserve design integrity New compound for transistor potting Dow Corning 18 semiconductor potting compound keeps junction temperatures at a minimum, cushions against shock, and absorbs moisture from within the cap to act as a moisture getter. These advantages... plus high centrifuge stability... combine to assure consistency and reliability of transistors, diodes, rectifiers, and other solid state devices. Supplied in sealed containers, this silicone-based, grease-like material combines high purity, low moisture content, high thermal conductivity, good dielectric properties, oxidation and evaporation resistance. These capabilities are maintained from -40 C to 400 C. CIRCLE 290 ON READER SERVICE CARD New resins protect device junctions Exceptionally high purity standards characterize three new Dow Corning transistor junction coating resins. Applied with a dropper or brush, they cure to form a tough protective film that seals out contaminants; assure junction integrity. Cure time is adjustable from 30 minutes at 250 C to 16 hours at 150 C, depending upon processing requirements and device heat limitations. Typical properties: viscosity, from 75 to 80 centistokes; specific gravity, 1.008; dielectric constant at 25 C, $10^2$ cycles per second, 2.23 to 3.22; dielectric strength, 2225 to 2380 volts per mil; volume resistivity, $1.0 \times 10^{18}$ to $5.27 \times 10^{15}$; surface resistivity, $1.38 \times 10^{13}$ to $9.33 \times 10^{13}$. CIRCLE 291 ON READER SERVICE CARD New fluid non-congealing at -110 F. Higher pumping rates, rapid heat transfer, and smaller, lighter pumps are among the design advantages made practical with Dow Corning 331 fluid coolant. This silicone fluid flows freely when other coolants are frozen solid, features an operating temperature range of from -130 F to 400 F, -90 C to 204 C. Other features: viscosity of 10 centistokes; low volatility; flash point of 420 F, 216 C, in a closed cup. Designed to meet MIL-S-27875, this new fluid also provides high dielectric strength, stable electric properties, high specific heat, hydrolytic stability, inertness, excellent thermal stability and oxidation resistance. CIRCLE 292 ON READER SERVICE CARD For information on these and other silicone materials, write Dept. 3931, Electronic Products Division, Dow Corning Corporation, Midland, Michigan. HOW WE SHRUNK TEFLON* Why Gudebrod’s Common Sense Approach to Lacing Problems Pays Dividends for Customers! Some years ago motor manufacturers had a problem! They required a high temperature lacing tape that would not deteriorate during the baking process of motor manufacture and would be practical in its application. Teflon offered the most practical solution to the problem since it provides a temperature range from -100°F to 500°F. We took teflon and flat braided it—we originated the process—but what about shrinkage? When teflon is baked it shrinks...it would cut thru fine motor wires! To meet this problem, we developed an exclusive pre-shrunk process for teflon. This patented process pre-shrinks teflon so that the maximum shrinkage is less than 3% after 16 hours at 425°F. We call this lacing tape Pre-Shrunk TEMP-LACE. Motor manufacturers use it in great quantities. Pre-shrinking teflon is but one of the many processes we have developed to meet the needs of customers. Whatever your lacing needs—nylon, glass, dacron, fungus proofing, color coding—Gudebrod’s common sense approach to the problem will pay dividends for you because 1. Gudebrod lacing tape increases production! 2. Gudebrod lacing tape reduces labor costs! 3. Gudebrod lacing tape means minimal maintenance after installation! 4. Gudebrod is quality—our standards for lacing tape are more exacting than those required for compliance with MIL-T! Write today for our Technical Products Data Book which explains the many advantages of Gudebrod lacing tape for both civilian and military use. Du Pont registered trademark for its TFE-fluorocarbon fiber. †Du Pont trade name for its polyester fiber. G U D E B R O D B R O S . S I L K C O . , I N C . Electronics Division 12 SOUTH 12th STREET, PHILADELPHIA 7, PENNA. Visit Gudebrod Booth #921 at the Wescon Show MEETINGS AHEAD MEDICAL ELECTRONICS INTERNATIONAL CONFERENCE, IFME, University of Liege, Liege, Belgium, July 22-26. ELECTROMAGNETIC MEASUREMENTS & STANDARDS SEMINAR, National Bureau of Standards; NBS Laboratory, Boulder, Colo., July 22-Aug. 9. AEROSPACE SUPPORT INTERNATIONAL CONFERENCE & EXHIBIT, IEEE, ASME; Sheraton-Park Hotel, Washington, D.C., Aug. 4-9. INTERNATIONAL ELECTRONICS CIRCUIT PACKING SYMPOSIUM, University of Colorado, et al; at the University, Boulder, Colo., Aug. 14-16. WESTERN ELECTRONICS SHOW AND CONFERENCE, WEMA, IEEE; Cow Palace San Francisco, Calif., August 20-23. DATA PROCESSING NATIONAL CONFERENCE & EXHIBITION, Association for Computing Machinery; Denver Hilton Hotel, Denver, Colo., Aug. 27-30. AUTOMATIC CONTROL INTERNATIONAL CONGRESS, International Federation of Automatic Control; Basle, Switzerland, Aug. 27-Sept. 4. MILITARY ELECTRONICS NATIONAL CONFERENCE, IEEE-PTGMIL; Shoreham Hotel, Washington, D.C., Sept. 9-11. ELECTRICAL INSULATION CONFERENCE, IEEE, NEMA; Conrad-Hilton Hotel, Chicago, Sept. 10-14. JOINT ENGINEERING MANAGEMENT CONFERENCE, IEEE, ASME, et al; Biltmore Hotel, Los Angeles, Sept. 12-13. INTERNATIONAL ASSOCIATION FOR ANALOG COMPUTING, AICA; Brighton College of Technology, Lewes Rd., Brighton, England, Sept. 14-18. INDUSTRIAL ELECTRONICS ANNUAL CONFERENCE, IEEE, ISA; Michigan State University, East Lansing, Mich., Sept. 18-19. PHYSICS OF FAILURE IN ELECTRONICS SYMPOSIUM, Armour Research Foundation and Rome Air Development Center, Illinois Institute of Technology, Chicago, Sept. 25-26. ADVANCE REPORT AEROSPACE ELECTRO-TECHNOLOGY INTERNATIONAL CONFERENCE, IEEE, et al; Phoenix, Ariz., April 19-25, 1964. Aug. 19 in the deadline for submitting a 250-word abstract to Aerospace & Electronics Technical Program Chairman, Mail 4016, The Martin Company, Baltimore 3, Md. Papers representing original contributions in the following fields of aerospace technology are invited: electronic systems (including mechanical and packaging design); electronics systems (including integrated electronics); energy conversion (including cryogenics and superconductivity); thermoelectricity (including device construction and performance). MTBF > 10,000 Hours That's the kind of reliability you can expect from this new Borg Frequency Standard. What's more... IT'S MODULAR—You can easily plug in extra modules, multipliers up to 100 mc or dividers down to 1 pulse per second . IT'S COMPACT—2 units fit 5¼" panel space. IT'S STABLE—Down to $5 \times 10^{-12}$ for short term (see specs below for long and short term stability). And it is designed to meet MIL-E-16400D. Fully transistorized, 1560 Series offers the highest demonstrated reliability of any commercially available frequency standard. Stability? Dual ovens hold the 5mc overtone crystal temperature to within ±0.005°C. All semiconductors are silicon or tetrode type. Rechargeable standby battery (good for +10 hrs) lets you carry unit to remote project sites or airship it anywhere without shutdown. Oven and oscillator available as a separate module for your system. If you need precision timing—for communications or weapons systems, navigation, radar, laboratory, or commercial use—you're a postage stamp away from the best source. Borg Equipment is the world's largest manufacturer of precision frequency standards. BORG EQUIPMENT DIVISION Amphenol-Borg Electronics Corporation, Janesville, Wisconsin SPECIFICATIONS: Transistorized Frequency Standard, Borg 1560 Series FREQUENCY STABILITY Long Term: $5 \times 10^{-10}$ per day after 21 days operation Short Term: $5 \times 10^{-12}$ rms average of 1 second counts OUTPUT FREQUENCY 5mc plus options (1 pps to 100mc) HARMONIC DISTORTION —40db FREQUENCY ADJUSTMENT Fine $1 \times 10^{-7}$ digital dial, 1 count equals $1 \times 10^{-10}$ Coarse $1 \times 10^{-8}$ METERING 12 selected circuits TEMPERATURE RANGE $-20^\circ C$ to $+50^\circ C$ ($-4^\circ F$ to $122^\circ F$) POWER CONSUMPTION 14 watts from 60 volt line DELIVERY in 30-60 days Some budget-minded project engineers, finding a gaping hole in their appropriations, have ordered Borg Frequency Standards, knowing that they can get plug-in multipliers, dividers, power supplies, and battery packs later. CIRCLE 41 ON READER SERVICE CARD cut your identification costs with modern methods like these Looking for faster identification... imprints that meet government specs... flexibility to meet changing requirements... a practical way to mark prototype or sample lots? Would direct printing or making your own labels cut costs, give higher print quality? Ask Markem to analyze your identification needs now — without obligation — and show you how and where you can save money. Frequently, a Markem in-plant identification system pays for itself in a few months. Contact the Markem Technical Representative near you, or write Markem Machine Co., Keene 5, New Hampshire. New 12-page Catalog describes Electrical/Electronic Identification Methods and Equipment. Copy on request. MARKEM TECHNICAL ASSISTANCE AND INDUSTRY-PROVEN EQUIPMENT WORLDWIDE... TO HELP YOUR PRODUCT SPEAK FOR ITSELF VARIAN KLYSTRON LOGS 42,000 HOURS SERVICE A Varian VA-220E klystron, Serial No. 59366, was installed in August, 1957, in a television repeater transmitter on Spruce Mountain near Wells, Nevada. Still operating after 37,000 hours, the tube was removed from service in November, 1961. Television Microwave, Inc., operators of the four-channel, three-repeater TV relay system, later installed this tube in its Wells transmitter, and obtained 5,000 more hours of unattended service. The tube is still good. ■ Long-life performance is quite usual for Varian tubes; Television Microwave, Inc., reports that its Varian tubes are seldom replaced before 30,000 hours of operation. ■ For assured long tube life, you can depend on Varian. MICROWAVE TUBE GROUP: PALO ALTO TUBE DIVISION • BOMAC DIVISION • S-F-D LABORATORIES, INC. • SEMICON OF CALIFORNIA, INC. VARIAN ASSOCIATES OF CANADA, LTD. • SEMICON ASSOCIATES, INC. EUROPEAN SALES HEADQUARTERS: VARIAN A.G., ZUG, SWITZERLAND electronics • July 19, 1963 CIRCLE 43 ON READER SERVICE CARD 43 8 SCIENTISTS AND ENGINEERS: From undersea navigation equipment for test on the research vessel Trieste to the development of non-inertial systems for deep space rendezvous, Motorola performance spans the broad spectrum of environments in advanced systems. In the area between, current programs include: advanced random access communications and side looking radar surveillance systems for the Army...air-to-air missile guidance and digital command systems for the Navy...data transfer and high speed teleprinting systems for the Air Force...satellite tracking, telemetry and instrumentation for NASA and extensive company funded R&D projects. Scientists and engineers interested in joining an electronics company with versatile interests unbounded by narrow specialization write today describing your background and training in: ZERO TO INFINITY Systems Design • operational and functional mission analysis, optimum time-bandwidth utilization, redundant system organization for reliability, phased arrays, digital and voice communications, and A-J secure communications. Equipment Design • solid state receivers, transmitters and transponders, distributed parameter microwave equipment, digital Modems, controls and displays, low-level switching circuits, and digital data processing circuitry. Familiarity with State-of-the-Art • spread spectrum techniques, visual spectrum intelligence transmission, statistical theory of communication, integrated circuit applications, multiple logic element techniques, organization of digital data handling systems, correlation and phase-lock techniques, and coding and modulation. We are particularly interested in programs on which your experience was obtained and the extent of your technical responsibility. Please address this information to our Manager of Engineering for immediate and confidential attention. MOTOROLA An equal opportunity employer Military Electronics Division Leach Balanced Armature Relays are the smallest relays on the market today that solve your really difficult application problems. For example, this relay (actually only 1.8 inches high) withstands 2000 foot lbs. of shock loading and vibration of 30 G’s to 2000 cps to meet Mil-S-901B for equipment-mounted components. Maximum benefits in minimum packaging! No smaller relays answer all these demands: radiation resistance, perfect performance at extreme altitudes, reliable contact switching, three phase rating, minimum voltage spike generation, high terminal strength, minimum current capability, no maintaining power required, grounded AC ratings or gold alloy contacts for dry circuit applications. Whatever your specifications, there’s a Leach Balanced Armature Relay to meet your requirements. If size is really the problem, ask about our subminiature relays. That’s why we urge you to look to Leach. You name your problem. We’ll take it from there. LEACH CORPORATION 405 Huntington Drive, San Marino, California • Export: Leach International S. A. Micropower circuits combining thin-film and diffusion technology are now limited to operating below 1-Mc, but higher speeds appear feasible with further size reduction. Circuit design for micropower—both analog and digital—promises to be on a par with conventional design. By W. W. GAERTNER and M. SCHULLER, CBS Laboratories, Stamford, Conn. BY BRINGING power consumption of individual circuit components down to microwatts, packaging densities of one million components per cubic inch become feasible since the heat generated is small and the circuits operate nearly at ambient temperature. Space electronics with a density of $\frac{1}{4}$ million components per cubic foot have already been delivered to NASA and much greater size reduction is possible. The technique developed to produce microwatt circuits integrates planar diffusion technology and deposited thin-film technology on the same silicon wafer. Microwatt power does not represent a lower limit for the process and even nanowatt power appears feasible. Nanowatt operation would bring micro-electronic circuits into the range of power of biological systems. TECHNOLOGY — Integration of planar diffused technology and deposited thin-film technology on the same silicon wafer is indicated in Fig. 1. All the conventional processes of epitaxial growth, diffusion and alloying are performed first and then the grown silicon oxide on the wafer surface is used as a substrate for depositing thin-film components. Finally, an inert passivation layer is deposited over the thin-film components to protect them from mechanical and chemical damage. The result is a fully passivated function block, as in conventional microelectronics, but with thin-film components included. The photograph on the cover of this issue shows such a surface-passivated thin-film semiconductor circuit wafer being subjected to a hot dichromic acid test. The circuit is unaffected. Thus the surface-passivation technique is an important step towards eliminating the present bulky package around the small semiconductor wafer. This increase in allowable packing density is not accompanied by a destructive temperature rise. The new technology brings several additional degrees of freedom to the microelectronics field and offers some distinct advantages. - High Resistor Values—Micropower operation requires resistances of the order of megohms. With a sheet resistance of only 400 ohms per square and a line width of 1 mil, a 1-megohm resistor covers an area of approximately $70 \times 70$ mils. A diffused 1-megohm resistor would take up twice this area. - Low RC Time Constants—All microelectronics is plagued by stray capacitance between the closely spaced circuit elements and the bulk of the wafer, resulting in undesirable coupling, spurious oscillations and R-C time delays that slow down the circuit significantly at all power and impedance levels. It is particularly noticeable, however, in low power, high resistance circuits. The capacitance of silicon $p-n$ junctions, which separate diffused resistors from the bulk and from other circuit elements, is typically about 0.2 pf per sq mil. But the capacitance across a 1-micron thick layer of grown silicon oxide that separates a deposited metal-film resistor from the substrate is only 0.024 pf per sq mil. Thus the time constant associated with a thin-film resistor of the same value and size as a diffused resistor is almost 10 times lower. In addition, the size of the deposited resistor itself will usually be at least 50 percent smaller. Furthermore, the silicon-oxide layer can still be made substantially thicker whereas there is a natural limitation to the thickness of a $p-n$ junction depletion layer. Therefore the combination thin-film semiconductor technology can produce faster microcircuits at all power-levels and not only in the microwatt range. Thin film resistors deposited on silicon oxide over silicon compare even favorably in response time with those deposited on ceramic substrates because the latter often have high dielectric constants which lead to high coupling capacitance between the closely spaced resistor lines. - Component Independence—Since the fabrication of transistors and resistors—and sometimes of capacitors—occurs at different times, their characteristics are largely independent and each can be changed and optimized without affecting the others. - Versatility—A direct result of the complete independence of transistor and resistor characteristics is design versatility. A number of universally useful transistors can be diffused into a silicon wafer and then various resistor and interconnection patterns can be added to produce a variety of different circuits for different speed and power ranges. When the transistors are fabricated it is not necessary to know whether a digital or analog circuit is to be built, or which package leads will be used for which purpose. Figure 2 shows six different resistor and interconnection patterns added to a basic transistor pair to form a variety of medium-power analog and digital circuits. Another example is given in Fig. 3, where several different NOR gates and a pair of emitter-followers have been evaporated onto the same transistor pattern. This not only aids efficient production but also allows quick reaction during circuit development. If a fast mask-making facility is available, a new circuit design can be ready for testing in microelectronic form in one day. New resistor and interconnection patterns are evaporated onto an existing universal transistor wafer, then tested after being scribed, mounted and bonded. - Tight Tolerances—In some applications tight tolerances on components must be achieved. Resistor tolerances are a function of line width and sheet resistance. Tight tolerances are achieved by monitoring the sheet resistance of a thin-film resistor during deposition and stopping the evaporation as close as possible to the desired value, as well as by being able to predict the small resistance changes which occur during high-temperature stabilization. Readily achievable tolerances range at the present time from ±5 percent with a line width of 1.5 mils and a sheet resistance of 400 ohms per square to fractions of 1 percent with a line width of 5 mils and a sheet resistance of 100 ohms per square. These tolerances are continuously being improved. Since there is a direct relationship between component tolerances and speed and power consumption of a system, performance superior to systems produced with diffused components alone is expected. The accuracy of resistors evaporated on the same substrate is within a few tenths of one percent. Figure 4 shows a high precision resistor ladder network for microelectronic analog-to-digital converters. - Temperature Stability—In applications such as f-m telemetry, temperature stability of components is an absolute necessity. In other applications it is desirable because it increases the speed and decreases the power consumption of the entire system. Figure 5 shows the temperature variation of a passivated thin-film resistor deposited on silicon oxide over silicon. Its temperature coefficient is 10 to 100 times better than diffused semiconductor resistors. - Resistor Materials—Any resistor material that can be evaporated or sputtered can be used in the thin-film semiconductor technology; the choice depends on sheet resistance, temperature coefficient or other property desired. Nickel-chrome, chrome and tantalum are useful for many applications. - Scaling—Many circuits depend on the ratios of resistances rather than on their absolute values. The speed, power consumption and fan-out capabilities of a thin-film semiconductor microcircuit can be changed over wide ranges by depositing different amounts of resistor materials onto the same pattern. - No High-Temperature Leakage—Unlike diffused semiconductor resistors, thin-film resistors on silicon oxide over silicon show no leakage currents between the resistor and the bulk of the wafer even at high temperatures and with large resistor areas. - Heat Conduction of Substrate—The high heat conductivity of the silicon substrate as compared to ceramic substrates equalizes hot spots in the microcircuit much better, so permissible power dissipation for a resistor covering a given area is much higher. Figure 6 shows the long term stability of these resistors under loading at elevated temperatures. - Extremely Pure Substrate—The extreme chemical purity of the grown silicon oxide substrate is considered a favorable factor for the long-range high-temperature stability of the thin film resistors. - Low Operating Temperature—In microwatt operation and at packing densities dictated by present packaging and interconnection schemes, the microelectronic function block remains at ambient temperature rather than experiencing the significant temperature rise typical in systems consisting of uncooled tightly packed conventional microelectronic packages. This low operating temperature is expected to contribute greatly to reliability. - Packing Density—Since one million transistor stages operating at one microwatt each will dissipate only 1 watt, micropower circuits alone will allow the size reductions previously predicted on the basis of size alone. Already delivered to NASA is microelectronic space hardware with a packing density of ½ million components per cubic foot, and an increase of many orders of magnitude is still feasible. - Secondary Advantages—Indirect benefits derive from microAPPLYING MICROPOWER Microwatt and nanowatt circuits should go a long way toward freeing the circuit designer from the problems of heat dissipation and circuit volume, at least so far as the signal processing and computing part of the circuit is concerned. One result is that highly sophisticated techniques can be developed for military and space applications—and eventually for more prosaic uses. PROBLEMS—Whereas the desirability of very low power operation and of complete integration of thin-film and semiconductor techniques is almost self-evident, the technological problems in the practical realization of these concepts have not been trivial. - Transistor Quality—To achieve microwatt operation the microelectronic function block must contain transistors with high current gain at an emitter current of 1 microampere and a collector voltage of less than 1 volt. Such transistors, with high capacitances, are now commercially available but are not necessary for and are not used in conventional milliwatt microelectronics. Since high-quality performance is required of adjacent transistors on the same function block, the overall yield of the microwatt transistor production process must be high. Furthermore, in the interest of speed the high gain at low current must be combined with low junction capacitance. Since decreasing size increases the surface-to-volume ratio of the junctions, resulting usually in lower current gain at low currents because of surface recombination at the junction, high precision in mask line-up and photoresist processing is required. The transistors shown in Fig. 2 and 3 have betas as high as 80 at one microampere collector current and junction capacitance of 5 pf at zero volts. - Substrate and Resistor Quality—In ordinary planar and microelectronic technology occasional pinholes and oxide flaws are not immediately harmful except near the junctions and may only later result in poor reliability. In the thin-film semiconductor technology, however, the oxide layer covering the silicon wafer is the substrate for the thin-film components and must be perfect and uniform over large areas, especially for resistors with long narrow lines closely spaced and for large capacitors. For resistors an irregularity may cause an immediate break in a line, or a weak spot which opens up under high load and temperature. Or, the resistor may have leakage or a short to the substrate. In capacitors, imperfections cause leakage and low breakdown voltage. Equally important to the quality of the silicon substrate are the controls on thin-film deposition. Since the substrate wafer contains transistors and diodes worth thousands of dollars, the yield of the deposition process must be near 100 percent. Subsequent trimming of resistors by abrasion, customary with some conventional thin-film resistors, is impossible and all components must be deposited within tolerances, and subsequent accelerated aging must not bring them outside tolerance. - Compatibility of Technologies—Thin-film deposition requires cleaning the substrate with highly reactive chemicals and heating it to high temperature for a prolonged time. Cleaning, however, must have no ill effects on the transistors and diodes in the substrate. For micropower operation the transistors must maintain their high current gain throughout the treatment. Thin-film components must withstand thermocompression bonding temperature. Both these steps present difficulties. - Effect on Cost—Several factors contribute to the present high cost of micropower microelectronics: the quality of the components required; the extra fabrication steps required by the addition of the thin films; the size of the megohm-range resistors which allows only a relatively small number of circuits on one wafer. With improved understanding and automation of all processes, however, and control of higher and higher sheet resistances, the cost of micropower microelectronics will come down close to that of conventional milliwatt. because of component independence—Fig. 3 electronic circuits. - Reliability—This is one of the main reasons for using microelectronic building blocks. Thin-film semiconductor microelectronic technology is at this point too new for extensive reliability data and the technology itself still changes rapidly to take advantage of possible improvements. An unusually high degree of reliability is expected, however. It is well established that an increase in operating temperature causes a decrease in reliability, and since micropower microelectronic blocks operate virtually at ambient, they should eventually have the highest reliability possible at a given ambient—higher than any type of circuit that raises its operating temperature significantly due to heat dissipation. The main source of reliability is in the fabrication processes. Although these are proprietary a few general rules can be mentioned. All work must be performed in white rooms with humidity, dust and temperature control; all fabrication should be carried out at the highest possible temperature to eliminate subsequent changes of component values in the operating range; the entire circuit is covered by a tough passivation layer such as silicon oxide, which allows the circuit to be submerged in hot dichromic acid $(\text{H}_2\text{SO}_4 + \text{H}_2\text{O} + \text{Na}_2\text{Cr}_2\text{O}_7)$ for almost an hour without damage to the thin-film components (see cover); each circuit chip can be monitored through all fabrication steps into final packaging. The letters and numbers on the circuit chips in Fig. 2 and 3 identify the location of the chip on a wafer, which may contain 40 to 1,000 circuits. These tight controls lead to high yield, which is another prerequisite for high reliability. NANOWATTS—Microwatt operation is not a limit dictated by semiconductor physics and a reduction of supply power by several orders of magnitude into the nanowatt range appears feasible. This would bring microelectronics down into the biological-supply power range. Several research projects are presently underway to penetrate the nanowatt range, with emphasis on the following. - Improved control on diffusion processes to achieve a high yield of transistors with sufficient current gain at emitter currents of 1 na - The exploration of stable thin-film resistors with a sheet resistance of many thousands of ohms per square - The replacement of visible light by electron and ion beams as the major technological tools to produce the minute geometries necessary for high speed at very low power. Transistors with a maximum area of $3 \times 3$ microns are forecast whose details could not even be observed under an optical microscope. Experimental electron and ion beam systems for such purposes are shown in the photographs. SYSTEMS—Micropower microelectronics has been applied to several subsystems of moderate complexity, some already delivered for operational systems use; one application involves over 300 transistors and 600 resistors and capacitors, all on micropower function blocks. The applications have led to some general design rules and procedures. - Micropower microelectronics lends itself to all types of analog and digital circuits. Transformers and inductances must be designed out of the circuits. Field-effect transistors and active R-C filters may sometimes be used. - The large resistance values required, together with the $p-n$ junction and stray capacitances in the circuit, limit the operating frequency to below 1 Mc. This limitation will be lifted as the size of the microcircuits decreases. Higher power circuits close to the frequency response of the transistors can be built. - Interfaces between micropower microelectronics and conventional- MONO-ENERGETIC* ion beam emerges from slit in top plate. Oven for generating neutral atomic beams is at lower right; evaporation well is under screw cap on top of oven component or microelectronic circuits operating at milliwatt level present no problem. Micropower function blocks can readily be driven by milliwatt circuitry and interface amplifiers (emitter-followers, see Fig. 3) can be provided wherever a micropower circuit must drive a milliwatt circuit. These amplifiers typically consume more power than all the microwatt circuits preceding; thus it is desirable to use microwatt circuits in as much of the system as possible. - Power consumption in each circuit of a system should be minimized to provide only the speed and fan-out required and not more. Thus in a typical system the resistance values and power consumption of circuit blocks of the same type may vary by more than a factor of ten. In a binary counter chain, for example, the power consumption of each successive stage can be reduced by 2 because less speed is required. In many systems most circuits require only a fan-out of about 4 and these circuits can be designed for much lower power consumption than the few high fan-out circuits. By changing the amount of metal deposited on the resistor pattern, and thus changing the sheet resistance and the resistance value, or by adding a different resistor pattern to the basic transistor pattern, speed, fan-out and power consumption can be varied continuously over 4 orders of magnitude. The exact minimum power circuit desired can thus be realized. - Micropower circuits can be used today where the reduced power and increased reliability justify present high prices. At the present time, with micropower components not yet generally available, circuit design should be done in collaboration with a fabricator of micropower circuits. The first step is to build a systems breadboard with conventional components but using the identical transistors that will later be used in the microelectronic function block, thus avoiding discrepancies in temperature, voltage and bias-current dependence of parameters. Next, the conventional resistors and capacitors are replaced by individual diffused and thin-film resistors and capacitors on a silicon substrate. This is the first step towards taking stray capacitance into account. Adjustments are made in the breadboard using conventional components for trimming. Then partial integration is accomplished by putting five to ten active and passive elements on the same silicon chip. Taps are provided on the resistors and capacitors and test points are brought outside the package in addition to the input, output, power supply and ground connections. Again conventional components or individual microelectronic components on silicon chips can be used for trimming. When the circuit design is frozen the circuit is fully integrated on a single silicon chip. Engineering samples of various micropower logic and analog stages will soon be available for systems design and will provide short cuts in circuit development. Experience in designing micropower systems accumulates rapidly; within a year circuit design will probably be no more difficult than with conventional components. C. Heizman, J. Kostelec, C. Levy, E. Littau, W. Meyer and M. Urban have been instrumental in the design and fabrication of the micro-watt function blocks. The electron-beam system and the ion gun have been designed and constructed by A. Andrews and F. Cook respectively, under sponsorship of the U.S. Naval Avionics Facility in Indianapolis and the U.S. Army Research and Development Laboratory, Fort Monmouth, New Jersey. Most of the other work has been sponsored by the National Aeronautics and Space Administration, and the U.S. Air Force. BIBLIOGRAPHY (1) W. W. Gaertner, M. Schuller, C. Heizman, C. Levy, Microwatt-Microelectronics, paper presented at the Northeast Electronics Research and Engineering meeting, Boston, Mass., Nov. 1961. (2) W. W. Gaertner, M. Schuller, C. Heizman, C. Levy, Microwatt-Microelectronics, lecture to Washington Chapter of PG&E, Washington, D. C., Nov. 27, 1961. (3) W. W. Gaertner, M. Schuller, C. Heizman, C. Levy, Microwatt-Microelectronics, lecture to Merrimack Valley Subsection of the IRE, North Andover, Mass., Jan. 15, 1962. (4) W. W. Gaertner, M. Schuller, C. Heizman, C. Levy, Microwatt-Microelectronics, Electronic Technology, Feb. 1962. (5) W. W. Gaertner, C. Helzma, M. Schuller, C. Levy, Microelectronics: Microwatt Power Amplifier and Interface Circuit Function Blocks for Space Applications, paper presented at 1962 National Symposium on Space Electromechanics and Telemetry, Miami Beach, Fla., Oct. 1962. (6) W. W. Gaertner, C. Helzma, C. Levy, M. Schuller, Microelectronics: Micropower Digital and Low Level Amplifiers for Space Applications, Spaceborne Computer Engineering Conference Proc., p 151, Anaheim, Calif., Oct. 1962. (7) W. W. Gaertner, Nanowatt Devices, paper presented at the 1962 Electron Devices Meeting of the IRE, Washington Special Group on Electron Devices, Wash., D. C., Oct. 1962. (Six descriptions of the CBS technology have also been given in the following Journals: Giants Jump Into Race for Tinier Electronics, Business Week, Aug. 25, 1962; Microelectronics: Computer Block, Electronic News, Oct. 1, 1962; Micromin, Communication Strive to Meet Space Needs, Electronic News, Nov. 25, 1962; Nanowatt Range Device Developed by CBS Labs, Electronic News, Oct. 29, 1962; Recap for Students, Engineers, Electronic Design, Nov. 22, 1962; and Now-A Nanowatt Circuit Design, Electronic Design, Nov. 22, 1962; Nanowatt Power: How Low Can It Go?, Electronic Design, Feb. 15, 1963; Advances in Microminiaturization, Electronics, Feb. 15, 1963. ANOTHER NEW COMPONENT HOT CARRIER DIODES Switch in Picoseconds Metal-semiconductor diodes increase switching speed now limited in p-n junctions by minority-carrier storage. Devices need very pure materials and improved epitaxy. By S. M. KRAKAUER, Applications Engineer S. W. SOSHEA, Project Leader HP Associates, Palo Alto, California EPITAXIAL SILICON hot carrier diodes were introduced by HP Associates only recently, but many of the principles of rectifying metal contacts have been known for decades. The great advances in germanium and silicon p-n junction devices in the last 15 years have tended to obscure the potential of metal-semiconductor contacts. The p-n junction diodes are, however, approaching the limit of their high-frequency performance, because of storage of minority carriers. Since minority carrier storage is virtually eliminated, metal-semiconductor diodes show renewed usefulness. The development of the modern hot carrier diode was made possible by the availability of very pure semiconductors, by improved techniques of surface cleaning and passivation, and by the epitaxial construction method. A hot carrier diode can be made in a variety of ways. A typical epitaxial type is shown in Fig. 1. Experimental models have been made on silicon using evaporated gold, platinum, palladium, silver and many other metals. Both hot electron (on n-type silicon) and hot hole (on p-type silicon) forms are possible, but the hot electron type is generally preferable because the higher electron mobility gives better high frequency performance. OPERATION—Hot carrier diode operation and the distinction between it and a p-n junction can be understood most clearly by means of the appropriate electron energy diagrams. Figure 2A shows the electron energy diagram of a hot electron diode with a Schottky-type barrier, and Fig. 2B shows the corresponding diagram for an abrupt p-n junction diode with the n-type region more heavily doped than the p-type region. When the hot carrier diode is forward biased, the electrons in the n-type semiconductor HOT ELECTRON DIODE construction differs from that of hot hole where substrate would be p-plus material and epitaxial layer p-type silicon—Fig. 1. diffuse over the barrier and are injected into the metal. The injected hot electrons interact with the lattice and the electrons of the metal and when the diode is reverse biased, these hot electrons are unable to surmount the barrier, so they do not contribute to the stored charge. When, however, the $p-n$ junction is forward biased, the electrons diffuse into the $p$-type region and build up to a concentration that is limited by the rate of carrier recombination, as depicted in Fig. 2B. When the $p-n$ junction is reverse biased, the stored electrons (minority carriers) flow back into the $n$-type region, thus lowering the rectification efficiency if the diode is used as a detector, or increasing the reverse recovery time if it is used as a switching diode. The current-voltage characteristics of hot carrier diodes can be described very closely by the ideal diode equation $$I_r = I_s [\exp(qV/kT) - 1]$$ in which the saturation current $I_s$ is proportional to $\exp(-qV_s/kT)$. The type of metal can be conveniently selected to have an internal barrier $V_s$ from 0.3 to 0.8 volt, corresponding to a saturation current, for a typical diode size (about $6 \times 10^{-4}$ cm$^2$), from $10^{-11}$ to $10^{-4}$ amp. The junction capacitance of the hot carrier diode varies as the inverse square root of voltage and is only slightly dependent on $V_s$. This combination of characteristics is analogous to a family of $p-n$ junctions of incrementally varying energy band gap and provides the circuit designer with an added degree of freedom that was not previously available. The reverse characteristics of hot carrier diodes appear very similar to those of $p-n$ junction diodes. The reverse leakage current increases with reverse voltage gradually, owing to the internal Schottky effect, until the avalanche multiplication voltage is reached. Hot carrier diodes are similar in concept and in operation to the ideal point-contact diode in which the contact is neither formed, alloyed nor bonded. Both the hot carrier diode and the ideal point contact diode employ a Schottky barrier, but there are many notable differences. Being of much larger area, the hot carrier diode has larger capacitance than the point contact, but it can handle greater power and is less sensitive to current transients than is the ideal point contact. The hot carrier diode, furthermore, is more stable mechanically and has more nearly ideal and reproducible electrical characteristics. PERFORMANCE—Recovery time as a function of minority carrier lifetime for these diodes is difficult to measure. The lifetime is so low that its influence is readily obscured by diode and circuit impedances and by transient response anomalies in the associated instrumentation. It has been found best to characterize the recovery time for these diodes relative to a sinusoidal excitation. The circuit is shown in Fig. 3, the resulting oscilloscope patterns in Fig. 4. The effective minority carrier storage can be related to the amplitude of the negative spike. Diode capacitance causes the baseline to tilt, and so capacitive conduction can be separated from storage conduction by measuring the spike amplitude from this tilted reference line, as shown in upper Fig. 4. Measurement using this technique is not completely quantitative, but it gives a convenient index of the diode recovery that corresponds to most applications. If, for example, the signal generator and amplifier are adjusted to 53 mc with sufficient output to produce a peak forward current flow of 20 ma and scope gain set to give a 5-cm deflection for the positive peak, then the amplitude of the negative spike (read as shown) will be related to lifetime as $\tau = 500$ ps/per cm for deflections less than 1.5 cm. This value is an effective rather than a true minority carrier lifetime. It is essentially the product of true minority carrier lifetime by the ratio of minority to majority carriers that is associated with forward conduction. This ratio is made smaller with reduced barrier height and reduced substrate resistivity. Currently available diodes have effective lifetimes below the resolution capability of this measurement (<50 Sufficiently high values of forward current can cause minority carrier injection, and storage. Hot carrier diodes now available are listed in the table. The static characteristics of these diodes, both in forward and reverse, are similar to conventional $p-n$ junction diodes. The type 2001 resembles a conventional silicon $p-n$ junction diode, and the type 2101 resembles germanium, as shown in Fig. 5. To take full advantage of their fast response capability, care is necessary in mounting these diodes. Minimum possible lead length will reduce to a minimum performance degradation owing to shunt capacitance and series inductance. The self-inductance of the present package is approximately 3 nh. A lower inductance package is under development. In general the same considerations that apply to the application of conventional $p-n$ junction diodes will apply to the hot carrier diode. The differences between them is confined to the lower storage and wider choice of barrier height that is associated with the hot carrier diode. Accordingly, hot carrier diodes might be substituted in many existing circuits without design modifications being required, and with a substantial gain in performance. NO DELAYS—Minority carrier storage in the hot carrier diode is so low that the turn-on and turn-off delays that are present in conventional $p-n$ junction diodes will be essentially eliminated. Accordingly, hot carrier diodes can be used effectively in those pulse and high frequency applications where lag-free response is required, such as detection, mixing and limiting at microwave and high frequencies. Within fractional nanosecond limits they can be used for clamping and gating rapidly. Freedom of choice in barrier height leads to applications that may or may not also require low storage. Low barrier, low storage diodes permit an approach to ideality for detection sensitivity, mixing efficiency, and harmonic generating capability because of the improved impedance matches. Also, tunnel diode logic circuits which require very low turn-on voltage for the associated diode may become possible. REFERENCE (1) S. M. Krakauer, Harmonic Generation, Rectification, and Lifetime Evaluation with the Step Recovery Diode, Proc. IRE, p 1668, July 1962. FIELD-STRENGTH-PROBE uses tunnel-diode frequency-modulated oscillator. Underwater testing requires the probe be potted. Antenna and coils occupy the most space. LATEST FIELD-MAPPING IMPROVEMENT Leadless Transceiver MEASURING the field pattern of an antenna is often difficult because the field-mapping probe disturbs the field being measured. Even if the probe is small, its connecting leads may upset field distribution. The problem is especially difficult in measuring the field distribution of parabolic disk antennas, microwave lenses, and many other fields that do not permit image-plane measuring methods. The need for lead wires to a probe for measuring the electromagnetic field may be eliminated with a miniature f-m wireless probe that has been built with readily available components. It can be used under water as well as in free space. Despite simple construction, the operation was satisfactory. Figure 1 shows the test setup, with the antenna operating at 114 Mc and using 1,000-cycle amplitude modulation. The probe detects the 114-Mc energy and uses it to frequency-modulate a 7-Mc tunnel-diode oscillator. Rebroadcast energy from the probe is picked up by the telemetering antenna and fed through a lock-in amplifier. Subsequent instrumentation extracts field intensity values from the frequency-deviation of the 7-Mc signal. TRANSCEIVER DESIGN — Overall size of the transceiver must be small to prevent the probe disturbing the original field, yet it should be stable and temperature insensitive. Figure 2 shows a circuit diagram of the transceiver tuned for a 114-Mc a-m input signal and a 7-Mc f-m output. A TD-2 tunnel diode (subminiature axial package) provided dual functions of 7-Mc r-f oscillator and frequency modulator, while a BD-7 backward diode detected the 114-Mc a-m signal. Tunnel diodes were selected because they are small, compact and relatively insensiMiniature probe senses field intensity, converts it to reradiated 7-Mc f-m signal. Frequency-deviation of reradiated signal gives strength of field; method eliminates leads, preventing interference with field pattern. Probe Works Underwater The upper part of Fig. 2 is a detector stage for the 114-Mc a-m signal from the short dipole antenna. The tank circuit tuned at 114-Mc consists of 8 turns of No. 18 tinned copper wire with a length of $ \frac{3}{4} $ in. and diameter of $ \frac{3}{8} $ in. (50 nh) and a 39 pf mica capacitor. The mica capacitor was preferred for its low loss tangent and temperature insensitivity, even though it is much larger than the smallest ceramic capacitors available. A backward diode BD-7 (having the advantage of low forward voltage drop and low stray capacitance) was connected to the tap at 1$ \frac{1}{2} $ turns of the coil, $ L_o $. The detected 1,000 cycle signal was then fed to the bias point of TD-2 oscillator through a 67-ohm coupling resistor. The resistor was used as a coupler instead of a capacitor because of the limited space, but a 25-$ \mu $f capacitor would be preferable if space were available. The tank circuit, $ C, L $, in the lower part (tuned to 7 Mc) is effectively a short circuit at 114 Mc, and, conversely, the tank circuit, $ C_o, L_o $, in the upper part (which is tuned to 114 Mc) is effectively a short circuit at 7 Mc. Thus, the two-tank circuit staggered and loaded with a short dipole antenna form a duplexer for receiving 114 Mc and transmitting 7 Mc. The lower part of the circuit diagram in Fig. 2 is the 7-Mc tunnel-diode series-parallel sinewave oscillator. Choice of tunnel diode is important if satisfactory results are to be obtained. Maximum power available from the tunnel diode can be estimated by \[ P_{\text{max}} = \frac{(I_p - I_v)^2}{8} \cdot \frac{1}{g_d} = \frac{(I_p)^2}{8g_d} \] where $ I_p $ and $ I_v $ are the peak and valley point currents, and $ g_d $ is the magnitude of the tunnel diode's negative conductance. As a rough approximation, the relation \[ I_p = V_v g_d \] where $ V_v $ is the valley point voltage, may be used. Introducing Eq. (2), Eq. (1) becomes \[ P_{\text{max}} = \frac{V_v^2 g_d}{8} \] The value of $ V_v $ is nearly the same for all types of tunnel diode, lying in the range 200 to 350 mv. From the viewpoint of available power, it is preferable to choose a tunnel diode with a large negative conductance, but $ g_d $ must not be too large because the maximum frequency of oscillation, $f_{vo}$, is related by $g_s$ by $$f_{vo} = \frac{g_s}{2\pi C_1} \sqrt{\frac{1}{R_T g_s} - 1}$$ For $f_{vo}$ to be real, it is necessary that $$R_T g_s < 1$$ The smaller $R_T g_s$ the greater the oscillator stability. The value of $g_s$ must not be chosen too large, or the value of $R_T$ will become too small for the restricted d-c power drain from an RM675 miniature mercury cell. Hence, it is necessary to choose a diode with $g_s$ near a middle value that satisfies both requirements. The TD-2 tunnel diode with the parameters $g_s = 16$ millimho and $V_r = 350$ mv was selected and gave about 250 microwatts r-f power. Efforts were made to reduce the number of elements as much as possible. Frequency modulation was accomplished by the 1,000-cycle signal through $R_s$, Fig. 2, which varied the operating point of the diode and modulated the value of $g_s$, which in turn modulated the oscillator frequency $f_{oo}$. $$f_{oo} = \frac{1}{2\pi} \sqrt{\frac{1}{L(C + C_1)} - \frac{g_s^2}{C_1(C + C_1)}}$$ The value of $R_s$ is critical. When it is too large, the waveform of the 7-Mc signal is distorted and the output is decreased; when it is too small, the deviation of the frequency modulation is not enough. A compromise has to be found by cut-and-try methods. The d-c current drain during the operation was about 4 to 5 ma and the lifetime of the stable operation of the probe with an RM675 mercury cell was about 15 to 20 hours. PROBE LENGTH—Since limited r-f power is available from the transceiver and one short dipole probe antenna is to be used both for receiving 114-Mc and sending 7-Mc, a careful design of the probe antenna is necessary. The driving-point impedance of such a short dipole is predominantly capacitive and its radiation resistance is small. The driving-point impedance $Z_o$ of a short antenna is $$Z_o \sqrt{\epsilon_r} = 18.3 \beta^2 h^2 - j \frac{396.0}{\beta h} [1 - 0.383 \beta^2 h^2]$$ where $\beta = \sqrt{\epsilon_r (2\pi/\lambda)}$, $\lambda$ = the wavelength in free space, $h =$ the half-length of the dipole = 3.5 cm, and $\epsilon_r =$ the relative dielectric constant = 1. Equation 7 gives $Z_o = 0.485 \times 10^{-4} - j 7.71 \times 10^4$ (for 7 Mc) $$Z_o = 128 \times 10^{-4} - j 0.473 \times 10^4$$ (for 114 Mc) To achieve better matching to the antenna circuit, this large capacitive component has to be neutralized with a series inductor $L_1 = 175 \mu h$ for the 7-Mc signal, and by $L_1 = 0.66 \mu h$ for the 114-Mc signal. These values give the same positive reactance. When the probe is to be operated in water of low loss tangent ($\epsilon_r = 78$), the reactive component is reduced by $1/\sqrt{\epsilon_r}$ (from Eq. 7). The values of the inductors needed then become $L_{w1} = 254 nh$ for 7 Mc and $L_{w1} = 0.957 nh$ for 114 Mc. The value of the loading inductor was chosen to meet the matching condition of the antenna for 7 Mc in water, rather than for 114 Mc, since the radiation resistance of the dipole increases with the square of the frequency, Eq. 7. The loading coil $L_{w1}/2$ was made of 4 turns of No. 18 tinned copper wire with the length of $\frac{1}{4}$ in. and diameter of $\frac{1}{8}$ in. The 7 Mc was chosen as the frequency of the oscillator because the matching condition of the short dipole for 7 Mc was not extremely different from that for 114 Mc, yet the two frequencies were separated far enough to ensure proper functioning of the duplexer. COMPONENT LAYOUT—Special care has to be taken with the layout of the components, since at vhf their physical positions are important. Components are mounted on a $\frac{1}{2}$-in. thick polystyrene H-shaped panel of 1$\frac{1}{2}$-in. long, $\frac{3}{4}$-in. high, and $\frac{1}{2}$-in. wide; they are self-supported by their own short-cut leads which pass through holes in the panel. The mercury cell and the biasing circuits were grouped above the panel; the oscillator and modulator circuits under the panel. Interaction between the H-field and the two loading coils located at the upper and lower parts of Fig. 2 may introduce errors in the measurement of the E-field. This can be suppressed by winding the two loading coils in opposite sense. The two coils were placed symmetrically about the center of the probe and separated to avoid coupling between them. Absence of interaction was checked by measuring the same field first with the probe's panel wrapped in a thin sheet of aluminum foil, then with it unwrapped. This arrangement reduced the interaction. MEASURED RESULTS—A near-zone field polarization pattern of a half-wave dipole immersed in a water solution of sodium chloride was measured with the wireless probe. The electric field in the immediate vicinity of the dipole is elliptically polarized. The measurement of the polarization pattern was performed by rotating the probe 360 degrees in the plane of the dipole at the position of $k_e = 2.0$, $k_h = 0.7$ in the conducting solution with the properties of $\epsilon_r = 78$, $\tan \delta = 0.71$. Coordinates $k_e$, $k_h$ specify location with respect to dipole position. Agreement between the results measured with the wireless probe and with a conventional thin electrically short dipole antenna supported by $\frac{1}{4}$ inch brass tubing is excellent. In a highly conducting solution the reflection of electromagnetic waves from the brass tubing which supports it is minimized, and the conventional dipole probe behaves like an isolated dipole. Consequently a comparison of the results measured by the wireless probe and by a conventional dipole probe was made with a conducting medium. No explanation is given for differences between measured and theoretical values near the minor axis. To demonstrate the superior characteristics of the wireless probe as compared with the conventional probe, the outside of the hollow polystyrene tubing support of the wireless probe was covered by an aluminum cylinder of various diameters to simulate a conventional dipole probe with a metal support (or lead wire). The polarization pattern set up by an electrically short dipole antenna in air was measured by the wireless probe with and without this aluminum cylinder. The results and the related electrical parameters are shown in Fig. 3. The waves scattered from the cylinder create a pair of extra ears in the direction of the minor axis as well as a distortion of the shape of the polarization pattern. When the cylinder becomes large in diameter, the antisymmetry of the pattern with respect to the center of the probe becomes more prominent. This is due to the scattered field components $E_s'$ and $E_s''$, which are associated with the induced circulating current $i_s$ on the surface of the aluminum cylinder and are antisymmetric with respect to probe center. In free space the probe was operated up to 30 feet from the receiver. Underwater range is shorter. The author thanks Professor R. W. P. King at Harvard University and Stephane Prévot. REFERENCES (1) K. Iizuka, A New Technique for Measuring an Electromagnetic Field by a Coil Spring, IRE Trans. PTGMTT (to be published). (2) U. Davidsohn, Y. Hwang, and G. Ober, Designing with Tunnel Diodes, Electronic Design, Feb. 3 and 17, 1960. (3) Purnell, Dipole Matching Generators, Electroic, p. 33, 1961. (4) R. W. P. King, Theory of Electrically Short Transmitting and Receiving Antennas, J. Appl. Phys., 23, No. 10, p. 1174, Oct., 1952. (5) K. Iizuka, How to Measure Field Patterns with Photosensitive Probes, Electronics, p. 39, Jan. 25, 1963. (6) R. W. P. King and K. Iizuka, The Complete Electromagnetic Field of a Half-Wave Dipole in a Dissipative Medium, Trans. IRE, PGAP-11, May 1963 ALUMINUM cylinders around probe (A) produce different degrees of field disturbance depending on cylinder size (B)—Fig. 3 Solving the AGC Dilemma SERVO SYSTEM USES ATTENUATOR Transistor fed with d-c control signal operates as attenuator in input circuit of amplifier. Technique is useful up to 150 Mc By FRED SUSI, Sylvania Electric Products, Inc., Sylvania Electronic Systems, Needham, Mass. AUTOMATIC GAIN CONTROL is extensively used in receivers to prevent variations in the audio output signal amplitude with changes in r-f input signal strength. To compensate for these fluctuations in signal strength, the gain of the i-f stages is usually varied either by continuously varying the collector voltage or emitter current, or both, of the transistors used in the i-f stages. In these forward and reverse agc methods, serious concessions in bandwidth, and shifts in center frequency of the i-f stages are usually made, owing to significant fluctuations in the transistor parameters with changes in quiescent conditions. One solution is to use buffering between the transistors and the tuned output, as in the cascade configuration. Another agc technique is the servo type system shown in Fig. 1A. This system uses an attenuator stage, an N stage amplifier in the transmission loop, and a feedback loop. This servo method, with emphasis on the attenuator stage, is described here. In the system of Fig. 1A, consider a sudden error signal $\Delta E$ developing at the input terminal, where $\Delta E$ is expressed in db above $E_{\text{in}}$. Since the attenuation stage (a) at this time is not activated by any new level of control current, the same $\Delta E$ appears at the input of the transmission amplifier, $A_T$. Assuming linear amplification in the transmission amplifier, $A_{\text{reb}} + \Delta E_{\text{eb}}$ appears at the system's output and, therefore, at the input of the feedback amplifier $B$. Again, assuming linear amplification in $B$, $\Delta E_{\text{eb}} + A_{\text{reb}} + B_{\text{eb}}$ will be impressed at the input of the detecting circuit. If the proportional direct control current produced at the output of the detector is sufficient to cause an attenuation of almost $\Delta E_{\text{eb}}$, that is $(\Delta E_{\text{eb}} - \delta_{\text{eb}})$, then $\delta_{\text{eb}}$ of error signal will be seen at the system's output at steady state conditions. A change in input signal of $\Delta E_{\text{eb}}$, therefore, causes a $\delta_{\text{eb}}$ change in output signal. This reasoning is also applicable when input error $\Delta E$ is below $E_{\text{in}}$. The compression or flatness factor $m^$ is defined as the ratio of change in input signal in db to a corresponding change in output signal in db at steady state; that is, $m = \Delta E / \delta$ for a final error $\delta$ appearing at the system's output for a corresponding $\Delta E$ at the input. The amount of error presented to the transmission amplifier for a given change in input signal determines the compression ratio of the system. If an $m$ value of 40 is specified for a 40 db change in signal at the system's input, then only 1 db change can be presented finally at the input of the amplifier. ATTENUATOR STAGE — Diode attenuators have been discussed extensively in current literature, and depending upon the application, have been successfully used up to frequencies of 70 mc. These attenuators, however, have serious limitations which can make the closed-loop system unwieldy. Such limitations include relatively low signal handling capabilities, large changes in attenuation with small fluctuations in control current, and high sensitivity to changes in temperature. A transistor attenuator, which overcomes these disadvantages, will be discussed in some detail. The transistor attenuator used in Fig. 1A is a grounded emitter stage whose collector is d-c isolated as shown in Fig. 1B. Figure 1C depicts the equivalent circuit of this attenuator, where the isolating capacitors $C_1$ and $C_2$ are made large enough to be disregarded at the operating frequencies. Also, from Fig. 1C, note that at relatively low frequencies ($f < 2$Mc) the transistor's collector-to-emitter voltage is $V_{ce} = R_{sp} E_{ce} / (R_s + R_{sp})$, where $R_{sp}$ is the shunt resistance of the attenuator for a given $I_{ce}$, $R_s$ is a series dropping resistor, and $C_s$ is the shunt capacitance over the entire operating frequency range. $C_p$ GAINING GROUND Automatic gain control using vacuum tubes is relatively easy to accomplish because with variable-mu tubes you merely make the grid negative and the stage-gain is reduced. It's not so simple with transistors. Although changing the bias level does alter the stage gain, the conditions under which this occurs are not always beneficial and the result is often instability, reduced bandwidth, or increased distortion. is approximately constant at 5 pf in the high frequency units. Typically, $R_{ap}$ changes exponentially from approximately 4,000 to 20 ohms for a 1.0-ma change in control current, or for a 200 $\mu$w change in control power. At relatively low frequencies, where the reactance of the shunt capacitance $C_p$ is negligible, this change in $R_{ap}$ represents a 45-db range of useful control. For higher frequencies where $f > 2$ Mc, the reactance of $C_p$ in quadrature with $R_{ap}$ governs the upper end (where $R_{ap}$ is large) of the control range. Consequently, at frequencies above 150 Mc, the high frequency transistors tested lost their usefulness as attenuators. Figure 2A is a plot of attenuation, $\alpha$ versus control voltage $V_{d-e}$, and control current, $I_{d-e}$, for the type 2N94 germanium alloy npn transistor. The $\alpha$ versus $V_{d-e}$ curve is linear in the operating range, while the $\alpha$ versus $I_{d-e}$ or $P_{d-e}$ is logarithmic in nature (see Fig. 2B) and follows the form $\alpha = K_1 \log P_{d-e} + K_2$ where $K_1$ and $K_2$ are constants. Solving for these constants from the plot of $\alpha$ versus $P_{d-e}$ on semilog paper (Fig. 2B), the following relationship is obtained $$\alpha = -\left[ \frac{\alpha_1 - \alpha_2}{\log(P_{\text{max}}/P_{\text{min}})} \right] \log \left( \frac{P_{d-e}}{P_{\text{min}}} \right) + (\alpha_1 - \alpha_2)$$ where $\alpha_1 =$ the $\alpha$ intercept (insertion loss of the attenuator in db), $\alpha_2 =$ the $P_{d-e}$ intercept, $\alpha =$ attenuation in db as a function of control power using $\alpha_2$ as the reference, $P_{d-e} =$ control power, $P_{\text{max}} =$ control power at $P_{d-e}$ intercept, and $P_{\text{min}} =$ control power at the $\alpha$ intercept. The left hand curves inset in Fig. 2A represent the V-I characteristics of a typical transistor used in the attenuator stage of Fig. 1A. The family of constant $I_b$ or constant $I_{d-e}$ curves crowd together at the high values of base drive and spreads at the low values. The end result of this phenomenon is shown in the graph of $\alpha$ versus $I_{d-e}$ in Fig. 2A. That is, the maximum rate of change of $R_{ap}$ with $I_{d-e}$ occurs at the lower base drives. In this range, however, the greatest distortion is found. This distortion can be attributed mainly to the relatively large non-linearity of the constant $I_b$ curves in this range. Also the constant $I_b$ curves are not coincident at the origin, but meet at a relatively small offset voltage and current. At different base drives, a corresponding V-I characteristic curve is traversed resulting in a marked change in collector-to-emitter impedance. Since the collector of the transistor attenuator is d-c isolated, its quiescent point shifts as indicated in the left-hand curves inset in Fig. 2A. In the attenuation stage for Fig. 2C, an $R_s$ of 10,000 ohm was used as the series dropping resistor. Inputs of 100 mw or less caused only slight distortion at near zero control currents. The distortion is mainly in the form of a large second harmonic, 90 degrees out of phase with the fundamental. A plot of the total harmonic content of the output signal versus control current is shown in Fig. 2C. As one might expect, the total harmonic distortion increases with decreasing $I_e$ and increasing $V_{in}$. Therefore, the distortion problem can be circumvented by increasing the quiescent control current slightly or by increasing the series dropping resistor, $R_s$. However, a trade is made between insertion loss and increasing $R_s$. For the transistors used in the attenuator stage, where $\alpha$ versus $I_{d-e}$ curves are shown in Fig. 2B, between 8 and 11 db insertion loss was realized. TEMPERATURE EFFECT* — The curve of attenuation versus control --- SEPARATE attenuator controlled by feedback signal adjusts amplitude of amplifier input (A). Rectified output from feedback amplifier sets attenuator impedance (B), attenuator equivalent circuit (C)—Fig. 1 ATTENUATOR control characteristics with transistor V-I characteristics inset (A), power control characteristics (B), harmonic distortion (C), and temperature characteristics for two different transistors (D)—Fig. 2 voltage shown in Fig. 2D for both the germanium and silicon units bear out the effect of decreasing $V_{ss}$ with increasing temperature. In closed loop operation, decreasing $V_{ss}$ decreases the delay of the agc. One way of compensating for this effect is to introduce a well by-passed resistor in the emitter circuit of the attenuator-transistor. In effect, this makes the input resistance of the stage more constant with the temperature. Also note from Fig. 2D that although the insertion loss of the silicon units is negligibly affected by large temperature changes, the germanium units are greatly affected. This is attributed to the low value of temperature-induced leakage current for the silicon units and a correspondingly high value for the germanium units. Increasing the temperature has relatively little effect on the value of $I_{d-e}$, since $I_{ss0}$ of the silicon units is negligible. H-F OPERATION—Assuming that the series dropping resistor of the attenuator is predominately resistive at high frequencies ($L$, and $C_p$ of the resistor are small), the high frequency equivalent circuit of the attenuator stage shown in Fig. 1C is valid. The $C_p$ of the transistor attenuator remains fairly constant (approximately 5 pf for a 2N781) over a wide control range. Therefore, when the system is used at a single frequency, as is the case in an i-f strip, $C_p$ can be tuned out. The range of agc in the transistor is extended by tuning out the $C_p$. The 2N781 can be useful as an attenuator at frequencies up to 150 Mc. AGC AMPLIFIER—The AGC system shown in Fig. 3A uses a low noise, wide-band general purpose amplifier in the transmission loop, and a direct-coupled grounded-emitter amplifier driving a detector circuit in the feedback loop. The power gain of the system with zero control current (the insertion loss of the attenuator stages included) is 28 db, while the conversion transconductance of the feedback loop is 31 $\mu A/mV$. The conversion transconductance is defined as the corresponding change of direct control current for a change of signal voltage at the input of the feedback loop. Figure 3B is a plot of $E_{out}$ versus $E_{in}$ for a 0 db/mw to 60 db/mw range of input signal at a typical i-f frequency of 455 Kc. The compression ratio exceeds 60 in the control range. REFERENCES (1) W. F. Chow, and A. P. Stern, Automatic Gain Control Of Translator Amplifiers, Proc. IRE., 43, p 1119, 1955. (2) J. R. James, Electronic Engineering, Amateur Radio, The Transistor Cascade Configuration, 32, p 44, Jan. 1960. (3) J. R. Perkins, Transistor Cascade Circuit Improves Automatic Gain Control In Amplifiers, Electronic, June 2, p 49, 1961. (4) A. W. Nolte, Adjustment Speed Of Automatic Volume Control Systems, Proc. IRE., p 911, July 1948. (5) J. S. Pegram, G. Pervault, A. F. Perkins, A Transistorized Compressor, Trans. A.I.E.E., p 791, June 1959. (6) Shirman, Designing A Stable Transistor AGC Amplifier, Electronic Design, May 11, p 44, 1960. (7) J. S. Brown, Semi-Conductor Gain Control Techniques, Proc. N.E.C., 16, p 541, 1960. (8) J. F. Gibbons, Super Saturated Transistor Switches, International Solid-State Circuits Conference Digest of Technical Papers. (9) R. B. Hurley, "Junction Transistor Electronics", Wiley and Sons, New York, NY, p 92, 1958. (10) T. Ormond, Feedback Pair Amplifiers, Unpublished Sylvania Applications Note. Swept, Marked, Logged, Calibrated... by KAY A sharp filter, swept and marked in frequency (fixed and variable) by 935-B and 990-A on the log amplitude display of the 1025-A, with calibrated level line set by the 432-C, switched in by the 255-A. \| Product \| Description \| Price \| \|--------------------------\|-----------------------------------------------------------------------------\|---------\| \| 935-B Sweeping Oscillator\| 50 cps to 220 mc<br>Audio Video, VHF<br>Price: $1295.00 \| \| \| 990-A CW Oscillator \| 4.5 to 220 mc<br>1.0V rms, AGC'd<br>Price: $373.00 \| \| \| 1025-A Log Amplifier \| 200 kc to 220 mc<br>80 db Dynamic Range<br>Price: $795.00 \| \| \| 432-C Attenuator \| DC to 500 mc<br>0 to 101 db in 1-db steps<br>Price: $110.00 \| \| \| 255-A Coaxial Electronic Switch \| DC to 500 mc<br>70 db "off" at 200 mc<br>Price: $295.00 \| \| Write for complete catalog information KAY ELECTRIC COMPANY Maple Ave, Pine Brook, Morris County, New Jersey Dept. E-7 • Capital 6-4000 MULTI-STAGE BLOWER THIS SMALL? New Globe multi-stage blowers drive cooling air against the extreme pressure resistance you encounter in heat exchangers, tightly packed black boxes, and crammed transistor circuitry. They deliver 45 cfm against 14" H₂O back pressure! (65 cfm free air.) With STAX-3-FC blowers you can design right past costly devices and heavy centrifugal air "pumps" (that use ¼ hp or more for comparable volume/pressure). And you can keep the whole package small. With a low specific speed (Ns = 15,000) STAX blowers perform far past the stall regions of other axial blowers. Unique dynamically balanced rotor design permits the use of one, two or three stages in the same size package. Motor operates on 200 v.a.c., or 115 v.a.c., 400 cycle, one or three phase power. Can also be designed for d.c. operation. Units are designed to meet applicable MIL specs; nominal continuous life is 1000 hours. Weight: 29 oz. (3-stage). Production tooling keeps the price within reason. If you need more performance, different power, etc., Globe will design the exact multi-stage blower you require. The powerful STAX is one of hundreds of miniature blowers made by Globe and engineered for your application. Request Bulletin STX, or call direct: Area Code 513 222-3741. Globe Industries, Inc., 1784 Stanley Avenue, Dayton 4, Ohio. Checking your components in braille? While other methods of non-destructive testing may indicate discontinuity or tiny imperfections, nothing brings proof to your eyes like a radiograph on Ansco Superay® 'H-D' Industrial X-ray film. This ultra fine grain Class I film makes you Ansco-sure of sending only faultless components to customers. For the ultimate in image sharpness and high contrast throughout the widest range of X-radiation, use Ansco Superay 'H-D'. Available in economical bulk packaging. Ask your Ansco Representative for technical literature, or write Ansco X-ray Sales, General Aniline & Film Corp., Binghamton, N. Y. NEW MODEL 430A POWER SUPPLY —a “first” for FLUKE in the high voltage field! 10.00-30.22 KV Line regulation, 0.005% Load regulation, 0.01% Stability, 0.005% Precision calibrated Model 430A fulfills exacting requirements for Electrostatic focusing in mass spectrometry Acceleration and focusing in high resolution cathode ray tubes Calibration of electrostatic voltmeters and high voltage dividers Pressurized ionization chamber experimentation High voltage research Charging high voltage capacitor banks Model 430A delivers calibrated DC from 10 kilovolts through 30.22 kilovolts, with an output current capability from 0 to 10 milliamperes. The supply may be operated with either a positive or negative output, with one side always grounded to the chassis. Polarity is easily changed by reversing a front panel polarity plug, which is interlocked for safety. No oil-filled tanks and associated heaters and thermostats are used. The sampling string features stress-free encapsulation, reed relay switching and 100 millivolt resolution at any output voltage. The high voltage output is developed from a conventional 50-60 cycle voltage doubler, free of interference associated with RF type supplies. Approximately 60 seconds after line power is switched on, the high voltage circuitry is armed, and the 430A is ready for use. BRIEF SPECIFICATIONS \| MODEL \| 430A \| MODELS 413C, 413D \| \|-----------\|---------------\|-------------------\| \| OUTPUT VOLTAGE \| 10 KV-30.22 KV \| 0-3111 VDC \| \| OUTPUT CURRENT \| 0-10 ma \| 0-20 ma \| \| LINE REGULATION, for 0% line change \| 0.005% \| 0.001% + 1 mv \| \| LOAD REGULATION, for full load change \| 0.01% \| 0.001% + 2 mv \| \| STABILITY Per hour \| ±0.005% \| ±0.005% \| \| \| Per day after warmup \| ±0.03% \| ±0.03% \| \| RESOLUTION, full range \| 100 mv \| 2 mv \| \| RIPPLE \| 5 mv RMS \| 150 uv RMS \| \| CALIBRATION ACCURACY \| ±0.25% \| ±0.25% \| \| SIZE \| 48" high x 19" wide x 18" deep \| 19" wide x 5¾" high x 16" deep \| \| WEIGHT \| 260 lbs. \| 50 lbs. \| \| PRICE \| $3900.00 \| $695.00 — 413C \| \| \| \| $595.00 — 413D \| Models 430A, 413C and 413D may be used over a wide range of environmental conditions OPERATING TEMPERATURE RANGE 0° C to 50° C TEMPERATURE COEFFICIENT, 10° C to 40° C ±0.002%/°C STORAGE TEMPERATURE RANGE -20° C to +60° C HUMIDITY 0 to 80% ALTITUDE 0 to 10,000 feet Other new Fluke power supplies: MODEL 413C, shown above, offers 0-3111 VDC; ±0.001% line and load regulation; front panel polarity switch; virtually complete absence of noise; low ripple; no voltage overshoot, and high resistance to mechanical shock and vibration. Same specifications apply to Model 413D (with front panel polarity plug; without meters). Before specifying any type of power supply, check the full range of Fluke models. Write for new Catalog Digest 3-63 showing complete line of Fluke test and measurement instruments, or ask your Fluke sales representative for complete technical data or demonstration. From industry's viewpoint, sound, thrifty government is a capital asset in Virginia's Capital Area. The Richmond-Petersburg-Hopewell-Colonial Heights area of Virginia is much more than a strategic rail, highway, air and deep-water transportation center. It is also the seat of one of the nation's soundest state governments. From industry's viewpoint, this represents long-range advantages that far outweigh any temporary plant location "gimmicks." They're enduring advantages such as freedom from state debt, freedom from overlapping taxes, outstanding labor legislation, and clean-cut, central administration of industrial water use and waste disposal privileges. For available site data, and a complete economic study of this key area, write, wire or phone. VIRGINIA ELECTRIC and POWER COMPANY J. Randolph Perrow, Manager—Area Development Electric Building, Richmond 9, Virginia • Milton 9-1411 Serving the Top-of-the-South with 2,540,000 kilowatts—due to reach 3,900,000 kilowatts by 1965. What Happens to Dielectrics in Space? New power systems, radiation resistances point way to space USE OF DIELECTRICS in space, a growing field of research in laboratories across the country, was discussed at the Dielectrics in Space symposium held in Pittsburgh under joint sponsorship of Westinghouse and NASA. New information about the behavior of existing dielectrics in space environments of temperature, vacuum and radiation extremes points the way to development of new power systems, improved equipment operation and better space-vehicle propulsion systems—such was the consensus of the meeting. The behavior of plasma as a dielectric, studied by Boeing Research Lab, was reported on by M. J. Kofoid. The study, connected with project Dynasoar, attempts to determine possible hazards during reentry due to abrupt changes in dielectric properties of the low-pressure plasma surrounding a space vehicle. For example, the Dynasoar vehicle may have exposed electric terminals on its skin where it was separated from its booster; in a reentry plasma a breakdown might occur between an exposed terminal and the vehicle skin. The study determined that breakdown is much less likely to occur if the exposed terminal has negative polarity with respect to the vehicle skin; also determined the optimum amount of recessing for the terminal as one diameter. A study of the effects of a simulated space environment on a large variety of dielectrics was summarized by L. J. Frisco of Johns Hopkins University. The NASA-sponsored project determined the properties of materials when subjected to vacuums of $10^{-4}$ Torr, 50-Kv X-ray radiation, and ultraviolet irradiation, with respect to flashover, a-c and d-c losses, dielectric strength, at a variety of frequencies to 18 megacycles. NEW POWER SYSTEM—A new type of power system suitable for space applications was described by B. H. Beam of NASA's Moffett Field, California. The system (ELECTRONICS Newsletter, July 5) is based on building a series of thin-film capacitors into the skin of a satellite. The capacitors are charged by radiation, and disFor precise control of low level signals from thermocouples, strain gages and resistance bridges, other millivolt-output transducers Low noise and drift, high linearity and freedom from ground loop problems are a few of many distinct advantages these stable, all-solid-state Sanborn amplifiers offer your low-level DC signal instrumentation. Typical inputs include millivolt signals from thermocouples and strain gages. Amplifier outputs can be connected to scopes, meters, magnetic tape, oscillographs, computers, and for high current output models, high frequency optical galvanometers. This choice from Sanborn lets you select the performance characteristics you need in your data acquisition system — you pay for only what you need. And the specs hold true in practice as well as on paper. - Consult Sanborn on your instrumentation requirements — your local Sanborn Sales-Engineering Representative will be glad to work out the details with you...and a phone call will get immediate action. Offices throughout the U.S., Canada and foreign countries. \| "FIFO" 860-4000 \| WIDEBAND 860-4200 \| NARROW BAND 860-4300 \| \|-----------------\|-------------------\|----------------------\| \| Bandwidth \| DC — 10 KC within 3 db \| DC — 50 KC within 3 db \| DC — 100 cps within 3 db \| \| Linearity \| ±0.1% of 10 V f.s. at DC \| ±0.01% of 10 V f.s. at DC \| ±0.03% of 5 V f.s. at DC \| \| Gain \| 1000, 500, 200, 100, 50. Smooth gain control covers intermediate ranges \| 1000, 500, 200, 100, 50, 20, 10. Does not phase invert \| 1000, 500, 200, 100, 50, 20, 10. (Gain of 10 to 20,000 in 12 fixed steps available on special order) \| \| Overload Recovery \| For 20 V, 1 ms to 1% of f.s. output \| \| For ±10 v, 200 ms to within 25 mv of original output \| \| Drift \| ±2 uv ref. to input, ±0.01% of f.s. at output at constant ambient for 40 hours \| ±0.02% of f.s. at constant ambient for 40 hours \| ±2 uv ref. to input, ±0.1 mv. ref. to output for constant ambient for 40 hours \| \| Noise \| 5 uv rms, DC-10 KC (ref. to input at gain of 1000) \| 7 uv rms, DC-50 KC (ref. to input) \| 1 uv p-p, DC-20 cps (ref. to input, at gain of 1000) \| \| Input \| Isolated from gnd. and output. Impedance 100 meg. min. at DC in parallel with 0.001 mfd. \| Impedance 100 meg. at DC in parallel with 0.001 mfd. \| Isolated from gnd. and output. Impedance 500K \| \| Output \| Isolated from input and ground. ±10 V at 10 ma. (-400D has grounded output, ±10 V at 100 ma.) \| ±10 V at ±100 ma. Sustained short across output will not cause damage to amplifier. \| Isolated from input and ground ±5V at 10 ma. Part or all of internal 2K in parallel with 15 mfd. may be removed, connected externally. \| \| Common Mode Characteristics \| 120 db rejection at 60 cps, 160 db rejection at DC (1000 ohms in either input lead). Tolerance =300 V DC or peak AC. \| Amplifier floats with respect to chassis. Isolation impedance is greater than 3000 megohms in parallel with 5 pf. \| 130 db rejection at 60 cps, 160 db rejection at DC (1000 ohms in either input lead). Tolerance =300 V DC or peak AC \| \| Price (F.O.B., Waltham, Mass.) \| $825 (860-4000P: $900) \| $650 (including internal power supply) \| $425 \| INDUSTRIAL DIVISION SANBORN COMPANY WALTHAM 54, MASS A Subsidiary of Hewlett-Packard Co. Fresnel Reflector to Orbit This Year FOUR-FOOT diameter Fresnel solar reflector will orbit the earth later this year in project EROS (Experimental Reflector Orbital Shot) to prove feasibility of using focussed sun's rays to run space equipment. The reflector has been delivered to Air Force's System Command, Aeronautical Systems Division. It was developed and built under $500,000-contract by GM's Allison Division. The flat-plate electroformed nickel reflector with aluminized surface will unfold in orbit to focus sunlight into a radiometer for measurement. It will orbit for thirty days, aligning with sun for about one minute each orbit. Energy conversion will not be tried, but officials predict the mirror could produce 100 to 500 watts. Larger reflector projects ranging up to 30 kilowatts are under study. Additional ASD solar energy collection programs are being established to simultaneously place two ten-foot reflectors in orbit for more than 90 days. Mort Mann can show you... how North Atlantic's Phase Angle Voltmeters solve tough ac measurement problems ...in the lab or in the field. Designed for critical tasks in circuit development, production and testing, North Atlantic's Phase Angle Voltmeters provide direct reading of phase angle, nulls, total, quadrature and in-phase voltages—with proven dependability even under field conditions. Your North Atlantic engineering representative can quickly demonstrate how they simplify ac measurement jobs from missile checkout to alignment of analog computers—from phasing servo motors to zeroing precision synchros and transducers. Shown below are condensed specifications for single-frequency Model VM-202. Other models include high sensitivity, three-frequency and broadband types. \| Specification \| Details \| \|--------------------------------\|----------------------------------------------\| \| Voltage Range \| 1 mv to 300 v f.s., 12 ranges \| \| Voltage Accuracy \| ±2% f.s. \| \| Phase Accuracy \| dial: ±1°; meter: ±3% of F.S. degrees \| \| Signal Frequency \| 1 Freq., 30 cps—10 kc \| \| Input Impedance \| 10 megohms \| \| Reference Input \| 100 K, 0.25 v min. \| \| Meter scale \| 3-0-3, 10-0-10 linear \| \| Phase Angle Dial \| 4 scales, 90° (elec.) apart \| \| Nulling Sensitivity \| 2 microvolts (phase sensitive) \| \| Harmonic Rejection \| .55db (with filters) \| \| Dimensions \| 5¼" h. x 19" w. x 7¾" d. \| The North Atlantic man in your area has full data on standard and special models for laboratory, production and ground support. Call today for his name, or request Bulletin VM-202. NORTH ATLANTIC industries, inc. TERMINAL DRIVE, PLAINVIEW, L. I., NEW YORK • OVERBROOK 1-8600 Trademark NOW from DEI Versatile ... Proven ... Modular VHF/UHF Telemetry Receiver - Video Amplifier Response DC to 1 mc with Impedences Selectable 75 or 600 ohms - AFC Optional on all RF Heads - Multiple Bandwidth Front Panel Switchable IF Amplifiers - Multi-Range Deviation Meter Calibrated Directly in KC Immediately available from Defense Electronics, Inc. is the versatile, field-proven, completely-modular TMR-5A telemetry receiver. This reliable unit will currently accept 14 plug-in tuning heads...14 plug-in IF strips and six demodulators...any one of which can be easily removed from the receiver in seconds. Additional heads, demodulators and IF amplifiers are available upon request. Plug-in RF heads can be instantly installed or removed from the receiver to provide the required frequency range by merely using the "single-action" pull out handle. The video drawer also can be speedily converted by the various plug-in IF strips, ranging from 3 KC to 2.4 MC, and plug-in demodulators for FM, AM, PM and FM phase-lock applications. This unit is particularly suitable for conical scan antenna tracking, dual diversity combining and predetection record/playback applications. Write for DEI bulletin TMR-5A...or call: Defense Electronics, Inc. Main Office: 5455 Randolph Rd. Rockville, Md. Phone: WH 6-2600 TWX: 301-949-6788 Regional Office: Sherman Oaks, Calif. Phone: 873-4322 Hungarians Develop 3-D X-ray Equipment VIENNA—Use of two X-ray tubes arranged to penetrate a human body from two different directions and thus produce a "plastic image" on a screen has been reported by Dr. Sandor Dekany, in Budapest. Special equipment senses differences between the two projections and locates targets exactly in three-dimensional manner, according to the inventor. Reported exposure time is one-fifth of a second; clear pictures can be obtained even of moving objects. The system is the subject of a patent application in several countries. High Intensity Gun Simulates Nuclear Blast PHYSICIST checks alignment as x-ray gun is fitted to specimen-chamber porthole HIGH-INTENSITY X-RAY GUN simulates the gamma-ray emission given off by a one-megaton nuclear blast 2 miles away, to test effects of nuclear explosion radiation on electronics equipment for 0.1 μsec. The unit, installed in General Precision's new Aerospace Research Center in Little Falls, New Thermistors?—Nobody makes a wider variety than Carborundum. Carborundum is turning out thermistors in varieties undreamed of a short time ago—from sub miniature discs to complete assemblies. Special assemblies can be manufactured to fit chassis and operating requirements. You can specify your electrical and dimensional needs and we'll design the assemblies to meet them. They meet precise tolerances for temperature coefficient, dissipation constant and resistance at reference temperature. Can be potted in epoxy. Provide long-term stability. Discs and washers are available in sizes from 1/10" to 1" in diameter. We supply them soldered to mounting plates or with special terminals to solve assembly problems. They're made of high-stability compositions. Beta values range from 2700 to 4800. Rods can be purchased from Carborundum in the widest range of body sizes and temperature coefficients available to industry—in lengths from 1/4" to 12" and diameters from .05" to 1". Beta values from 1100 to 4600. PTC Thermistors are new, positive-temperature-coefficient units with high sensitivity. Sensitivities as high as 12 percent per °C can be achieved. Use PTC's in combination with NTC's (negative coefficient units) and with linear resistors to construct an astonishing variety of temperature compensation curves for your circuits. We also manufacture PTC assemblies and probes for special needs. The graph shows resistance of three different PTC thermistors over a range of temperatures as a multiple of resistance at room temperature. For more information on this complete thermistor line, write to: Electronics Division, Globar Plant, Dept. ED-8T, Niagara Falls, N.Y. Magnesium oxide crushable preforms now in longer lengths High-purity (99.4%) MgO tubing is now available in lengths up to six inches. This is double the previous size available. Now you can produce swaged, high-precision thermocouple assemblies faster. Cut costs at the same time. These ceramic preforms are available in 2", 4" and 6" lengths depending upon O.D. size. In addition, O.D., hole size, and camber tolerances have been tightened. They have an AQL (acceptable quality level) of 2.5% for O.D., hole size and camber. Using the longer lengths, you can produce thermocouples with fewer air gaps, better resistivity and lower fatigue factor. They're just what you need wherever rigidity, resistance to vibration and high reliability are a must. For our new literature on ceramic insulating tubing, write to: Electronics Division, Latrobe Plant, Dept. EL-7, Latrobe, Pa. NOW-PRECISION VOLTAGE AND PHASE COMPARISONS AUTOMATICALLY-WITH A GERTSCH COMPLEX RATIO BRIDGE This self-nulling AC bridge displays both in-phase and quadrature ratios on a 5-place Nixie readout. You get fast, accurate measurements without the need for a skilled operator. In-phase ratio accuracy is better than .005% of range—quadrature accuracy: better than .05% of range, plus calibration. Average nulling time is less than 10 seconds. Instrument is ideal for production testing of transformers, tachometers, rate gyros, all types of transducers, AC amplifiers, AC networks, and AC systems. Measures step-up ratios as high as 1:6 to 3-, 4-, or 5-place resolution. Automatic quadrant selection and indication. Choice of operating frequencies. Either 400 or 800 cps models available. Other frequencies furnished on request. Completely self-contained—requires no external calibration sources. Output provided for printer. Both cabinet and rack-mounted models are available. Write for complete literature. Bulletin CRB-3. Gertsch Products, Inc. 3211 S. La Cienega Blvd., Los Angeles 16, Calif. • Upton 0-2761 • VErmont 9-2201 IONOSPHERE STUDIES, with a view to eventual control of ionosphere for long-distance radio communication, are being conducted by University of Illinois with an instrumented Aerobee 150 rocket. The rocket's transmitter will heat the electrons of the upper atmosphere; the resulting glow will be observed by photocells in the rocket. The effect of this on polarization or direction of radio wave propagation through the ionosphere will be then determined. Special interest is the cross-modulation effect of the heated ionosphere; this causes intermodulation between radio signals of different frequencies. Want to talk to a salesman who says, "My Company Makes it Better--to Save You Money"? The woods are full of hermetic terminal manufacturers selling price. Exactly how many no one knows. They come and go too fast to keep track of. We won’t pretend that Fusite is invulnerable against their allure for some of our customers. But nearly always, a manufacturer learns the hard way that the most expensive terminal he can use is a cheap one that fails on the job. The very best Fusite Terminal you can use is a small fraction of the total cost of your assembled product. In the 20 years since Fusite pioneered in its field we’ve watched the “price sellers” come and go, while we continued to make the best terminals we know how. If you want the least expensive terminals you can buy we’d like to do business with you now. If you’re looking for the cheapest price—we’ll still be around later. We hope you will. THE FUSITE CORPORATION DEPT. J2, 6000 FERNVIEW AVE., CINCINNATI 12, OHIO Gentlemen: Please send me, without obligation, the items checked below. ☐ An enlargement of this cartoon, without advertising, suitable for framing, ☐ The Fusite catalog ☐ A sample terminal of ________ type ☐ A sales representative Name ____________________________________________ Company ___________________________________________ Address ___________________________________________ electronics • July 19, 1963 STC LOW NOISE TRAVELLING WAVE TUBES EXTEND RADAR RANGE! STC offers two tubes of eminently suitable design for use in /S band TYPE W9/2E FOR BROADBAND COVERAGE with a gain of 40 dB and noise factor of about 8-5 dB. It is intended for operation over the whole frequency range 2-5 to 4-1 Gc/s with fixed voltages. An aluminium foil mount is available with coaxial r.f. connectors. TYPE W10/3E FOR NARROW BAND OPERATION with about 23 dB gain and 6-5 dB noise factor with the grid voltages set for optimum noise factor at the appropriate centre frequency. W10/3E has a frequency range 2-7 to 3-3 Gc/s in solenoid circuit 495—LVA—003 with waveguide r.f. connectors or frequency range 2-8 to 3-8 Gc/s in solenoid circuit 495—LVA—006 with coaxial r.f. connectors. COMPONENTS GROUP VALVE DIVISION Standard Telephones and Cables Limited FOOTSCRAY SIDCUP KENT USA enquiries to ITT COMPONENTS DIVISION · P.O. BOX 412 · CLIFTON N.J. now there are 2 new 260® VOMs WITH MIRROR SCALE Series 4M $50.95 WITH STANDARD 260 SCALE Series 4...Still Only $48.95 with 6 new features Now the world-famous 260° volt-ohm-milliammeter is a better buy than ever before. Call your distributor now for immediate delivery on either of these two new 260's, or on any of the other 50 Simpson testers—the world's largest selection. - New Self-Shielded Movement - New Higher Accuracy: 2% DC, 3% AC - New Ruggedness From Spring-Backed Jewels - Greater Repeatability - Increased Linearity and Stability - Mirror Scale (260, Series 4M) SIMPSON INSTRUMENTS STAY ACCURATE Both new 260's can be used with Simpson's popular "Add-A-Tester" adapters. Write for Revised Stock Catalog 2064A. SIMPSON ELECTRIC COMPANY 5203 West Kinzie Street, Chicago 44, Ill. Representatives in Principal Cities ...See Telephone Yellow Pages Phone: (312) Estebrook 9-1121 In Canada: Bach-Simpson Ltd., London, Ontario CIRCLE 77 ON READER SERVICE CARD WHAT MAKES AN INSTRUMENTATION CABLE FAIL? It can pass inspection perfectly one minute and fail miserably the next. Simply manufacturing it to spec isn't good enough. Insurance against failure must be built into the cable at every step from diagram to installation. Where can it go wrong? At almost any point not adequately safeguarded. Here are four of the most common trouble spots: (1) Incompatible Plasticizers (2) Filler Material (3) Component lay-factors (4) Shielding INCOMPATIBLE PLASTICIZERS A unique form of chemical warfare within cable materials has fouled more than one missile program. Plasticizer materials have to be added to compounds to obtain the required flexibility. These additives are seldom compatible with each other. Incompatible plasticizers used in systems in contact with each other without control may attack each other with disastrous effects. (As a prime example, additives in low temperature neoprene jackets are not always compatible with the insulating materials.) Manufacturers can control plasticizer migration problems by selecting proper materials and by using suitable barriers between components. Many specifications make the use of barrier material optional and a manufacturer whose only concern is price will leave it out. Rome-Alcoa, as a result of its wide experience with materials, always uses barriers where migration could be a problem. FILLER MATERIALS When spurious signals arrive at your display, recording or control panel, the fault could be in the improper selection of filler material. Compatibility between insulations and filler materials is of prime importance. In the case of some plastics or rubbers, the material's "memory" can cause it to shrink disproportionately, creating undue stresses internally in the cable. This can cause kinking of the insulated conductors; electrical failures follow. Only experience can tell a cable manufacturer how to compensate for "memory" and how to control compatibility in filler materials. Experience in areas such as this has given Rome-Alcoa its remarkable record of instrumentation cable reliability. COMPONENT LAY-FACTORS Conductor kinking can also be a result of mistakes in the twisting of component conductors. Inconsistent tensions and improper sequence of lay-up can create uneven tensions in the assembled conductors. In such cases, individual conductors may actually push through their insulations, causing electrical failures. Obviously, these mistakes should be avoided during cabling. At this stage in cable construction careful, experienced workmanship can provide safeguards against possible trouble later on. Such careful craftsmanship sometimes costs a little more, but it can make the difference between success and failure. SHIELDING Constructed of many ends of fine strands, shielding braids are prone to having broken and loose ends. These can break through insulations and short out component conductors. Improperly treated, they are the most common cause of shielding failures. It's cheaper to let such loose ends remain in the braid—but it can also be disastrous. Experience on thousands of such shieldings has taught Rome-Alcoa the exact tensions which must be maintained, as well as methods of protecting and treating loose ends. HOW TO AVOID FAILURES No manufacturer can promise you 100% reliability at every development stage. But it's only logical that the one way to be sure of maximum reliability is to have your cable planned and manufactured by a company with depth of experience and a record of reliability in the field. Rome-Alcoa is, frankly, one of the few companies that qualify. We've been designing and constructing these cables since their first conception—long enough to know what can cause a cable failure, and how to avoid it. If you're planning to design or install instrumentation cable soon, call us. As a starter, send for our 24-page booklet titled "Instrumentation Cables, Cable Assemblies and Hook-up Wires." In it, we describe instrumentation cable constructions, production, military specifications and our qualifications. For your copy, write Rome Cable Division of Alcoa, Dept.27-73,Rome, N.Y. ALCOA ROME CABLE DIVISION July 19, 1963 • electronics NEW! volt-pac® variable transformers 10.0 Amperes DOLLAR SAVINGS— 19 Basic Ratings—Wider Selection Industry's Most Complete Line LONGER LIFE— 30-100% Greater Overload Capability SPACE SAVINGS— Up to 33% Greater Over-all Capacity in Same Physical Size You benefit three big ways with General Electric's new volt-pac variable transformer line. First, G.E.'s wider selection offers big dollar savings. You can choose from nineteen basic G-E 120- and 240-volt, single-core ratings—more than ever before—to more precisely meet your applications. By making smaller jumps between ratings, you save the difference in dollars. And G.E.'s 2-65 amp single-core current range, 15 amps more than previously available, means one volt-pac unit can replace two stacked units in these higher ratings. You lower costs and space requirements too! Second, you get longer life—the result of three exclusive volt-pac features. General Electric's ① new unique heat-sink ring located beneath the gold-plated brush track quickly dissipates heat from the hot spot. The ② solid-carbon, grain-oriented brush and the ③ quick-transfer current collector* provide more reliable operation. Total effect: up to 100% greater overload capability than previously available. Third, you get up to 33% more over-all capacity—in existing core sizes. And in the popular sizes through 10 amps, G-E units are directly interchangeable with your present mounting arrangements. For the full story on G.E.'s manual and motor-operated volt-pac lines, and the automatic line with solid-state SCR control, see your G-E Sales Engineer or authorized G-E Electronics Distributor. Or write for Bulletin GEA-7751 to Section M458-02, General Electric Company, Schenectady 5, New York. Patent Pending Progress Is Our Most Important Product GENERAL ELECTRIC CIRCLE 79 ON READER SERVICE CARD How Ceramic Benders Control Light Rays Here's piezoelectricity with a new twist. Ceramic acts as light shutter By F. W. KANTOR, Consultant Silver Spring, Maryland WHEN VOLTAGE is applied to two different thin strips of piezoelectric materials, bonded together, one strip increases in length and the other becomes shorter. The composite element curls, according to the polarity of the applied voltage. Considerable mechanical travel can be produced. For example, the Clevite bimorph bender 1.75-in. long × 0.125-in. wide × 0.0021-in. thick bends about 0.07-in. at 150 v. This curling motion can control the passage of light from light source(s) to photoconductor(s). With the ceramic element deflected in an open position, relays can be constructed, driven by 10 to 100 nanowatts and capable of controlling several watts. If the bender is initially mounted in the open position, and bent closed, normally-closed switching operation is obtained. Figure 1A shows a possible arrangement for a single-pole single-throw normally-open relay. The input resistance of the ceramic bender element is greater than $10^{14}$ ohms. Capacitance is less than 0.005 $\mu$F. Benders operate with stored energy less than 50 microjoules. The output characteristics depend mostly on the photoconductor. This permits construction of relays with widely-varying contact characteristics in the same unit. Photoconductive switching provides very low noise operation for small-signal switching, and does not require spark suppression when handling moderate loads. With present photoconductors, maximum output can be as high as 25 w, but is more typically 0.25 to 2 watts. The operating time depends primarily on the rise and fall time of the photoconductors, and is in the order of 10 to 20 ms, depending on the design. With any design, the lamp serves as a control element, providing two isolated inputs. THE DESIGNS—Bender elements can be arranged to move a fiber light pipe to different photoconductors, Fig. 1B. Such a device makes a small, low-output multiple-throw switch for polarity sensing. In another design, light from the lamp is reflected on a surface—to the photoconductor, Fig. 1C. The bender carries a light absorber which conforms to the reflecting surface. When voltage is applied to the bender, the total internal reflection is destroyed. When the absorber moves a few wavelengths of BOTH THESE MAGNETIC TAPES HAVE A POLYESTER BASE ...BUT ONLY ONE IS MYLAR® (8 YEARS PROVEN) Eight years ago instrumentation tape of Du Pont MYLAR® polyester film appeared on the scene and set new standards of reliability. Naturally enough, people whose needs called for a magnetic tape of highest performance couldn't risk a tape other than MYLAR. Now, other polyester films are beginning to appear. They are not all the same: MYLAR is a polyester film, but other polyester films are not MYLAR. In the past you could safely assume you were getting MYLAR when you specified "polyester base". Today you cannot. There's only one way to be sure you're getting the MYLAR you've used and trusted for magnetic tapes of proven reliability: specify MYLAR by name. E. I. du Pont de Nemours & Co. (Inc.), 10452 Nemours Bldg., Wilmington 98, Delaware. Du Pont's registered trademark for its polyester film. DU PONT BETTER THINGS FOR BETTER LIVING...THROUGH CHEMISTRY only DU PONT makes MYLAR® POLYESTER FILM CIRCLE 81 ON READER SERVICE CARD light away from the surface, reflection is restored. The short travel gives great sensitivity. The on state is sensitive to mechanical shock and vibration, and the off state does not have a high resistance, due to slight surface irregularities in the absorber. The large area available for reflection permits great efficiency in transferring light from the lamp to the photoconductor, and makes uniform illumination easier. The increased capacity of recent photoconductors (ELECTRONICS, Apr. 12, p 100) can control small lamps, motors, and heavy relays. OPTICAL LEVER — A mirror, mounted on the end of the bender, can be made for 10-v spacing between photoconductors, Fig. 1D. Photoconductors, facing a calibrated screen, act as an relay with adjustable trip points. In principle, it is possible to mount a resistor-photoconductor-conductor sandwich in place of discrete photoconductors. Thus, continuous output can operate as an electrometer amplifier. With all designs, the output photoconductor can be removed and light can be used as a visual readout for human consumption. CIVIL DEFENSE—High input resistance of this proposed family of photoconductive switches can make them useful in RC time-delay circuits, for monitoring inputs which are integrated by a capacitor, as short-term radiation-dosage meters and controls, or in RC running integral circuits used in radiation-level monitors. One configuration could greatly simplify mass-produced radiation monitors, because it can operate without amplification from the output of a geiger tube. When combined with a small rectifier to prolong discharge of their internal capacitance, such relays can be operated by a single pulse in the microsecond range in a type of latching mode, freeing control-switching circuits. Relays can be held closed for up to 15 minutes. The high power gain and small differential make feedback operation attractive: 70 to 90 db overall gain can be obtained. Aging effects occur in the lamp, the photoconductor, and the bender elements. Lamps with life of one million hours are available. Photoconductor aging does not particularly affect the performance. Of more significance is the aging of the bender, which requires 10 to 15 percent more voltage for the same travel after 2 to 3 years. This effect is well behaved, however, and as such does not effect reliability. Preaging of bender elements can reduce the voltage effect considerably. --- Easy Way to Test Accelerometers Flapper circuits are simple. Just energize and monitor system By H. W. HOSSFELD, McDonnell Aircraft Corp. St. Louis, Mo. A DESIGN modification has been developed to provide a self-test feature for accelerometers. The circuit allows the accelerometer to be tested for proper operation without removal from its mount and was developed for accelerometers used in the autopilot for F4B and F4C aircraft. Two basic types of accelerometers are used in the aircraft autopilot. The G limit and trim cutout accelerometers are essentially G sensitive switches. Switch action occurs at the limit acceleration values. The lateral accelerometer includes an inductive pickoff and provides an output signal proportional to applied acceleration. The self test assembly can be applied to both the switch types and proportional type. Construction of the accelerometers is very similar, the only difference being in the type of output pickoff. In the trim cut out accelerometers, the sensing mass rotates about a pivot point just under the mounting flange. Movement of the BIDIRECTIONAL flapper is shown mounted on G-limit accelerometer—Fig. 1 SIMPLE hookup deflects sensing mass when coils are deenergized—Fig. 2 mass carries the extension arm which supports the switch wiper. The switch is designed for a G range of ±½ G. Beyond this range, the wiper moves off the conductive material and the circuit is broken. Operation of the limit accelerometer is similar, except that the circuit is maintained for a range from −1 to 4 G. The lateral accelerometer has an E core pickoff to give a proportional output. The center coil is excited with a 400 cycle voltage and the signal is taken from the series-connected outside coils. FLAPPER—The self-test mechanism is a bidirectional flapper. Energizing either set of coils will develop a magnetic field through their iron cores, an end cap, and one side of the armature. The force developed by the action of the magnetic field on the armature will rotate it and the attached yoke about the pivot. A compression spring mounted in a hole in the armature and working against a flat on the pivot shaft returns the yoke to a center position on removal of power from the coils. Figure 1 shows the self test as- new ILo® crystal filters obsolete narrow band image parameter designs Obsolete is a very strong word. We wouldn’t use it lightly. But compared with conventional image parameter crystal filters, new ILo® crystal filters by Midland have much lower insertion losses, better symmetry, tighter phase control, better approximations of Chebishev or Butterworth responses, tighter control over passband characteristics, better spurious control, and are far more easily reproduced in quantity. What word would you use? Ilo® crystal filters are no more expensive than conventional designs, and there are no technical tradeoffs. They are equal or superior to image parameter designs in every other characteristic. New ILo® crystal filters are designed by a technique radically different from conventional filters, and may represent the most important advance in filter design in the past five years. Write for facts from world’s largest producer of crystals and crystal filters. this is an ILo® filter this is not MIDLAND MANUFACTURING COMPANY 3155 FIBERGLAS ROAD KANSAS CITY 15, KANSAS An electronics division of Pacific Industries, Inc. all sizes for all jobs Waugh FREQUENCY TO DC CONVERTERS One Frequency-to-DC Converter (handling input frequencies from 5 to 10 Kc) is a printed circuit card measuring three by four inches. Our solid state FR-500 series features modular construction for multi-channel operation in your instrument control room. In addition, a variety of Frequency-to-DC Converters are available that have been shaped, sized and engineered for virtually every type of application or environment—airborne, shipboard, field, plant or laboratory. Typical applications include Frequency-to-DC conversion of symmetrical pulse trains from turbine flowmeters, tachometer generators, electro-magnetic coils, photocells and nuclear detection devices. Wherever incoming AC signals must be converted to a triggering or recording DC output with exceptional linearity, sensitivity and reliability, specify FOXBORO. Where and How to Use Gallium Arsenide THE POTENTIAL of gallium arsenide devices to surpass either silicon or germanium devices for certain applications has been outlined by Texas Instruments. Company has demonstrated the potential for gallium arsenide for higher power and increased frequency response. Before this potential can be realized, however, advances in GaAs materials and device technology will have to be made. Selects Zig-Zag Antenna To Meet UHF Changeover MODIFIED cardioid pattern of new GE antenna teams with GE's newly designed 25 Kw uhf transmitter to provide KERO-TV Bakersfield California with coverage for the rich southern San Joaquin Valley. Television station changes over from vhf Channel 10 to uhf Channel 23 under FCC deintermixture order. Transmitter-antenna site is 22 miles east of Bakersfield, on Mt. Breckenridge. LEVIN now offers a complete line of Instrument Lathes mounted on a welded steel cabinet. These machines have the advantages of greater rigidity, higher spindle speeds, increased operator efficiency due to cabinet design, multi-vee belt drive, and for the first time, operating and alignment tolerances heretofore available only on specially built lathes. Your LEVIN dealer or the factory will be glad to furnish complete details. These machines are on display in dealer showrooms in New York and Los Angeles. Send for complete catalog. LOUIS LEVIN & SON, INC., 3573 HAYDEN AVENUE CULVER CITY, CALIFORNIA DEALERS: RUSSELL HOLBROOK & HENDERSON, INC. NEW YORK and LOS ANGELES INDUSTRIAL TOOL & SUPPLY COMPANY — SAN JOSE MARSHALL TOOL & SUPPLY CORPORATION — LOS ANGELES ROYAL TSCHANTRÉ — BALTIMORE BRISTOL sub-miniature chopper 0.1 CU. IN. (shown actual size) Featuring □ airborne environmental ratings and complete shielding for military applications. □ Unmatched noise level for low level instrument applications. Write for detailed spec sheet. The Bristol Company, Aircraft Equipment Division, 152 Bristol Road, Waterbury 20, Conn. A Subsidiary of American Chain & Cable Company, Inc. In Canada: The Bristol Company of Canada Ltd. 71-79 Duchess Street, Toronto 2, Ontario. BRISTOL ...engineers for precision, builds for reliability. comparable with present technology attained for silicon and germanium, company says. Fabrication feasibility of high-temperature medium-power and low-power gallium arsenide transistors were only partially fulfilled because no satisfactory high-temperature and high-injection efficiency emitter has been found. Principal objectives of the TI program were to carry out studies for the Navy which would lead to the development of gallium arsenide transistors. Using Microcircuits In Micromodules SEVERAL recommendations were made by Servomechanisms for microelectronic circuits in micromodules. Where required, larger substrates containing more riser notches will not only permit more circuits per wafer, but will also permit more wafers per module. Multiples such as 0.310-in. × 0.620-in., and 0.620-in. square would be compatible with present dimensions. Servomechanism says glass is preferred over alumina as a substrate material when vacuum-deposited components are to be made, because of superior surface quality. Peripheral terminals, formed simultaneously with the vacuum-deposition of conductors, would save the expense of an extra process and eliminate a possible source of contamination. The gold terminals should have an undercoat of copper so that the terminals will not dissolve during soldering, company says. Other recommendations include protection of thin-film capacitors against shock if they are soldered, company says. Conductive epoxy serves well for attaching riser wires, according to findings. Servomechanisms says semiconductor manufacturers should supply transistors and diodes with 0.002-in., or 0.003-in. diameter gold leads which can be thermocompression bonded directly to thin films. Recommendations were submitted to the government as part of a program that investigated possible problem areas of micromodule circuits. Ravenous Tiger It's late evening. All the Tiger pickup trucks are empty. In five minutes, the last shipment will be fed into the huge cargo hold of the Swingtail-44 and the dock will be clear. Three minutes later, everybody's shipment will be on its way. Everybody's. That's the beauty of the Swingtail-44. It has a big appetite—33 tons. It carries off every package on the dock. Nothing waits to go out. Nothing gets left behind. And when you've got a whole fleet of 44's...and a fleet of trucks for door-to-door service...and a gang of detail-minded people to expedite your shipment all along the way...then you've really got airfreight service. The fast kind. The kind you can depend on. The Flying Tiger kind. first in airfreight with airfreight first FLYING TIGER LINE New low price is a big reason to use IBM Wire Contact Relays Compare IBM relay price/performance with the relay you are now using. Four PDT IBM relays start at a new low of $2.90, 12 PDT start as low as $5.40. And IBM latch relays start at $8.45. Add all this to the performance shown above and they can’t be beat. Delivery is un-beatable, too. Production shipments within 24 hours if requested. Design-in or plug-in an IBM relay. Here’s what you get: - Life: Up to 200 million operations. - Operate Speed: As fast as 4 ms. - Release: 5 ms. Performance is an even bigger reason to use IBM Wire Contact Relays at 48 VDC attainable with these relays. Coil Voltages: Up to 100 VDC. Integrated Package: Solderless connections, multiple coils, compactness and standardized mountings give you manufacturing savings and lower product costs. Applications. Wire contact relays are ideal for counting, logic switching, shift registers, flip-flop and timing circuits in numerical control, communications, test equipment, process control, data processing, supervisory control systems, and many other applications. Send for Catalog that tells you the relay story—performance, reliability, economy, and applications. NEW HIGH-AVERAGE-POWER LASER FLASH TUBES FX-65 Water-cooled tube designed to handle 8 kw average power input with water flow of 1 gpm. Capable of flash rates up to 50 fps. Arc length is 6½", overall length 12⅞", O.D. 15 mm overall, discharge bore 7 mm. A shorter tube with 3" arc length and 4 kw rating will soon be available. EG&G is in a position to quote on a power supply capable of driving the FX-65. FX-52 Air-cooled flashtube rated at 600 watts average power input with a minimum air velocity of 40 linear ft/sec. Electrical ratings are similar to those for EG&G FX-42. Arc length is 3", overall length 6⅜/16", O.D. 0.945" overall, discharge bore 7 mm. FX-55 Similar configuration to FX-52 but with 6" arc length, conservatively rated at 1000 watts average power input. FX-56 Very high energy flashtube designed for operation at up to 3000 joules per inch of arc length with pulse duration of 3 milliseconds. Arc length 6½", overall length 12⅞", O.D. 31 mm, discharge bore 28 mm. Also available on order in arc lengths from 3" to 36". FX-57 A higher rated version of the popular FX-47. Rating: 13,000 joules @ 3 millisecond pulse duration. Arc length 6½", overall length 12⅞", O.D. 15 mm overall, discharge bore 12 mm. Available on order in any arc length to 36". EG&G has the widest range of standardized Xenon flashtubes available, plus an unequaled capability for development and manufacture of special tubes. Write or phone Marketing Department for detailed information. Edgerton, Germeshausen & Grier, Inc., 176 Brookline Avenue, Boston 15, Massachusetts. ANTI-BACKLASH GEARING assures smooth drive... HANSEN SYNCHRON® GIVES "DIMENSIONAL STABILITY" for those tricky fourth-dimension design areas Sometimes it takes more than a quality motor to solve drive problems. That's why Hansen has teamed its reliable SYNCHRON motor with anti-backlash gearing for a chart drive application. Two identical gears are spring-loaded in opposing directions, both meshed with the same pinion. One gear holds in one direction while the other gear holds in the opposite direction. This allows minimum movement or "pickup" of the output shaft during rotation, insuring 1/2" maximum backlash at most speeds. And this is only one gearing adaptation — Hansen engineers design many special applications. ON ALL HANSEN SYNCHRON® MOTORS, pinions are machined from one piece of steel, carefully tested with "go"—"no go" gauges. Other gears may be of steel, brass, or special materials. They're double-staked for permanence and operate reliably with minimum noise. Outputs are 8, 20, or 30 inch ounces at 1 RPM. 90 inch ounces of static torque may be applied to output shafts without damaging the motor or stripping gears. If you have a special requirement, particularly one demanding precision, call on Hansen — today. APPROXIMATE SPACE REQUIREMENTS Round Mount — 2" diameter, 1 1/4" deep Pear Shape — 2" wide, 2 1/2" high, 1 3/4" deep OUTPUT DRIVES Pinion drives, flatted, threaded, cross-drilled, knurled, or slatted shafts; adapters, or crank assemblies. OR HANSEN WILL INTEGRATE PINIONS, GEARS, OR SPECIAL DRIVE ASSEMBLIES OF YOUR OWN MANUFACTURE. Information on D.C. motors available on request. HANSEN REPRESENTATIVES: R. S. HOPKINS COMPANY Sherman Oaks, Calif. EICHORN & MELCHIOR, INC. San Carlos, Calif. THE FROMM COMPANY, Chicago, Ill. H. C. JOHNSON AGENCY, INC. Binghamton, N.Y., Fayetteville, N.Y., Rochester, N.Y., Schenectady, N.Y., Williamsville, N.Y. WINSLOW ELECTRIC COMPANY Essex, Connecticut, New York City, Philadelphia, Pa. EXPORT DEPARTMENT 15 Moore Street, New York City General RF Fittings, Inc. 702 BEACON STREET, BOSTON 15, MASSACHUSETTS Telephone: 617 267-6120 CIRCLE 204 ON READER SERVICE CARD July 19, 1963 • electronics WILL IT WORK AT 10 MC? FIND OUT FAST WITH AN EECO HIGH-SPEED DIGITAL SYSTEM BREADBOARD It looks good on paper. Now—what will be the effect of clock duty cycle? Wiring? Propagation delay? You can get the answers fast with EECo system breadboard equipment—an exceptionally simple means of formulating and testing digital electronic circuits at clock speeds to 10 Mpps. This versatile transistor equipment lets you patch up trial circuit combinations with the same catalog modules that go into the final system, perform tests at operating frequencies by pushing a button on the control panel, and get a "stop-action" look at the over-all logic flow. The built-in clock generator permits fast set-up of a wide range of test conditions, and indicators give you an immediate reading on any part of the circuit. Plastic symbol cards further speed your work by giving you a road map of the system as you put it together. You may operate the system slowly to check individual operations or at end-system speeds, introducing high-speed pulses either singly or in pairs. If you want to try an alternative design idea, a few minutes of patching will make your brainstorm a reality. Discover for yourself this speedy route from paper to prototype. Write for our new breadboard brochure today. ENGINEERED ELECTRONICS Company 1441 East Chestnut Avenue, Santa Ana, California Telephone: 547-5651 Cable Address: ENGELEX Module Production Quality Assured by Thermal Cycling and Life Testing Checks stringently on flow soldering and special construction techniques By T. V. GORE, W. V. LANE U. S. Army Electronics R & D Lab. Fort Monmouth, N. J. THERMAL CYCLING tests give a close and immediate check on reliability levels of cordwood printed circuit modules manufactured by Republic Aviation Corp. (Farmingdale, N. Y.). Operating under a program sponsored by the Army's Electronic Research & Development Laboratory, Republic employed the tests as an integral part of production process. Tests testified to the effectiveness of special techniques in manufacturing modules: lead compliance configuration, pre-tinning leads for reliable soldering, two-stage packaging (coating and encapsulation). It was definitely shown that component failure rates are less than that estimated for military parts in units assembled by conventional means. ASSEMBLY—The disciplined layout of subminiature parts in cordwood modules with packaging densities as high as 100,000 parts/cu. ft. demands use of sophisticated assembly techniques. These, in turn, demand close checking to weed out any production flaws occurring during the making of the complex module structure: axial-lead parts are stacked between two parallel printed wiring end plates (boards), while multi-lead parts such as transistors are nested one-over-the-other with leads attached to the appropriate plate. When assembled modules are subsequently loaded with a silica-filled epoxy, stresses are introduced to soldered joints at both lead-part interfaces and lead-printed wiring interfaces. These stresses are sufficient not only to cause failures of soldered joints but can also result in the failure of components. SOLDERED JOINTS — To obtain maximum strength in the soldered joint, plates having plated-through holes were used, giving a large soldering surface area. Also, pre-tinning of all part leads provided optimum surface conditions. Following the above requirements, excellent joint uniformity has been obtained with hand soldering. However, even further improvement is expected with a flow soldering (wave soldering) process developed by Republic for large production runs. FLOW SOLDERING—Flow soldering is also expected to minimize soldering-heat effect on parts. ReIn our photocell, too—the Blue Dot says it works Sylvania photocells and flashbulbs have one obvious thing in common: Blue Dot insurance against a common enemy, moisture. Reliable protection for the cadmium sulfide in Sylvania photocells comes from hermetic sealing in a one-piece, glass enclosure. For extra safety, the moisture-sensitive Blue Dot is put there. If the hermetic seal is ever broken and moisture enters from any cause whatsoever, the Blue Dot turns pink. As little as .02% water vapor triggers this warning, enabling you to spot the defective cell, in most cases, days or weeks before it actually becomes inoperative. If you're in the market for photoconductors, consider the Sylvania T-4 line. Rated at 400 volts and featuring 300 mW dissipation, the selection ranges from 750 ohms to 16,000—widest in the industry. One of them is probably right for your streetlighting, furnace alarm, remote control or other application. See your Sylvania sales engineer or write to Electronic Tube Division, Sylvania Electric Products Inc., Box 87, Buffalo, New York. AT WESCON: BOOTHS 2901-06 AND 2917-22 SYLVANIA SUBSIDIARY OF GENERAL TELEPHONE & ELECTRONICS NEW CAPABILITIES IN: ELECTRONIC TUBES • SEMICONDUCTORS • MICROWAVE DEVICES • SPECIAL COMPONENTS • DISPLAY DEVICES 222 Steps to the first plateau Tantalum powder travels a long road until it assumes final form in a solid or wet slug capacitor. A minimum of 222 individual steps are required for complete electrical, chemical and physical evaluation of the powder before it is transformed into its usable capacitor form, the pellet. Someone is watching, measuring or testing...before, during and after each operation. When the powder is pressed into pellets only the first plateau has been reached. Throughout the entire manufacturing process, inspections and tests are conducted and recorded on every production lot of tantalum capacitors. If we seem fanatical about testing, it's because we know from over twelve years experience in making tantalum capacitors that this is essential to consistently high quality. And that's what you're always sure of getting in Mallory tantalum capacitors. They're the product of the most modern tantalum capacitor facility in the world. Visit us and see for yourself. Mallory Capacitor Company, Indianapolis 6, Indiana—a division of P. R. Mallory & Co. Inc. SOLID, WET SLUG AND FOIL TANTALUM CAPACITORS CIRCLE 94 ON READER SERVICE CARD Mallary tantalum capacitors delivered from stock at factory prices by these distributors: - Baltimore, Md. Radio Electric Service - Binghamton, N.Y. West Electronics - Birmingham, Ala. MG Electronics & Equipment Co. - Boston, Mass. Crown Electronics DeMambro Radio Supply Co. Lafayette Radio QPL Sales, Inc. - Buffalo, N.Y. Summit Distributors, Inc. - Chicago, Ill. Allied Electronics Corp. Newark Electronics Corp. - Cincinnati, Ohio United Radio - Cleveland, Ohio Pioneer Electronics - Dallas, Texas Engineering Supply Co. Hall-Mark Electronic Corp. - Dayton, Ohio Stotts-Friedman Co. - Denver, Colo. Newark Denver Electronic Supply - Houston, Texas Harris Radio Equipment Co., Inc. Lenert Company - Indianapolis, Ind. Graham Electronics Kansas City, Mo. Radio Lab - Los Angeles, Calif. California Electronics Kieruff Electronics, Inc. Lynch Electronics Radio Product Sales - Minneapolis, Minn. Northern Electronics Corp. - Montreal, Que. Canadian Electrical Supply Co. - Muskegon, Mich. Fitzpatrick Electric Co. - Nashville, Tenn. Electra Dist. Co. - New York, N.Y. Lafayette Radio - New York, N.Y. Harrison Electronics Corp. Harvey Radio Co., Inc. Lafayette Radio Mills Electronics Terminal Division Electronics. - Oakland, Calif. Elmar Electronics, Inc. - Oklahoma City, Okla. Radio Lab - Orlando, Fla. Hammond Electronics, Inc. - Ottawa, Ont. Ward's Radio-TV Lab. - Palo Alto, Calif. Zack Electronics - Perth Amboy, N.J. Atlas Electronics - Philadelphia, Pa. Herban & Reisman Philco-Milta Electronics - Phoenix, Ariz. Kieruff Electronics, Inc. - Pittsburgh, Pa. Radio Lab, Inc. - Salt Lake City, Utah Kimball Electronics - San Antonio, Texas Parry Sliwinski - San Francisco, Calif. Kieruff Electronics, Inc. - St. Louis, Mo. Olive Electronics - Seattle, Wash. F. Connolly Co. - Springfield, Ill. Federated Purchaser, Inc. - Toronto, Ont. Alpha Aracan Radio Co. Electronic Supply Wholesale Radio & Electronics - Tulsa, Okla. Empire Radio Supply Co. - Washington, D.C. Capitol Radio Wholesalers Electronic Industrial Sales - White Plains, N.Y. Westchester Electronic Supply Co., Inc. - Winston-Salem, N.C. Electronic Wholesalers Inc. MANUFACTURING FLOW DIAGRAM indicates use of tests during and after production of modules public set up the following process specifications: Solder used conforms to QQ-S-571, having a composition of S, 60. Solder bath temperature is maintained between 500 degrees F (260 degrees C) and 520 degrees F (272 degrees C). Pump speed is set so that the wave is approximately $\frac{1}{4}$ inch above the level at underside of board; traversing track carrying board is arranged so that this does not require a wave height above $\frac{1}{4}$ inch over discharge nozzle level (see accompanying diagram) unless there are projections on underside of circuit which must have higher wave to clear nozzle. In order that no portion of assembly be in tangential contact with solder bath (having above-indicated temperatures) no longer than 5 seconds, the transfer mechanism is made to carry assemblies through solder at speeds of from 2 to 4 feet per minute. LEAD COMPLIANCE—Linear displacement of the two end plates can be as much as 0.006 inch when modules encounter military equipment temperature changes ranging from $-55$ to 85 degrees C. To offset resulting lead stresses, bell-shaped kinks with dimensions of approximately 0.03 by 0.06 inch are included in part leads of all diodes, jumpers and ¼-watt resistors. This has proven an adequate substitute for finding an encapsulant having an expansion coefficient matching the accumulative coefficient of the various part materials. ENCAPSULATION — After assembly and soldering, parts and leads are surrounded with a compliant coating that prevents later adherence of epoxy potting compound to permit lead movement when axial forces are applied at lead terminal joints. Dow Corning DC-271, for example, prevents adherence because of its resilient silicon base material. After coating, the assembly is placed in an open mold and cast in a silica-filled epoxy (HYSOL 6020 or equal). Casting is then cured, and excess resin trimmed off. THERMAL CYCLES—As part of the production process, all completed modules were subjected to 5 thermal cycles. Each cycle was sequenced as follows: - A temperature reduction to $-55$ degrees C from room temperaCircuit breaker selection made easier! New fact file describes distinctive advantages of all four types of KLIXON circuit breakers For the first time, you can get all the data you need to evaluate thermal, thermal-magnetic, magnetic and magnetic-electronic types of circuit breakers... in one convenient package. Only TI makes all four... and favors no particular type. So, you get unbiased help in selecting the type that best protects your circuits and/or systems from faults due to overcurrent or overvoltage. Compare! Our SP-40 fact file shows that New KLIXON Magnetic-Electronic Circuit Breakers, with nanosecond response, are the fastest acting devices ever made for protecting transistorized equipment circuitry against transient and prolonged overloads. You'll also find that New KLIXON Magnetic Circuit Breakers combine trip-free protection, fast response time and more design options than comparable units. As for KLIXON Thermal and Thermal-Magnetic Circuit Breakers, you'll see that their high rupture capacity and resistance to shock and vibration have no equals for protecting wire and cable in airborne applications. Write today for our new SP-40 fact file... the quickest, easiest means of selecting the right circuit breaker from all four types. METALS & CONTROLS INC. 5007 FOREST ST., ATTLEBORO, MASS. A CORPORATE DIVISION OF Texas Instruments Incorporated LIFE TESTS—Modules surviving thermal tests were subjected to various environmental tests without any failures: shock, vibration, moisture resistance, altitude. These demonstrated that military-quality resistors, capacitors, transistors and diodes could be packaged in the printed circuit cordwood module without degrading their electrical performance. Storage life tests exposing modules to 71 degrees C for 2,600 hours gave the same answer. The biggest process hurdle for modules was an operational life test. Modules numbering 274 were test-circuit operated in an oven ambient temperature of 85 degrees for a total of 3,750 hours. Six capacitors failed. But 5 of these were in read-amplifiers, in which as mentioned previously an assembly process error resulted in capacitors being installed with reverse polarity. Since only very low voltage was applied to capacitors, this indicated effectiveness of life test in checking-out production. In view of above, only one valid component failure occurred after the 3,750 hours. This duration together with the exposure of 274 modules each containing at least 539 components, established a confidence level of 75 percent. At the end of that time, an additional 2,750 hours was accumulated for a total of 6,500 hours to provide 1.78 million unit hours for modules as a group. Unit hours for parts ranged from 3.1 million (capacitors) to 76.7 million (solder To Order Reprints Fill in, Cut out coupon below, insert in envelope and mail to: electronics Reprint Dept. 330 West 42nd Street, New York, N.Y.—10036 REPRINT ORDER FORM For Listing of Reprints Available See Reader Service Card (To help expedite mailing of your reprints please send cash, check or money order with your order.) For Reprints of the latest Special Report: Radio Frequency Interference Send me Reprints of Key No. R-39 1-10 copies 75¢ each, 11-24 copies 60¢ each, 25 or more 50¢ each. For Reprints of previous Special Reports or Feature Articles fill in below: Send me Reprints of Key No.(s) @ $ each. (For prices, see chart above.) For orders of Bulk Reprints of other editorial articles in this issue or past issues: Send me Reprints of page (no.) of issue of article entitled. Minimum bulk order 100 copies. You will be advised of costs by return mail. Name Number and Street City, Zone No., State Inland Gearless Torquers help make pinhole detection fool-proof! Bethlehem Steel Company runs tinplate at high speed through a photo inspection device. Purpose is to detect pinholes. Side-to-side movement of the continuous strip creates a problem. The least light entering at the edges of the strip can cause false pinhole indications. Fast-response positioning of edge guides by two Inland Gearless Torquers on signal from photoelectric sensors blocks out "false-alarm" light. These direct-drive d-c torque motors have peak torque of 60 ounce-inches. Rapid, high-resolution response to servo-position error signal has earned for Inland Gearless Torquers a place in all major missile and space programs to date, as well as in an increasing number of industrial applications. Their superior performance comes from torque-to-inertia ratios 10 times higher than equivalent gear-train servo motors. Moreover, their compact, pancake configuration overcomes space and weight limitations. What's your problem? If you're currently planning a servo system calling for output torque between 20 ounce-inches and 3000 pound-feet, compare Inland Gearless Torquers with any alternative. Write for all the facts today, 347 King Street, Northampton, Mass. Higher torque output levels can be provided on special order. --- Pinhole detector for continuous inspection of tinplate has an automatic edge shield system developed by Bethlehem Steel Co. --- Calculated on basis of established 75 percent confidence level mean time to failure for parts was as follows: - Transistors—1.9 million hours - Diodes—7.3 million hours - Resistors—6.6 million hours - Capacitors—1.2 million hours - All parts—9.3 million hours - Solder joints—55.5 million hours Translating above meantime to failure data to failure rate per 1,000 hours at a 75 percent confidence level, we have: - Transistors—0.052 percent per 1,000 hours - Diodes—0.013 percent per 1,000 hours - Resistors—0.015 percent per 1,000 hours - Capacitors—0.087 percent per 1,000 hours - All parts—0.01 percent per 1,000 hours - Solder joints—0.002 percent per 1,000 hours Thus, tests indicate and provide assurance that the printed circuit modular configuration manufactured with techniques described, can result in component reliability levels equal to or greater than levels reported for similar components in a "free-air" (non-encapsulated) environment. --- Heliax Cable Goes to Five Inches Communication cable measuring 5-inches in diameter has convoluted inner and outer copper tube conductors and air dielectric to permit easy bending. Addition of a 5-inch diameter cable to its heliax line is being made by Andrew Corp. of Chicago. For use in high power 2-way communications, the cables are manufactured from oxygen-free high conductivity copper. Newest addition to the flexible-cable line, uses... What new high speed tape transport means less down time? Here's how it's done. The highly stable, solid state servo requires less adjustment. Capstan rollers are quickly changed. Settings are quickly made. The head assembly is isolated. (This makes the tape path highly stable. In fact, dynamic skew at start time is less than dynamic skew continuous.) And precision practices reduce static skew. There's no tape flap; an erase head isn't needed. Positive safety interlocks eliminate tape damage. Instantaneous speed variation has been reduced. Plus: up to 150 ips tape speeds; 2 ms maximum start time; 1.5 ms maximum stop time; 800 bpi capability with or without clock track; available as a complete tape memory system with transfer rates up to 240 kc. And above all, there's Ampex reliability and ruggedness throughout. The TM-5 is made by the Ampex Computer Products Co., Culver City, Calif. For information write Ampex Corp., Redwood City, Calif. Worldwide sales, service. CIRCLE 101 ON READER SERVICE CARD 101 Hall Device Built for Production Line PRODUCTION LINE permanent magnet testers using a Hall-effect probe is being introduced by Indiana General Corp. of Valparaiso, Indiana. Testers are claimed to be ideal for on-line use, quality control, and incoming inspection because of a better than 1-percent repeatability. Colored panel lights give visual indication of whether magnets in a given run are in either of two magnetic-quality categories. A readout meter checks precise quality level values. Provision is made for attachment of external bells, buzzers, switches, gating circuits which can be set to operate in conjunction with the quality-indicating lights. Constant-current supply provided for probe has a voltage stabilizer. Cleaner Than Clean WALL of absolute filters (rear) draws air into clean room at General Dynamics/Pomona. Air is drawn-out through a facing wall of return air registers. Reportedly, room is 200 times cleaner than hospital operating rooms and exceeds maximum military standards for clean rooms. Time after time engineers specify Johnson sockets! Whatever the choice...a miniature 7-pin steatite wafer or a low-loss Kel-F socket for high power transmitting tubes...time and time again design and development engineers specify Johnson tube sockets! All sockets have been categorized under a socket standardization program, reducing the number of variations in each socket type. Standardization and immediately available stock shortens delivery cycles—permits fast selection of a Johnson socket for almost any application! Standard: Commercial grade for general requirements. Bases are glazed porcelain or steatite. Etched aluminum shields or bayonet shells. Industrial: Superior in quality to "Standard" Grade. Glazed steatite bases, DC-200 treated. Phosphor bronze or beryllium copper contacts .0005 silver-plated. Aluminum shells and shields are iridite No. 14 treated. Fungus resistant cushion washers under contacts. Military: Top quality to meet military requirements. Glazed L-4 steatite bases, DC-200 treated. Phosphor bronze or beryllium copper contacts heavily silver-plated. Hot tin dipped solder terminals. Brass bayonet shells .0003 nickel-plated. Aluminum shells and shields are iridite No. 14 treated. Fungus resistant cushion washers under contacts. Wafer sockets protected for 200 hour salt spray test. KEL-F SERIES—Molded of low dielectric loss-factor Kel-F plastic—designed for use with a wide selection of high power transmitting tubes such as: 6CX250B; 6CX250A; 4CX250B; 4CX250B; 4X250F; 7034; 7035. Basic sockets are available in several designs—with or without screen grid by-pass capacitors, mounting saddles, or steatite chimney to direct air flow through tube envelope. A molded-in "guide" is machined for greater alignment accuracy, and tabs for 6-32 machine screw. All contacts are low resistance silver-plated beryllium copper. Tube pin contacts are heat treated to provide positive contact pressure as well as extended life. Annealed soldering tabs may be easily bent or formed. BAYONET TYPES—Includes Medium and Heavy Duty Medium, Jumbo and Super Jumbo 4-pin types. For use with tubes such as: 866A or 811A, E.I.A. Base No. A4-10; 872A, 211, and others with E.I.A. Base No. A4-29; and tubes such as: 8008, SC22, FG104, GL146 and others with E.I.A. Base No. A4-18. STEATITE WAFER TYPES—Available in 4, 5, 6, 7, and 8-pin standard socket types, as well as Super Jumbo 3-pin for tubes with E.I.A. Base Nos. A4-15, A4-16, and A4-18. Giant 5 and 7-pin models for tubes with E.I.A. Base Nos. A5-19 and A7-17. Special sockets for tubes such as the 7094 with E.I.A. Base No. E7-2; and VHF Power Sockets for tubes such as: 5894, 6524, 6252 with E.I.A. Base No. E7-20; and 826, 832, 4D32 with E.I.A. Base No. E7-2. MINIATURE TYPES—All steatite, available in Standard Wafer Type or Shield Base Type for 7-pin miniatures such as the: 1RS, 1S5, 6C86, etc., with E.I.A. Base No. E7-1. SPECIAL PURPOSE TYPES—Includes sockets for special purpose tubes such as the: 204A and 849; the 833 and 833A; 152TL; 304TL; 750TL; 1500T; 2-2000A; 5D21, 705A and others. NOTE: Detailed specifications on all Johnson tube sockets have been prepared for engineering department use in Socket Standardization Booklet 536. Should you wish a copy—please make your request on company letterhead. E. F. JOHNSON COMPANY 2821 TENTH AVENUE S.W. • WASECA, MINNESOTA STANDARD, INDUSTRIAL, MILITARY TUBE SOCKETS DETAILED COMPONENTS CATALOG AVAILABLE—Write today on company letterhead • CAPACITORS • TUBE SOCKETS • CONNECTORS • PILOT LIGHTS • INSULATORS • KNOBS AND DIALS • INDUCTORS • HARDWARE electronics • July 19, 1963 CIRCLE 103 ON READER SERVICE CARD 103 Synchronizer Accepts PCM At Rates To 1 Mc Signal conditioner will operate over a 40 db dynamic range MODULATION bit synchronizer and signal conditioner manufactured by Defense Electronics, Inc., 5455 Randolph Road, Rockville, Md., is designed for use in shipboard, aircraft and ground station telemetry systems. Called the model BRS-1, the unit will accept a serial pcm signal at rates up to 1 Mc and will operate over a 40 db dynamic range, allowing the signal to originate from the output of a telemetry receiver, tape recorder or other source. Moreover, the BRS-1 will generate a stable clock that is phase-coherent with data output. The synchronizer also accomplishes functions of data reconstruction and format conversion. Its output is a reshaped, noise-free, serial pcm signal plus a clock signal that is phase-coherent with output data. Bit rate/clock frequency is accurately established by means of thumbwheel switches and push-button range selectors so that a known bit rate can be established within 0.1 percent prior to actual reception. This feature permits the unit to be operated without the bevy of test equipment usually required for setup. A completely digital loop filter automatically establishes a loop bandwidth that is proportional to bit rate at all frequencies. According to the manufacturer, the integrate-and-dump-bit-detection circuit uses a unique correlating function generator that causes the bit detection performance to approach that of a perfectly matched filter. CIRCLE 301, READER SERVICE CARD Unique Bridge Measures RF Operating Impedance MODEL OIB2 high-frequency operating impedance bridge can be inserted directly into any part of a 2-30 Mc antenna system to measure the operating impedance of individual radiators, network inputs or transmission line terminals under r-f conditions. According to the manufacturer, Delta Electronics, 4206 Wheeler Ave., Alexandria, Va., the OIB-2 will handle powers up to 1,000 watts and can be excited either directly from the operating r-f power or a high-power signal generator. Moreover, they point out that the device is particularly useful in the adjustment of multi-element high-frequency antenna systems, where the insertion of conventional units might upset the circuit. Insertion effect of the bridge approximates that of 5 inches of 150-ohm transmission line. An external detector jack permits the unit to be used with a low-power signal generator and sensitive communications receiver as a normal impedance bridge for antenna or laboratory r-f measurements. The OIB-2 has a resistance range of ±500 ohms and a reactance range MINCOM SERIES CM-100 RECORDER/REPRODUCER Built-in reliability inspires confidence everywhere in this Mincom Wideband Instrumentation Recorder/Reproducer. CM-100's reputation for minimum downtime stems from its rugged simplicity—only twelve moving parts, four easy adjustments, dynamic braking. Plus, of course, overall electronic craftsmanship and the extreme precision achieved by Mincom's longitudinal recording with fixed heads. Versatile, too: 1.5 megacycles at 120 ips for analog data storage or simultaneous post- and predetection capabilities in FM/FM modulation, PCM, PCM/FM and other FM-type carrier systems. Seven or fourteen tracks. Write today for complete specifications. Mincom Division 3M Company 2049 South Barrington Avenue, Los Angeles 25 425 13th Street N. W., Washington 4, D. C. BUSS Sub-Miniature Fuse-Holder Combination A light weight, protective device for space-tight applications in multiple circuit apparatus. Fuse has transparent window for visual inspection of element. Fuse may be mounted alone or used in holder on printed circuit boards. HWA holder can also be panel mounted with or without use of knob. Knob makes holder water proof for front of panel. For full details write for BUSS bulletin SFB. BUSS BUSSMANN MFG. DIVISION, McGraw-Edison Co., St. Louis 7, Mo. BUSS Miniature Fuses Made To Foreign Standards 5MM x 20 MM (.197 x .769 INCHES, GLASS TUBE) Designed for protection of miniaturized circuits or equipment. Commonly used in equipment of foreign make. Write for BUSS Bulletin SFB. BUSS BUSSMANN MFG. DIVISION, McGraw-Edison Co., St. Louis 7, Mo. BUSS: the complete line of fuses. AROUND THE WORLD IT'S KEW MODEL P-25 MODEL P-40 MODEL PV-202 MODEL TK-90A MODEL KHE-505 MODEL EW-8 MODEL EW-16 KYORITSU ELECTRICAL INST. WORKS, LTD. NO. 120, NAKANE-CHO, MEGURO-KU, TOKYO, JAPAN CABLE ADDRESS: "KYORITSUKEIKI TOKYO" TEL (717) 0131-5 - 0151-3 FAST DELIVERY ANYWHERE MIDWEC INSTRUMENT GRADE MYLAR* DIELECTRIC Ca-pac-i-tors Best Shipping Interval In The Industry—3 Weeks Standard High Reliability and Quality Competitively Priced Specialists In Low Tolerance Units Approved for use in Talos, Minute Man, Titan, Typhon, Telephone Companies ■ 100% Test for dielectric strength, capacitance, insulation resistance and dissipation factor MIDWEC OSHKOSH, NEBRASKA write for data sheets and prices SALES OFFICE: 601 So. Jason St., Denver 23, Colo. TWX 792-3891—Telephone SM 4-3481—DDD 303 DuPont TM for Polyester Film. of ±500 ohms and is contained in a heavy deep-drawn aluminum instrument case with detachable cover and carrying handle. The unit sells for $695 and is constructed to ensure dependable performance under rugged field conditions. CIRCLE 302, READER SERVICE CARD Regulated Supplies Have Mode Indication MANUFACTURED by Kepco, Inc., 131-38 Sanford Avenue, Flushing, New York, Model VIX indicators provide visual evidence of power supply mode and exact point of crossover from constant voltage to constant current regulation. According to the company's engineers, VIX indication greatly simplifies testing and provides substantial savings in time and engineering. Supplies using the mode indicators use two control bridge circuits, one for voltage and one for current. Between these and a control pass device, an exclusive-OR circuit acts as a trigger or switch without ambiguity or back-lash. The OR circuit both selects the current or voltage bridge and switches the lights that provide visual indication of operating state. Aside from the external indicators, Kepco supplies, embodying VIX, feature front-panel pilots for the same purpose. Moreover, VIX supplies can be stacked in either series or parallel configurations without auxiliary or slaving connections. In this way, the mode of the individual units is readily shown along with the added versatility available with stacked units. The parallel configuration along with operational characteristics is shown in the diagram along with a schematic of the basic system. (303) Tiny Frequency Meter Has 1/4% Accuracy CRYDOM LABORATORIES, Inc., 12850 Western Ave., Garden Grove, Calif., has just announced the availability of a new precise frequency meter that achieves an accuracy of ±¼ ... of unquestioned high quality BUSS Sub-Miniature PIGTAIL TRON FUSES Body Size Only .145 x .300 Inches Tron fuses are so small they can be used as an integral part of circuit—to protect miniaturized devices—or gigantic multi-circuit electronic devices, without sacrifice of space. They are hermetically sealed for potting without danger of sealing material affecting operation and have high resistance to shock or vibration. Operate without exterior venting. May be teamed with other components in replaceable unit. BUSS BUSSMANN MFG. DIVISION, McGraw-Edison Co., St. Louis 7, Mo. If you should have a special problem in electrical protection... ... we welcome your request either to quote or to help in designing or selecting the special type of fuse or fuse mounting best suited to your particular conditions. Submit description or sketch, showing type of fuse to be used, number of circuits, type of terminal, etc. If your protection problem is still in the engineering state, tell us current, voltage, load characteristics, etc. Be sure to get the latest information BEFORE final design is crystallized. At any time our staff of fuse engineers is at your service to help solve your problems in electrical protection. BUSS BUSSMANN MFG. DIVISION, McGraw-Edison Co., St. Louis 7, Mo. NOW! RECORD Volts, Ohms, Milliamps with ONE RECORDER ... NO EXTRAS! New Bausch & Lomb V.O.M.-5 RECORDER ... an all-new, complete 5-inch strip-chart recorder that breaks all precedent in the field ... brings you the finest features of potentiometric recorders for one low price. Compare these exclusive advantages, all these "extras" at no extra cost, with any other recorder in its class. - Five voltage ranges, 10 millivolts to 500 volts D.C.—full scale deflection. - Six linear ohms scales, 1-to-100,000 ohms full scale, with zener diode D.C. supply. - Four D.C. current ranges—10 microamperes to 100 milliamperes. - Off balance input impedance—over 10 megohms. - Five chart speeds, 400-to-1 range. - Event marker, with interchangeable pens. - Function switch with mechanical pen letdown. - Operates in flat, 30° tilt, or wall-mounted position. - Compact—only 4¾" x 14½" x 11¾". - Portable—only 16 lbs. And more. Lots more! Mail the coupon now for the whole story on this new 5-speed recorder with versatility-plus! BAUSCH & LOMB Kilovoltmeter Is Fully Transistorized DUAL INSTRUMENT has 0 to 5 Kv/10Kv/50 Kv and 100 Kv d-c and 60 cycle a-c ranges, large 6 in. precision mirror scale meter. Completely transistorized with virtually zero drift amplification, results in Here's a valuable new find in your endless quest for significant design advantages — Hitemp's "TWR" Ribbon Cable. Lighter than round conductor cable, "TWR" separation of individual conductors is controlled to 15-20 mils spacing, with its flat configuration offering unlimited design possibilities in terms of the number of conductors, types and sizes of wires and cables, and color coding, among other specifications. In such difficult applications, for example, as situations where lateral movement is unavoidable, "TWR" Ribbon Cable has exceptional durability because the unsintered Teflon tape covers 90% or more of the outer surface — reducing wear to an absolute minimum. And Hitemp also makes other ribbon cables with Teflon insulated wires — one where individual insulated conductors are heat bonded together, Type "ER", another where individual insulated conductors are woven with either Dacron, Rayon, Nylon, or Teflon yarn, Type "TR". TWR, ER, and TR ribbon configurations are perfectly suited to miniaturized components, restricted packaging, and point-to-point contact for hook-up of control, signal, and power circuits. So, now you can relax whenever size, weight, abrasion resistance, flexibility, and reliability are critical — because Hitemp's Teflon insulated Ribbon Cables provide the safe, sound solution. SEND FOR NEW BULLETIN No. 100 for the full details on HITEMP Ribbon Cables. Or, send along a piece of what you're currently using — we'll evaluate it carefully and advise you of any savings you might be able to achieve with Hitemp's Teflon insulated Ribbon Cables. Take advantage of either or both offers right now. HITEMP WIRES CO. a Division of Simplex Wire & Cable Co. 1200 SHAMES DRIVE, WESTBURY, NEW YORK 1532 S. CALIFORNIA AVE., MONROVIA, CALIF. only a few microamperes drain on voltage being measured. Features a novel compensator which automatically provides for voltage coefficient correction of multiplier resistors. Model 50 has special spike bypass circuitry to make it insensitive to overloads up to 100 percent. Kalpa Scientific Laboratories, Inc., P. O. Box 172, Flemington, N. J. CIRCLE 305, READER SERVICE CARD Switch Assemblies Wired to Customer Order NEW FEATURES in the series L7700 illuminated push button switch include alternate action stations (push-lock, push-release), lock-out, or "no-two operate" feature between rows as well as within each row of a two row interlocking assembly, and square plexiglas buttons, white or a choice of several colors. Single or double row assemblies are available with from 2 to 20 stations in each row. An escutcheon plate simplifies mounting since only one panel opening is required. Donald P. Mossman, Inc., Box 265, Brewster, N. Y. (306) D-C Supplies Use Silicon Transistors REGULATED d-c power supplies are built with silicon transistors, have a 75 C ambient temperature range and sell for $89. Mean time between failure is over 25,000 hours. Models are available with output voltages from 6 to 31 v d-c and current capacities from 600 ma to 1.5 amp. Input is 105-125 v a-c, 47 to Compact View of Hart Relays SERIES R&S Miniature, Hermetically Sealed. 4PD'T. Contact ratings from microamperes to 10 amps. Meet or exceed MIL-R-5757D. A-c coil version available. SERIES P High Speed Polarized. SPDT. Operating response to 200 microseconds. No contact bounce. SERIES W General Purpose. DPDT, double break, a-c, d-c relays. Plug-in type or quick-disconnect terminals. Rated up to 25 amps, yet more compact than most 10 amp relays. Holding contact available. For complete information write to: HART MANUFACTURING CO. 202 BARTHOLOMEW AVENUE HARTFORD 1, CONNECTICUT Telephone: Area Code 203 525-3491 A SUBSIDIARY OF OAK MANUFACTURING CO. Digital Voltmeter Is Highly Stable MODEL 600 series Decameters are precision 5 digit differential instruments designed for measurement of a-c and d-c voltages and resistance. They offer 0.01 percent accuracy for a-c voltage measurements. Extremely high input resistance on all ranges, Zener references, complete overload protection and transistorized circuitry provide an extremely reliable instrument. A recorder output is optionally provided to drive any conventional T-Y, strip-chart or X-Y recorder or digital voltmeter equipped with automatic printout. The Decameter output is a conventional 100 mv for a deviation as low as ±100 µv or ±10 milliohms. Houston Instrument Corp., Auto Data Sales Div., 4950 Terminal Ave., Bellaire, Texas. (308) Pressure Controller Works Automatically INSTRUMENT controls pressure or partial pressures from $10^{-11}$ mm Hg up to one atmosphere and is bakeable to 450 C for minimum contamination. The throughput of a single unit is large enough to swamp a 50,000 liter/sec pump at $10^{-2}$ mm Hg and small enough to control laboratory systems operated at the lowest pressures attainable. Applications are in ion sources of mass spectrometers and high energy particle accelerators, in space DYNAMICS VARIABLE ELECTRONIC FILTERS - 3rd to 6th order Tchebysheff Butterworth Linear Phase - Up to 36 db/octave slope - Low-pass or high-pass - Low dc drift - Source or load independent - Drive long lines - No insertion loss - Variable frequency — 4-decade range Noise and ripple: below 1 millivolt peak-to-peak, at the output Linearity: 0.1% Output voltage and current: ±10 volts; ±100 ma Drift: ±2 millivolts at output Output impedance: below 1 ohm Input resistance: 10 megohms 10th decade steps over a 4-decade range Model 6370-L6 (illustrated) is a linear phase, low-pass, 6th order filter. Phase is linear in the range dc to 2½ times the cut-off frequency. All Dynamics filters are compatible with other Dynamics signal conditioning equipment. Write for literature on this model, or on the entire line of electronic filters. DYNAMICS INSTRUMENTATION COMPANY 583 Monterey Pass Rd., Monterey Park, Calif. • Phone: CU. 3-7773 CIRCLE 21D ON READER SERVICE CARD NEW BETTER-THAN-EVER RELIABILITY for long-distance point-to-point communications NORTHERN RADIO NEW 16-CHANNEL TRANSISTORIZED VOICE FREQUENCY DIVERSITY CARRIER TELEGRAPH TERMINAL TYPE 235 MODEL 3 MIL DESIGNATION AN/FGC-61A ... All units militarized: components and design approved by U.S. Military. ... Converters have equalized gain and adjustable time delay in each channel for better diversity performance and interchangeability. ... Switching Panels provide "local" or "remote" selection of 2-channel or 4-channel diversity modes. ... Combiners have adjustable gains in each channel, for complete switching flexibility, and the combining follows an ideally modified square law function for both 2-channel space or frequency and 4-channel space plus frequency diversity. ... Keyers have adjustable "threshold" sensitivity control and simplified input circuit selection. ... Detter and Delay Indicator provides test keying signal source for keyers and delay equalizers in all channels. Write for complete literature. Pace-Setters in Quality Communication Equipment. NORTHERN RADIO COMPANY, inc. 147 WEST 22nd ST., NEW YORK 11, NEW YORK In Canada: Northern Radio Mfg. Co., Ltd., 1950 Bank St., Billings Bridge, Ottawa, Ontario. CIRCLE 112 ON READER SERVICE CARD simulators, backfilling, process control or wherever manual control of pressure is now used. Granville-Phillips Co., 5675 East Arapahoe, Boulder, Colo. CIRCLE 309, READER SERVICE CARD Volt-Ohm-Milliammeter In Two New Types MODEL 260 vom is now available in two new types—series 4 with standard scale and series 4M with mirror scale. They feature self-shielded movement; higher accuracy (2 percent d-c, 3 percent a-c); ruggedness from spring-backed jewels; greater repeatability; and increased linearity and stability. Price of the series 4 is $48.95; series 4M, $50.95. Simpson Electric Co., 5205 West Kinzie St., Chicago 44, Ill. (310) SPDT Diode Switch Spans 100-2,500 Mc DEVELOPMENT of model 7762 spdt solid state microwave switch is announced. It covers from 100 to 2,500 Mc and accomplishes high speed channel switching while having a high isolation and low insertion loss. Termination of a third port allows the unit to be used as an absorptive type voltage controlled attenuator, r-f modulator, or an Electronic Systems Engineers INTERESTED IN broad-gauge projects These are rare opportunities for youngish, aggressive EE's in that the work is both interesting and important, the future very much up to you, and the locale delightful. You will be responsible for the systems design—in terms of adaptation and compatibility—the installation, test, and maintenance of ALL the Navy's shore-based electronics systems in several Middle-Atlantic states, Iceland, Bermuda, and the Azores. Most of your work will involve the Naval Tactical Data and Communications System in the Norfolk-Portsmouth-Virginia Beach area, and include contact with contractors and engineers of other Navy activities. Specific staff openings include: • COMMUNICATIONS ENGINEER, for all Naval shore-based radio systems • SYSTEMS ENGINEER, to concentrate on the NTDS facility at the Fleet Programming Center (Both of the above positions start at $9,475, plus the benefits of Career Civil Service, and will be increased to $9,980 on January 1st) • GENERAL ELECTRONICS ENGINEERS to work on special projects such as Command Control Centers, AIRNAVAIDS, Radar, Communications, etc. (These positions start at $8,575—$8,970 after January 1st—and include the complete benefits of Career Civil Service.) If you have your BS degree, plus three or more years' experience, and are capable of working independently on electronics systems...from concept to conclusion...get in touch with: INDUSTRIAL MANAGER FIFTH Naval District P. O. Box 215, Portsmouth, Virginia THE DECI-CAP New Subminiature Ceramic Capacitor—0.100" Diameter by 0.250" Molded Envelope—24 Hour Delivery 5 pf to 470 pf in 19 values, 200 WVDC Epoxy molded for highest reliability and performance—less than 7½% capacitance change from −55°C to +125°C FEATURES: Standardized size for high density cordwood packaging Designed to meet all the requirements of MIL-C-11015 The DECI-CAP is the latest addition to Nytronics' DECI Series—a series that does consist of inductors, capacitors and resistors in a uniform envelope to facilitate point-to-point assembly in cordwood, printed circuit and other high density module assemblies. For complete engineering data, write Dept. WL-55, or phone 201-464-9300. NYTRONICS, INC. 550 Springfield Ave., Berkeley Heights, N.J. Design Leaders STANDARD components to meet CUSTOM requirements CIRCLE 211 ON READER SERVICE CARD New! MINIATURE AXIAL FANS with up to 4 times greater cooling efficiency! FULLY GUARANTEED FOR 10,000 HRS. UNDER NORMAL OPERATING CONDITIONS \| Characteristic \| PAMOTOR Model 1000 \| Conventional Fan \| \|--------------------------------\|--------------------\|------------------\| \| Type of Motor \| induction (capacitor type squirrel cage) \| shaded-pole \| \| Housing \| die cast warp-free zinc \| plastic \| \| Output @ 60 cps \| 125 cfm \| 100 cfm \| \| (0 back pressure) \| 75 cfm \| 20 cfm \| \| (.25" back pressure) \| 50 cfm \| 0 \| \| Output @ 50 cps \| 100 cfm \| 75 cfm \| \| (0 back pressure) \| 62.5 cfm \| 5 cfm \| \| (.25" back pressure) \| \| \| \| Operating Temp. Range \| −55°C to +85°C \| −18°C to +44°C \| Model 1000 meets MIL-T-5422E (Class 2), and MIL-STD-202 Environmental specs. Inside-outside rotating motor design gives flywheel effect, resulting in constant, quiet fan speed. Large surface sleeve bearings mean minimum maintenance, maximum reliability. For complete specifications and name of nearest stocking distributor, write to: PAMOTOR, Inc. 312 Seventh Street • San Francisco 3, Calif. CIRCLE 113 ON READER SERVICE CARD adaptability (of a triple-A work force) Electronics firms locate and expand in WESTern PENNsylvania because of a triple-A work force: Available...Adaptable...and with a good Attitude. For example—one company, as a result of only one help-wanted ad, obtained 500 job applicants; another firm traveled 20,000 miles before it found its needs satisfied in WESTern PENNsylvania. This triple-A work force has the ability to be readily trained in the skills YOU need and at no cost to you through Pennsylvania's unique pre-production training program. This adaptability...plus a favorable tax climate...plus 100% financing of industrial plant space gives WESTern PENNsylvania a top combination of plant location values. Learn more—write, wire or call collect. WEST PENN POWER an operating unit of ALLEGHENY POWER SYSTEM INDUSTRIAL SHELL BUILDINGS available at $2.95 to $3.25 a sq. ft. can be completed to your specs in 60-90 days. WEST PENN POWER Area Development Department (Area Code 412) Greensburg, Pennsylvania Phone: Temple 7-3000 E 3-10-7 Please send me data on: ☐ Labor Availability ☐ Pre-production Training ☐ Favorable Tax Climate ☐ 100% Industrial Plant Financing ☐ Industrial Properties Name ________________________________________ Title _________________________________________ Company ______________________________________ Address _______________________________________ City ___________________________ Zone ________ State ________________ Phone _________________ Transistor Headers of Metalized Ceramic METALIZED ceramic transistor headers are offered to high reliability semiconductor manufacturers. They provide improved physical strength, high and low temperature characteristics and spst switch. Blocking capacitors are provided in the enclosed transmission line to isolate the biasing or modulation circuitry from the microwave network. Temperature range is -55 C to +90 C. Unit cost is $365. Antenna and Radome Research Associates, 27 Bond St., Waterbury, N. Y. CIRCLE 311, READER SERVICE CARD Flow Monitor Has Simplified Design SOLID-STATE, completely-integrated flow monitor is designed for use in oil and gas pipeline applications to provide high-accuracy remote flow totalizing, meter proving and flow rate read-out. Only the controls and displays used for flow monitoring appear on the front panel, thus eliminating the selector switches, adjustment knobs and interconnections of several instruments previously necessary with such equipment. Input power requirements: 115 v a-c ±10 percent; power, 30 w. Input signal requirements: 20 mv rms minimum; 15 v rms max; 2500 cps max. Operating temperature range: 0-50 C. 3 M Co. Instrument Dept., 12909 So. Cerise Ave., Hawthorne, Calif. (312) POWER SUPPLIES AND LIGHT PUMPS FOR LASERS Standard model Power Supplies, Light Pumps and Special Devices are available for prompt delivery. They are designed and built to be readily adaptable to the requirements of almost every type of solid state laser material. New equipments with enlarged capabilities and new features (including the CW solid state laser) are under development. The models listed below merely indicate the variety of units now in production. POWER SUPPLIES All models with continuously variable voltages from 0 to maximum rating. Fast recycling. Console Models 325 — 5,000 volt with output to 20,000 joules. 330 — 10,000 volt with output to 20,000 joules. Modular Units Model 265 — 5,000 volt with output to 20,000 joules. Model 320 — 10,000 volt with output to 20,000 joules. PORTABLE LASER POWER SUPPLY with self-contained energy storage. Model 322—Complete, self-contained Power Supply and Energy Storage. Voltage continuously variable 0 to 3,000. Output to 800 joules. SOLID STATE LASER MOUNTS 6 models available. Choice of air or liquid (nitrogen) cooling for helix or straight arc flash tubes. Maximum powers, 2,000 to 20,000 joules. ENERGY STORAGE BANKS... available in units of 1,000 joules for use with the Power Supplies listed above. SPECIAL DEVICES... Model 3C-4000 Rotary Arc Quench for fast quenching. Model 3C-4039 External Reflector Model 3C-4040 Rotary Disc and Mirror Assembly for "Q" spoiling or "Q" dumping. Write for further information ELECTRO POWERPACS, INC. A subsidiary of Hydra-Power Corp. 16 Hadley Street Cambridge 40, Mass. CIRCLE 212 ON READER SERVICE CARD resistance to thermal shock. Standard packages are presently inventoried for TO-5, TO-18-46 transistors. The ceramic is 94 percent alumina with flanges available in Kovar or OFHC copper. Leads are supplied in either Kovar or copper cored No. 52 alloy material. Metalized pads are finished for soldering in sizes and shapes to meet customer requirements. Metalized Ceramics Corp., 25 Acorn St., Providence 3, R. I. (313) Pulse Transformer Fits in Small Space MINIATURE, ceramic-fitted, high voltage modulation pulse transformer is designed for use in airborne equipment which must meet critical space limitations. The custom designed unit delivers an output pulse of 2,000 v from a 90 v scr source. It operates in altitudes up to 55,000 ft through temperatures from +170 C to -55 C. PCA Electronics, Inc., 16799 Schoenborn St., Sepulveda, Calif. (314) Solid State Pulser Is Overload Protected MODEL 522 can be used to provide fast-rise pulses from input waveforms of any type. It can also act as a frequency divider or a pulse shaper. With no input signal a single pulse can be generated by pushing the button on the front panel. Output pulse width may be varied from less than 0.5 μsec to greater than 0.1 sec. Rise time is less than 50 nsec for a negative pulse and 200 OVER 100 DIFFERENT TYPES NOW IN PRODUCTION JENNINGS CERAMIC VACUUM CAPACITORS ... have been accorded an enthusiastic vote of approval from users for their superior performance in the field. Now we've added many new styles to accommodate the demand for these capacitors in an even wider variety of size, capacitance, voltage and current levels. Ceramic vacuum capacitors combine the inherent advantages of vacuum with a high strength ceramic envelope to form the most advanced high voltage capacitor ever devised. The low loss ceramic allows operation in excess of 400 megacycles. It also provides better vibration characteristics, greater shock resistance, higher current ratings, and smaller size. A few of the many ceramic vacuum capacitors available from Jennings are illustrated below. \| Type CVFA-450 \| Capacity Range: 25-450 PF \| Voltage Rating: 40 kv pk \| RF Current Rating: 100 amps rms \| Length: 9½ inches \| Width: 5½ inches \| \|---------------\|--------------------------\|-------------------------\|-------------------------------\|------------------\|----------------\| \| \| \| \| \| \| \| \| Type CVA-7 \| Capacity Range: 3.5-7 PF \| Voltage Rating: 35 kv pk \| RF Current Rating: 60 amps rms \| Length: 4½ inches \| Width: 3¼ inches \| \|---------------\|--------------------------\|-------------------------\|-------------------------------\|------------------\|----------------\| \| \| \| \| \| \| \| \| Type CFHA-1000\| Capacity: 1000 PF \| Voltage Rating: 50 kv pk \| RF Current Rating: 200 amps rms \| Length: 6½ inches \| Width: 7 inches \| \|---------------\|--------------------------\|-------------------------\|-------------------------------\|------------------\|----------------\| \| \| \| \| \| \| \| \| Type CVHA-650 \| Capacity Range: 30-650 PF \| Voltage Rating: 55 kv pk \| RF Current Rating: 150 amps rms \| Length: 10½ inches \| Width: 7 inches \| \|---------------\|--------------------------\|-------------------------\|-------------------------------\|------------------\|----------------\| \| \| \| \| \| \| \| Our radio frequency laboratory with 12 functioning transmitters ranging from 17 kc to 600 mc and up to 100 kw cw power is at your service to test our products under your particular circuit conditions. Write for more detailed information regarding these capacitors. RELIABILITY MEANS VACUUM / VACUUM MEANS Jennings JENNINGS RADIO MFG. CORP., 970 McLAUGHLIN AVE., SAN JOSE 8, CALIF., PHONE CYPRESS 2-4025 CIRCLE 115 ON READER SERVICE CARD Water Soluble Fluxes Save Labor Costs SERIES of water soluble fluxes that can be easily removed by water rinse have been developed. Company claims that these fluxes are weakly acidic or slightly basic, stable compounds with no toxic agents. Their salts have resistivities on the order of 10' ohms. Nearly all of the fluxes have temperatures of evaporation low enough so that with proper temperature control, residues can be eliminated even without water rinse. Company says basic theories of acids, bases, reductants, and carbon linkages were analyzed and more than a hundred possibilities were tried before their present family of fluxes was created. Connecticut Valley Chemicals, Inc., 178 Prince St., New York 12, N. Y. (316) Transient Detector Aids Test Engineers PORTABLE precision instrument for monitoring h-f transients in electronic/electrical systems is now in production. Model 2601 is designed to aid test engineers in detecting and eliminating transient impulses which create rfi and reliability problems with transistor electronics. Impulses as short as 0.1 μsec in duration can be detected. Adjustable detection levels measure the voltage amplitude of positive or negative transients occurring up to 3,000 per sec. High input impedance (9.0 meghoms at 2.5 pf) prenew low cost precision THERMISTORS match standard curves -40° to 150° C. Resistance Tolerance 30K thermistor No. 44008 - YSI precision thermistors can now be stocked by the thousands, used interchangeably. The high cost problems of matching, padding, auxiliary resistances or individual calibration have been eliminated. - Stock base resistances at 25° C. of: 100 Ω 1K 10K 300 Ω 3K 30K 100K - For 5 years YSI has manufactured precise, interchangeable thermistors for laboratory instrumentation. - Now we offer as components a family of precision thermistors which match the same Resistance-Temperature curves to within ± 1% over a wide temperature range. - $4.90 each, with substantial discounts on quantity orders. - Quantities under 100 available from stock at Newark Electronics Corporation and its branches. For complete specifications and details write: YSI—COMPONENTS DIVISION Yellow Springs, Ohio CIRCLE 215 ON READER SERVICE CARD Band-Pass Filter For 215-260 Mc Use LIGHT-WEIGHT band-pass filter is designed to remove spurious radiation inherent in solid state telemetry transmitters. Operating in the 215-260 Mc telemetry band, model FV-211 contains three resonant circuits which provide a narrow pass band of low insertion loss and steep attenuation skirts to remove undesired signals. Insertion loss is 1.0 db (max), (typically 0.75 db), at factory-tuned frequency. At ±20 Mc of the tuned frequency, loss is 20 db min, (typically 25 db). Vswr is 1.3:1 max (typically 1.2:1), and 3 db bandwidth is about 15 Mc. Dimensions are ½ by 1 by 4 in. Rantec Corp., Calabasas, Calif. (318) Solid State Relay Needs No Maintenance MINIATURIZED neutral-to-neutral solid state relay for signal communications networks has been developed. All maintenance and mechanical adjustment of relay contacts is eliminated. Added savings may be realized since the need for a power supply for bias is unnecessary. Input signals are: current—15, 20 or 60 ma neutral; sampling—mid-point of pulse with appropriate shunt for 60 ma; maximum rating—300 ma forward current surge. Input to output isolation is 50 megohm to ground and output at PULSE POINTERS Twin Pulse Feature Now Standard On General Purpose Pulse Generator Model 3450D Pulse Generator Modules expand capability of basic test instrument—Need a variety of special purpose design/test instrumentation? Investigate modular flexibility of Servo Corporation's SERVOPULSE™ Model 3450D. With twin pulse feature now standard, it can readily be modified to perform the function of six special purpose instruments. Rep rate is 2 mc-2 cps; rise time is 15 nanosec-0.5 μsec, continuously variable. Modular Concept Applied To Test Instrumentation Plug-in module extended for service Standard sub-assemblies reduce cost of special purpose equipment— Plug-in modular construction is the basis for low-cost diversification in Servo Corporation's SERVOPULSE™ instruments line. Thirty-three cataloged instruments and 200 standard pulse and digital circuit modules afford unmatched instrument flexibility. Their advanced pulse techniques and circuitry are proved on major missile programs. For complete details, write: SERVOPULSE™ PRODUCTS SERVO CORPORATION OF AMERICA 111 New South Road Hicksville, L.I., N.Y. WElis 8-9700 500 v. Keying rate is in excess of 1,000 baud. Ortronix, Inc., P.O. Drawer 8217, Orlando, Fla. CIRCLE 319, READER SERVICE CARD Pulse Generator Has Fast Rise Time PULSE GENERATOR features a rise time of 400 picoseconds. Model 126 uses transistorized circuitry and a high speed coaxially mounted mercury switch to produce its fast rise times. The output waveform is very clean with negligible overshoot and ringing. Frequency is variable from 40 to 300 cps. The pulser will be useful in nuclear applications and for checking transient response of wide band systems, fast solid state switching circuitry and computer devices. Price is $370. E-H Research Laboratories, Inc., 163 Adeline St., Oakland 20, Calif. (320) Reed Relays for High Speed Switching THIRTEEN different models of reed relays are suited for use in computers, telemetry, automatic controls and similar sophisticated circuitry. The switching mechanism consists of magnetic reeds hermetically sealed in a glass capsule filled with an inert gas that inhibits oxidation. Life expectancies extend into the millions of cycles at maximum rating and into the hundreds of millions at low load conditions. Several styles are available with a variety of contact arrangements from spst to 4pdt and in octal, plug-in and subminiature models. Dormeyer Industries, 3418 Milwaukee Ave., Chicago 41, Ill. (321) Breadboard Solves Workbench Problem NEW Mark V breadboard is a solution to the electronic engineers' workbench problem. Circuit diagrams can either be drawn or placed on the schematic transferal board and seen from the work surface above. There is no need for soldering components for the spring connectors are very sturdy and gold plated to assure good electrical contact. The Mark V measures 14 by 17 in., weighs 4$\frac{1}{2}$ lb and costs $29.95 net. Phillips Control Co., 59 West Washington, Joliet, Ill. (322) Miniature Capacitors Encapsulated in Epoxy CERAMIC capacitor in model EA-4 case measures 0.125 diameter by 0.250 long. Type 2042 capacitors are encapsulated in machined epoxy cases, and are designed to operate over a temperature range of $-55$ C to $+125$ C at full working voltage (100 v d-c). Capacitance values range from 10 pf to 56,000 pf, with standard capacity tolerance of $\pm 10$ percent. Total capacitance drift over the temperature range is guaranteed less than $\pm 15$ percent. The Potter Co., 7351 North Lawndale, Skokie, Ill. (323) Certifying data with respect to test qualifications available. All data compiled by a government approved testing laboratory. Paktron MOLDED MINIATURE MYLAR CAPACITORS—MR330 and MR700—provide the widest possible capacitance range and most useful voltage ratings in the smallest physical size consistent with HIGH PERFORMANCE AND RELIABILITY. For specific information regarding performance characteristics, write today. Dupont Polyester Film PAKTRON DIVISION ILLINOIS TOOL WORKS INC. 1321 LESLIE AVENUE • ALEXANDRIA, VIRGINIA CIRCLE 119 ON READER SERVICE CARD 119 "WHERE'S THAT TAPE OF 'THE ALLIGATOR GLEE CLUB AT CARNEGIE HALL'?" Oh, hello Rip! You got here just in time. Drop that demo on the bench and listen to this tape. It's part of a new batch that Station 16 just sent in — even worse than the ones I was telling you about. Worse? I'll say! Sounds like a sped-up playback of "Concerto for Seagulls and Fish Pier"! But I thought you said you were getting groans and burps? That's what's rough — the stations NEVER know what kind of interference they'll get next! You told me I don't need two separate filters — how is this one Krohn-Hite black box going to clear up the confusion? Because the 315-A is two filters . . . matter of fact, three, on one chassis. As I get it, your radio-telephone transmissions are being loused up by all kinds of noise and interference — above, below, or right in the middle of the intelligence band, and never in the same place twice. Now start that "Screaming Meemy" tape again, while I plug the 315-A into the monitor output and listen through the filter with these earphones. At the stations, they'd do just about the same on live transmission, except that when they had set the filter to maximize the intelligence, they would just switch it right into the line at any convenient a-f stage. . . . I see what you mean — I can barely make out the voice, with a horrible hash above it and below it too. Now let me switch to band-pass, and move in from the ends with both cut-offs independently. I'll spin through that top decade below 200 kc fast, since for this work you'd never hear the difference. But I just dropped out a thumb somewhere down around 30 cycles — probably someone chopping liver! Here you are . . . listen to this . . . clear as a bell! So far so good. But keep listening. Just about here I think a pig got stuck — skewered real good at about 2 kc. Watch the gain! Owwwww — I just found it! Quick — let me find a real DEEP null for my aching ears! We turn to band-reject, sneak in from the sides with both dials, and . . . I think somebody just told that pig "down boy! 60 db down!!" Listen for yourself. That makes both types we've cleaned up! I think you've just made yourself a sale. But wait a minute — you said the 315-A is THREE filters. What's the third function? High-pass! ALSO tuneable all the way from 20 cps to 200 kc, with the same 24 db per octave attenuation outside the pass band. And if you ever get squawk patterns in the same spots, don't forget — the dials are direct reading and calibrated to 10%. Log 'em and kill 'em fast. Now — how about lunch to celebrate, at a low-decibel restaurant? I'm with you! KROHN-HITE CORPORATION 580 Massachusetts Avenue, Cambridge 39, Mass. Area Code 617 491-3211 Pioneering in Quality Electronic Instruments Literature of the Week MILLIMETER TUBES Amperex Electronic Corp., 230 Duffy Ave., Hicksville, L. I., N. Y., has published a bulletin covering millimeter tubes and components for industry and research. CIRCLE 324, READER SERVICE CARD D-C POWER SUPPLIES Power Devices, Inc., 9760 Cozycroft, Chatsworth, Calif. Double regulated d-c power supplies for bench or system use are described in a two-page catalog sheet. (325) WIRE & CABLE Belden Mfg. Co., 415 S. Kilpatrick Ave., Chicago 44, Ill., announces availability of its 1963 Electronic Wire Catalog. (326) POWER-AGING EQUIPMENT Wyle Manufacturing, 133 Center St., El Segundo, Calif. A 12-page booklet describes various types of power-aging equipment for semiconductors and other small components. (327) INSTRUMENT TRANSFORMERS James Electronics, Inc., 4050 N. Rockwell St., Chicago 18, Ill., offers a catalog covering 68 different instrument transformers for use in analog acquisition, d-c amplifiers, a-c instrumentation and geophysical and medical fields. (328) AIR-MOVING DEVICE Rotron Mfg. Co., Inc., Woodstock, N. Y. Complete technical information on the new Feather Fan is available in a 6-page catalog. (329) VIDEO VOLTMETER Ballantine Laboratories, Boonton, N. J., has published a technical bulletin that describes model 314A sensitive video voltmeter. (330) LOGIC MODULES Raytheon Co., Norwood, Mass. A 60-page logic module handbook provides input and output loading data, performance characteristics, and new low price schedules. (331) R-F INSTRUMENTS Ferris Instrument Co., Boonton, N. J., has published a six-page catalog illustrating and describing a complete line of precision r-f instruments. (332) AUTOMATIC DATA PROCESSING General Electric Co., Computer Department, Phoenix, Ariz. Brochure CPB-257P, entitled "Assembly Line Balancing On the GE-225 Computer—For Sound Manpower Planning", is available. (333) PCM TELEMETRY EQUIPMENT Monitor Systems, Inc., Fort Washington, Pa. Comprehensive series of five brochures, in attractive folder, describes pcm telemetry equipment currently in production. (334) BOBBIN WINDER Gorman Machine Corp., 480 South Main St., Randolph, Mass. Bulletin describes the Spin Winder, a versatile dual head bobbin winder that can be set up easily for any size bobbin up to 3 in. diameter and 2$\frac{1}{4}$ in. long with minimum tooling. (335) ZENER DIODES* National Transistor, a subsidiary of International Telephone and Telegraph Corp., 500 Broadway, Lawrence, Mass., has released B-106, a new Zener diode cross reference chart. (336) ULTRASONIC SOLDER POTS The Redford Corp., Schenectady 3, N. Y., has released two bulletins containing information on two types of ultrasonic solder pot. (337) CROSS-LINKED POLYSTYRENE The Polymer Corp., 2120 Fairmont Ave., Reading, Pa. Bulletin describes the properties and uses of Polypenco cross-linked polystyrene, a rigid, clear, transparent insulating material. (338) DAC CALIBRATOR Telemetrics, Inc., 12927 So. Budlong Ave., Gardena, Calif. Data sheet 9 describes model 6206 calibrator, which powers, displays and calibrates up to 50 digital-to-analog converters. (339) BALANCED ARMATURE RELAYS C. P. Clare & Co., 3101 W. Pratt Blvd., Chicago 45, Ill. Three high reliability balanced armature relays are described in detail in data sheet No. 753. (340) H-F POWER AMPLIFIER Collins Radio Co., Cedar Rapids, Iowa. Six-page folder is devoted to a 10-Kw h-f linear power amplifier. (341) SUPERCONDUCTING WIRE Westinghouse Materials Mfg. Division, Blairsville, Pa., has published a bulletin that tells the current availability and price of niobium-zirconium superconducting wire. (342) D/A CONVERTER Scientific Data Systems, 1649 Seventeenth St., Santa Monica, Calif. Information bulletin describes a low cost digital-to-analog converter. (343) INDUCTIVE POT CORES Ferroxcube Corp. of America, Saugerties, N. Y. A new design and selection chart for inductive pot cores is now available. (344) COMPUTER PRODUCTS Ampex Corp., 934 Charter St., Redwood City, Calif., has released an 8-page catalog on its complete line of computer products. (345) SILVER-ZINC BATTERIES Epic Inc., 150 Nassau St., New York 38, N. Y. Two-page bulletin illustrates and describes miniature silver-zinc chargeable batteries. (346) TRANSFORMERS Central Transformer Co., 900 W. Jackson Blvd., Chicago 7, Ill. Bulletin G100A contains practical data for estimating and specifying transformers. (347) SCOPE CAMERA SYSTEMS Analab Instrument Corp., Cedar Grove, N. J. Catalog AN-TDS-505 describes a line of oscilloscope camera systems and accessories for Polaroid, and the new continuous motion 35-mm fast developing film. (348) Push a button... follow the operation of a Navcor 400 Series system function module one logical step at a time on the neon indicator lights. Each module is a complete system function, grouping many logic elements. Push button and lights permit positive checkout in a fraction of the time required with less sophisticated flip-flops. Write for data on complete MIL Standard line of Transistorized Digital System Function Modules NAVCOR VALLEY FORGE INDUSTRIAL PARK 930 RITTENHOUSE ROAD, NORRISTOWN, PA. • GL 2-6531 CIRCLE 121 ON READER SERVICE CARD 121 Airpax Completes $1 Million Plant AIRPAX ELECTRONICS Incorporated has announced the completion of its new $1 million air-conditioned, fire-proof plant in Cambridge, Md. The facility is devoted entirely to the manufacture of miniature choppers and circuit breakers. Interior design is such as to permit efficient "flow through" of materials and parts to finished products for stock or immediate shipment. The new plant has an area of 94,000 square feet and is designed to permit future expansion. The production area is over 13,000 square feet and features special "clean" assembly lines equipped with modern filtering facilities. Other areas, completely isolated from the production area, house a centralized machine shop, impregnating, plating, paint, shipping, receiving, and maintenance departments. Engineering, model shop, drafting and environmental testing facilities are so integrated as to permit new design, prototypes, and complete testing to be efficiently done in a single area. The new plant replaces the one destroyed by fire in July, 1962. Airpax also has a plant in Fort Lauderdale, Florida, which specializes in the manufacture of telemetry equipment, discriminators and transformers. Thornton Accepts Elgin Position RICHARD P. THORNTON has been appointed general manager of Elgin Laboratories, Waterford, Pa., a subsidiary of Erie Resistor Corp. Prior to joining Elgin, Thornton was with the Armour Research Foundation as a management consultant. Elect Bartholomew Vice President JOSEPH F. BARTHOLOMEW has been elected vice president of Epsco, Inc. and appointed marketing department manager of the newly formed Cambridge division of the Cambridge, Mass., analog and digital electronics manufacturer. Bartholomew has been elevated to this position after several years with Epsco in varied marketing positions. Wabash Magnetics Promotes Executives WABASH MAGNETICS, Inc., Wabash, Ind., has announced three executive promotions. Bernard F. Forrest, vice president of manufacturing, has been appointed vice president and assistant to the president. The newly created position of vice president of operations will be filled by Shethar Davis, presently director of marketing. Robert O. Jefferson, assistant vice president of manufacturing, has been promoted to vice president and general manager of the company's Deluxe Coils and Hi-Voltage divisions. Elect Thomas Nast To Top Post THE BOARD of directors of Robinson Technical Products, Inc., Teterboro, N. J., has announced the election of Thomas Nast as president and chief executive officer of the corporation. He had been executive vice president of the corporation and president of the Kensico Tube division, Mt. Kisco, N. Y. Nast replaces C. S. Robinson, founder, who becomes chairman of DON'T OVERLOOK THE VARIETY OF CIRCUITRY AVAILABLE IN SMALL SWITCHES Small switches are providing simple answers to an increasing variety of design problems. These space savers are available in a wide variety of contact arrangements—often eliminating the need for complicated wiring and extra components—greatly simplifying system design. And, all are quickly available from stock. This broad MICRO SWITCH line of small switches also offers many different actuator and terminal types. For complete application information, contact our Branch Office (see Yellow Pages), or write for Catalog 62. Single-pole single-throw normally open. Single-pole single-throw normally closed. Single-pole double-throw. SPDT reset switches, either "trip-free" (non-recycling) or recycling type. Two SPDT circuits with a single actuator. Double-pole double-throw. Split-contact double throw. Double-break. Make-before-make action closes one circuit before closing the second circuit. Make-before-break action transfers circuit without interruption. Two moving contacts provide single electrical pulse. Drawings indicate action. MICRO SWITCH FREEPORT, ILLINOIS A DIVISION OF HONEYWELL IN CANADA: HONEYWELL CONTROLS LIMITED, TORONTO 17, ONTARIO HONEYWELL INTERNATIONAL—SALES AND SERVICE OFFICES IN ALL PRINCIPAL CITIES OF THE WORLD. MANUFACTURING IN UNITED STATES, UNITED KINGDOM, CANADA, NETHERLANDS, GERMANY, FRANCE, JAPAN. electronics • July 19, 1963 the board. Robinson will continue as president of the Robinson Vibra-shock division. JAMES L. PALMER has been named manager of the electron tube section of The Bendix Corporation's Red Bank division, Eatontown, N.J. He succeeds John H. Wyman who will continue with the division in a staff position. Since 1961, Palmer has been manager of research and development for the Huggins Company, Sunnydale, Calif. SANFORD C. SIGOLOFF has been elected a vice president of Electro-Optical Systems, Inc., Pasadena, Calif. He also continues in his present position as corporate manager of operations, according to A. M. Zarem, president. Prior to joining EOS, Sigoloff was manager of operations for the Santa Barbara division of Edgerton, Germeshausen and Grier, Inc. ELECTION of Stanley Luke as a vice president of International Telephone and Telegraph Corporation has been announced by Harold S. Geneen, president. Luke was also named director of business development for the North American Area of ITT, which is headed by vice president John J. Graham. Formerly an assistant vice president, Luke has been associated with ITT in various executive capacities both in the United States and abroad for the past 18 years. WILLIAM R. PETRICONE has been appointed general manager of the Western division of The Torrington Manufacturing Company in Van Nuys, Calif. He will be in charge of manufacturing and all staff functions in the production of Torrington's fans, blowers and specialty blower units in the western states. Petricone joined Torrington Manufacturing in 1958. FAIRCHILD SEMICONDUCTOR has appointed Charles Askanas to the position of plant manager, InstruHere is MEASURED RELIABILITY! Ten thousand EL-MENCO high reliability dipped mica capacitors were put on life test at 85°C with 225% of the rated DC voltage applied—After 26,500,000 actual test unit-hours no failures of any type occurred. The accumulated 26.5 x 10^6 test unit-hours without any failures can be used to calculate many different failure rates depending upon the confidence level desired. However, we shall explore the meaning of the results at a 90% confidence level. Assuming no acceleration factor for either temperature or voltage, we have verified a failure rate of less than 0.01% per 1000 hours. (Actually, there is a temperature effect and it has been found that, with the DC voltage stress remaining constant, the life decreases approximately 50% for every 10°C rise in temperature. There is also a voltage effect such that, with the temperature stress remaining constant, the life is inversely proportional to the 8th power of the applied DC voltage.) Assuming no temperature acceleration factor and assuming the voltage acceleration exponent is such as to yield an acceleration factor as low as 100, we have nevertheless verified a failure rate of less than 0.0001% per 1000 hours. Assuming no temperature acceleration factor and assuming the voltage acceleration factor is on the order of 250 (test results are available to confirm this) we have accumulated sufficient unit-hours to verify a failure rate of less than 0.00004% per 1000 hours! Note that all the above failure rates are calculated at a 90% confidence level! Only 1 Failure in 14,336,000 Unit-Hours for 0.1 MFD Capacitors Life tests have proved that El-Menco Mylar-Paper Dipped Capacitors — tested at 105°C with rated voltage applied — have yielded a failure rate of only 1 per 1,433,600 unit-hours for 1.0 MFD. Since the number of unit-hours of these capacitors is inversely proportional to the capacitance, 0.1 MFD El-Menco Mylar-Paper Dipped Capacitors will yield ONLY 1 FAILURE IN 14,336,000 UNIT-HOURS. MINIMUM LIFE EXPECTANCY FOR 1.0 MFD MYLAR-PAPER DIPPED CAPACITORS AS A FUNCTION OF VOLTAGE & TEMPERATURE THE NUMBER OF UNIT-HOURS IS INVERSELY PROPORTIONAL TO THE CAPACITY IN MFD UNIT-HOURS FOR ONE FAILURE * Registered Trade Mark of DuPont Co. Write for Reliability Study and technical brochures. THE ELECTRO MOTIVE MFG. CO., INC. WILLIMANTIC, CONNECTICUT Dipped Mica • Molded Mica • Silvered Mica Films • Mica Trimmers & Padders Mylar-Paper Dipped • Paper Dipped • Mylar Dipped • Tubular Paper ARCO ELECTRONICS, INC., Community Drive, Great Neck, L. I., New York Exclusive Supplier to Jobbers and Distributors in the U. S. and Canada West Coast Manufacturers Contact: COLLINS & HYDE CO., 535 Middlefield Road, Palo Alto, California 8380 Whittier Blvd., Los Angeles, Calif. achieve STABLE FREQUENCY with Varo miniature tuning forks and transistorized fork oscillators - Computer Clock Pulses - Rotary and Static Power Supplies - Tape Recorders - Narrow Bandwidth Filters - Facsimile Equipment - Lab Frequency Standards Frequency stability .1% to .01% in a temperature range from -65 C. to 125 C. Frequencies up to 4kc. Missile Environment. For additional information write: VARO INC ELECTROKINETICS DIV. P.O. Box 1500, Santa Barbara, Calif. Area Code 805 966-3116 CIRCLE 219 ON READER SERVICE CARD HERE IS THE WORLD'S SMALLEST MOTOR YET IT'S SO POWERFUL... MITSUMI MICRO MOTOR MODEL MI-100 Less than 20mm in diameter, the new Mitsumi Micromotor provides a startling efficiency of over 50%, the barrier which miniature motors are not allowed to pass. A novel construction principle helped to make this accomplishment possible. The form is more simplified by setting all the terminals at one position. Because the entire mechanism is given full protection against irregular revolution and above all, electrical noise is entirely eliminated, you may call this the most perfect micromotor yet devised. Please write for complete information on Mitsumi Micromotor, and we will send you specifications and data. MITSUMI PARTS MITSUMI ELECTRIC CO., LTD. TOKYO • OSAKA • NEW YORK CIRCLE 126 ON READER SERVICE CARD PEOPLE IN BRIEF Daniel A. Worsham leaves Fairchild Semiconductor Corp. to join Siliconix Inc. as mfg. mgr. Ira Zames, formerly with Polarad Electronics Corp., named administrative coordinator at Pilot Radio Corp. William E. Baird promoted by GE to mgr. of mfg. for the Process Computer Section. John J. McKenna advances to mgr. of service engineering for the Aero-Space div. of Walter Kidde & Co., Inc. Burroughs Corp. ups F. R. Caudle, Jr., to g-m of its new Norwegian subsidiary, Burroughs S/A. Frank L. Weston, previously with Acme Steel Co., appointed director of industrial engineering for Oak Mfg. Co. Joseph F. Fischer moves up to chief engineer of Genistron, Inc. Eugene J. Bourque promoted to works mgr. of the Diode and Rectifier dept. of Motorola's Semiconductor Products div. Henry N. Bowes, ex-General Dynamics Corp., named mgr., training devices project, at Lockheed Electronics Co. General Technology Corp. upgrades three staff members to the positions of v-p: George A. Bronson, Stuart L. Ridgway, and Donald J. Farmer. E. Robert Kebbon, from Curtiss-Wright to Elizabeth div. of Elastic Stop Nut Corp. of America as AGASTAT chief engineer. Billy M. McCormac, formerly with the Defense Atomic Support Agency, named scientific advisor to the Physics div. at IIT Research Institute. Floyd D. Shipley advances to chief engineer at Decibel Products, Inc. ATTENTION: ENGINEERS, SCIENTISTS, PHYSICISTS This Qualification Form is designed to help you advance in the electronics industry. It is unique and compact. Designed with the assistance of professional personnel management, it isolates specific experience in electronics and deals only in essential background information. The advertisers listed here are seeking professional experience. Fill in the Qualification Form below. STRICTLY CONFIDENTIAL Your Qualification form will be handled as "Strictly Confidential" by ELECTRONICS. Our processing system is such that your form will be forwarded within 24 hours to the proper executives in the companies you select. You will be contacted at your home by the interested companies. WHAT TO DO 1. Review the positions in the advertisements. 2. Select those for which you qualify. 3. Notice the key numbers. 4. Circle the corresponding key number below the Qualification Form. 5. Fill out the form completely. Please print clearly. 6. Mail to: Classified Advertising Div., ELECTRONICS, Box 12, New York 36, N. Y. (No charge, of course). \| COMPANY \| SEE PAGE \| KEY # \| \|---------\|----------\|-------\| \| ATOMIC PERSONNEL INC. \| 72* \| 1 \| \| Philadelphia, Pa. \| \| \| \| BUREAU OF SHIPS \| 70* \| 2 \| \| Department of the Navy \| \| \| \| Washington, D. C. \| \| \| \| 5th NAVAL DISTRICT \| 113 \| 3 \| \| Dept. of the Navy \| \| \| \| Portsmouth, Virginia \| \| \| \| HEWLETT-PACKARD CO. \| 128 \| 4 \| \| 1501 Page Mill Rd. \| \| \| \| Palo Alto, Calif. \| \| \| \| INTERNATIONAL BUSINESS MACHINES CORP. \| 72* \| 5 \| \| Space Guidance Center \| \| \| \| Owego, New York \| \| \| \| SOCONY MOBILE OIL CO. INC. \| 72* \| 6 \| \| Dallas 21, Texas \| \| \| \| SPACE & INFORMATION SYSTEMS \| 66* \| 7 \| \| Div. of North American Aviation, Inc. \| \| \| \| Downey, Calif. \| \| \| \| SPACE TECHNOLOGY LABORATORIES, INC. \| 13 \| 8 \| \| Sub. of Thompson Ramo Wooldridge Inc. \| \| \| \| Redondo Beach, California \| \| \| \| UNITED AIRLINES \| 128 \| 9 \| \| San Francisco 28, California \| \| \| * These advertisements appeared in the July 12, issue. (please type or print clearly. Necessary for reproduction.) Personal Background NAME ............................................................... HOME ADDRESS ........................................................ CITY .................. ZONE ........... STATE .............. HOME TELEPHONE .................................................. PROFESSIONAL DEGREE(S) ......................................... MAJOR(S) ..................................................................... UNIVERSITY ................................................................. DATE(S) ...................................................................... FIELDS OF EXPERIENCE (Please Check) ☐ Aerospace ☐ Antennas ☐ ASW ☐ Circuits ☐ Communications ☐ Components ☐ Computers ☐ ECM ☐ Electron Tubes ☐ Engineering Writing ☐ Fire Control ☐ Human Factors ☐ Infrared ☐ Instrumentation ☐ Medicine ☐ Microwave ☐ Navigation ☐ Operations Research ☐ Optics ☐ Packaging ☐ Radar ☐ Radio-TV ☐ Simulators ☐ Solid State ☐ Telemetry ☐ Transformers ☐ Other ................ CATEGORY OF SPECIALIZATION Please indicate number of months experience on proper lines. Technical Experience (Months) Supervisor Experience (Months) RESEARCH (pure, fundamental, basic) ....... ....... RESEARCH (Applied) .............. ....... SYSTEMS (New Concepts) ............ ....... DEVELOPMENT (Model) ............. ....... DESIGN (Product) ............... ....... MANUFACTURING (Product) .......... ....... FIELD (Service) ............... ....... SALES (Proposals & Products) ........ ....... CIRCLE KEY NUMBERS OF ABOVE COMPANIES' POSITIONS THAT INTEREST YOU 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 WHAT DO YOU WANT LACING CORDS and TAPES TO DO FOR YOU? NYLON and DACRON CORDS FLAT BRAIDED TAPES and TEFLON COATED FIBERGLAS TAPES GIVE YOU ALL THESE ADVANTAGES Available in wax-coated, wax-free or "G.E." finish Write for free samples and specification book. THE HEMINWAY & BARTLETT MFG. CO. Electronics Division: 500 Fifth Avenue, New York 38, N.Y. CIRCLE 220 ON READER SERVICE CARD DIRECT VACUUM READINGS to less than 10-13 torr with NRC'S REDHEAD MAGNETRON GAUGE The NRC Model 752 ultra-high vacuum gauge is a cold cathode linear output ionization tube which measures pressure several decades lower than any other commercially available gauge. The Model 752 has several unique advantages - High Sensitivity of 4.5 amperes per millimeter of mercury - 50 times that of hot wire ionization gauges - Cold Cathode Operation which keeps the vapor pressure of gauge elements below operation pressures - eliminates contamination by metallic vapors ... No Outgassing Circuit required which simplifies operation ... Self Regulating Emission which keeps X-ray background current always much lower than the ionization current. Write today for data sheet and prices for the Model 752. A Subsidiary of National Research Corp. 160 Charlemont Street. Newton 61, Massachusetts MANUFACTURING PLANTS IN NEWTON, MASS., AND PALO ALTO, CALIF. CIRCLE 221 ON READER SERVICE CARD CLASSIFIED SEARCHLIGHT SECTION ADVERTISING EMPLOYMENT • BUSINESS • OPPORTUNITIES • EQUIPMENT—USED or RESALE ELECTRONIC MARKETING MANAGER—FRANCE French manager wanted to assume complete responsibility of new Hewlett-Packard sales office in Paris. Applicant must be French National who desires to live permanently in Paris. Technical background in electronics, plus sales engineering or sales management experience essential. Salary & benefits commensurate with French industry. Starting salary between N.F. 50,000 and 60,000. Growth opportunity unlimited. Write in confidence to: Mr. N. O. Williams, Professional Employment Supervisor. Hewlett-Packard Company 1501 Page Mill Road, Palo Alto, California An equal opportunity employer. ENGINEERS Electrical, Aeronautical, Mechanical Engineers wanted in San Francisco to conduct the technical engineering required to provide safety, reliability, and efficiency of all aircraft operations. Aircraft engineers develop basic standards, specifications, and procedures for the installation, operation, maintenance, inspection, and testing of aircraft and allied equipment. They evaluate service performance and develop and design modifications, working closely with component manufacturers. Experience desired but not required in Structures, Hydraulics, Pneumatics, Mechanical Systems, Radio, Electric Power, Servo Controls, Autopilots, Instrumentation, Turbine Power Plants, and Airplane Components. Salary commensurate with experience. Permanent career with excellent company benefits including family travel privileges here and abroad. Write today, enclosing resume and salary history to Mr. R. W. Ladd, Regional Employment Manager, Dept. E, United Air Lines, San Francisco International Airport, San Francisco 28, California. UNITED AIR LINES An Equal Opportunity Employer TO SEARCHLIGHT ADVERTISERS Your Inquiries Mean OPPORTUNITY! 100% for the advertiser, or the publisher, but FOR YOU! When you mention this publication in inquiries to advertisers, you enable them to value the inquiry more highly. This advertisers' satisfaction means we have an easier time serving more of your advertisers' requirements — meaning MORE information, MORE choice of products, MORE value — FOR YOU! July 19, 1963 • electronics INDEX TO ADVERTISERS - Ampex Corporation ............ 101 - Ansco, General Aniline & Film Corp. .................. 65 - Arco Electronics, Inc. ....... 124 - Avnet Electronics Corp. ...... 116, 117 - Bausch & Lomb, Inc. .......... 108 - Boonton Radio Co. ............ 16 - Borg Equipment Division, Amphenol-Borg Electronics Corp. 41 - Bristol Company, The ........ 86 - Bussmann Mfg. Co., Div. of McGraw Edison Co. .. 106, 107 - Carborundum Company, The ... 73 - Chester Cable Corp. .......... 110, 111 - Corning Electronic Components . 21 - Defense Electronics, Inc. ..... 72 - Digital Electronics ............ 116 - Dow Corning Corp. ............ 38, 39 - Du Pont de Nemours & Co., Inc. E. I. .................. 81 - Durant Manufacturing Co. ..... 90 - Dynamics Instrumentation Co. .. 112 - Edgerton, Germeshausen & Grier, Inc. .................. 89 - Electro Mative Mfg. Co., Inc. .. 125 - Electro Powerpacks, Inc. ...... 114 - Engineered Electronics Co. .... 91 - Fairchild Semiconductor Corp. .. 130 - Fifth Naval District .......... 113 - Fluke Mfg. Co., Inc., John. .... 66 - Flying Tiger Line .............. 87 - Fusite Corp., The ............. 75 - General Electric Co. .......... 79 - Receiving Tube Dept. ......... 14, 15 - Silicone Products Dept. ....... 82 - Rectifier Components Dept. .... 34, 35 - General R. F. Fittings, Inc. ... 90 - Gertsch Products, Inc. ......... 74 - Glabe Industries, Inc. ......... 64 - Gudebrod Bros. Silk Co., Inc. .. 40 - Hansen Mfg. Co., Inc. .......... 90 - Hart Manufacturing Co. ........ 110 - Heminway & Bartlett Mfg. Co., The ........................................ 128 - Hewlett-Packard Company ....... 10, 11 - Hitemp Wires Inc. ............. 109 - Honeywell, Denver Division .... 20 - IBM Corp. ..................... 88 - Inland Motor Corp. ............ 100 - Sub. of Kollmorgen ............. 100 - Jennings Radio Manufacturing Corp. .................. 115 - Jerrold Electronics Corp. ...... 4 - Johnson Company, E. F. ........ 103 - Kay Electric Co. ............... 63 - Krohn-Hite Corp. ............... 120 - Kyoritsu Electrical Instruments Works, Ltd. ............ 106 - Leach Corporation .............. 46 - Levin and Son, Inc., Louis. .... 85 - Machlett Laboratories Inc., The ........................................ 23 - Mallory and Co., Inc., P. R. .... 94, 95 - Markem Machine Co. ............ 42 - McLean Engineering Laboratories. 111 - Microswitch, Division of Honeywell Midland Mfg. Co. ............ 83 - Midwec ................................ 106 - Minnesota Mining & Mfg. Co., Mincom Division ............ 105 - Mitsumi Electric Co., Ltd. ..... 126 - Motorola Semiconductor Products Inc. .................. 27 - Motorola, Military Electronics Div. .................. 44, 45 - NRC Equipment Corp. ........... 128 - NAVCOR .......................... 121 - North Atlantic Industries, Inc. 71 - Northern Radio Co., Inc. ....... 112 - Nytronics Inc. .................. 113 - Ohmite Mfg. Co. ................ 32, 33 - Paktron, Div. of Illinois Tool Works, Inc. ............ 119 - Pamotor Inc. .................... 113 - Potter Instrument Co., Inc. .... 19 - Radio Corporation of America Rome Cable Div. of Alcoa. ....... 78 - Sanborn Company ............... 69 - Sankaisha Co., Ltd. ............ 118 - Servo Corporation of America. .. 118 - Shizuki Electric Works Co., Ltd. 117 - Sierra Electronics Div. of Philco. Simpson Electric Company. .... 77 - Southern Electronics Corp. ..... 36 - Space Technology Laboratories, Inc. .................. 13 - Spectra-Physics .................. 31 - Speer Carbon Co. ............... 28, 29 - Sprague Electric Co. .......... 9, 24 - Stackpole Carbon Co. ........... 25 - Standard Telephone & Cables, Ltd. Sylvania Electric Products, Inc. Electron Tube Div. ............ 93 - Tektronix, Inc. .................. 5 - Texas Instruments Incorporated Industrial Products Group. .... 37 - Texas Instruments Incorporated Metals & Controls Division. .... 96 - Trak Microwave Corp. ........... 116 - Transitron Electronic Corp. ..... 17 - United Shee Machinery Corp. .... 70 - United Transformer Corp. ....... 2nd cover - Utica Drop Forge & Herbrand Tool Division, Kelsey-Hayes Co. .... 102 - Varian Associates .............. 43 - Varo Inc. ....................... 126 - Virginia Electric & Power Co. .. 67 - Waugh Engineering Div. ........ 84 - West Penn Power .................. 114 - Yellow Springs Instrument Co., Inc. 117 - Yokogawa Electric Works Inc. ... 118 CLASSIFIED ADVERTISING F. J. Eberle, Business Mgr. (2357) EMPLOYMENT OPPORTUNITIES .... 128 EQUIPMENT (Used or Surplus New) For Sale ................................ 128 CLASSIFIED ADVERTISERS INDEX Hewlett-Packard Co. ............. 128 Radio Research Instrument Co. .... 128 Telephone Engineering Co. ....... 128 United Airlines ................... 128 See advertisement in the July 25, 1962 issue of Electronics Buyers' Guide for complete line of products or services. This index and our Reader Service Numbers are published as a service. Every precaution is taken to make them accurate, but electronics assumes no responsibilities for errors or omissions. ADVERTISING REPRESENTATIVES ATLANTA (9): Michael H. Miller, Robert C. Johnson 1375 Peachtree St. N.E., Trinity 3-0323 (area code 404) BOSTON (16): William S. Hodgkinson McGraw-Hill Building, Copley Square, Congress 2-1160 (area code 617) CHICAGO (11): Harvey W. Wernecke, Robert M. Denmead 645 North Michigan Avenue, Mohawk 4-3800 (area code 312) CLEVELAND (13): Paul T. Fogley 55 Public Square, Superior 1-7000 (area code 216) DALLAS (1): Frank Le Beau The Valley Bldg., 1712 Commerce St., Riverside 7-9721 (area code 214) DENVER (2): John L. Patton Tower Bldg., 1700 Broadway, Alpine 5-2981 (area code 303) HOUSTON (25): Kenneth George Prudential Bldg., Malcombe Blvd., Riverside 8-1280 (area code 713) LOS ANGELES (17): Ashley P. Hartman, John O. Zisch, William C. Gries 1125 W. 6th St., Huntley 2-5450 (area code 213) NEW YORK (36): Donald A. Miller (212) 971 3615 George F. Werner (212) 971 3617 Donald R. Furth (212) 971 3616 500 Fifth Avenue PHILADELPHIA (3): Warren M. Gardner, William J. Boyle 6 Penn Center Plaza, LOust 8-4336 (area code 215) SAN FRANCISCO (11): Richard C. Alcarn 235 California Street, Douglas 2-4600 (area code 415) LONDON W1: Edwin S. Murphy Jr. 34 Dover St. FRANKFURT/Main: Matthae Herfurth 85 Westendstrasse GENEVA: Michael R. Zeynel 2 Place du Port TOKYO: George Olcott, 1, Katahiracho, Shiba, Minato-ku A Fairchild First: 2N2884 HIGH FREQUENCY POWER AMPLIFIER / OSCILLATOR .75 W min $P_o$ @ 500 mc • 1.75 W min amplifier output @ 200 mc • 5.5 db min power gain @ 200 mc The new striped structure of Fairchild's 2N2884 is divided into multiple areas interconnected by thin film metallization. This configuration — made possible by advanced Planar® epitaxial techniques — has two major advantages: it allows rapid dissipation of generated heat, and reduces parasitic lead inductance to a minimum. As a result, the 2N2884 has high power capacity at high frequency. In just one frequency doubling step, starting at 500 mc, it is possible to generate 0.5 watt at one kilomegacycle. Used as an oscillator, the 2N2884 has a minimum output of .75 watt at 500 mc. Typical amplifier output is two watts with a 6 decibel gain at 200 mc. Intended primarily for use in VHF and UHF bands, the 2N2884 has excellent characteristics for microwave applications. Planar: a patented Fairchild process AVAILABLE DIRECTLY FROM DISTRIBUTOR STOCKS FAIRCHILD SEMICONDUCTOR CIRCLE 901 ON READER SERVICE CARD THE PICTURE IS CLEAR! CURB "SNIVETS" HERE WITH RCA NOVAR TUBES New RCA NOVAR Beam Power Tubes Minimize the Picture-Spoiling Effects of TV "Snivets" New FCC Regulations require that television sets, manufactured after April 30, 1964, and shipped in interstate commerce, shall be capable of adequately receiving all channels allocated by the FCC to the television broadcast service, including UHF channels. Since "snivets" are most likely to occur in UHF TV receivers, the set designer must, more than ever, be prepared for this problem. There is a simple solution. The 9-pin construction of RCA NOVAR tubes provides a separate base-pin connection for grid-No. 3. This basing arrangement permits application of positive voltage to grid No. 3 to minimize "snivet"-type interference originating in the horizontal-deflection-amplifier circuit. Three new families of RCA NOVAR beam power tubes are specifically designed to curb "snivets" in both UHF and VHF TV receivers, color and black-and-white: - RCA-6JB6, -12JB6* and -17JB6 NOVAR tubes for black-and-white TV - RCA-22JG6 NOVAR tubes for low-B+ black-and-white TV receivers - RCA-6JE6 NOVAR tubes for color TV These new beam power tubes provide the established NOVAR benefits of: - Large pin circle, long pins, rugged cage structure, integral-based all-glass envelope to assure firm socket seating and very effective heat dissipation. - Wide pin spacing (0.172") to guard against high-voltage breakdown and interelectrode leakage. - Low initial cost. - Plus RCA's highly-efficient "Dark Heater" for cooler heater operation, longer tube life and more stable performance. Specify applicable RCA NOVAR tubes for your horizontal-deflection-amplifier stages. And for added dependability, there are NOVAR types for TV damper service, audio-output stages and low voltage rectifier use. Ask your RCA Field Representative about them, or write: Commercial Engineering, Section G-19-DE-3, RCA Electronic Components and Devices, Harrison, N. J. The Most Trusted Name in Electronics Field Offices - EAST: 32-36 Green Street, Newark 2, N. J., (201) 485-3900 • MIDWEST: Suite 1154, Merchandise Mart Plaza, Chicago 54, Ill., (312) 527-2900 • WEST: 6801 E. Washington Blvd., Los Angeles 22, Calif., (213) RAymond 3-8361 • 1838 El Camino Real, Burlingame, Calif., (415) 697-1620.
The Sequoia 2000 Electronic Repository A major effort in the Sequoia 2000 project was to build a very large database of earth science information. Without providing the means for scientists to efficiently and effectively locate required information and to browse its contents, however, this vast database would rapidly become unmanageable and eventually unusable. The Sequoia 2000 Electronic Repository addresses these problems through indexing and retrieval software that is incorporated into the POSTGRES database management system. The Electronic Repository effort involved the design of probabilistic indexing and retrieval for text documents in POSTGRES, and the development of algorithms for automatic georeferencing of text documents and segmentation of full texts into topically coherent segments for improved retrieval. Various graphical interfaces support these retrieval features. Global change researchers, who study phenomena that include the Greenhouse Effect, ozone depletion, global climate modeling, and ocean dynamics, have found serious problems in attempting to use current information systems to manage and manipulate the diverse information sources crucial to their research. These information sources include remote sensing data and images from satellites and aircraft, databases of measurements (e.g., temperature, wind speed, salinity, and snow depth) from specific geographic locations, complex vector information such as topographic maps, and large amounts of text from a variety of sources. These textual documents range from environmental impact reports on various regions to journal articles and technical reports documenting research results. The Sequoia 2000 project brought together computer and information scientists from the University of California (UC), Digital Equipment Corporation, and the San Diego Supercomputer Center (SDSC), and global change researchers from UC campuses to develop practical solutions to some of these problems. One goal of this collaboration was the development of a large-scale (i.e., multiterabyte) storage system that would be available to the researchers over high-speed network links. In addition to storing massive amounts of data in this system, global change researchers needed to be able to share its contents, to search for specific known items in it, and to retrieve relevant unknown items based on various criteria. This sharing, searching, and retrieving had to be done efficiently and effectively, even when the scale of the database reached the multiterabyte range. The goal of the Electronic Repository portion of the Sequoia 2000 project was to design and evaluate methods to meet these needs for sharing, searching, and retrieving database objects (primarily text documents). The Sequoia 2000 Electronic Repository is the precursor of several ongoing projects at the University of California, Berkeley, that address the development of digital libraries. For repository objects to be effectively shared and retrieved, they must be indexed by content. User interfaces must allow researchers to both search for items based on specific characteristics and browse the repository for desired information. This paper summarizes the research conducted in these areas by the Sequoia 2000 project participants. In particular, the paper describes the Lassen text indexing and retrieval methods developed for the POSTGRES database system, the GIPSY system for automatic indexing of texts using geographic coordinates based on locations mentioned in the text, and the TextTiling method for improving access to full-text documents. Indexing and Retrieval in the Electronic Repository The primary engine for information storage and retrieval in the Sequoia 2000 Electronic Repository is the POSTGRES next-generation database management system (DBMS). POSTGRES is the core of the DBMS-centric Sequoia 2000 system design. All the data used in the project was stored in POSTGRES, including complex multidimensional arrays of data, spatial objects such as raster and vector maps, satellite images, and sets of measurements, as well as all the full-text documents available. The POSTGRES DBMS supports user-defined abstract data types, user-defined functions, a rules system, and many features of object-oriented DBMSs, including inheritance and methods, through functions in both the query language, called POSTQUEL, and conventional programming languages. The POSTQUEL query language provides all the features found in relational query languages like SQL and also supports the nonrelational features of POSTGRES. These features give POSTGRES the ability to support advanced information retrieval methods. We used these features of POSTGRES to develop prototype versions of advanced indexing and retrieval techniques for the Electronic Repository. We chose this approach rather than adopting a separate retrieval system for full-text indexing and retrieval for the following reasons: 1. Text elements are pervasive in the database, ranging in size from short descriptions or comments on other data items to the complete text of large documents, such as environmental impact reports. 2. Text elements are often associated with other data items (e.g., maps, remote sensing measurements, and aerial photographs), and the system must support complex queries involving multiple data types and functions on data. 3. Many text-only systems lack support for concurrent access, crash recovery, data integrity, and security of the database, which are features of the DBMS. 4. Unlike many text retrieval systems, DBMSs permit ad hoc querying of any element of the database, whether or not a predefined index exists for that element. Moreover, there are a number of interesting research issues involved in the integration of methods of text retrieval derived from information retrieval research with the access methods and facilities of a DBMS. Information retrieval has dealt primarily with imprecise queries and results that require human interpretation to determine success or failure based on some specified notion of relevance. Database systems have dealt with precise queries and exact matching of the query specification. Proposals exist to add probabilistic weights to tuples in relations and to extend the relational model and query language to deal with the characteristics of text databases. Our approach to designing this prototype was to use the features of the POSTGRES DBMS to add information retrieval methods to the existing functionality of the DBMS. This section describes the processes used in the prototype version of the Lassen indexing and retrieval system and also discusses some of the ongoing development work directed toward generalizing the inclusion of advanced information retrieval methods in the DBMS. Indexing The Lassen indexing method operates as a daemon invoked whenever a new text item is appended to the database. Several POSTGRES database relations (i.e., classes, in POSTGRES terminology) provide support for the indexing and retrieval processes. Figure 1 shows these classes and their logical linkages. These classes are intended to be treated as system-level classes, which are usually not seen by users. The wn_index class contains the complete WordNet dictionary and thesaurus. It provides the normalizing basis for terms used in indexing text elements of the database. That is, all terms extracted from data elements in the database are converted to the word form used in this class. The POSTQUEL statement defining the class is ```plaintext create wn_index ( termid = int4, /* unique term ID / word = text, / the term or phrase / pos = char, / WordNet part of speech information / sense_cnt = int2, / number of senses of word / ptruse_cnt = int2, / types and locations of / offset_cnt = int2, / related terms in WordNet/ ptruse = int2[], / database are stored in / offset = int4[]) / these arrays / ``` All other references to terms in the indexing process are actually references to the unique term identifiers (termid) assigned to words in this class. The wn_index dictionary contains individual words and common phrases, although in the prototype implementation, only single words are used for indexing purposes. The other parts of the record include WordNet database information such as the part of speech (pos) and an array of pointers to the different senses of the word. The kw_term_doc_rel class provides a linkage between a particular text item in any class or text large object (we will refer to either as documents) and Figure 1 The Lassen POSTGRES Classes for Indexing and Their Linkages a particular term from the wn_index class. The POSTQUEL definition of this class is ```sql create kw_term_doc_rel ( termid = int4, / WordNet termid number / synset = int4, / WordNet sense number / docid = int4, / document ID / termfreq = int4) / term frequency within the document / ``` The raw frequency of occurrence of the term in the document (termfreq) is included in the kw_term_doc_rel tuple. This information is used in the retrieval process for calculating the probability of relevance for each document that contains the term. The kw_doc_index class stores information on individual documents in the database. This information includes a unique document identifier (docid), the location of the document (the class, the attribute, and the tuple in which it is contained), and whether it is a simple attribute or a large object (with effectively unlimited size). The kw_doc_index class also maintains additional statistical information, such as the number of unique terms found in the document. The POSTQUEL definition is as follows: ```sql create kw_doc_index ( docid = int4, / document ID / reloid = oid, / oid of relation containing it / attroid = oid, / attribute definition of attr containing it / attrnum = int2, / attribute number of attr containing it / tupleid = oid, / tuple oid of tuple containing it / sourcetype = int4, / type of object -- attribute or large object / doc_len = int4, / document length in words / doc_ulen = int4) / number of unique words in document / ``` The kw_sources class contains information about the classes and attributes indexed at the class level, as well as statistics such as the number of items indexed from any given class. The following POSTQUEL statement defines this class: ```sql create kw_sources ( relname = char16, / name of indexed relation / reloid = oid, / oid of indexed relation / attrname = char16, / name of indexed attribute / attroid = oid, / object ID of indexed attribute / attrnum = int2, / number of indexed attribute / attrtype = int4, / attribute type -- large object or otherwise / num_indexed = int4, / number of items indexed / last_tid = oid, / oid and time for last / last_time = abstime, / tuple added / tot_terms = int4, / total terms from all items / tot_uterm = int4, / total unique terms from all items / include_pat = text, / simple patterns to / exclude_pat = text) / match for indexable items / ``` The other classes shown in Figure 1 relate to the indexing and retrieval processes. The Lassen prototype uses the POSTGRES rules system to perform such tasks as storing the elements of the bibliographic records in an appropriate normalized form and to trigger the indexing daemon. Defining an attribute in the database as indexable for information retrieval purposes (i.e., by appending a new tuple to the kw_sources definition) creates a rule that appends the class name and attribute name to the The indexing process extracts each unique keyword from the indexed attributes of the database and stores it along with pointers to its source document and its frequency of occurrence in kw_term_doc_rel. This process is shown in Figure 3. The indexing daemon and the rules system maintain other global frequency information. For example, the overall frequency of occurrence of terms in the database and the total number of indexed items are maintained for retrieval processing. The indexing daemon attempts to perform any outstanding indexing tasks before it shuts down. It also updates the kw_doc_index tuple for a given indexable class and attribute with a time stamp for the last item indexed (last_tid and last_time). This permits ongoing incremental indexing without having to reindex older tuples. Retrieval* The prototype version of Lassen provides ranked retrieval of the documents indexed by the indexing daemon using a probabilistic retrieval algorithm. This algorithm estimates the probability of relevance for each document based on statistical information on term usage in a user’s natural language query and in the database. The algorithm used in the prototype is based on the staged logistic regression method. A POSTGRES user-defined function invokes ranked retrieval processing. That is, from a user’s perspective, ranked retrieval is performed by a simple function call (kwsearch) in a POSTQUEL query language statement. Information from the classes created and maintained by the indexing daemon are used to estimate the probability of relevance for each indexed document. (Note that the full power of the POSTQUEL query language can also be used to perform conventional Boolean retrieval using the classes created by the indexing process and to combine the results of ranked retrieval with other search criteria.) Figure 4 shows the process involved in the probabilistic ranked retrieval from the repository database. The actual query to the Lassen ranked retrieval process consists simply of a natural language statement of the searcher’s interests. The query goes through the same processing steps as documents in the indexing process. The individual words of the query are extracted and located in the wn_index dictionary (after removing common words or “stopwords”). The termids for matching words from wn_index are then used to retrieve all the tuples in kw_term_doc_rel that contain the term. For each unique document identifier in this list of tuples, the matching kw_doc_index tuple is retrieved. With the frequency information contained in kw_term_doc_rel and kw_doc_index, the estimated probability of relevance is calculated for each document that contains at least one term in common with the query. The formulae used in the calculation are based on experiments with full-text retrieval.$^8$ The basic equation for the probabilistic model used in Lassen states the following: The probability of the event that a document is relevant $R$, given that there is a set of $N$ “clues” associated with that document, $A_i$ for $i = 1, 2, \ldots, N$, is \[ \log O(R\|A_1, \ldots, A_N) = \log O(R) + \sum_{i=1}^{N} [\log O(R\|A_i) - \log O(R)], \] where for any events $E$ and $E'$, the odds $O(E\|E')$ is $P(E\|E')/P(\bar{E}\|E')$, i.e., a simple transformation of the probabilities. Because there is not enough information to compute the exact probability of relevance for any user and any document, an estimation is derived based on logistic regression of a set of clues (usually terms or words) contained in some sample of queries and the documents previously judged to be relevant to those queries. For a set of $M$ terms that occur in both a query and a given document, the regression equation is of the form \[ \log O(R\|A_1, \ldots, A_M) \approx c_0 + c_1 \cdot f(M) \sum_{i=1}^{M} X_{m,i} + \cdots \\ + c_K \cdot f(M) \sum_{i=1}^{M} X_{m,K} + c_{K+1} M + c_{K+2} M^2, \] where there are $K$ retrieval variables $X_{m,K}$ used to characterize each term or clue, and the $c_j$ coefficients are constant for a given training set of queries and documents. The coefficients used in the prototype were derived from analysis of full-text documents. and queries (with relevance judgments) from the TIPSTER information retrieval test collection. The derivation of this formula is given in “Probabilistic Retrieval Based on Staged Logistic Regression.” The full retrieval equation used for the prototype version of retrieval described in this section is \[ \log O(R \| A_1, \ldots, A_M) \approx -3.51 \\ + \frac{1}{\sqrt{M} + 1} \left[ 374 \sum_{i=1}^{M} X_{m,1} + 0.330 \sum_{i=1}^{M} X_{m,2} \\ - 0.1937 \sum_{i=1}^{M} X_{m,3} \right] + 0.0929 M, \] (3) where $X_{m,1}$ is the quotient of the number of times the $m$th term occurs in the query and the sum of the total number of terms in the query plus 35; $X_{m,2}$ is the logarithm of the quotient arrived at by dividing the number of times the $m$th term occurs in the document by the sum of the total number of terms in the document plus 80; $X_{m,3}$ is the logarithm of the quotient arrived at by dividing the number of times the $m$th term occurs in the database (i.e., in all documents) by the total number of terms in the collection; $M$ is the number of terms held in common by the query and the document. Note that the $M^2$ term called for in Equation 2 was not found to provide any significant difference in the results and was omitted from Equation 3. The constants 35 and 80, which were used in $X_{m,1}$ and $X_{m,2}$, are arbitrary but appear to offer the best results when set to the average size of a query and the average size of a document for the particular database. The sequence of operations performed to calculate the probability of relevance is shown in Figure 5. Note that in the figure, $k1, \ldots, k5$ represent the constants of Equation 3. The probability of relevance is calculated for each document (by converting the logarithmic odds to a probability) and is stored along with a unique query identifier, the document identifier, and some location information in the kw_retrieval class. The query itself Figure 5 The Calculation for the Staged Logistic Regression Probabilistic Ranking Process and its unique identifier are stored in the kw_query class. To see the results of the retrieval operation, the query identifier is used to retrieve the appropriate kw_retrieval tuples, ranked in order according to the estimated probability of relevance. The kw_retrieval and kw_query classes have the following POSTQUEL definitions: ```sql create kw_query ( query_id = int4, /* ID number / query_user = char16, / POSTGRES user name / query_text = text) / the actual query / create kw_retrieval ( query_id = int4, / link to the query / doc_id = int4, / document ID number / rel_oid = oid, / location of doc / attr_oid = oid, attr_num = int2, tuple_id = oid, doc_len = int4, / size of document / doc_match_terms = int4, / number of query terms in the document / doc_prob_rel = float8) / estimated probability of relevance / ``` The algorithm used for ranked retrieval in the Lassen prototype was tested against a number of other systems and algorithms as part of the TREC competition and provided excellent retrieval performance.$^{10}$ We have found that the retrieval coefficients used in the formula derived from analysis of the TIPSTER collection appear to work well for a variety of document types. In principle, the staged logistic regression retrieval coefficients should be adapted to the particular characteristics of the database by collecting relevance judgments from actual users and reapplying the staged logistic regression analysis to derive new coefficients. This activity has not been performed for this prototype implementation. The primary contribution of the Lassen prototype has been as a proof-of-concept for the integration of full-text indexing and ranked retrieval operations in a relational database management system. The prototype implementation that we have described in this section has a number of problems. For example, in the prototype design for indexing and retrieval operations, all the information used is visible in user-accessible classes in the database. Also, the overhead is fairly high, in terms of storage and processing time, for maintaining the indexing and retrieval information in this way. For example, POSTGRES allocates 40 bytes of system information for each tuple in a class, and indexing can take several seconds per document. Currently, we are investigating a class of new access methods to support indexing and retrieval in a more efficient fashion. The class of methods involves declaring some POSTGRES functions that can extract subelements of a given type of attribute (such as words in a text document) and generate indexes for each of the subelements extracted. Other types of data might also benefit from this class of access methods. For example, functions that extract subelements like geometric shapes from images might be used to generate subelement indexes of image collections. Particular index element extraction methods can be of great value in providing access to the sort of information stored in the Sequoia 2000 Electronic Repository. The following section describes one such index extraction method developed for the special needs of Sequoia 2000 data. GIPSY: Automatic Georeferencing of Text* Environmental Impact Reports (EIRs), journal articles, technical reports, and myriad other text items related to global change research that might be included in the Sequoia 2000 database are examples of a class of documents that discuss or refer to particular places or regions. A common retrieval task is to find the items that refer to or concentrate on a specific geographic region. Although it is possible to have a human catalog each item for location, one goal of the Electronic Repository was to make all indexing and retrieval automatic, thus eliminating the requirement for human analysis and classification of documents in the database. Therefore, part of our research involved developing methods to perform automatic georeferencing of text documents, that is, to automatically index and retrieve a document according to the geographic locations discussed or displayed in or otherwise associated with its content. In Lassen and most other full-text information retrieval systems, searches with a geographical component, such as “Find all documents whose contents pertain to location X,” are not supported directly by indexing, query, or display functions. Instead, these searches work only by references to named places, essentially as side effects of keyword indexing. Whereas human indexers are usually able to understand and apply correct references to a document, the costs in time and money of using geographically trained human indexers to read and index the entire contents of a large full-text collection are prohibitive. Even in cases where a document is meticulously indexed manually, geographic index terms consisting of keywords (text strings) have several well-documented problems with ambiguity, synonymy, and name changes over time.$^{11,12}$ Advantages of the GIPSY Model To deal with these problems, we developed a new model for supporting geographically based access to text.$^{13}$ In this model, words and phrases that contain geographic place names or geographic characteristics are extracted from documents and used as input to certain database functions. These functions use spatial reasoning and statistical methods to approximate the geographic position being referenced in the text. The actual index terms assigned to a document are a set of coordinate polygons that describe an area on the earth’s surface in a standard geographical projection system. Using coordinates instead of names for the place or geographic characteristic offers a number of advantages. - Uniqueness. Place names are not unique, e.g., Venice, California, and Venice, Italy, are not apparently different without the qualifying larger region to differentiate them. Using coordinates removes this ambiguity. - Immunity to spatial boundary changes. Political boundaries change over time, leading to confusion about the precise area being referred to. Coordinates do not depend on political boundaries. - Immunity to name changes. Geographic names change over time, making it difficult for a user to retrieve all information that has been written about an area during any extended time period. Coordinates remove this ambiguity. - Immunity to spatial, naming, and spelling variation. Names and terms vary not only over time but also in contemporary usage. Geographic names vary in spelling over time and by language. Areas of interest to the user will often be given place names designated only in the context of a specific document or project. Such variations occur frequently for studies done in oceanic locations. Names associated with these studies are unknown to most users. Coordinates are not subject to these kinds of verbal variations. Indexing texts and other objects (e.g., photographs, videos, and remote sensing data sets) by coordinates also permits the use of a graphical interface to the information in the database, where representations of the objects are plotted on a map. A map-based graphical interface has several advantages over one that uses text terms or one that simply uses numerical access to coordinates. As Furnas suggests, humans use different cognitive structures for graphical information than for verbal information, and spatial queries cannot be fully simulated by verbal queries.\textsuperscript{14} Because many geographical queries are inherently spatial, a graphical model is more intuitive. This is supported by Morris’ observation that users given the choice between menu and graphical interfaces to a geographic database preferred the graphical mode.\textsuperscript{15} A graphical interface, such as a map, also allows for a dense presentation of information.\textsuperscript{16} To address the needs of global change scientists, the Sequoia 2000 project team proposed a new browser paradigm.\textsuperscript{17} This system, called Tioga, displays information topologically according to continuous characteristics that are attributes of the data.\textsuperscript{18} For example, documents may be displayed on a map according to their latitude and longitude. Documents may also be displayed according to the time at which they were generated and the time to which they refer, as well as by more abstract functions such as the reading level of the document and the author’s attitudes as expressed in the document. A prototype of the geographical browsing component was included in the Lassen Geographic Browser, which is shown in Figure 6. This browser allows any georeferenced object in the database to be indicated by an icon on the map. The user employs the mouse to center the map on any location and to zoom in or out for more or less map detail. Icons can be made to appear at any coordinates and for any range of magnification values. When an icon is selected by the user, a menu of the objects georeferenced at the icon coordinates and detail level are displayed for selection. An Algorithm to Georeference Text The advantages of georeferencing are apparent. Not so apparent is how to perform such a task automatically. We developed the following three-part thesaurus-based algorithm to explore this task; the algorithm provides the basis for georeferencing in GIPSY.\textsuperscript{19} 1. Identify geographic place names and phrases. This step attempts to recognize all relevant content-bearing geographic words and phrases. The parser for this step must “understand” how to identify geographic terminology of two types: a. Terms that match objects or attributes in the data set. This step requires a large thesaurus of geographic names and terms, partially hand built and partially automatically generated. b. Lexical constructs that contain spatial information, e.g., “adjacent to the coast,” “south of the delta,” and “between the river and the highway.” To implement this part of the algorithm, a list of the most commonly occurring constructs must be created and integrated into a thesaurus. 2. Locate pertinent data. The output of the parser is passed to a function that retrieves geographic coordinate data pertinent to the extracted terms and phrases. Spatially indexed data used in this step can include, for example, name, size, and location of cities and states; name and location of endangered species; and name, location, and bioregional characteristics of different climatic regions. The system must identify the spatial locations that most closely match the geographic terms extracted by the parser and, when geographic modifiers are used, heuristically modify the area of coverage. For example, the phrase “south of Lake Tahoe” will map to the area south of Lake Tahoe, covering approximately the same volume. This spatial representation is, by necessity, the result of an arbitrary assumption of size, but its purpose is to provide only partial evidence to be used in determining locations as described below. Since geopositional data for land use (e.g., cities, schools, and industrial areas) and habitats (e.g., wetlands, rivers, forests, and indigenous species) is also available, extracted keywords and phrases for these types of data must be recognized. The thesaurus entries for this data should incorporate several other types of information, such as synonymy (e.g., Latin and common names of species) and membership (e.g., wetlands contain cattails, but geopositional data on cattails may not exist, so we must use their mention as weak evidence of a discussion of wetlands and use that data instead). For our implementation of GIPSY, we used two primary data sets to construct the thesaurus. The first was a subset of the United States Geological Survey’s Geographic Names Information System (GNIS). This data set contains latitude/longitude point coordinates associated with over 60,000 geographic place names in California. To facilitate comparison with other data sets, the GNIS latitude/longitude coordinates were converted to the Lambert-Azimuthal projection. Examples of place names with associated points include University of California Davis: $-1867878 -471379$ Redding: $-1863339 -234894$ Data for land use and habitat data was derived in the United States Geological Survey’s Geographic Information Retrieval and Analysis System (GIRAS). Each identified name, phrase, or region description is associated with one or more polygons that may be the place discussed in the text. Weights can be assigned to each of these polygons based on the frequency of use of its associated term or phrase in the text being indexed and in the thesaurus. Many relevant terms do not exactly match place names or the feature and land use types listed above. For example, alfalfa is a crop grown in California and should be associated with the crop data from the GIRAS land use data set. The thesaurus was therefore extended, both manually and by extraction of relationships from the WordNet thesaurus, to include the following types of terms: synonymy \[ \begin{align} = & : = \text{synonym} \\ \sim & : = \text{hyponym (maple is a } \sim \text{ of tree)} \\ @ & : = \text{hypernym (tree is a } @ \text{ of maple)} \end{align} \] kind-of relationships \[ \begin{align} # & : = \text{meronym (finger is a } # \text{ of hand)} \\ % & : = \text{holonym (hand is a } \% \text{ of finger)} \\ & & : = \text{evidonym (pine is a } & \text{ of shortleaf pine)} \end{align} \] 3. Overlay polygons to estimate approximate locations. The objective of this step is to combine the evidence accumulated in the preceding step and infer a set of polygons that provides a reasonable approximation of the geographical locations mentioned in the text. Each geophrase, weight, polygon tuple can be represented as a three-dimensional “extruded” polygon whose base is in the plane of the $x$– and $z$–axes and whose height extends upward on the $y$–axis a distance proportional to its weight (see Figure 7a). As new polygons are added, several cases may arise. a. If the base of a polygon being added does not intersect with the base of any other polygons, it is simply laid on the base map beginning at $y = 0$ (see Figure 7b). b. If the polygon being added is completely contained within a polygon that already exists on the geopositional skyline, it is laid on top of that extruded polygon, i.e., its base plane is positioned higher on the $y$–axis (see Figure 7c). c. If the polygon being added intersects but is not wholly contained by one or more polygons, the polygon being added is split. The intersecting portion is laid on top of the existing polygon and the nonintersecting portion is positioned at a lower level (i.e., at $y = 0$). To minimize fragmentation in this case, polygons are sorted by size prior to being positioned on the skyline (see Figure 7d). In effect, the extruded polygons, when laid together, are “summed” by weight to form a geopositional skyline whose peaks approximate the geographical locations being referenced in the text. The geographic coordinates assigned to the text segment being indexed are determined by choosing a threshold of elevation $z$ in the skyline, taking the $x$-$z$ plane at $z$, and using the polygons at the selected elevation. Raising the elevation of the threshold will tend to increase the accuracy of the retrieval, whereas lowering the elevation tends to include other similar regions. To see the results of this process in the GIPSY prototype, consider the following text from a publication of the California Department of Water Resources: The proposed project is the construction of a new State Water Project (SWP) facility, the Coastal Branch, Phase II, by the Department of Water Resources (DWR) and a local distribution facility, the Mission Hills Extension, by water purveyors of northern Santa Barbara County. This proposed buried pipeline would deliver 25,000 acre-feet per year (AF/YR) of SWP water to San Luis Obispo County Flood Control and Water Conservation District (SLOCECWCD) and 27,723 AF/YR to Santa Barbara County Flood Control and Water Conservation District (SBCFCWCD).... This extension would serve the South Coast and Upper Santa Ynez Valley. DWR and the Santa Barbara Water Purveyors Agency are jointly producing an EIR for the Santa Ynez Extension. The Santa Ynez Extension Draft EIR is scheduled for release in spring 1991.$^{22}$ The resulting surface plot appears in Figure 8. The figure contains a gridded representation of the state of California, which is elevated to distinguish it from the base of the grid. The northern part of the state is on the left-hand side of the image. The towers rising over the state’s shape represent polygons in the skyline generated by GIPSY’s interpretation of the text. The largest towers occur in the area referred to by the text, primarily centered on Santa Barbara County, San Luis Obispo, and the Santa Ynez Valley area. The surface plots generated in this fashion can also be used for browsing and retrieval. For example, the two-dimensional base of a polygon with a thickness above a certain threshold can be assigned as a coordinate index to a document. These two-dimensional polygons might then be displayed as icons on a map browser such as the one shown in Figure 6. Future Work Research remains to be done on several extensions to the existing GIPSY implementation. Because a geographic knowledge base and spatial reasoning are fundamental to the georeferencing process, they have been the focus of initial research efforts. The existing prototype can be complemented by the addition of more sophisticated natural language processing. For example, spatial reasoning and geographic data could be combined with parsing techniques to develop semantic representations of the text. Adjacency indicators, such as “south of” or “between,” should be recognized by a parser. Also, the work on document segmentation described below could be used to explore the locality of reference to geographic entities within full-text documents. GIPSY’s technique may be most effective when applied to a paragraph or section level of a text instead of to the entire document. (a) The “weight” of a polygon, indicated by the vertical arrow, is interpreted as “thickness.” (b) Two adjacent polygons do not affect each other; each is merely assigned its appropriate “thickness.” (c) When one polygon subsumes another, their “thicknesses” in the area of overlap are summed. (d) When two polygons intersect, their “thicknesses” are summed in the area of overlap. Figure 7 Overlaying Polygons to Estimate Approximate Locations TextTiling: Enhancing Retrieval through Automatic Subtopic Identification Full-length documents have only recently become available on-line in large quantities, although technical abstracts, short newswire texts, and legal documents have been accessible for many years.\textsuperscript{23} The large majority of on-line information has been bibliographic (e.g., authors, titles, and abstracts) instead of the full text of the document. For this reason, most information retrieval methods are better suited for accessing abstracts than for accessing longer documents. Part of the repository research was an exploration of new approaches to information retrieval particularly suited to full-length texts, such as those expected in the Sequoia 2000 database. A problem with applying traditional information retrieval methods to full-length text documents is that the structure of full-length documents is quite different from that of abstracts. (In this paper, “full-length document” refers to expository text of any length. Typical examples are a short magazine article and a 50-page technical report. We exclude documents composed of headlines, short advertisements, and any other disjointed texts of whatever length. We also assume that the document does not have detailed orthographically marked structure. Croft, Krovetz, and Turtle describe work that takes advantage of this kind of information.\textsuperscript{24}) One way to view an expository text is as a sequence of subtopics set against a backdrop of one or two main topics. A long text comprises many different subtopics that may be related to one another and to the backdrop in many different ways. The main topics of a text are discussed in its abstract, if one exists, but subtopics are usually not mentioned. Therefore, instead of querying against the entire content of a document, a user should be able to issue a query about a coherent subpart, or subtopic, of a full-length document, and that subtopic should be specificable with respect to the document’s main topic(s). Consider a \textit{Discover} magazine article about the Magellan space probe’s exploration of Venus.\textsuperscript{25} A reader divided this 23-paragraph article into the following segments with the labels shown, where the numbers indicate paragraph numbers: 1–2 Intro to Magellan space probe 3–4 Intro to Venus 5–7 Lack of craters 8–11 Evidence of volcanic action 12–15 River Styx 16–18 Crustal spreading 19–21 Recent volcanism 22–23 Future of Magellan Assume that the topic of volcanic activity is of interest to a user. Crucial to a system’s decision to retrieve this document is the knowledge that a dense discussion of volcanic activity, rather than a passing reference, appears. Since volcanism is not one of the text’s two main topics, the number of references to this term will probably not dominate the statistics of term frequency. On the other hand, document selection should not necessarily be based on the number of references to the target terms. The goal should be to determine whether or not a relevant discussion of a concept or topic appears. A simple approach to distinguishing between a true discussion and a passing reference is to determine the locality of the references. In the computer science operating systems literature, locality refers to the fact that over time, memory access patterns tend to concentrate in localized clusters rather than be distributed evenly throughout memory. Similarly, in full-length texts, the close proximity of members of a set of references to a particular concept is a good indicator of topicality. For example, the term volcanism occurs 5 times in the Magellan article, the first four instances of which occur in four adjacent paragraphs, along with accompanying discussion. In contrast, the term scientists, which is not a valid subtopic, occurs 13 times, distributed somewhat evenly throughout. By its very nature, a subtopic will not be discussed throughout an entire text. Similarly, true subtopics are not indicated by only passing references. The term belly dancer occurs only once, and its related terms are confined to the one sentence it appears in. As its usage is only a passing reference, belly dancing is not a true subtopic of this text. Our solution to the problem of retaining valid subtopical discussions while at the same time avoiding being fooled by passing references is to make use of locality information and to partition documents according to their subtopical structure. This approach’s capacity for improving a standard information retrieval task has been verified by information retrieval experiments using full-text test collections from the TIPSTER database.\textsuperscript{26,27} One way to get an approximation of the subtopic structure is to break the document into paragraphs, or for very long documents, sections. In both cases, this entails using the orthographic marking supplied by the author to determine topic boundaries. Another way to approximate local structure in long documents is to divide the documents into even-sized pieces, without regard for any boundaries. This approach is not practical, however, because we are interested in exploring the performance of motivated segmentation, i.e., segmentation that reflects the text’s true underlying subtopic structure, which often spans paragraph boundaries. Toward this end, we have developed TextTiling, a method for partitioning full-length text documents into coherent multiparagraph units called tiles.\textsuperscript{26,28,29} TextTiling approximates the subtopic structure of a document by using patterns of lexical connectivity to find coherent subdiscussions. The layout of the tiles is meant to reflect the pattern of subtopics contained in an expository text. The approach uses quantitative lexical analyses to determine the extent of the tiles and to classify them with respect to a general knowledge base. The tiles have been found to correspond well to human judgments of the major subtopic boundaries of science magazine articles. The algorithm is a two-step process. First, all pairs of adjacent blocks of text (where blocks are usually three to five sentences long) are compared and assigned a similarity value. Second, the resulting sequence of similarity values, after being graphed and smoothed, is examined for peaks and valleys. High similarity values, which imply that the adjacent blocks cohere well, tend to form peaks, whereas low similarity values, which indicate a potential boundary between tiles, create valleys. Figure 9 shows such a graph for the Discover magazine article mentioned earlier. The vertical lines indicate where human judges thought the topic boundaries should be placed. The graph shows the computed similarity of adjacent blocks of text. Peaks indicate coherency, and valleys indicate potential breaks between tiles. The one adjustable parameter is the size of the block used for comparison. This value, $k$, varies slightly from text to text. As a heuristic, it is assigned the average paragraph length (in sentences), although the block size that best matches the human judgment data is sometimes one sentence greater or smaller. Actual paragraphs are not used because their lengths can be highly irregular, leading to unbalanced comparisons. Similarity is measured by using a variation of the tf.idf (term frequency times inverse document frequency) measurement.\textsuperscript{30} In standard tf.idf, terms that are frequent in an individual document but relatively infrequent throughout the corpus are considered to be good distinguishers of the contents of the individual document. In TextTiling, each block of $k$ sentences is treated as a unit, and the frequency of a term within each block is compared to its frequency in the entire document. (Note that the algorithm uses a large stop list; i.e., closed class words and other very frequent terms are omitted from the calculation.) This approach helps bring out a distinction between local and global extent of terms. A term that is discussed frequently within a localized cluster (thus indicating a cohesive passage) will be weighted more heavily than a term that appears frequently but scattered evenly throughout the entire document, or infrequently within one block. Thus if adjacent blocks share many terms, and those shared terms are weighted heavily, there is strong evidence that the adjacent blocks cohere with one another. ![Graph showing similarity values for sentence gaps](image) Figure 9 Results of TextTiling a 77-sentence Science Article Similarity between blocks is calculated by the following cosine measure: Given two text blocks $b1$ and $b2$, $$\cos(b1, b2) = \frac{\sum_{t=1}^{n} w_{t,b1} w_{t,b2}}{\sqrt{\sum_{t=1}^{n} w_{t,b1}^2 \sum_{t=1}^{n} w_{t,b2}^2}},$$ where $t$ ranges over all the terms in the document, and $w_{t,b1}$ is the tf.idf weight assigned to term $t$ in block $b1$. Thus, if the similarity score between two blocks is high, then not only do the blocks have terms in common, but the common terms are relatively rare with respect to the rest of the document. The evidence in the reverse is not as conclusive. If adjacent blocks have a low similarity measure, this does not necessarily mean that the blocks cohere. In practice, however, this negative evidence is often justified. The graph is then smoothed using a discrete convolution\textsuperscript{31} of the similarity function with the function $h_k(.)$, where $$b_k(i) = \begin{cases} \frac{1}{k^2} (k - \|i\|), & \|i\| \leq k - 1 \\ 0, & \text{otherwise}. \end{cases}$$ The result is smoothed further with a simple median smoothing algorithm to eliminate small local minima.\textsuperscript{32} Tile boundaries are determined by locating the lowermost portions of valleys in the resulting plot. The actual values of the similarity measures are not taken into account; the relative differences are what are of consequence. Retrieval processing should reflect the assumption that full-length text is meaningfully different in structure from abstracts and short articles. We have conducted retrieval experiments that demonstrate that taking text structure into account can produce better results than using full-length documents in the standard way.\textsuperscript{29,28,29} By working within this paradigm, we have developed an approach to vector-space-based retrieval that appears to work better than retrieving against entire documents or against segments or paragraphs alone. The resulting retrieval method matches a query against motivated segments and then sums the scores from the top segments for each document. The highest resulting sums indicate which documents should be retrieved. In our test set, this method produced higher precision and recall than retrieving against entire documents or against segments or paragraphs alone.\textsuperscript{26} Although the vector-space model of retrieval was used for these experiments, probabilistic models such as the one used in Lassen are equally applicable, and the method should provide similar improvement in retrieval performance. We believe that recognizing the structure of full-length text for the purposes of information retrieval is very important and will produce considerable improvement in retrieval effectiveness over most existing similarity-based techniques. Conclusion The Sequoia 2000 Electronic Repository project has provided a test bed for developing and evaluating technologies required for effective and efficient access to the digital libraries of the future. We can expect that as digital libraries proliferate and include vast databases of information linked together by high-bandwidth networks, they must support all current and future media in an easily accessible and content-addressable fashion. The work begun on the Sequoia 2000 Electronic Repository is continuing under UC Berkeley’s digital library project sponsored jointly by the National Science Foundation (NSF), the National Aeronautics and Space Administration (NASA), and the Defense Advanced Research Projects Agency (DARPA). Digital libraries are a fledgling technology with no firm standards, architectures, or even consensus notions of what they are and how they are to work. Our goal in this ongoing research is to develop the means of placing the contents of this developing global virtual library at the fingertips of a worldwide clientele. Achieving this goal will require the application of advanced techniques for information retrieval, information filtering, resource discovery, and the application of new techniques for automatically analyzing and characterizing data sources ranging from texts to videos. Much of the work needed to enable our vision of these new technologies was pioneered in the Sequoia 2000 Electronic Repository project. References 1. J. Dozier, “How Sequoia 2000 Addresses Issues in Data and Information Systems for Global Change,” Sequoia 2000 Technical Report 92/14 (S2K-92-14) (Berkeley, Calif.: University of California, Berkeley, 1992) (ftp://s2k-ftp.cs.berkeley.edu/pub/sequoia/tech-reports/s2k-9 2-14/s2k-92-14.ps). 2. M. Stonebraker, “An Overview of the Sequoia 2000 Project,” Digital Technical Journal, vol. 7, no. 3 (1995, this issue): 39–49. 3. M. Stonebraker and G. Kemnitz, “The POSTGRES Next-generation Database Management System,” Communications of the ACM, vol. 34, no. 10 (1991): 78–92. 4. N. Fuhr, “A Probabilistic Relational Model for the Integration of IR and Databases,” Proceedings of the Sixteenth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’93), Pittsburgh, June 27–July 1, 1993 (New York: Association for Computing Machinery, 1993): 309–317. 5. D. Blair, “An Extended Relational Document Retrieval Model,” Information Processing and Management, vol. 24 (1988): 349–371. 6. R. Larson, “Design and Development of a Network-Based Electronic Library,” Navigating the Networks: Proceedings of the ASIS Midyear Meeting, Portland, Oregon, May 21–25, 1994 (Medford, N.J.: Learned Information, Inc., 1994): 95–114. Also available as Sequoia 2000 Technical Report 94/54, July 1994 (ftp://s2k-ftp.cs.berkeley.edu/pub/sequoia/tech-reports/s2k-94/54/s2k-94-54.ps). 7. G. Miller, R. Beckwith, C. Fellbaum, D. Gross, and K. Miller, “Five Papers on WordNet,” CSL Report 43 (Princeton, N.J.: Princeton University: Cognitive Science Laboratory, 1990). 8. W. Cooper, F. Gey, and D. Dabney, “Probabilistic Retrieval Based on Staged Logistic Regression,” Proceedings of the Fifteenth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’92), Copenhagen, Denmark, June 21–24, 1992 (New York: Association for Computing Machinery, 1992): 198–210. 9. D. Harman, “The DARPA TIPSTER Project,” SIGIR Forum, vol. 26, no. 2 (1992): 26–28. 10. W. Cooper, A. Chen, and F. Gey, “Experiments in the Probabilistic Retrieval of Full Text Documents,” Text Retrieval Conference (TREC-3) Draft Conference Papers (Gaithersburg, Md.: National Institute of Standards and Technology, 1994). 11. A. Griffiths, “SAGIS: A Proposal for a Sardinian Geographical Information System and an Assessment of Alternative Implementation Strategies,” Journal of Information Science, vol. 15 (1989): 261–267. 12. D. Holmes, “Computers and Geographic Information Access,” Meridian, vol. 4 (1990): 37–49. 13. A. Woodruff and C. Plaunt, “GIPSY: Georeferenced Information Processing SYstem,” Journal of the American Society for Information Science, vol. 45, no. 9 (1994): 645–655. 14. G. Furnas, “New Graphic Reasoning Models for Understanding Graphical Interfaces,” Human Factors in Computing Systems: Reaching Through Technology Proceedings (CHI ’91 Conference), New Orleans, April–May 1991 (New York: Association for Computing Machinery, 1991): 71–78. 15. B. Morris, “CARTO-NET: Graphic Retrieval and Management in an Automated Map Library,” Special Libraries Association, Geography and Map Division Bulletin, vol. 152 (1988): 19–35. 16. C. McCann, M. Taylor, and M. Tuori, “ISIS: The Interactive Spatial Information System,” International Journal of Man-Machine Studies, vol. 28 (1988): 101–138. 17. J. Chen, R. Larson, and M. Stonebraker, “Sequoia 2000 Object Browser,” Digest of Papers, Thirty-seventh IEEE Computer Society International Conference (COMPCON Spring 1992), San Francisco, February 24–28, 1992 (Los Alamitos, Calif.: Computer Society Press, February 1992): 389–394. 18. M. Stonebraker, J. Chen, N. Nathan, C. Paxson, and J. Wu, “Tioga: Providing Data Management Support for Scientific Visualization Applications,” Proceedings of the Nineteenth International Conference on Very Large Data Bases, Dublin, Ireland (August 1993): 25–38. 19. A. Woodruff and C. Plaunt, “Automated Geographic Indexing of Text Documents,” Sequoia 2000 Technical Report 94/41 (S2K-94-41) (Berkeley, Calif.: University of California, Berkeley, 1994) (ftp://s2k-ftp.cs.berkeley.edu/pub/sequoia/tech-reports/s2k-94-41/s2k-94-41.ps). 20. Geographic Names Information System/United States Department of the Interior, United States Geological Survey, rev. ed., Data User’s Guide, vol. 6 (Reston, Va.: United States Geological Survey, 1987). 21. J. Anderson, E. Hardy, J. Roach, and R. Witmer, “A Land Use and Land Cover Classification System for Use with Remote Sensor Data,” United States Geological Survey Professional Paper #964 (Washington, D.C.: United States Government Printing Office, 1976). 22. State Water Project, Coastal Branch, Phase II, and Mission Hills Extension (Sacramento, Calif.: California Department of Water Resources, 1991). 23. C. Tenopir and J. Ro, Full Text Databases (New York: Greenwood Press, 1990). 24. W. Croft, R. Krovetz, and H. Turtle, “Interactive Retrieval of Complex Documents,” Information Processing and Management, vol. 26, no. 5 (1990): 593–616. 25. A. Chaikin, “Magellan Pierces the Venusian Veil,” Discover, vol. 13, no. 1 (January 1992). 26. M. Hearst and C. Plaunt, “Subtopic Structuring for Full-Length Document Access,” Proceedings of the Sixteenth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’93), Pittsburgh, June 1993 (New York: Association for Computing Machinery, 1993): 59–68. 27. M. Hearst, “Context and Structure in Automated Full-Text Information Access,” Ph.D. dissertation, Report No. UCB/CSD-94/836 (Berkeley, Calif.: University of California, Berkeley, Computer Science Division, 1994). 28. M. Hearst, “TextTiling: A Quantitative Approach to Discourse Segmentation,” Sequoia 2000 Technical Report 93/24 (S2K-93-24) (Berkeley, Calif.: University of California, Berkeley, 1993) (ftp://s2k-ftp.cs.berkeley.edu/pub/sequoia/tech-reports/s2k-93-24/s2k-93-24.ps). 29. M. Hearst, “Multi-Paragraph Segmentation of Expository Text,” Proceedings of the Thirty-second Meeting of the Association for Computational Linguistics, Los Cruces, New Mexico, June 1994. 30. G. Salton, Automatic Text Processing: The Transformation, Analysis, and Retrieval of Information by Computer (Reading, Mass.: Addison-Wesley, 1989). 31. The authors are grateful to Michael Braverman for proving that the smoothing algorithm is equivalent to this convolution. 32. L. Rabiner and R. Schafer, Digital Processing of Speech Signals (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1978). Biographies Ray R. Larson Ray Larson is an Associate Professor at the University of California, Berkeley, in the School of Information Management and Systems (formerly the School of Library and Information Studies). He teaches courses and conducts research on the design and evaluation of information retrieval systems. Ray received his Ph.D. from the University of California. He is a member of the American Society for Information Science (ASIS), the Association for Computing Machinery (ACM), the IEEE Computer Society, the American Association for the Advancement of Science, and the American Library Association. He is the Associate Editor for ACM Transactions on Information Systems and received the ASIS Journal Best Paper Award in 1993. Allison G. Woodruff Allison Woodruff is a Ph.D. student in the Electrical Engineering and Computer Science Department at the University of California, Berkeley. Her research interests include spatial information systems, multimedia databases, visual programming languages, and user interfaces. Previously, she worked as a geographic information systems specialist for the California Department of Water Resources. Allison holds a B.A. in English from California State University, Chico, and an M.A. in linguistics and an M.S. in computer science from the University of California, Davis. Marti A. Hearst Currently a member of the research staff at Xerox Palo Alto Research Center, Marti Hearst completed her Ph.D. in computer science at the University of California, Berkeley, in April 1994. Her dissertation examined context and structure of full-text documents for information access. Her current research interests include intelligent information access, corpus-based computational linguistics, user interfaces, and psycholinguistics. Christian Plaunt Christian Plaunt is a doctoral student and graduate research assistant at the University of California, Berkeley, School of Information Management and Systems. His interests include experimental information retrieval system modeling, simulation, design, and evaluation; artificial intelligence techniques for information retrieval; multistage retrieval techniques; information filtering; and music. Chris holds master’s degrees in library and information studies and in music (composition). In his spare time, he composes, plays the piano, and works in the Music Library at California State University, Fresno, near which he lives with his wife and their three Siamese cats.
A Resolution amending the Rules of Council for the Harrisburg City Council. (Additions indicated by underscoring; deletions by [bracketing].) RULES OF COUNCIL GENERAL RULES: ELECTION OF OFFICERS RULE NO. 1. At its organizational meeting required by law on the first Monday of January following the regular municipal election, the members of City Council shall assemble in Council Chambers at 12:00 p.m. (noon) and organize and elect a President and Vice President of the Council, by majority vote, from among its members to serve a two (2) year term. If the first Monday is a legal holiday, the meeting shall be held on the first day following. In the event the President of City Council is unable to complete his/her term, for whatever reason, City Council shall, within thirty (30) days of notification, duly elect a new President to serve the remainder of the term. Pending said election, the Vice President shall preside. In the event that the Vice President of City Council is unable to complete his/her term, for whatever reason, City Council shall, within thirty (30) days of notification, duly elect a new Vice President to serve the remainder of the term. MEETINGS OF COUNCIL RULE NO. 2. A. VOTING SESSIONS. Council shall meet in Legislative Session at 6:00 p.m. on the second and fourth Tuesday of each month in Council Chambers, preceded by a non-voting review session at 5:30 p.m., but Council may adjourn from time to time. When such days fall on legal holidays, the regularly stated meeting for such day shall be held at the stated time and place on the next day. Special meetings may be called by the President or upon request of four members for special business, and no other business shall be considered. Each member shall have at least twenty-four (24) hours' notice of special meetings, or they may waive the same. If a member is unable to attend a Legislative Session or Special Legislative Session, said member shall notify the City Clerk and the President of Council of his/her absence as soon as practicable. B. WORK SESSIONS. Council shall meet in a Work Session at 5:30 p.m. on the first and third Tuesday of each month in Council Chambers or at such other locations designated by the Council President, which meetings Council may adjourn from time to time. In Work Sessions, Council shall meet as a COMMITTEE OF THE WHOLE, with the Council President serving ex-officio to all committees and presiding. The Work Sessions should be informal in nature and flexible. An Agenda shall be required to identify the matters anticipated to be reviewed by Council and shall include an opportunity for public comment. No formal action shall be taken. CALLING MEETINGS TO ORDER RULE NO. 3. The President shall call the meeting to order and shall preside. In the absence of the President, the Vice President shall serve as the Presiding Officer. If both the President and Vice President are absent, the meeting shall not be called to order unless the President has designated a member of Council, in writing, to serve as Presiding Officer of that session. No Legislative or Special Legislative Session of Harrisburg City Council shall be called to order with or without a quorum in the absence of the City Solicitor or a staff attorney from the Law Bureau. CONFLICT OF INTEREST RULE NO. 4. City Council members must avoid conflicts of interest between the member and the business of the city of Harrisburg. Any situation that involves, or may be reasonably expected to involve a conflict of interest with his/her duties and decisions as a City Council member should be promptly disclosed to the President and all other Council members. A conflict of interest can occur when a member's private interest interferes or influences in any way, or even appears to interfere with or influence, the interests/decisions of the Council body. The President shall consult with the City Solicitor as to what is the appropriate action to be taken to deal with the conflict of interest and the member shall abide with the recommendation of the City Solicitor. PROCEDURE RULE NO. 5. The Standard Code of Parliamentary Procedure by Alice Sturgis (most current edition) shall govern the proceedings of the Council on all matters not specifically provided for herein. The Solicitor shall serve as the Parliamentarian. AMENDMENTS RULE NO. 6. Proposed amendments to these Rules must be presented in and shall be adopted only by the affirmative vote of two-thirds (2/3) of all members of Council. SUSPENSION OF RULES RULE NO. 7. These rules may be suspended by the affirmative vote of the majority of all members of Council present, but such suspension shall not extend beyond adjournment. No Rule of Council which is a subject of legislative enactment shall be suspended. LEGISLATIVE SESSIONS RULE NO. 8. The order of business for legislative meetings shall be as follows: - Call to Order - Roll Call - Invocation/Moment of Silence - Pledge of Allegiance - Courtesy of the Floor - Approval of Minutes - Communications - Reports of Committees - Ordinances for First Reading - Ordinances for Amendment - Ordinances for Final Passage - Resolutions - Old Business - New Business - Adjournment COURTESY OF THE FLOOR RULE NO. 9. Courtesy of the Floor is the order of business during which residents or taxpayers of the city of Harrisburg may address Council on any matter of concern, official action or deliberation which is or may be before City Council, prior to taking action. A sign-in sheet shall be prominently displayed, and all persons wishing to address Council must print their name, address and telephone number or e-mail address. The President shall set guidelines for such courtesy and has the option of establishing time limits consistent with the number of people who will address the members. The President shall refer all questions/issues to the appropriate committee for further review, if necessary. The President has sole discretion as to which committee any matter is assigned. AGENDA RULE NO. 10. No Resolution or Bill shall be considered by Council unless the same is delivered to the City Clerk by noon of the Friday preceding the Council meeting. The City Clerk shall furnish all members of Council and the Mayor with copies of the proposed agenda after it is approved by the Council President. All legislation must be reviewed by the City Solicitor, or his/her designee, and such review will be indicated by the signature or initials of the same. DUTIES OF PRESIDING OFFICER RULE NO. 11. A. President 1.) On the first Monday of January following the regular municipal election, the members of Council shall assemble in Council Chambers at 12:00 noon and organize and elect a president and vice president of Council from among its members. If the first Monday is a holiday, the meeting shall be held at the same time and place on the day thereafter. The Parliamentarian shall preside at the meeting until the President is elected. 2.) The President of Council shall have general supervision of all Legislative Sessions and Special Legislative Sessions. The President shall be responsible for overseeing the general administrative and fiduciary functions of the office of the City Clerk. The City Clerk shall be responsible for all administrative, fiduciary, and day-to-day operations of the office of the City Clerk, including employee supervision, work assignment, leave requests, and evaluations. The President shall be responsible for approving the leave requests of the City Clerk, reviewing and approving the office Budget prepared by the City Clerk, signing contracts, bonds, and the annual financial statement provided to the Commonwealth of Pennsylvania. B. Vice President 1.) The Vice President shall preside over all Legislative and Special Legislative Sessions in the absence of the President. The Vice President shall represent Harrisburg City Council at functions and perform other requirements normally done by the President, in the President’s absence. ORDINANCES AND RESOLUTIONS RULE NO. 12. A. Every legislative act of City Council shall be by Ordinance and every act of policy shall be by Resolution, both having passed Council, signed by the President of Council, and attested to by the City Clerk. B. All Bills to be given first reading and Resolutions to be considered shall be introduced in written form as Bills or Resolutions and shall be numbered serially for the calendar year. No Ordinances shall be passed containing more than one subject, except for general appropriation Ordinances. C. The title of every Bill shall be read when introduced, if amended, and on final passage. A complete copy of every Bill introduced shall be available for public inspection at the City Clerk’s Office during regular office hours. Upon final passage, Ordinances shall be numbered serially for the calendar year. D. A majority of the whole number of members of City Council shall constitute a quorum. No Resolution or Ordinance shall be adopted by City Council without the affirmative vote of a majority of all members of Council. E. The vote upon every Motion, Resolution, or Bill shall be taken by roll call, and the yeas and nays shall be entered in the Minutes. The Presiding Officer shall vote last on all questions and may take part in debate without leaving the Chair. F. All Ordinances shall be adopted and published in the manner required by general law. G. The City Clerk shall record all Ordinances and Resolutions adopted by Council and at the close of each year, with the advice and assistance of the City Solicitor, as required by law, shall bind, compile, or codify all of the Ordinances and Resolutions, or true copies thereof, of the City which then remain in force and effect. The City Clerk shall also properly index the record books, compilation or codification of Ordinances and Resolutions. H. No rule or regulation made by any department, officer, agency, or authority of the City, except such as relates to the organization or internal management of the City government or a part thereof, shall take effect until it is filed either with the City Clerk or in such other manner as may be provided by Ordinance. The Council shall provide for the prompt publication of such rules and regulations. I. Ordinances adopted by the Council shall be submitted by the City Clerk to the Mayor within seventy-two (72) hours after passage, or as otherwise required by law. RECONSIDERATION RULE NO. 13. A motion to reconsider shall be entertained only on Motion of a member who voted in the affirmative on the matter and may only be brought at the next legislative meeting (regular or special) following the one at which the vote was taken. RECEIVING REPORTS RULE NO. 14. The Presiding Officer at a meeting of City Council may announce that the Minutes, a report, or communication is received and accepted and that this action shall be so noted in the Minutes. This shall not exclude the right of any member of City Council to move that a report or communication be referred to a committee or other appropriate body. COMMITTEES RULE NO. 15. The President of City Council shall appoint the following standing committees with duties that include, but are not limited to, the matters as indicated. When a matter can [me] be placed in more than one committee, the President shall solely determine the appropriate committee to which that matter shall be assigned, on a non-precedent setting basis. A. ADMINISTRATION: All matters relating to nominations, appointments, personnel policy, personnel administration, Department of Administration, and related City agencies. This Committee shall review all legislation being considered to support matters at federal, state and regional levels. B. BUDGET AND FINANCE: All matters relating to contracts, budget, finance, taxes, indebtedness, bond issues, penalties, the operating and capital budgets, water and sewer rates and fees. C. BUILDING AND HOUSING: All matters related to the activities of the Department of Building and Housing Development, or its successor, related municipal authorities, communications & technology, the rehabilitation of the housing stock within the city, and matters concerning the Harrisburg Architectural Review Board (HARB), Planning Commission and Redevelopment Authority. D. COMMUNITY AND ECONOMIC DEVELOPMENT: Specific Matters related to subdivision and land development plan approvals, the Federal Community Development Block Grant Program and the administration of the city’s entitlement of CDBG funds. E. PARKS, [AND] RECREATION & ENRICHMENT: All matters falling within the jurisdiction of the Department of Parks and Recreation, including but not limited to: Art Education, Special Events and Festivals. The Youth Commission shall report to the Chair of Parks, Recreation & Enrichment. F. PUBLIC SAFETY: All matters falling within the jurisdiction of the Department of Public Safety; the Bureaus of Fire and Police and City Codes Department. G. PUBLIC WORKS: All matters falling within the jurisdiction of the Department of Public Works and its related authorities; transportation issues and projects, sustainability and livability objectives. The Environmental Advisory Council shall report to the Chair of Public Works. Council may establish special committees, as needed. The President of City Council shall serve as an ex-officio voting member of all committees. No committee meeting will be held prior to 5:00 p.m. and all committee meetings will be held from Monday through Thursday only, unless a majority number of Harrisburg City Council agree to change the date and/or time. ESTABLISHMENT OF FEES RULE NO. 16. A. Property owners and/or interested parties who make a request of City Council for street vacation, opening, widening, easements, rights-of-way, special exceptions, appeals, or any other waiver or action which requires public advertisement, stenographic services or other costs must pay a fee equal to the costs incurred by the Office of the City Clerk prior to the final passage and receipt of a certified copy of City Council’s actions. B. Applicants for such actions as noted in (A.) above shall be notified of the fees at the time the action is requested of City Council. C. Upon written request of the applicant, fees may be waived by City Council upon a two-thirds (2/3) majority vote. COMMITTEE MATTERS RULE NO. 17. No items shall be removed from a committee and placed on City Council’s agenda without the approval of the Committee Chair, unless it is approved by a two-thirds (2/3) majority vote of City Council. PROFESSIONAL SERVICES AGREEMENTS RULE NO. 18. A. Each step in the process, from interviewing to setting final agreement terms, conditions, and restrictions, shall only be authorized by passage of a resolution of Harrisburg City Council, approved by a minimum of four votes. B. No individuals or professional firms shall be contacted or approached by any member of City Council or staff, in order to discuss the terms, conditions, general or specific nature of representation, reviews related to financial requirements, at any time, unless authorized by passage of a Resolution of Harrisburg City Council. C. The President of Harrisburg City Council shall be the designated representative of the body, who shall communicate the intentions of City Council with respect to professional service contracts for “as needed” services, which can only be authorized upon a vote called by the President with a minimum of four votes for approval. D. Termination of any professional services contract for “as needed” services can only be authorized by passage of a Resolution by four members of City Council. E. After all individuals/firms have been contacted in writing and all requested materials have been furnished, City Council shall meet in the Administration Committee and follow all rules pertaining to the further disposition of business. ELECTRONIC RECORDINGS OF CITY COUNCIL MEETINGS RULE NO. 19. No video taped session of City Council shall be edited by a videographer for any reason without or unless a majority of City Council members agree that the deletion is warranted by law and that the content will not otherwise be altered by said edit. Before any said edit is made, the videographer must submit a written explanation for this edit to all members of City Council and it must be approved by the majority of City Council members. PLEDGE OF ALLEGIANCE RULE NO. 20. All Legislative Sessions of Harrisburg City Council shall include the Pledge of Allegiance to the flag of the United States of America as the first order of business after the Invocation/Moment of Silence. Seconded by: Passed the City Council April 24, 2018 Wanda R.D. Williams President of City Council Kathryn City Clerk \| YEAS \| NAYS \| \|------\|------\| \| MR. ALLATT \| \| \| MS. DANIELS \| \| \| MS. GREEN \| \| \| MR. JOHNSON \| \| \| MR. MADSEN \| \| \| MR. MAJORS \| \| \| MS. WILLIAMS \| \| Yea: 7 Nay: 0
Chains of Length 2 in Fillings of Layer Polyominoes Mitch Phillipson Catherine H. Yan Jean Yeh Department of Mathematics Texas A&M University College Station, Texas, U.S.A. {phillipson,cyan,email@example.com Submitted: Apr 11, 2013; Accepted: Sep 16, 2013; Published: Sep 26, 2013 Mathematics Subject Classification: 05A15, 05B50, 05E10 Abstract The symmetry of the joint distribution of the numbers of crossings and nestings of length 2 has been observed in many combinatorial structures, including permutations, matchings, set partitions, linked partitions, and certain families of graphs. These results have been unified in the larger context of enumeration of northeast and southeast chains of length 2 in 01-fillings of moon polyominoes. In this paper we extend this symmetry to fillings of a more general family—layer polyominoes, which are intersection-free and row-convex, but not necessarily column-convex. Our main result is that the joint distribution of the numbers of northeast and southeast chains of length 2 over 01-fillings is symmetric and invariant under an arbitrary permutation of rows. Keywords: chains of length 2, polyomino, symmetric distributions 1 Introduction Recently there have been many interesting results on the combinatorics of crossings and nestings over discrete structures. In particular, it is observed that the numbers of crossings and nestings of length 2 often have a symmetric distribution. The first examples are the classical permutation statistics that count the numbers of inversions and coinversions. Another example comes from complete matchings, which are partitions of the set $\{1, 2, \ldots, 2n\}$ into $n$ blocks of size 2. In a matching, two blocks $(i_1, j_1)$ and $(i_2, j_2)$ form a crossing if $i_1 < i_2 < j_1 < j_2$; they form a nesting if $i_1 < i_2 < j_2 < j_1$. The symmetry of the joint distribution of crossings and nestings over complete matchings follows from a bijection of de Sainte-Catherine [8], who also found the generating function for the number of crossings (or nestings). This result was extended to set partitions by Kasraoui and Zeng [12], to linked set partitions by Chen, Wu and Yan [3], and to a certain diagram associated with permutations by Corteel [5]. Klazar [13], and Poznanovikj and Yan [15] studied further the distributions of crossings and nestings over the generating trees of matchings and set partitions, respectively. Kasraoui unified the symmetry of crossings and nestings in [11] by using the model of fillings of moon polyominoes, where a moon polyomino is a collection of cells in $\mathbb{Z}^2$ that are convex and intersection-free, (see Section 2 for exact definitions). Crossings and nestings become northeast and southeast chains of length 2 in fillings of polyominoes. Kasraoui proved that the statistic $(ne, se)$, the numbers of northeast and southeast chains of length 2, has a symmetric joint distribution over the set of 01-fillings of moon polyominoes where either every row has at most one 1, or every column has at most one 1. In both cases, the joint distribution of $(ne, se)$ can be expressed as a product of $(p, q)$-Gaussian coefficients. This symmetry was further strengthened in [2] which showed that the northeast and southeast chains can be mixed according to the positions of the 1-cells in a chain, yet the new statistics still have the same joint distribution as $(ne, se)$. From the bi-variate generating function of $(ne, se)$ one notes that these statistics (and their mixed variants) are invariant under permutations of rows and columns of the underlying polyomino, provided that the resulting polyomino is still convex and intersection-free. Such an invariant property also appears in other combinatorial statistics of fillings of moon polyominoes, for example, in the sizes of longest northeast chains and southeast chains [9, 10, 16], and in the major index for 01-fillings of moon polyominoes [4]. The motivation of the present paper is to understand this invariant property for the pair $(ne, se)$. To keep the convexity, rows (or columns) of a moon polyomino can not be permuted arbitrarily. It is natural to ask whether there is a larger family of polyominoes that allows a free permutation of rows while preserving the distribution of $(ne, se)$ over 01-fillings. As observed by Kasraoui [11], the condition of intersection-free is necessary. In this paper we propose the notion of layer polyominoes, which are polyominoes that are intersection-free and row-convex, but not necessarily column-convex. The family of layer polyominoes is closed under permutations of rows. In fillings of layer polyominoes the northeast and southeast chains are still well-defined. Our main result is that over all 01-fillings of a layer polyomino where every row has at most one 1, or every column has at most one 1, the distribution of $(ne, se)$ depends only on the sets of the rows, but not how the rows are arranged. The paper is organized as follows. Section 2 contains necessary notation and the main result that the distribution of $(ne, se)$ is invariant under the action of the symmetric group $S_n$ on the rows of the underlying layer polyomino. Then we give new proofs for the symmetry of the mixed statistics for fillings of moon polyominoes, and describe how to obtain the joint distribution of $(ne, se)$ over 01-fillings of a layer polyomino. This is the content of Sections 3 and 4. In Section 5 we prove the symmetric distribution of a variation of NE/SE chains. We conclude the paper with some comments and counterexamples to a few seemingly natural generalizations. 2 Northeast and southeast chains in layer polyominoes We begin by giving necessary notations in the language of fillings of polyominoes. A polyomino is a finite subset of $\mathbb{Z}^2$, where every element of $\mathbb{Z}^2$ is represented by a square cell. The polyomino is row-convex (column-convex) if its intersection with any row (column) of $\mathbb{Z}^2$ is connected. It is convex if it is both row-convex and column convex. A polyomino is said to be intersection-free if every two rows are comparable, i.e., the column-coordinates of one row form a subset of those of the other row. It is easily checked that this is equivalent to saying that every two columns are comparable. Definition 1. A polyomino is called a moon polyomino if it is convex and intersection-free; it is called a layer polyomino if it is row-convex and intersection-free, but not necessarily column-convex. The term of “moon polyomino” was first introduced by Jonsson [9] in the study of generalized triangulations and diagonal-free subsets of polyominoes. Jonsson also suggested a family of stalactite polyominoes, which are column-convex, intersection-free, and the columns are aligned on the top. By definition, a moon polyomino is a layer polyomino whose rows are arranged in a unimodal order; while a stalactite polyomino, after a $90^\circ$ rotation, becomes a layer polyomino that is aligned on the left (or right). Figure 1 shows a moon polyomino and a general layer polyomino. We remark that all the results in this paper apply to polyominoes which are intersection-free and column-convex, but not necessarily row-convex. ![Figure 1: A moon polyomino and a layer polyomino](image) Given a layer polyomino $\mathcal{L}$, we assign 0 or 1 to each cell of $\mathcal{M}$ so that there is at most one 1 in each column. Throughout this paper we will simply use the term filling to denote such 01-fillings. In a filling a cell is empty if it is assigned 0, and is a 1-cell otherwise. We label the rows from top to bottom, and the columns from left to right. The cell $(i,j)$ refers to the cell lying in the $i$-th row and the $j$-th column. A $2 \times 2$ submatrix $S$ of $\mathcal{L}$ is a set of four cells in $\mathcal{L}$ with the coordinates $$S = \{(i_1, j_1), (i_1, j_2), (i_2, j_1), (i_2, j_2) \in \mathcal{L} : 1 \leq i_1 < i_2, 1 \leq j_1 < j_2\}. \quad (1)$$ Note that if $S$ is a submatrix of $\mathcal{L}$, then all the cells $\{(i_1, k), (i_2, k) : j_1 \leq k \leq j_2\}$ are also in $\mathcal{L}$; if furthermore $\mathcal{L}$ is a moon polyomino, then the rectangle $\{(l, k) : i_1 \leq l \leq i_2, j_1 \leq k \leq j_2\}$ is contained in $\mathcal{L}$. A northeast (NE) chain of length 2 in a filling $L$ consists of a $2 \times 2$ submatrix $S$ as in (1) where $(i_1, j_2)$ and $(i_2, j_1)$ are 1-cells. There is no constraint on the filling of the other two cells. Similarly, a southeast (SE) chain of length 2 consists of a submatrix $S$ where $(i_1, j_1)$ and $(i_2, j_2)$ are 1-cells. In both cases we say that the submatrix $S$ is the support of the chain. Unless otherwise specified, all the chains in this paper are of length 2. The numbers of NE-chains and SE-chains of $L$ are denoted by $\text{ne}(L)$ and $\text{se}(L)$, respectively. Example 2. Figure 2 shows a 01-filling of a layer polyomino, where $a, b, c, d, e, f$ are 1-cells. This filling has four NE-chains $\{ad, bd, cd, ef\}$, and two SE-chains $\{df, de\}$. ![Figure 2: A filling of a layer polyomino](image) Assume a polyomino $\mathcal{L}$ has $n$ rows and $m$ columns. Let $\mathbf{s} = (s_1, \ldots, s_n) \in \mathbb{N}^n$ and $\mathbf{e} = (\varepsilon_1, \ldots, \varepsilon_m) \in \{0, 1\}^m$ with $$\sum_{i=1}^{n} s_i = \sum_{j=1}^{m} \varepsilon_j.$$ We denote by $\mathbf{F}(\mathcal{L}, \mathbf{s}, \mathbf{e})$ the set of fillings $L$ of $\mathcal{L}$ such that the $i$-th row has exactly $s_i$ 1’s, and the $j$-th column has exactly $\varepsilon_j$ 1’s, for $1 \leq i \leq n$ and $1 \leq j \leq m$. For example, the filling in Figure 2 has $\mathbf{s} = (0, 1, 1, 1, 1, 2)$ and $\mathbf{e} = (1, 0, 1, 1, 1, 1, 1, 1)$. Our main result is that the distribution of $(\text{ne}, \text{se})$ is invariant under any permutation of rows of the underlying layer polyomino. Theorem 3. Let $\mathcal{L}$ be a layer polyomino with rows $R_1, \ldots, R_n$ from top to bottom. For a permutation $\sigma \in \mathfrak{S}_n$, let $\mathcal{L}' = \sigma(\mathcal{L})$ be the layer polyomino whose rows are $R_{\sigma(1)}, \ldots, R_{\sigma(n)}$, and $\mathbf{s}' = \sigma(\mathbf{s}) = (s_{\sigma(1)}, \ldots, s_{\sigma(n)})$. Then $$\sum_{L \in \mathbf{F}(\mathcal{L}, \mathbf{s}, \mathbf{e})} p^{\text{ne}(L)} q^{\text{se}(L)} = \sum_{L' \in \mathbf{F}(\mathcal{L}', \mathbf{s}', \mathbf{e})} p^{\text{ne}(L')} q^{\text{se}(L')}.$$ Proof. It is sufficient to prove Theorem 3 for adjacent transpositions, that is, when $\sigma = (k, k+1)$ where $k \in \{1, \ldots, n-1\}$. Explicitly, let $\mathcal{L}'$ be the layer polyomino obtained from $\mathcal{L}$ by exchanging the rows $R_k$ and $R_{k+1}$. We construct a bijection $$\phi : \mathbf{F}(\mathcal{L}, \mathbf{s}, \mathbf{e}) \to \mathbf{F}(\mathcal{L}', \mathbf{s}', \mathbf{e})$$ such that $\text{ne}(L) = \text{ne}(\phi(L))$ and $\text{se}(L) = \text{se}(\phi(L))$ for every $L \in F(L, s, e)$. Let $R$ be the largest rectangle with two rows that is contained in $R_k \cup R_{k+1}$. For a filling $L$ of $L$, we define $L' = \phi(L)$ by the following steps. 1. Any 1 of $L$ in $L - (R_k \cup R_{k+1})$ stays in the same cell. 2. Any 1 of $L$ in the cell $(i, r)$ where $(i, r) \in (R_k \cup R_{k+1}) - R$ moves to the position $(\sigma(i), r)$ in the polyomino $L'$. 3. For the 1’s in the cells in $R$, let $R' \subseteq R$ consist of the non-empty columns of $L \cap R$. All the cells in $R - R'$ remain empty in $L'$. Assume $R'$ contains $l$ columns. Identify fillings of $R'$ with binary words $(w_1 \cdots w_l) \in \{0, 1\}^l$ by letting $w_i = 0$ if the 1-cell in the $i$-th column of $R'$ is in row $R_k$; otherwise $w_i = 1$. Let $w(L) = (w_1 \cdots w_l)$ be the word corresponding to the filling $L \cap R'$. Then the filling of $L'$ on $R'$ is the one given by the word $(u_1 \ldots u_l)$, where $u_i = 1 - w_{l+1-i}$. That is, $(u_1 \ldots u_l)$ is obtained from $w(L)$ by first reversing the word, then exchanging 0’s with 1’s. See Figure 3 for an example of Step 3, where the rectangle $R$ is indicated by black lines, and $R'$ contains the unshaded cells in $R$. We have $w(L) = (0, 1, 0, 0)$, and $w(L') = (1, 1, 0, 1)$. ![Figure 3: Step 3 of $\phi$](image) Claim: $\text{ne}(L) = \text{ne}(L')$ and $\text{se}(L) = \text{se}(L')$. Proof of the claim. It is clear that $L' \in F(L', s', e)$. We prove the first equation only. The second one can be treated similarly. First note that any NE-chain formed by two 1-cells outside $R_k \cup R_{k+1}$ is not changed. Next we consider the NE-chains with exactly one 1-cell in $R_k \cup R_{k+1}$. Let $X = (a, b)$ be a cell of $L$ outside $R_k \cup R_{k+1}$. An NE-chain formed by $X$ and $Y = (i, r) \in R_k \cup R_{k+1} - R$ in $L$ is replaced by the NE-chain with 1-cells $X$ and $(\sigma(i), r)$ in $L'$. For the NE-chains in $L$ formed by $X$ and 1-cells in $R$, let $C_b$ be the column of $X$. There are two possibilities. 1. $C_b \cap R = \emptyset$. Without loss of generality, assume $\|R_k\| \geq \|R_{k+1}\|$. Then the number of NE-chains formed by $X$ and 1-cells in $R$ is 0 unless $(k, b) \in L$, in which case the number equals the number of 1s in $R \cap R_k$. In $L'$, row $R_k$ is moved to the row $k+1$, the cell $(k, b)$ is moved to $(k+1, b)$, and $L'$ has the same number of 1’s in $R \cap R_{k+1}$ as that of $L$ in $R \cap R_k$. 2. $ C_b \cap R \neq \emptyset $. Then the number of NE-chains formed by $ X $ and 1-cells in $ R $ equals the number of columns in \[ \begin{cases} \{C_t \cap R' : (a, t) \text{ a cell in the underlying polyomino, } t < b\} & \text{if } a < k, \\ \{C_t \cap R' : (a, t) \text{ a cell in the underlying polyomino, } t > b\} & \text{if } a > k + 1. \end{cases} \] In both cases, this number remains same in $ L $ and $ L' $. Now we consider the rows $ R_k \cup R_{k+1} $. Let $ w(L) $ be the word defined as in Step 3. Then the number of NE-chains formed by two 1-cells in $ R_k \cup R_{k+1} $ is exactly the number of inversions of $ w(L) $. But for any sequence $ w_1w_2\ldots w_n \in \{0, 1\}^n $, \[ \text{inv}(w_1w_2\ldots w_l) = \text{inv}(u_1u_2\ldots u_l) \quad \text{if } u_i = 1 - w_{l+1-i}. \] Hence there is an equal number of NE-chains of this type in $ L $ and $ L' $. Combining the above cases, we have $ \text{ne}(L) = \text{ne}(L') $. \[\blacksquare\] Corollary 4. The joint distribution of $ (\text{ne}(L), \text{se}(L)) $ is symmetric, that is, \[ \sum_{L \in F(\mathcal{L}, s, e)} p^{\text{ne}(L)} q^{\text{se}(L)} = \sum_{L \in F(\mathcal{L}, s, e)} p^{\text{se}(L)} q^{\text{ne}(L)}. \] Proof. Let $ \tilde{\mathcal{L}} $ be the polyomino obtained by reversing the rows of $ \mathcal{L} $, that is, $ \tilde{\mathcal{L}} = \sigma(\mathcal{L}) $ where $ \sigma = n \cdots 21 $. For any filling $ L $ of $ \mathcal{L} $, let $ \tilde{L} $ be obtained from $ L $ by keeping every 1 in the same cell while re-arranging rows from $ \mathcal{L} $ to $ \tilde{\mathcal{L}} $. Then $ \text{ne}(\tilde{L}) = \text{se}(L) $, $ \text{se}(\tilde{L}) = \text{ne}(L) $ and the symmetry follows from Theorem 3. We remark that Corollary 4 implies immediately the symmetry of the distribution of $ (\text{ne}, \text{se}) $ over moon polyominoes, and that the bijection given here is different from Kasraoui’s [11]. In addition to the set $ F(\mathcal{L}, s, e) $, we can also consider 01-fillings of a layer polyomino with arbitrary column sum, but every row has at most one 1. Explicitly, let $ \mathcal{L} $ be a layer polyomino with $ n $ rows and $ m $ columns. Given $ e = (\varepsilon_1, \ldots, \varepsilon_n) \in \{0, 1\}^n $, $ s = (s_1, \ldots, s_m) \in \mathbb{N}^m $ with \[ \sum_{i=1}^{n} \varepsilon_i = \sum_{j=1}^{m} s_j, \] let $ F^c(\mathcal{L}, e, s) $ be the set of fillings of $ \mathcal{L} $ with the row-sum $ e $ and column-sum $ s $. We have Theorem 5. Let $ \mathcal{L} $ be a layer polyomino with rows $ R_1, \ldots, R_n $ from top to bottom. For a permutation $ \sigma \in S_n $, let $ \mathcal{L}' = \sigma(\mathcal{L}) $ be the layer polyomino whose rows are $ R_{\sigma(1)}, \ldots, R_{\sigma(n)} $, and $ e' = \sigma(e) = (\varepsilon_{\sigma(1)}, \ldots, \varepsilon_{\sigma(n)}) $. Then \[ \sum_{L \in F^c(\mathcal{L}, e, s)} p^{\text{ne}(L)} q^{\text{se}(L)} = \sum_{L' \in F^c(\mathcal{L}', e', s)} p^{\text{ne}(L')} q^{\text{se}(L')}. \] Proof. Again we prove that the equation holds for any adjacent transposition $ \sigma = (k, k+1) $ by constructing a bijection $ \phi^c : F^c(L, e, s) \to F^c(L', e', s) $ that preserves the statistic $ (\text{ne}, \text{se}) $. The map $ \phi^c $ is essentially the same as $ \phi $, but it can be described in a much simpler way since for fillings in $ F^c(L, e, s) $ every row has at most one 1. Again let $ R $ be the largest rectangle with two rows that is contained in $ R_k \cup R_{k+1} $. Let $ L $ be a filling in $ F^c(L, e, s) $. Then $ L' = \phi^c(L) $ is the filling of $ F^c(L', e', s) $ obtained from $ L $ by the following steps. (1) Any 1 in a cell $(i, j)$ with $ i \notin \{k, k+1\} $ stays in the same cell; (2) Any 1 in a cell $(i, j) \in (R_k \cup R_{k+1}) - R$ moves to the cell $(\sigma(i), j)$. (3) If there is only one 1-cell in $ R $, say $(i, j)$, then $(\sigma(i), j)$ becomes a 1-cell in $ L' $. Otherwise (i.e., there are two 1-cells in $ R $), the fillings $ L $ and $ L' $ agree on $ R $. The proof that $ \phi^c $ preserves $ (\text{ne}, \text{se}) $ is similar to that of Theorem 3, and is omitted here. $ \blacksquare $ ### 3 Mixed statistics in fillings of moon polyominoes There is a stronger symmetry between the northeast and southeast chains of length 2 in fillings of moon polyominoes. In [2] Chen et al. introduced four families of mixed statistics with respect to a bipartition of rows or columns of the underlying moon polyomino, and proved that the symmetric joint distribution holds for each family of mixed statistics. In this section we explain such a symmetry for moon polyominoes from Theorems 3 and 5. Let $ S $ be a subset of rows of a moon polyomino $ M $. An NE/SE chain is called a top $ S $-NE/SE chain if its top 1-cell is in $ S $; Similarly, it is a bottom $ S $-NE/SE chain if the lower 1-cell of the chain is in $ S $. Let $ \bar{S} = M \setminus S $ be the complement of $ S $. Given a filling $ M \in F(M, s, e) $, define the top-mixed statistic $ \alpha(S; M) $ and the bottom-mixed statistic $ \beta(S; M) $ with respect to $ S $ as \[ \alpha(S; M) = \#\{\text{top } S\text{-NE chains of } M\} + \#\{\text{top } \bar{S}\text{-SE chains of } M\}, \] \[ \beta(S; M) = \#\{\text{bottom } S\text{-NE chains of } M\} + \#\{\text{bottom } \bar{S}\text{-SE chains of } M\}. \] Analogously one defines the left-mixed statistic $ \gamma(T; M) $ and right-mixed statistic $ \delta(T; M) $ with respect to a subset $ T $ of columns. The main result of [2] can be stated as follows: Let $ \lambda(A; M) $ be any of the four statistics $ \alpha(S; M), \beta(S; M), \gamma(T; M), \delta(T; M) $, then the joint distribution of the pair $ (\lambda(A; M), \lambda(\bar{A}; M)) $ over $ F(M, s, e) $ is symmetric and independent of the subsets $ S, T $. In particular, the joint distribution is the same as the distribution of $ (\text{ne}(M), \text{se}(M)) $. We show that the stable distribution of mixed statistics can be derived from results in Section 2. We discuss the case of top-mixed statistics only. Other mixed-statistics can be treated similarly. In a moon polyomino $ M $, assume the rows in $ S $ are $ R_{i_1}, R_{i_2}, \ldots, R_{i_s} $, We discuss briefly how to get the explicit formula of $F_{\mathcal{L}}(p, q)$ for a layer polyomino $\mathcal{L}$. There are two approaches. The first one is to transform the layer polyomino $\mathcal{L}$ to a moon polyomino by a permutation of rows, then apply Theorem 2.2 of Kasraoui’s [11] which expressed the distribution of (ne, se) for fillings of moon polyominoes as a product of $(p, q)$-Gaussian coefficients. Alternatively, using Theorems 3 and 5, we can transform the layer polyomino $\mathcal{L}$ to a Ferrers diagram (partition shape) by first rearranging the rows from longest to shortest, then rearranging the columns from large to small, see Figure 5 for an illustration. This reduces the problem to computing the distribution of (ne, se) over 01-fillings of a Ferrers shape, which, by a bijection of de Mier [7], is equivalent to the enumeration of crossings and nestings of length 2 over linked partitions. The latter has been studied and explicit formula was given in [3, Theorem 3.5]. This in turn provides a new proof for Kasraoui’s formula for 01-fillings of moon polyominoes. ![Figure 5: Transformation from layer polyomino to a Ferrers diagram.](image) We remark that in [17] the authors provided another way to transform the enumeration of (ne, se) over fillings of moon polyominoes to those of a Ferrers diagram. The idea is to associate with a moon polyomino $\mathcal{M}$ a charge function $C : \mathcal{M} \to \{\pm 1\}$ which induces a sign on the NE/SE-chains of length 2, and then prove that the numbers of positive and negative chains have a stable distribution which is independent of the charge function. The connection between charge functions and layer polyominoes is currently under investigation. ## 5 A variation of NE/SE chains In this section we consider a variation of the NE/SE chains of length 2 in fillings of layer polyominoes. Let $\mathcal{L}$ be a layer polyomino and $L$ a 01-filling of $\mathcal{L}$. For an NE/SE chain with the support $$S = \{(i_1, j_1), (i_1, j_2), (i_2, j_1), (i_2, j_2) \in \mathcal{L} : 1 \leq i_1 < i_2, 1 \leq j_1 < j_2\},$$ we say that it is a strong chain if the whole rectangle $\{(l, k) : i_1 \leq l \leq i_2, j_1 \leq k \leq j_2\}$ is also contained in $\mathcal{L}$. For example, in the filling of Figure 2, there are three strong NE-chains $\{bd, cd, ef\}$, and no strong SE-chains. On the other hand, for moon polyominoes any NE/SE chain is strong. Denote by ne$\Box(L)$ and se$\Box(L)$ the numbers of strong NE-chains and strong SE-chains of $L$ respectively. It is easy to see that for a general layer polyomino, the distribution of $(\text{ne}^\square, \text{se}^\square)$ is not preserved under permutations of rows, and not necessarily equal to the distribution of $(\text{ne}, \text{se})$. Nevertheless, we prove that the distribution of $(\text{ne}^\square, \text{se}^\square)$ is still symmetric over fillings in $F(\mathcal{L}, s, e)$. Theorem 7. The statistic $(\text{ne}^\square, \text{se}^\square)$ has a symmetric joint distribution over the set $F(\mathcal{L}, s, e)$, where $s = (s_1, \ldots, s_n) \in \mathbb{N}^n$ and $e = (\epsilon_1, \ldots, \epsilon_m) \in \{0, 1\}^m$. That is, \[ \sum_{L \in F(\mathcal{L}, s, e)} p^{\text{ne}^\square(L)} q^{\text{se}^\square(L)} = \sum_{L \in F(\mathcal{L}, s, e)} p^{\text{se}^\square(L)} q^{\text{ne}^\square(L)}. \] (2) Proof. We prove by induction on $n$, the number of rows of $\mathcal{L}$. For $n = 1, 2$ the claim is true since every layer polyomino of no more than two rows is a moon polyomino, and hence $(\text{ne}^\square, \text{se}^\square)$ coincides with $(\text{ne}, \text{se})$. Assume the claim is true for all layer polyominoes with less than $n$ rows, and with arbitrary row sum in $\mathbb{N}$ and column sum in $\{0, 1\}$. Let $\mathcal{L}$ be a layer polyomino with $n$ rows and $m$ columns, $s = (s_1, \ldots, s_n) \in \mathbb{N}^n$, and $e = (\epsilon_1, \ldots, \epsilon_m) \in \{0, 1\}^m$. We shall prove the theorem for $\mathcal{L}$. First we need some notations. Let $R_i$ be a row of $\mathcal{L}$ with maximal number of cells. If there are more than one such rows, take the one with maximal index. Among $R_{i-1}$ and $R_{i+1}$, pick the one with more cells. Without loss of generality, we assume that it is $R_{i-1}$. Let $\mathcal{R}$ be the maximal rectangle of two rows that is contained in $R_{i-1} \cup R_i$, and let $\mathcal{T} = R_i - \mathcal{R}$. For $L \in F(\mathcal{L}, s, e)$, let \[ \Gamma(L) = \{L' \in F(\mathcal{L}, s, e) : L' \text{ and } L \text{ agree on } \mathcal{T}\}. \] We shall prove the following equation for arbitrary $L$, which then implies (2): \[ \sum_{L' \in \Gamma(L)} p^{\text{ne}^\square(L')} q^{\text{se}^\square(L')} = \sum_{L' \in \Gamma(L)} p^{\text{se}^\square(L')} q^{\text{ne}^\square(L')}. \] (3) Assume in all fillings of $\Gamma(L)$, there are $k$ 1-cells in $\mathcal{T}$ in columns $t_1, t_2, \ldots, t_k$. These 1-cells do not appear in any strong NE/SE chains. Hence removing $\mathcal{T}$ from $\mathcal{L}$ does not change $(\text{ne}^\square, \text{se}^\square)$ for any filling. Let $\mathcal{L}_1 = \mathcal{L} - \mathcal{T}$, $s_1 = (s_1, \ldots, s_{i-1}, s_i - k, s_{i+1}, \ldots, s_n)$ and $e_1$ be obtained from $e$ by changing $\epsilon_{t_1}, \ldots, \epsilon_{t_k}$ from 1 to 0. Then the map \[ \pi : \Gamma(L) &\rightarrow& F(\mathcal{L}_1, s_1, e_1) \\ L' &\mapsto& L' \text{ restricted to } \mathcal{L}_1 \] is a bijection that preserves the statistic $(\text{ne}^\square, \text{se}^\square)$. Define $\mathcal{L}_2$ with $n - 1$ rows $(R_1, \ldots, R_{i-1}, R_{i+1}, \ldots, R_n)$, $s_2 = (s_1, \ldots, s_{i-2}, s_{i-1} + s_i - k, s_{i+1}, \ldots, s_n)$ and $e_2 = e_1$. Let $\rho : F(\mathcal{L}_1, s_1, e_1) \rightarrow F(\mathcal{L}_2, s_2, e_2)$ by defined as follows. For $L_1 \in F(\mathcal{L}_1, s_1, e_1)$ perform the following operations. 1. Move all 1s from row $i$ into row $i - 1$ preserving their column, 2. Delete row $i$, 3. Shift rows $i+1, i+2, \ldots, n$ up one, so row $R_i$ becomes $R_{j-1}$ for $j = i+1, i+2, \ldots, n$. The map $\rho$ is surjective, but not injective. In fact for any $L_2 \in F(L_2, s_2, e_2)$, there are $\binom{s_{i-1} + s_i - k}{s_{i-1}}$ fillings in $F(L_1, s_1, e_1)$ that map to it. This is obtained by splitting the $s_{i-1} + s_i - k$ 1-cells in the $(i-1)$st row of $L_2$ into two subsets of sizes $s_{i-1}$ and $s_i - k$. Thus $$\sum_{L_1 \in F(L_1, s_1, e_1: \rho(L_1) = L_2} p^{\text{ne}}(L_1) q^{\text{se}}(L_1) = p^{\text{ne}}(L_2) q^{\text{se}}(L_2) \left[ \frac{s_{i-1} + s_i - k}{s_{i-1}} \right]_{p,q},$$ where $$\left[ \frac{n}{s} \right]_{p,q} = \frac{[n]_{p,q}!}{[s]_{p,q}![n-s]_{p,q}!}.$$ is the $(p,q)$-Gaussian coefficient with the $(p,q)$-integers $[n]_{p,q} = p^{n-1} + p^{n-2}q + \cdots + pq^{n-2} + q^{n-1}$ and $(p,q)$-factorial $[n]_{p,q}! = \prod_{i=1}^{n} [i]_{p,q}$. It follows that $$\sum_{L' \in \Gamma(L)} p^{\text{ne}}(L') q^{\text{se}}(L') = \sum_{L_1 \in F(L_1, s_1, e_1)} p^{\text{ne}}(L) q^{\text{se}}(L) = \left[ \frac{s_{i-1} + s_i - k}{s_{i-1}} \right]_{p,q} \cdot \sum_{L_2 \in F(L_2, s_2, e_2)} p^{\text{ne}}(L) q^{\text{se}}(L).$$ Equation (3) follows from the inductive hypothesis on $F(L_2, s_2, e_2)$. 6 Concluding remarks We conclude this paper with some comments and counterexamples to a few seemingly natural generalizations of Theorem 3 and 5. First, it was observed by Kasraoui [11] that with no restrictions on the row sum or the column sum, the joint distribution of the statistics (ne, se) is not symmetric over arbitrary 01-fillings of moon polyominoes, and hence layer polyominoes. A similar problem can be considered for N-fillings, i.e., each cell $(i,j)$ of the polyomino is assigned a non-negative integer $n_{i,j}$. For a submatrix $S$ as in (1) in the polyomino, it contributes $n_{i_2,j_1} n_{i_1,j_2}$ NE-chains, and $n_{i_1,j_1} n_{i_2,j_2}$ SE-chains. It is natural to ask whether the joint distribution of (ne, se) is symmetric over arbitrary N-fillings with a fixed total sum of entries. The answer is negative, as shown by the following counterexample. Example 8. Consider the Ferrers diagram of shape $(3,3,2)$ and N-fillings with sum of entries equal to 5. We compute the number of fillings with ne = 6 or se = 6. It is easy to check that to get six NE-chains or six SE-chains, the set of entries in the fillings must be $(3,2), (2,2,1),$ or $(2,1,1,1)$. With the entries $(3,2)$, there are ten fillings with ne = 6 and ten with se = 6; With the entries $(2,2,1)$, there are six fillings with ne = 6 and six with se = 6; while with the entries $(2,1,1,1)$, there is one filling with se = 6 (see Figure 6), but none with ne = 6. Notice that the difference between a moon polyomino and a layer polyomino is that the latter is only required to be row-convex, but not necessary column-convex. Another natural question is whether we can get rid of the convexity completely? That is, for an arbitrary intersection-free polyomino, whether the pair (ne, se) is symmetric or preserved under permutation of rows. Example 9 gives a negative answer. Example 9. The left polyomino $\mathcal{L}$ in Figure 7 is intersection-free but not row-convex nor column-convex. (The cell at the 3rd row and 2nd column does not belong to $\mathcal{L}$. ) Consider 01-fillings $L$ such that every row has exactly one 1, and the column sum is $(2, 1, 1)$. There are six possible 01-fillings. We obtain that $\sum_L p^{\text{ne}(L)} q^{\text{se}(L)} = p^3 + 3p^2q + 2q^3$. On the other hand, the right polyomino $\mathcal{L}'$ is obtained from the left one by exchanging the middle two rows, for which we have $\sum_{L'} p^{\text{ne}(L')} q^{\text{se}(L')} = p^3 + 2p^2q + 2pq^2 + q^3$. ![Figure 7: Non-convex polyomino](image) Kasraoui also provided examples [11, Sec.6] showing that the distribution of (ne, se) may not be symmetric if the polyomino is convex but not intersection-free. Finally we consider the length of maximal NE-chains in a filling. A $k$-NE chain in a filling $L$ on $\mathcal{L}$ is a $k \times k$ submatrix consisting of cells $\{(i_s, j_t) \in \mathcal{L} : 1 \leq s, t \leq k, i_1 < i_2 < \cdots < i_k, \text{ and } j_1 < j_2 < \cdots < j_k\}$ such that the cells $(i_r, j_r)$ are filled with 1 for $r = 1, \ldots, k$. Let $\text{maxne}(L)$ be the maximal $k$ such that $L$ contains a $k$-NE chain. The statistic $\text{maxne}(L)$ has been studied for fillings of Ferrers diagrams, stack polyominoes, and moon polyominoes, for example, see [1, 6, 7, 9, 10, 14, 16]. In particular, [16] showed that in 01-fillings of a moon polyomino with given row sum, (but no constraints on the column sum), the distribution of $\text{maxne}$ depends only on the set of columns, but not how the columns are arranged. The proofs of [16] imply that in 01-fillings of a moon polyomino where every row and every column has at most one 1, the distribution of $\text{maxne}$ depends only on the set of rows (or columns). This is not true for layer polyominoes. We found two polyominoes, shown in Example 10, that differ by exchanging the position of two adjacent rows, yet the distributions of the length of the longest NE-chains are different. Example 10. In the following two layer polyominoes, consider the 01-fillings where every row and every column has exactly one 1. We obtained the following generating function for the statistic maxne: For the left polyomino $ L_1 $, \[ \sum_{L_1} p^{\text{maxne}(L_1)} = p + 37p^2 + 31p^3 + 3p^4, \] while for the right polyomino $ L_2 $, \[ \sum_{L_2} p^{\text{maxne}(L_2)} = p + 36p^2 + 32p^3 + 3p^4. \] Furthermore, for $ L_1 $, \[ \sum_{L_1} p^{\text{maxse}(L_1)} = p + 36p^2 + 32p^3 + 3p^4 \] where $ \text{maxse}(L) $ is the maximal $ k $ such that $ L $ contains a $ k $-SE chain. Hence in general layer polyominoes maxne and maxse do not have the same distribution. Acknowledgments This publication was made possible by NPRP grant # [5-101-1-025] from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors. The authors would like to thank an anonymous referee for carefully reading the manuscript and for giving many valuable comments and suggestions which substantially helped improving the quality of the paper. In particular, the variation of NE/SE chains studied in Section 5 was suggested by the referee. References [1] W. Y. C. Chen, E. Y. P. Deng, R. R. X. Du, R. P. Stanley and C. H. Yan, Crossings and nestings of matchings and partitions, Trans. Amer. Math. Soc. 359 (2007) 1555–1575. [2] W.Y.C Chen, A. Y. Z. Wang, C.H.Yan and A. F. Y. Zhao. Mixed statistics on 01-fillings of moon polyominoes, SIAM J. Discrete Math. 24 (2010) no. 4, 1272–1290. [3] W. Y. C. Chen, S. Y. Wu and C. H. Yan, Linked partitions and linked cycles, European J. Combin. 29 (2008) 1408–1426. [4] W. Y. C. Chen, S. Poznanović, C. H. Yan and A. L. B. Yang, Major index for 01-fillings of moon polyominoes, J. Combin. Theory, Ser. A. Vol. 117 (2010), no.8, 1058–1081. [5] S. Corteel. Crossings and alignments of permutations. Adv. Appl. Math. 38(2007), 149–163. [6] A. de Mier, On the symmetry of the distribution of crossings and nestings in graphs, Electron. J. Combin. 13 (2006), N21. [7] A. de Mier, $k$-noncrossing and $k$-nonnesting graphs and fillings of Ferrers diagrams, Combinatorica 27 (2007) 699–720. [8] M. de Sainte-Catherine, Couplages et Pfaffiens en Combinatoire, Physique et Informatique, Ph.D. Thesis, University of Bordeaux I, Talence, France, 1983. [9] J. Jonsson, Generalized triangulations and diagonal-free subsets of stack polyominoes, J. Combin. Theory Ser. A 112 (2005), 117–142. [10] J. Jonsson and V. Welker. A spherical initial ideal for Pfaffians. Illinois J. Math. 51(4):1397–1407, 2007. [11] A. Kasraoui, Ascents and descents in 01-fillings of moon polyominoes, European J. Combin. 31 (2010), no.1, 87–105 [12] A. Kasraoui and J. Zeng, Distributions of crossings, nestings and alignments of two edges in matchings and partitions, Electron. J. Combin. 13 (2006) R33. [13] M. Klazar, On identities concerning the numbers of crossings and nestings of two edges in matchings, SIAM J. Discrete Math. 20 (2006) 960–976. [14] C. Krattenthaler, Growth diagrams, and increasing and decreasing chains in fillings of Ferrers shapes, Adv. in Appl. Math. 37 (2006), 404–431. [15] S. Poznanović, C. H. Yan, Crossings and nestings of two edges in set partitions, SIAM J. Discrete Math. 23 (2009), no.2, 787–804. [16] M. Rubey, Increasing and decreasing sequences in fillings of moon polyominoes, Adv. in Appl. Math. 47 (2011), 57–87. [17] A.Y.Z. Wang and C.H. Yan, Positive and negative chains in charged moon polyominoes. Adv. in Appl. Math. 51 (2013), 467–482.
GENERAL MEETING Randall Museum . 199 Museum Way . San Francisco 7:00 pm Doors Open . 7:30 pm Announcements . 8:00 pm Speaker Effective February 17, 2015: SFAA’s General Meetings occur on the 3rd Tuesday of each month (except January) A NEW HOME FOR SAN FRANCISCO AMATEUR ASTRONOMERS NEW MEETING LOCATION Effective February 17, 2015: SFAA’s General Meetings occur on the 3rd Tuesday of each month (except January) !!! TUESDAY !!! FEBRUARY 17, 2015 – 7:30 pm THE PRESIDIO OBSERVATION POST - BUILDING 211 Additional Information Links Observation Post/Building 211 layout http://www.presidio.gov/venues/Documents/Bldg%200211%20Floor%20Plan.pdf Driving Directions http://www.presidio.gov/venues/Pages/observation-post-at-the-presidio-driving-directions.aspx Public transportation information link http://www.presidio.gov/transportation/Pages/default.aspx FRANCK MARCHIS, PH.D., SETI Institute & UC Berkeley Breaking the Seeing Barriers for Planetary Astronomy Once monthly, the San Francisco Amateur Astronomers hosts distinguished guest speakers who are leaders in the fields of astronomy, physics, and related disciplines, to present the latest developments from cutting-edge scientific programs. Dr. Franck Marchis, currently a Principal Investigator at the Carl Sagan Center of the SETI Institute, opens the 2015 Monthly Lecture Series with a presentation on Breaking the Seeing Barriers for Planetary Astronomy. In the past four centuries planetary astronomy, the study of our solar system bodies using telescopes, has increased our knowledge of the environment of Earth, the evolution of planets, the origin of comets and asteroids, and the formation of our solar system. In this presentation, Dr. Marchis will discuss the contributions of telescopic observation over the past 50 years, highlighting the most recent discoveries and looking into the future of space-based astronomy, particularly for the search and study of planets around other stars in our galaxy. About the Speaker: Franck Marchis is a Principal Investigator at the SETI Institute, and also an Assistant Research Astronomer at UC Berkeley. He received his Ph.D. in 2000 from University of Toulouse, France in planetary science, then moved to California shortly after through a postdoctoral position at UC Berkeley. In 2007, he was appointed as a Planetary Scientist at the Carl Sagan Center of the SETI Institute where he expanded his research on multiple asteroids using space-telescope facilities. In June 2011, he took a full-time position at the Carl Sagan Center to lead the development of space mission concepts and new high-resolution & high contrast instruments for ground-based telescopes. He is an associate astronomer at Observatoire de Paris since June 2003. Ages: Geared for adults; All ages welcome Cost: FREE – Donations encouraged Info: www.sfaa-astronomy.org 2015 STAR PARTY DATES Anthony Barreiro Here’s the current schedule for SFAA star parties during 2015. Please mark your calendar and plan to attend! Thanks to Paul Salazar and Michael Patrick of SFAA, Tinka Ross of the Friends of Mount Tamalpais, and Supervising Ranger Roberto Walton of Mt. Tam State Park for helping to plan and schedule these events. Mount Tamalpais members nights are subject to final approval of our State Park permit application to use the Rock Springs parking area. Members nights are scheduled for the Saturday nearest the new Moon, and give us a chance to observe deep sky objects in a relatively dark sky. Make sure your membership is up to date and you have a current SFAA parking pass. (You can renew your membership at www.sfaa-astronomy.org/membership/. If you need a parking pass, please pick one up at a meeting or send an email to email@example.com with your mailing address.) March 21 will be our annual Messier marathon, an opportunity to try to observe all 109 Messier objects in one night. From April through October SFAA presents monthly public astronomy programs in collaboration with the Friends of Mount Tamalpais. Friends of Mt. Tam presents a lecture by a professional astronomer in the Mountain Theater, followed by telescope viewing in the Rock Springs parking area. Public nights are generally on the Saturday nearest the waxing quarter Moon, so the Moon, planets, and brighter deep sky objects like double stars and star clusters are good things to share with the visitors. After the visitors leave we get a few hours of observing time to ourselves. You need a parking pass to stay after the visitors leave. During 2014 SFAA held our monthly City Star Parties at Point Lobos / Lands End, the Embarcadero outside the Exploratorium, and the Presidio Parade Ground. Locations for 2015 City Star Parties will be announced month by month. With urban light pollution and a bright Moon, the Moon, planets, and the very brightest deep sky objects work well for City Star Parties. Each star party give us a chance to hang out with one another, check out other members’ equipment, and learn more about the sky and the cosmos. The Mount Tam public nights and City Star Parties are excellent opportunities to share our love of astronomy and the night sky with the public and to let people know about SFAA and our other programs. If you’re able to bring a telescope to a star party that’s great, but even without a telescope you can still enjoy the sky and share your knowledge with others. It’s best to arrive before sunset, so you’ll have light to set up your equipment and so you won’t be shining your headlights into other people’s dark-adapted eyes. If you’ve been to a few star parties and you’re willing to serve as a contact person for an upcoming event, please let me know -- email firstname.lastname@example.org. January 17 Mt. Tam members night January 31 City star party February 14 Mt. Tam members night February 21 City star party March 21 Mt. Tam members night -- Messier marathon March 28 City star party April 18 Mt. Tam members night April 25 Mt. Tam public night May 2 City star party May 16 Mt. Tam members night May 23 Mt. Tam public night May 30 City star party June 13 Mt. Tam members night June 20 Mt. Tam public night June 27 City star party July 18 Mt. Tam members night July 25 Mt. Tam public night August 1 City star party August 15 Mt. Tam members night August 22 Mt. Tam public night August 29 City star party September 12 Mt. Tam members night September 19 Mt. Tam public night September 26 City star party October 10 Mt. Tam members night October 17 Mt. Tam public night October 24 City star party November 14 Mt. Tam members night November 21 City star party December 12 Mt. Tam members night December 19 City star party \| EVERY FRIDAY NIGHT 7:00 PM – 10:00 PM excluding major holidays \| THE TELESCOPE MAKERS’ WORKSHOP is held every Friday night from 7pm - 10pm, excluding major holidays (e.g. Christmas Day and New Year’s Day) that fall on Fridays. The Workshop is always closed on Memorial Day Weekend. Attendance every Friday night is not mandatory, and members work at their own pace. The Workshop meets at Chabot Space & Science Center, 10000 Skyline Blvd., Oakland. Chabot’s TMW is one of only a handful of regularly scheduled telescope making workshops in the U.S., and probably the world; it meets every Friday evening throughout the year, except Memorial Day weekend. It has been in operation since December of 1930, founded by Franklin B. Wright, and is currently run by Eastbay Astronomical Society member Rich Ozer, with help from other EAS members, Dave Barosso, Barry Leska, and others. The price of admission is FREE. All you have to do is show up, buy a mirror blank and a “tool” (typically around $100 - $200 depending on the size of the mirror) and start ‘pushin’ glass!’ We supply you with instruction, the various grits you’ll need to first grind, and then polish and figure your mirror, and all the testing equipment needed. With a small bit of luck, you could wind up with a telescope that costs 1/3 or 1/4 the cost of a store-bought telescope, that is yet optically superior! It does take time - depending on how much time you put in on it, and other factors, it could take a few months or several months. But, it’s a fun project, great for kids, and at the end you get a great telescope! For more information call or email Richard Ozer at email@example.com or phone (510) 406-1914. \| \| --- \| --- \| \| EVERY FRIDAY & SATURDAY EVENING, weather permitting 7:30 PM – 10:30 PM CHABOT SPACE AND SCIENCE CENTER 10000 Skyline Boulevard Oakland CA 94619-2450 (510) 336-7300 \| EXPLORE THE NIGHT SKIES AT THE CHABOT OBSERVATORIES For more information: http://www.chabotspace.org/ Free Telescope Viewing Regular hours are every Friday & Saturday evening, weather permitting: 7:30pm - 10:30pm Come for spectacular night sky viewing the best kept secret in the Bay Area and see the magnificence of our telescopes in action! Daytime Telescope Viewing On Saturday and Sunday afternoons come view the sun, moon, or Venus through Chabot’s telescopes. Free with General Admission. (weather permitting) 12pm - 5pm: Observatories Open \| Sunset – 5:11 PM (TWICE MONTHLY) Inclement weather (clouds, excessive wind and showers) will cause the event to be canceled without notice. SAN MATEO COUNTY ASTRONOMICAL SOCIETY STAR PARTY STAR PARTIES AT CRESTVIEW PARK, SAN CARLOS Come out and bring the kids for a mind expanding look at the universe The City of San Carlos Parks and Recreation Department and the San Mateo County Astronomical Society has open Star Parties twice a month. These events are held in Crestview Park, San Carlos California. Note that inclement weather (clouds, excessive wind and showers) will cause the event to be canceled without notice. For more information call Bob Black, (650)592-2166, or send an email to firstname.lastname@example.org or call Ed Pieret at (650)862-9602. Reasons to Attend If you have kids interested in space or planets bring them here for a real life view of planets, nebula, star clusters and galaxies. If you are thinking of buying a telescope or want help using a telescope you own, come here to talk with experienced users. If you think you might have an interest in astronomy come and talk to experienced amateur astronomers. Cautions Dress warmly and wear a hat. Visitors should park on the street and walk into the park so your headlights don't affect the observer's dark adaptation. Only park in the parking lot if you are arriving before dark and plan to stay until the end of the event. You shouldn't need lights but if you feel you do, only bring a small flashlight with the lens covered using red cellophane or red balloon. Please respect the telescopes and ask permission from the owner if you wish to touch. Parents, please watch your children. The park is residential, and adjacent to homes and backyards, please keep noise to a minimum. Schedule Time Astronomers arrive to set up at around sunset. Observing starts at about one hour after sunset and continues for two to three hours. EVERY CLEAR SATURDAY MORNING OBSERVATORY 10:00 AM – 12:00 PM FOOTHILL COMMUNITY COLLEGE 12345 Moody Road Los Altos Hills Cost: Free Solar observing with a Hydrogen alpha solar telescope every clear Saturday morning. This allows spectacular views of solar prominences and unusual surface features on the Sun not otherwise visible with regular white light telescopes. Admission is free. Foothill Observatory is located on the campus of Foothill College in Los Altos Hills, CA. Take Highway 280 to the El Monte Rd. exit. The observatory is next to parking lot 4. Parking at the college requires visitor parking permits that are available from the machines in the parking lots for $ 3.00. EVERY CLEAR FRIDAY EVENING 9:00 PM – 11:00 PM FOOTHILL COMMUNITY COLLEGE OBSERVATORY Foothill Observatory is open for public viewing every clear Friday evening from 9:00 p.m. until 11:00 p.m. Visitors can view the wonders of the universe through the observatory's computer-controlled 16-inch Schmidt-Cassegrain telescope. Views of objects in our solar system may include craters and mountains on the moon, the moons and cloud-bands of Jupiter, the rings of Saturn, etc. Deep space objects including star clusters, nebulae, and distant galaxies also provide dramatic Foothill Observatory is located on the campus of Foothill College in Los Altos Hills, CA. Take Highway 280 to the El Monte Rd exit. The observatory is next to parking lot 4. Parking at the college requires visitor parking permits that are available from the machines in the parking lots for $3.00. Come to Foothill Observatory and join us in the exploration of our Universe! Foothill Observatory is located on the campus of Foothill College in Los Altos Hills, CA. Take Highway 280 to the El Monte Rd exit. The observatory is next to parking lot 4. Parking at the college requires visitor parking permits that are available from the machines in the parking lots for $3.00. --- NASA SCIENCE NEWS Hubble: Pillars of Creation are also Pillars of Destruction Jan. 7, 2015: Although NASA’s Hubble Space Telescope has taken many breathtaking images of the universe, one snapshot stands out from the rest: the iconic view of the so-called “Pillars of Creation.” The jaw-dropping photo, taken in 1995, revealed never-before-seen details of three giant columns of cold gas bathed in the scorching ultraviolet light from a cluster of young, massive stars in a small region of the Eagle Nebula, or M16. In celebration of its upcoming 25th anniversary in April, Hubble has revisited the famous pillars, providing astronomers with a sharper and wider view. Although the original image was dubbed the Pillars of Creation, the new image hints that they are also “pillars of destruction.” Astronomers using NASA's Hubble Space Telescope have assembled a bigger and sharper photograph of the iconic Eagle Nebula's "Pillars of Creation." Credit: NASA/ESA/Hubble Heritage Team (STScI/AURA)/J. Hester, P. Scowen (Arizona State U.) "I'm impressed by how transitory these structures are," explains Paul Scowen of Arizona State University in Tempe. "They are actively being ablated away before our very eyes. The ghostly bluish haze around the dense..." edges of the pillars is material getting heated up and evaporating away into space. We have caught these pillars at a very unique and short-lived moment in their evolution.” Scowen and astronomer Jeff Hester, formerly of Arizona State University, led the original Hubble observations of the Eagle Nebula. The original 1995 images were taken in visible light. The new image includes near-infrared light as well. The infrared view transforms the pillars into eerie, wispy silhouettes seen against a background of myriad stars. That’s because the infrared light penetrates much of the gas and dust, except for the densest regions of the pillars. Newborn stars can be seen hidden away inside the pillars. The infrared image shows that the very ends of the pillars are dense knots of dust and gas. They shadow the gas below them, keeping the gas cool and creating the long, column-like structures. The material in between the pillars has long since been evaporated away by the ionizing radiation from the central star cluster located above the pillars. At the top edge of the left-hand pillar, a gaseous fragment has been heated up and is flying away from the structure, underscoring the violent nature of star-forming regions. “These pillars represent a very dynamic, active process,” Scowen said. “The gas is not being passively heated up and gently wafting away into space. The gaseous pillars are actually getting ionized, a process by which electrons are stripped off of atoms, and heated up by radiation from the massive stars. And then they are being eroded by the stars’ strong winds and barrage of charged particles, which are literally sandblasting away the tops of these pillars.” When Scowen and Hester used Hubble to make the initial observations of the Eagle Nebula in 1995, astronomers had seen the pillar-like structures in ground-based images, but not in detail. They knew that the physical processes are not unique to the Eagle Nebula because star birth takes place across the universe. But at a distance of just 6,500 light-years, M16 is the most dramatic nearby example – as the team soon realized. The original 1995 image was beautiful. Compare this view to the 2014 image in a side-by-side montage As Scowen was piecing together the Hubble exposures of the Eagle, he was amazed at what he saw. “I called Jeff Hester on his phone and said, ‘You need to get here now,’” Scowen recalled. “We laid the pictures out on the table, and we were just gushing because of all the incredible detail that we were seeing for the very first time.” The first features that jumped out at the team in 1995 were the streamers of gas seemingly floating away from the columns. Astronomers had previously debated what effect nearby massive stars would have on the surrounding gas in stellar nurseries. “There is the only one thing that can light up a neighborhood like this: massive stars kicking out enough horsepower in ultraviolet light to ionize the gas clouds and make them glow,” Scowen said. “Nebulous star-forming regions like M16 are the interstellar neon signs that say, ‘We just made a bunch of massive stars here.’ This was the first time we had directly seen observational evidence that the erosionary process, not only the radiation but the mechanical stripping away of the gas from the columns, was actually being seen.” By comparing the 1995 and 2014 pictures, astronomers also noticed a lengthening of a narrow jet-like feature that may have been ejected from a newly forming star. The jet looks like a stream of water from a garden hose. Over the intervening 19 years, this jet has stretched farther into space, across an additional 60 billion miles, at an estimated speed of about 450,000 miles per hour. Our sun probably formed in a similar turbulent star-forming region. There is evidence that the forming solar system was seasoned with radioactive shrapnel from a nearby supernova. That means that our sun was formed as part of a cluster that included stars massive enough to produce powerful ionizing radiation, such as is seen in the Eagle Nebula. “That’s the only way the nebula from which the sun was born could have been exposed to a supernova that quickly, in the short period of time that represents, because supernovae only come from massive stars, and those stars only live a few tens of millions of years," Scowen explained. "What that means is when you look at the environment of the Eagle Nebula or other star-forming regions, you're looking at exactly the kind of nascent environment that our sun formed in." Credits: Production editor: [Dr. Tony Phillips](http://science.nasa.gov) \| Credit: [Science@NASA](http://science.nasa.gov) More information: The Hubble Space Telescope is a project of international cooperation between NASA and the European Space Agency. NASA's Goddard Space Flight Center in Greenbelt, Maryland, manages the telescope. The Space Telescope Science Institute (STScI) in Baltimore conducts Hubble science operations. STScI is operated for NASA by the Association of Universities for Research in Astronomy, Inc., in Washington. NASA is exploring our solar system and beyond to understand the universe and our place in it. We seek to unravel the secrets of our universe, its origins and evolution, and search for life among the stars. Today's announcement shares the discovery of our ever-changing cosmos, and brings us closer to learning whether we are alone in the universe. Kepler Discovers 1000th Exoplanet Jan. 6, 2015: How many stars like our sun host planets like our Earth? NASA's Kepler Space Telescope continuously monitored more than 150,000 stars beyond our solar system, and to date has offered scientists an assortment of more than 4,000 candidate planets for further study -- the 1,000th of which was recently verified. Using Kepler data, scientists reached this millenary milestone after validating that eight more candidates spotted by the planet-hunting telescope are, in fact, planets. The Kepler team also has added another 554 candidates to the roll of potential planets, six of which are near-Earth-size and orbit in the habitable zone of stars similar to our sun. NASA Kepler's Hall of Fame: Of the more than 1,000 verified planets found by NASA's Kepler Space Telescope, eight are less than twice Earth-size and in their stars' habitable zone. All eight orbit stars cooler and smaller than our sun. Three of the newly-validated planets are located in their distant suns' habitable zone, the range of distances from the host star where liquid water might exist on the surface of an orbiting planet. Of the three, two are likely made of rock, like Earth. "Each result from the planet-hunting Kepler mission's treasure trove of data takes us another step closer to answering the question of whether we are alone in the Universe," said John Grunsfeld, associate administrator of NASA's Science Mission Directorate at the agency's headquarters in Washington. "The Kepler team and its science community continue to produce impressive results with the data from this venerable explorer." To determine whether a planet is made of rock, water or gas, scientists must know its size and mass. When its mass can’t be directly determined, scientists can infer what the planet is made of based on its size. Two of the newly validated planets, Kepler-438b and Kepler-442b, are less than 1.5 times the diameter of Earth. Kepler-438b, 475 light-years away, is 12 percent bigger than Earth and orbits its star once every 35.2 days. Kepler-442b, 1,100 light-years away, is 33 percent bigger than Earth and orbits its star once every 112 days. Both Kepler-438b and Kepler-442b orbit stars smaller and cooler than our sun, making the habitable zone closer to their parent star, in the direction of the constellation Lyra. The research paper reporting this finding has been accepted for publication in The Astrophysical Journal. "With each new discovery of these small, possibly rocky worlds, our confidence strengthens in the determination of the true frequency of planets like Earth," said co-author Doug Caldwell, SETI Institute Kepler scientist at NASA's Ames Research Center at Moffett Field, California. "The day is on the horizon when we'll know how common temperate, rocky planets like Earth are." With the detection of 554 more planet candidates from Kepler observations conducted May 2009 to April 2013, the Kepler team has raised the candidate count to 4,175. Eight of these new candidates are between one to two times the size of Earth, and orbit in their sun's habitable zone. Of these eight, six orbit stars that are similar to our sun in size and temperature. All candidates require follow-up observations and analysis to verify they are actual planets. "Kepler collected data for four years -- long enough that we can now tease out the Earth-size candidates in one Earth-year orbits", said Fergal Mullally, SETI Institute Kepler scientist at Ames who led the analysis of a new candidate catalog. "We're closer than we've ever been to finding Earth twins around other sun-like stars. These are the planets we're looking for". These findings also have been submitted for publication in The Astrophysical Journal Supplement. Work is underway to translate these recent discoveries into estimates of how often rocky planets appear in the habitable zones of stars like our sun, a key step toward NASA's goal of understanding our place in the universe. Credits: Production editor: [Dr. Tony Phillips](http://www.nasa.gov/kepler) \| Credit: [Science@NASA](http://www.nasa.gov/kepler) More information: Ames is responsible for Kepler's mission operations, ground system development and science data analysis. NASA's Jet Propulsion Laboratory in Pasadena, California, managed Kepler mission development. Ball Aerospace & Technologies Corp. in Boulder, Colorado, developed the Kepler flight system and supports mission operations with the Laboratory for Atmospheric and Space Physics at the University of Colorado in Boulder. The Space Telescope Science Institute in Baltimore archives, hosts and distributes Kepler science data. Kepler is NASA's 10th Discovery Mission and was funded by the agency's Science Mission Directorate in Washington. For more information about the Kepler mission, visit: [http://www.nasa.gov/kepler](http://www.nasa.gov/kepler) San Francisco Amateur Astronomers Application for New or Renewing Membership 1. Memberships, with dues payment, are for one year and standard renewal dates of 1 July to 30 June and 1 January to 31 December. 2. Submitting appropriate dues in April, May, June, July, August or September, membership will run to 30 June of the next year. 3. Submitting appropriate dues in October, November, December, January, February or March, membership will run to 31 December of the next year; submitting appropriate dues in January, February or March, membership will run to 31 December of the same year. 4. Renewals are maintained at the original membership date unless the renewal is made later than the original cutoff date (e.g., September or March as described in 3). In such cases the membership date is shifted to the next renewal date 30 June or 31 December. 5. New or renewal memberships sent in via USPS mail will have membership start date based on postmark date. This application is for: ☐ New ☐ Renewing Name: ____________________________________________________________ Address: _________________________________________________________ _______________________________________________________________ Email: ___________________________________________________________ Home Telephone (optional): _________________________________________ Cell Phone (optional): _____________________________________________ Membership Type: ☐ Individual $25.00 / ☐ Family $30.00 / ☐ Student $10.00 / ☐ Supporting $75.00 ☐ Please mail to me a Mt. Tamalpais Parking Permit To complete the membership process: A. Print and fill out this form B. Make check or money order payable to San Francisco Amateur Astronomers C. Mail this form and payment to: Treasurer, SFAA PO Box 15097 San Francisco, CA 94115 New members will be entered onto the SFAA roster on the Night Sky Network (NSN) and will receive a verifying email from the NSN with username and password for the NSN. Renewing members will have their information updated but will not receive an email from the NSN. Both new and renewing members will receive a verifying email from the SFAA Treasurer upon completion of the membership process.
Treatment of Non-Repeatable Residual Delay The present form of the specification (Section 5.6.2 attached for convenience) seems to state that some contributions to the path errors, specifically the “slowly varying ones” can be doubly differenced before inclusion in the error budgets. By “doubly differenced” I mean that the contribution to be counted is only the change over some time $\Delta t$ (given as 3 minutes) in the difference in the path that occurs due to an angular movement $\Delta \theta$ (2 degrees). I believe that this double differencing should not have been permitted by the specifications. My basic argument is as follows: Consider that a particular antenna has a path length residual, $p(t,\theta)$, which is a function of time, $t$, and some generic angular parameter, $\theta$. (I’ll outline the proper expressions later.) The reason that $p$ is described as a residual is that it is the deviation of the path length from that for the nominal antenna, e.g. one with no deformations due to temperature, wind, etc. Note that we do not take any account here of the fact that we are really only concerned with differences between antennas: the path errors are supposed to be considered on each antenna individually. (I shall return to this point later.) Now suppose that we are doing some form of switching and that this is done so rapidly that the change in $p$ with time over a switching cycle is negligible. The relevant path error is therefore $$\Delta p_\theta = \left(\frac{dp}{d\theta}\right) \Delta \theta.$$ Obviously this shows up as a phase error – effectively we measure the wrong distance between our calibrator and the object of interest. If $p$ were completely constant with time, then this term would only change as a result of the changing geometry as the source moves across the sky – that is to say that $\Delta p_\theta$ is just determined by the details of the dependence on the elevation and azimuth angles. In general this would look the same as an error in the baseline. It would therefore be taken out as one of the parameters fitted in finding the baseline. Strictly speaking, however, this would only apply to errors that were constant right through from the time of the baseline solution until the time of the observation. If this term $\Delta p_\theta$ does in fact vary with time, then clearly all measurements will be effected. This is obvious in the case of a position measurement – we get a error of order $\Delta p_\theta / \text{baseline}$ – and presumably variations on timescales of an hour or two would show up as spurious structure on synthesis maps in the way that phase errors usually do. I see no place here for saying that we are only interested in variations of this term on periods of less than three minutes. It seems to me that the time period of three minutes is intended to represent the time for which we may wish to integrate coherently on a source without moving to a reference source. It is true that for many observations we will be switching more frequently than that, but we certainly do not want to be constrained to do that all of the time. Recall also that these specifications are intended to cover the ACA as well as the antennas for the 64-element array, and that the ACA will not usually be used in a fast switching mode. Any changes in the path delay will cause a loss of coherence. The figure of 15 microns is comparable to the specifications in other areas (water vapour radiometer, local oscillator stability, etc.) The term to be included here is presumably \[ \Delta p_t = (\frac{dp}{dt}) \Delta t . \] This will at least give us the first order effect on the change in path for a slowly varying quantity. This would appear directly as a phase error that accumulates over the observing period $ \Delta t $. Again, I see no argument for saying that we can difference this term with respect to angle. Instead it seems to me consistent with our general policy that both these terms, $ \Delta p_\theta $ and $ \Delta p_t $, should appear in our error budget. They are in some sense orthogonal (see below) so the full residual error should be the RSS of these two terms, plus of course the ones due to wind, etc., to which these differencing arguments do not apply. Various mitigating arguments can be made. For example, in the case of time variation, one could say that it is actually only the deviation from the mean path error during the three-minute period that matters, or even that one can interpolate linearly between the phase calibration measurements at either end of the period so only the second and higher time derivatives are important. Similarly, for the angular dependence one could use several calibration sources around the object and fit a function for $ \Delta p_\theta $. In general, however, I think that our policy is to make the antennas good enough that we do not have to do these complicated things. Obviously we can’t consider changing the specification at the moment, but if others agree with this analysis we will have to fix it up at the earliest opportunity. More details: For completeness I have run through the derivation of the terms, replacing the generic angle $ \theta $ by the Elevation and Azimuth pointing directions $ E $ and $ A $. For convenience, the path can be broken down into three components: $ p_1 $, the part above the elevation axis; $ p_2 $, the component due to movements of the elevation bearing with respect to the azimuth axis; and $ p_3 $, any movement of the azimuth axis. Now $ p_1 $ will contain things like the overall expansion of the dish, together thermal effects in the part of the cabin which lie between the elevation bearing and the BUS. This term will have no direct angular dependence, so it subtracts out completely as we move from source to reference. Note that there could be an indirect dependence if there were for example some gravitational terms in the dish deformations that were not being taken out correctly. The temperature distribution could also be affected by a change in pointing, but one would not expect this to be a large effect for 2 degrees. Neglecting these finer points, the derivative of $ p_1 $ with respect to the angles $ E $ and $ A $ is zero so it contributes only to $ \Delta p_t $ and not to $ \Delta p_\theta $. For $ p_2 $ we will have a vertical and a horizontal components $ \Delta z $ and $ \Delta h $ due to movements of the elevation axis where $ h $ is the direction along the Azimuth where the antenna is pointing. Note that these would include things like expansion and bending of the yoke but also the effects of tilts in the base and its expansion. Resolving along the line of sight we get: \[ p_2 = \Delta z \sin E + \Delta h \cos E \] and clearly the derivative is \[ \frac{dp_2}{dE} = \Delta z \cos E - \Delta h \sin E, \] with no direct dependence on Azimuth. (The suggestion that the terms $\Delta p_\theta$ and $\Delta p_t$ are orthogonal, in some loose sense, comes just from the fact that the sin and cos terms are interchanged between the two equations above. Obviously this is not supposed to be anything rigorous but it is weak justification for applying the usual RSS treatment of the terms.) Finally we need to consider movements in the base that shift the azimuth axis by amounts $\Delta x$ and $\Delta y$. These will not rotate with the source so they produce a term like $$p_3 = (\Delta x \cos A + \Delta y \sin A) \cos E$$ This does have an azimuth dependence so we get non-zero derivatives with respect to both azimuth and elevation. In practice however these lateral shifts in the base should be extremely small. As already noted the forms of the terms $p_2$ and $p_3$ are the same as baseline errors. Returning briefly to the fact that we are really only interested in the differences in these effects between different designs, it is clear that we are to some extent being conservative in insisting that the antennas meet the requirement individually. On the other hand some effects will not be the same at different antennas across the array. This is certainly true for the wind contributions and one can imagine cases where it is true for some thermal effects – for example where one antenna shadows another or when cloud covers part of the site. I would comment that this is a prime example of where having two different antenna designs, with different levels of thermal response and different time constants, would cause a lot of problems. I would re-emphasize the point made above that the movements of the location of the axis crossing point will look like a baseline error and will only be properly removed by performing a baseline fit. I would also draw attention to the “Repeatable Residual Delay” (5.6.1). This is specified at 20 microns and is again only thought to be of concern over an angular step of 2 degrees. I am concerned that in principle this could allow much larger errors on wider angles and that these would make it extremely difficult to find a proper baseline solution. The implication of the way this is written is that these effects could be measured and stored, perhaps in a look-up table. Is this really what we intend to do? Do the software people know about this? I note also that this is another example of something that would not cancel with non-identical antennas. Richard Hills 9th June 2004 5.6 PATH LENGTH ERROR Path length errors must be considered since the antennas will be used in an array. Path length (also called delay) errors are defined as follows. Consider a plane wave arriving at the antenna from the direction of the boresight. Define the “excess delay” of the antenna to be the difference between the arrival time of that wave at the secondary focus (via the main reflector and subreflector) and its arrival time at an arbitrary reference point fixed with respect to the ground as if the antenna were not present. (It is convenient to choose the reference point along the azimuth axis. If the boresight axis, elevation axis and azimuth axis all intersect, then choosing that intersection as the reference results in an excess delay that is nearly constant.) The excess delay (expressed as a path length) for the nominal antenna in the absence of environmental perturbations, shall be computed as a function of boresight direction over the full range of azimuth and elevation. This shall be defined as the “nominal excess delay function.” Now define the “residual delay” as the difference between the actual excess delay of a particular antenna under existing conditions and the nominal excess delay. The residual delay is limited by the specifications of this section. The residual delay has a repeatable and a non-repeatable component. 5.6.1 REPEATABLE RESIDUAL DELAY The repeatable residual delay is caused by the difference in gravity deformation between an antenna and the nominal antenna (for example, this could be caused by differences in the material properties of an antenna compared to the nominal material properties), axis alignment errors, bearing runout, bearing alignment, and similar errors, which repeat as a function of antenna position and can be corrected using a computer delay model. It applies: The repeatable residual delay for an antenna shall not change by more than 20 micrometers when the antenna moves between any two points 2 degrees apart in the sky. 5.6.2 NON-REPEATABLE RESIDUAL DELAY The non-repeatable residual delay is the delay component that varies with time or is not repeatable as a function of antenna position. It is caused by wind, effects of temperature differences and temperature changes, acceleration forces, bearing non-repeatability and other sources of non-repeatable errors. The Contractor may include metrology equipment in the antenna design that can be used to estimate the residual delay in real time. In that case, both the measured values and the results of a calculation estimating the residual delay shall be provided to ALMA via the digital interface (see Section 220.127.116.11). If this is done, then the estimate shall be subtracted from the actual residual delay for the purpose of meeting the specifications of this section. Further, for slowly varying sources of residual delay, but not for wind induced residual delay, the contribution to the residual delay budget may be limited to the differential residual delay over a solid angle of 2 degrees radius on the sky and then only the change in that differential delay over a 3 minute period when tracking at the sidereal rate. The non-repeatable residual delay under Primary Operating Conditions (section 4.4.3) must be less than 15 micrometers RSS when tracking an astronomical source at sidereal rate. 5.6.3 COMPUTATION OF THE NON-REPEATABLE RESIDUAL DELAY The non-repeatable residual delay shall be computed both for nighttime and daytime conditions by use of error budgets. a) Computation of the wind induced delay: Calculate the quasistatic delay error for each of the eleven wind directions shown in Table 18.104.22.168-a, with the average wind speed defined for the Primary Operating conditions. Compute the weighted RMS of these eleven wind directions with the weighting factors defined in the table to obtain the steady state component of the wind. The gust component of the wind can be computed by scaling the quasistatic values obtained for the steady state. b) Computation of the thermal effects (daytime) For daytime Primary Operating Conditions a computer thermal model of the antenna shall be used to determine the worst case non-repeatable delay error due to temperature differences and temperature changes in the structure. The structural temperature differences shall be calculated using a computer thermal model of the antenna, assuming a wind speed of < 5 m/sec. c) Nighttime residual delay The nighttime residual delay shall be the quadratic sum (RSS) of all the individual sources of delay including the steady state and gust wind contribution according to the Primary Operating Conditions of Section 4.4.3. No thermal effect contribution shall be included in the nighttime error budget. The effect of the steady state and gust wind shall be computed according to the methodology and with the weighting factors of Section 22.214.171.124. d) Daytime residual delay The daytime pointing residual delay error shall be the quadratic sum (RSS) of all the individual sources of delay error including the steady state and gust wind contribution according to the Primary Operating Conditions of Section 4.4.3 and the thermal effects contribution. The effect of the steady state and gust wind shall be computed according to the methodology and with the weighting factors of Section 126.96.36.199. The daytime thermal residual delay is defined to be 75 percent of the thermal effects on delay computed for the worst case under point b) above.
Minutes of the Stansbury Progress Association Inc. Executive Committee Meeting held in the Stansbury Institute on Monday 12th March at 7.30pm. PRESENT: Don McPhee (Chairperson), Shaun Toft (Vice Chairperson), Ted Buttfield, Sue Oldfield, Beth Faulkner, Graham Klose, Doug Carruthers, Andrew Pisani, Cr Trevor Clerke, Michelle Hoffmann and Leon Cartledge (Park Managers), Carley Crouch (minute taker). MEETING OPENED: 7.30pm 1. WELCOME: Don McPhee welcomed everyone to the meeting. 2. CONFLICT OF INTEREST: The Chairperson reminded all members of their requirement to disclose any conflict of interest, and provide details of the relevant interest to the Committee prior to consideration of that item on the Agenda. 3. APOLOGIES: Kevin Forrest, Julie Harris 4. CARAVAN PARK REPORT: 5.1 Caravan Park Manager’s Report: A written report was tabled. Leon and Michelle reported the following: - The new Speedqueen washing machine is having some issues, Leon is sorting these out. The two new Speedqueen dryers need to be hardwired. James Moss will carry out this work on Wednesday morning. One of the dryer stacks will be taken over to OPD to replace the old machines which are no longer working. - A return guest on site #88, Foreshore Park, reversed into the powerhead which had the water pipe next to it. The water pipe has been snapped below the isolation valve. A temporary fix has been carried out, however, the powerhead will need to be moved further away so the water pipe can be properly repaired. - The power outage over the weekend threw out all the boom gate passcodes. The boom gates are currently up, and will be put down once the weekend visitors have left, and the codes can be reset. An option may be to install a Tesla power wall which can act as a back up power supply during outages. - Having difficulty getting in contact with G & J East, Kadina to arrange them to come back and finish the install of the tractor door. They did not have all the parts when they came last time, and need to return to finish the job. - The mesh on the roof of the Foreshore amenities block has been fixed today by Andrew Johnson. Corey, Yorke Solar, will attend in the next couple of weeks to clean out under the solar panels, and put a guard around them to stop the pigeons getting under the panels. There are a few pigeons inside the roof space which did not fly out before the mesh was fixed. The door to the void will be opened Wednesday morning to allow them to get out. Rapid Detail will assess the void in between the blocks, and advise if they are able to clean up the mess made by the pigeons. - Overall the long weekend went smoothly. However, a complaint was received regarding the unpowered sites at OPD. A group of repeat guests took up a large area, more than the allocated site allowance. The maximum number of guests per site is 6, with one vehicle. If there is a second vehicle it is to be parked outside of the park. Discussion took place regarding whether sites can be marked out ahead of the busy periods. Further consideration is required as to the best way to go about this. - Peninsula Home Improvements came out to install the blinds at the Foreshore camp kitchen. However one was the wrong size. They are yet to return to install the final blind. - A quote (QT01600) was received from Peninsula Home Improvements for the installation of three PVC blinds on the decks of the new villas to help stop the wind. They would go from the fascia to the top of the balustrade. Cost for all three is $8,850.00 Moved: Ted Buttfield Seconded: Doug Carruthers To obtain a second comparison quote for the villa blinds. Carried: #1 12/03/2024 - The second reception computer failed last Wednesday (SSD died). This has been sent over to Adelaide for repair. The computer which was set up for Leon to use, has been moved to the front desk in the interim. - An old laptop which was used for the OPD boom gate software many years ago, has been located. Leon is testing to see if he can get it to work. - The quote for the surveillance cameras has not yet been received. - Peter Hancock is still working on an idea about how the two sets of boom gates “talk to each other”. He is also still waiting on the main central unit part, which is required to complete the OPD wifi upgrade. - Stansbury were the top performing park in the SA Parks “Road to a Million Campaign” for February. The prize is some additional promotion via SA Parks. - Facebook posts do not have a lot of reach, in regards to marketing and improving occupancy. Gday Parks have been contacted to see if they are able to promote the park, however they generally only promote parks they own. - Will try using Campermate for last minute deals to fill vacancies. Will also investigate whether it’s possible to do any advertising on Wikicamps, as this is popular with our demographic. - It was noted that Travel Auctions have been tried in the past, but it is difficult to make money from this model of advertising. - Currently there is a large group of guests (15 sites) in the park who are part of an Island Star caravan group. - The tree near site 33 looks dead, and will need to be removed. - Common feedback about the Foreshore Park from interstate visitors booking online, is that the sites are small. It is noted that the park was developed at a time when vans and cars were a lot smaller. Now everyone has very large, long vans, and large cars and boats which take up all the space. Many parks will experience the same issues. Unfortunately, unless sites are merged together, there is no way to make the sites larger. Office staff will try and manage the online bookings and move them to larger sites if possible. - A septic smell has been reported in the OPD park. Leon has contacted Minlaton Septic to book them in to come and assess/empty the septic. - A snake was sited on site 18, but not reported to the office until the next day. - Rat and mice bait stations have been set along the foreshore. It was noted that vegetation along the foreshore cannot be removed. - Some of the trees along the foreshore nature strip look like they have been poisoned. Don will follow up with Council. Moved: Graham Klose Seconded: Beth Faulkner To accept the Caravan Park Manager’s Report for February. Carried: #2 12/03/2024 Michelle and Leon left at 8.31pm and did not return. 5. ACCEPTANCE OF MINUTES: 5.1 Acceptance of the Minutes Re matter 6.1, Music Festival 2024 in the 12th February minutes: Sue Oldfield requested an amendment regarding the proposed music event. Minutes to read that “the Stansbury Sports and Community Club suggested moving the music event from June to the October long weekend”. Don McPhee advised that the original idea of having an event in June, was to bring people into the town during the quieter winter months, whereas the October long weekend already has a high number of visitors in the town. Andrew Pisani reported that the Yacht Squadron in Adelaide have approached him as to whether Stansbury would be interested in hosting them over the October long weekend. In the past they have stayed at Port Vincent, however due to various unresolved issues (the shags, cleaning, infrastructure), they are now considering coming to Stansbury instead. There would be approximately 120 yachts and require a marquee, dinner and music to be provided. Depending on weather, there is anchorage in the bay, however it is not enough for 120 yachts. This event would potentially occur every two years on the opposite year to the Paskeville Field Days. It was noted that there are a lot of logistics to work through in order to hold an event of this nature, and that the Progress volunteers are already involved in other town activities over the October long weekend. With last year’s 150th event held on the October long weekend, Progress resolved that a sub committee of community members was required to co-ordinate the event which streamlined the administration process and accounting for expenditure. If town members are prepared to support the event by being involved in a subcommittee tasked with co-ordinating the event, then Progress would support the idea. Andrew Pisani advised he believes there are possibly 12 people, including two from the yacht squadron, and some from the Stansbury Sports and Community Club, who are prepared to be involved in a committee to organise the event. A representative from Progress would need to be involved, and assist with liaising between the sub committee and Progress. Leon Cartledge is assisting with a music budget. It is recognised that more members from the community need to be involved in arranging local events which draw visitors to the town. It was confirmed that the smaller event that was proposed for the June long weekend has been cancelled. Moved: Andrew Pisani Seconded: Beth Faulkner To form a sub committee tasked with investigating the logistics of holding an event for the Yacht Squadron over the October long weekend. Carried: #3 12/03/2024 Moved: Shaun Toft Seconded: Sue Oldfield That the minutes of the Stansbury Progress Association Inc. Executive Committee meeting held on Monday 12th February 2024 be accepted as read and confirmed as true records with the amendment to item 6.1 as noted. Carried: #4 12/03/2024 6. MATTERS ARISING FROM MINUTES: 6.1 Pigeons in the foreshore amenities block: As per item 5.1 6.2 Foreshore Park front office update: Andrew Davidson, AD Design, is unable to assist with drawings at this time due to other commitments. However, he has suggested investigating the option of installing a transportable building which is fit for purpose. There was concern that a different style of building (transportable) next to the existing brick house/office could be unattractive. It was queried whether building a second story on the existing house was an option. However, the existing house/office building is owned by Council, whereas a transportable building would belong to Progress. The opportunity of purchasing a stand alone house outside of the park as a manager’s residence, has passed. At this stage, information is being gathered for consideration by the committee. 6.3 AED donation from ADBRI update: Daniel Baille, ADBRI, has confirmed that the three AED machines and cabinets have arrived. They are waiting on the plaques which will be attached to the cabinet stating they were donated to the community by ADBRI. Once these are ready, a handover to Progress will be scheduled. The ongoing maintenance and replacement of parts will be the responsibility of the recipient of the machines. Stansbury Progress Association were tasked with the decision of where to install the machines, and it was unanimously decided to install a machine at each of the following locations: - The Stansbury Institute - The Stansbury Visitor Information Centre - The Stansbury boat ramp (subject to Council approval). This machine will also require a structure to protect it from the weather. The Progress Association will be responsible for the ongoing maintenance of all three machines. It was noted that the AED cabinet installed at the Stansbury Pharmacy has deteriorated in the weather, as has the box that the AED is housed in. Sue Oldfield advised that it is her understanding that St Johns will replace used AED pads at no charge. A letter from Stansbury Bowling Club, 20th February, 2024, was tabled. The new legislation (Automated External Defibrillators (Public Access) Bill 2022) doesn’t currently adequately address all aspects regarding the installation and upkeep of AEDs. Until the details are finalised, it is difficult to develop a policy around AEDs. Moved: Ted Buttfield Seconded: Graham Klose That due to the new legislation requiring further detail, it is difficult to develop a policy in regard to AEDs. Therefore no further action will be taken at this time, re taking on the responsibility of additional AEDs in the town. Carried: #5 12/03/2024 7. BUSINESS MANAGER AND FINANCE REPORT: 7.1 A written report for February was tabled. Moved: Graham Klose Seconded: Shaun Toft To increase the Visitor Information Officer’s hours by six hours per week until 12th May 2024. Carried: #6 12/03/2024 8. MARKET CO-ORDINATOR/ADMINISTRATION REPORT: 8.1 A written report for February was tabled. It was noted that the 9th March, Stansbury Seaside Market was cancelled due to the forecasted weather of 27 knott North/North Easterly winds and very high temperatures. These conditions would have been terrible for stallholders, and likely resulted in stallholders cancelling on the day. The forecast weather would likely have also impacted visitor numbers. In actuality, the wind came in later than predicated, however notice needed to be given to stallholders in advance, particularly those with fresh produce, so they could make other arrangements. Holding an unpleasant event would likely have adversely effected goodwill, and been counter productive. Safety and risk of damage to stock and equipment, also had to be taken into account. The decision was made with sufficient notice, and based on the best information available at the time. Due to the cancellation of the 9th March market, the Stansbury Museum were unable to hold their annual raffle. It was agreed to offer the Museum the option of holding the raffle on the 13th April, at the Salt Water Classic market. Ted will offer this option at the next Museum meeting on Friday 15th March. Moved: Andrew Pisani Seconded: Sue Oldfield To accept the Business Manager and Administration reports February. Carried: #7 12/03/2024 9. SUB COMMITTEE REPORTS: 9.1 Institute: No meeting has been held. Nil to report. 9.2 Tidy Towns: Don McPhee reported the following: - The remaining feral vegetation on the corner of West Terrace and St Vincent Street has been removed. The vegetation was impacting visibility at the corner. - Some vegetation had to be cut off a ground level and covered with dirt, due to fibre optic cable being run underneath them. - A lot of watering, weeding and gardening has been carried out around town. 9.3 Town Projects: Nil to report. 9.4 Access Advisory Committee: Nil to report. Moved: Graham Klose Seconded: Ted Buttfield To accept the sub committee reports for February. Carried: #8 12/03/2024 10. CORRESPONDENCE: 10.1 Letter – Stansbury Bowling Club, tabled. (refer item 6.3) 10.2 Letter – Southern Yorke Peninsula, Royal Flying Doctor Service, tabled. Moved: Graham Klose Seconded: Sue Oldfield To accept the correspondence for February. Carried: #9 12/03/2024 11. GENERAL BUSINESS: 11.1 New vape and smoking regulations: Trevor Clerke reported that the new vape and smoking regulations commenced from 1st March 2024. All public buildings need to have a sticker displayed at the entrance indicating it is a smoke and vape free area. Stickers are free and can be ordered online from SA Health. 11.2 Change in RSL President Doug Carruthers reported that after many years of leadership, Reg Ricketts has retired as President of the RSL. Mick Strawbridge is the new President. 11.3 New start time for Executive Committee Meetings Shaun Toft proposed an earlier start time for the monthly Executive Committee Meetings. Moved: Shaun Toft Seconded: Graham Klose To start the Executive Committee Meetings at 7.00pm. Carried: #10 12/03/2024 11.4 Elanora grounds Graham Klose reported that the lower limbs of another tree at Elanora have been cut off for firewood. It is unclear as to who is responsible. The grounds in general, are looking very untidy. 12. CLOSURE: Meeting closed at 9.37 pm. Next meeting: Executive Committee Meeting, Monday 8th April, 2024, 7.00pm. Don McPhee Chairperson Date: 8/4/2024 \| 1 \| 2 \| 3 \| 4 \| 5 \| \|---\|---\|---\|---\|---\| \| 6 \| 7 \| 8 \| 9 \| 10 \| \| 11 \| 12 \| 13 \| 14 \| 15 \| \| 16 \| 17 \| 18 \| 19 \| 20 \| \| 21 \| 22 \| 23 \| 24 \| 25 \| \| 26 \| 27 \| 28 \| 29 \| 30 \| \| 31 \| 32 \| 33 \| 34 \| 35 \| \| 36 \| 37 \| 38 \| 39 \| 40 \| \| 41 \| 42 \| 43 \| 44 \| 45 \| \| 46 \| 47 \| 48 \| 49 \| 50 \| \| 51 \| 52 \| 53 \| 54 \| 55 \| \| 56 \| 57 \| 58 \| 59 \| 60 \| \| 61 \| 62 \| 63 \| 64 \| 65 \| \| 66 \| 67 \| 68 \| 69 \| 70 \| \| 71 \| 72 \| 73 \| 74 \| 75 \| \| 76 \| 77 \| 78 \| 79 \| 80 \| \| 81 \| 82 \| 83 \| 84 \| 85 \| \| 86 \| 87 \| 88 \| 89 \| 90 \| \| 91 \| 92 \| 93 \| 94 \| 95 \| \| 96 \| 97 \| 98 \| 99 \| 100 \| --- Figure 1. A schematic diagram of the experimental setup for the study of the effect of the magnetic field on the flow of a nanofluid in a microchannel.
The Black Hole Idealization Paradox Dominic Ryder\textsuperscript{1} \textsuperscript{1}Department of Philosophy, Logic and Scientific Method, London School of Economics. firstname.lastname@example.org April 15, 2024 Abstract Stephen Hawking’s derivation of Hawking radiation relied on one particular spacetime model, that of a star collapsing into a black hole which then remains in existence forever. He then argued that Hawking radiation implies this model should be thrown away in favour of a different model, that of an evaporating black hole. This aspect of Hawking’s argument is an example of an idealization that is pervasive in the literature on black hole thermodynamics, but which has not yet been widely discussed by philosophers. The aim of this paper is to clarify the nature of Hawking’s idealization, and to show a sense in which it leads to a paradox. After identifying this idealization paradox in classic derivations of Hawking radiation, I go on to show how various research programmes in black hole thermodynamics can be viewed as possible resolutions to the paradox. I give an initial analysis of the prospects for success of these various resolutions, and show how they shed light on both the philosophical foundations of both Hawking radiation on the nature of idealizations in physics. 1 Introduction “He must, so to speak, throw away the ladder after he has climbed up it.” – Ludwig Wittgenstein, Tractatus Logico-Philosophicus Derivations of Hawking radiation are cornerstones of modern physics. The consensus view is that Hawking radiation leads to the black hole information paradox, and huge amounts of work in physics has been dedicated to understanding and resolving it (Page, 1994; Raju, 2022). Philosophers have also analysed various aspects of Hawking radiation, including: the black hole information paradox (Belot, Earman, & Ruetsche, 1999; Maudlin, 2017; Manchak & Weatherall, 2018; Wallace, 2020); black hole thermodynamics (Dougherty & Callender, 2016; Wüthrich, 2019; Wallace, 2018, 2019; Prunkl & Timpson, 2019); and the universality of Hawking radiation (Gryb et al., 2019). In this paper I will argue that there is another problem, distinct from those listed above, that arises because a seemingly essential idealization is used in three mainstream derivations of Hawking radiation: Hawking’s original derivation (1975)$^1$, Fredenhagen and Haag’s (1990) “watertight” derivation, and algebraic approaches such as Dimock and Kay (1987) and Dappiaggi et al. (2011). This paper establishes the paradox for these derivations, categorises its possible resolutions, and offers an initial analysis of the success of various resolutions. The resolution of this problem, which I call the idealization paradox, can teach us about the kinds of idealizations used in science, how global spacetime structure encodes local spacetime structure, and the nature of Hawking radiation. The paradox arises out of an argument of Hawking (1975), who derived the eponymous radiation in a spacetime which represents a star that collapses into a black hole which, once formed, is unchanging and exists for the rest of time. I will call this spacetime collapse-Schwarzschild. In the same paper, Hawking also presented the first arguments that the backreaction of the radiation on the spacetime will lead to a negative energy flux into the black hole, thus causing the black hole to lose mass and evaporate. Given that a black hole evaporates, Hawking reasoned, it is not well represented by collapse-Schwarzschild. Instead, we should represent the black hole using a spacetime that models an evaporating black hole. I will this spacetime evaporation-Schwarzschild. In other words, the use of collapse-Schwarzschild in the derivation of Hawking radiation is an idealization. Significantly, evaporation-Schwarzschild does not exhibit the same properties as collapse-Schwarzschild, and as I show in section 3.2, Hawking’s derivation cannot be carried out in evaporation-Schwarzschild. We throw away the spacetime we were using as a ladder to Hawking radiation, collapse-Schwarzschild, in favour of evaporation-Schwarzschild, but the original derivation is not consistent with our new spacetime. The thesis of this paper is that, according to Hawking’s and other mainstream derivations, Hawking radiation is inconsistent with black hole evaporation. It is possible to state a sketch of the paradox (the details of which I will complete in the next sections) for a derivation of Hawking radiation based $^1$And a fortiori Wald (1975), as this is just a more mathematically rigorous version of Hawking’s derivation. upon a certain set of properties $X$: The Idealization Paradox 1. (Hawking Radiation Derivation) If spacetime exhibits the set of properties $X$, then Hawking radiation occurs. 2. (Backreaction Arguments) If Hawking radiation occurs, then black hole evaporation occurs. 3. (Inconsistency Claim) If black hole evaporation occurs, then spacetime does not exhibit the set of properties $X$. 4. (Spacetime Postulate) Spacetime exhibits the set of properties $X$. This set of premises is inconsistent. What justifies the first premise? Of Hawking’s original calculations, Unruh (2014) writes they are “mathematically unimpeachable”, and the other derivations discussed in this paper only improve upon the degree of mathematical rigour. Thus, the secure mathematical status of the derivations in question means the first premise is hard to challenge.$^2$ What about the second premise? Using global definitions of energy one can derive a positive energy flux out toward infinity in the black hole spacetimes, as I discuss further in section 2.2. Hence, assuming global energy conservation, one recovers a negative energy flux into the black hole, which is strong motivation for black hole evaporation. The third premise, Inconsistency Claim, is defended in the bulk of this paper. So what about the fourth premise? It is hard to reject Spacetime Postulate, because then we can’t use the derivation of the first premise to derive Hawking radiation. So without the fourth premise, we lose our motivation for believing in Hawking radiation. Hence, according to the derivation used in the first premise, Hawking radiation is inconsistent with evaporation. Notice that Hawking Radiation Derivation makes reference to a particular derivation.$^3$ I call derivations to which the idealization paradox applies evaporation-inconsistent derivations, and conversely those to which it does not evaporation-consistent derivations. Thus, I argue Hawking (1975), Fredenhagen and Haag (1990) and algebraic approaches such as Dimock and Kay (1987), Dappiaggi et al. (2011) are evaporation-inconsistent. Due to the possibility of evaporation-consistent derivations (and resolutions to the paradox for evaporation-inconsistent derivations), the idealization paradox does not imply we ought to be skeptical about the existence of Hawking radiation or black hole evaporation. To protect these phenomena from the paradox, one may claim that there exists a derivation of Hawking radiation that uses physically reasonable properties and is evaporation-consistent. Call this existence claim the consistency conjecture. The ‘physically reasonable’ qualification is necessary because a physically implausible evaporation-consistent derivation (e.g. a derivation in a two-dimensional spacetime) should not al- $^2$Unruh also calls the calculations “nonsense physically” due to the trans-Planckian problem, recently discussed by philosophers Gryb et al. (2019). However, I ignore the trans-Planckian problem for the purpose of this paper. $^3$More precisely, a particular set of properties assumed in a derivation. If two derivations of Hawking radiation assume the exact same properties, then we consider them equivalent for the purpose of this paper. leviate our concerns. I expect the consistency conjecture is true.\footnote{In the literature there are derivations which are plausible candidates for evaporation-consistency (e.g. Visser (2003); Parikh and Wilczek (2000)). I discuss these in section 6, but a full analysis requires another paper, which the author intends to undertake in the future of the project. See Curiel (2023) for an overview of the plethora of Hawking radiation derivations.} However, I will argue that even if the consistency conjecture is true, the idealization paradox is still paradoxical and must be resolved; the paradox identifies a mystery about how and why evaporation-inconsistent derivations were successful. Some philosophers, such as Batterman (2002, 2005, 2011, 2017) and Morrison (2012), have argued that idealizations (construed as false descriptions) are essential for our scientific theories and models to represent and explain reality: there is “something deeply correct about the ‘unrealistic’ idealization” (Batterman, 2005, p. 237). Conversely, many have defended the view that idealizations are dispensable (for example, Norton (2012); Butterfield (2011); Menon and Callender (2013) and Palacios (2019, 2022)). This attitude is captured in what Jones (2006)\footnote{See also Landsman (2013); Fletcher (2020).} has called \textit{Earman’s principle}: “no effect can be counted as a genuine physical effect if it disappears when the idealizations are removed” (Earman, 2004, p. 191).\footnote{A widely discussed example in this literature is the unrealistic use of infinite limits in statistical mechanics to recover singularities in the thermodynamic theory of phase transitions. For a topical introduction to the debate and bibliography see Sheeh (2018, 2023). Fletcher et al. (2019); Sheeh (2018, 2023) catalogue some of the philosophical issues that arise from the use of idealizations in physics; and see Frigg and Hartmann (2020); Potochnik (2017) and Frigg (2022, chapter 11) for general overviews on idealization in science.} Applying Earman’s principle to Hawking radiation, dispensabilists will presumably demand that the collapse-Schwarzschild idealization must be removed.\footnote{Earman’s principle has also been applied to Hawking radiation by Gryb et al. (2019), in which the authors note that the response to the trans-Planckian problem which models Hawking radiation as Goldstone bosons has only been carried out in stationary spacetimes, and it is an important open question whether these models can be deidealized.} This would have the further benefit of helping to explain why new derivations of Hawking radiation continue to be produced, despite Hawking’s original derivation being widely viewed as successfully establishing the phenomenon. However, as we shall soon see, deidealizing Hawking’s argument is not conceptually straightforward, lending some initial plausibility to the essentialist claim. Nonetheless, a more careful look at recent research programmes in the foundations of Hawking radiation also reveals several distinct options for the dispensabilist. My aim in this paper will be to establish the paradox for the three derivations and then categorise possible dispensabilist responses to the paradox, each of which seeks to deidealize the derivations. The plausible resolutions are associated with prominent research programmes in the foundations of Hawking radiation, including an appeal to quantum gravity, the approximation regime proposed by Hawking (1975), and what I call “essential structure” derivations. I give an initial analysis of these approaches and find their prospects of success vary significantly. In particular Hawking’s approximation regime fails for his own derivation, but essential structure derivations represent a very promising possible resolution. The lessons of the paradox vary across possible resolutions, but initial hints suggest insight into: the nature of Hawking radiation, how global physical properties encode local physical properties, and what sort of idealizations are used in our best physical theories. In section 2 I introduce the theory of black holes and quantum field theory on black hole spacetimes that will be required. In section 3 I show that the idealization paradox applies to Hawking’s derivation, and in sections 4 and 5 I show that the idealization paradox bites for Fredenhagen and Haag’s derivation and algebraic approaches respectively. Finally, in section 6 I categorise and analyse resolutions to the paradox. 2 Primer on Quantum Field Theory in Black Hole Spacetimes This section reviews the background material important to the claim that the derivations I analyse are evaporation-inconsistent. I begin with black hole physics treated from the global perspective (Hawking and Ellis (1973) and Wald (1984)). I then introduce quantum field theory on curved spacetimes which underwrites the particle concept in Hawking radiation, before sketching the black hole evaporation heuristic. Readers familiar with quantum field theory on curved spacetime may wish to skip to section 3. 2.1 Black Hole Spacetimes and Conformal Diagrams For my purposes, a black hole spacetime is one that is asymptotically flat at past and future null infinity $(\mathcal{I}^\pm)$ and for which there is a region of the spacetime causally isolated from the rest of the spacetime for all time $(J^-(\mathcal{I}^+) \neq M).$$^8$ The black hole region is the causally isolated region $(\mathcal{B} = M - J^-(\mathcal{I}^+))$, and the event horizon bounds this region $(\mathcal{H}^E = \partial \mathcal{B})$. There are in fact multiple inequivalent ways to define a black hole (Curiel, 2019), but the one given here is standard in the formal and philosophical foundations of general relativity. Birkhoff’s theorem states that any solution of Einstein’s vacuum $(R_{ab} = T_{ab} = 0)$ equations which is spherically symmetric$^9$ in an open set $V$, is isometric in $V$ to part of the inextendible Schwarzschild solution (Hawking & Ellis, 1973). This solution describes an uncharged, non-rotating black hole of mass $m$ with the event horizon at the Schwarzschild radius, $r = 2m$. A spacetime is said to be \textit{inextendible} if there does not exist a ‘larger’ spacetime into which there is a proper isometric embedding. A spacetime is \textit{stationary} if it admits a global timelike Killing vector field.$^{10}$ Roughly, a stationary $^8$Heuristically, a spacetime is asymptotically flat at $\mathcal{I}^\pm$ iff it is approximately Minkowski at infinity, approaches Minkowski smoothly, and is complete. For a formal definition see Wald (1984, chapter 11.1), and see Landsman (2021, sec. 10.3) for in depth discussion of these conditions. $^9$Admits the group $SO(3)$ as a group of isometries, with the group orbits spacelike two-surfaces. $^{10}$A Killing vector field is a vector field whose flow is a one-parameter group of isometries $\phi_t : \phi_t : M \rightarrow M$ such that $\phi_t^* g = g$. spacetime does not change if one follows the integral curves of the Killing vector field. Schwarzschild spacetime is stationary. However, physical black holes are formed by some astrophysical process, such as stellar collapse, so the physical spacetime will be neither stationary nor vacuum. Therefore, for a more realistic representation, we analyse a spacetime that includes spherically symmetric, non-rotating and uncharged matter that collapses into a black hole. The collapse-Schwarzschild conformal diagram is the resulting model, depicted in figure 1. This diagram will be important, so I identify some of its distinctive features. Outside the matter the spacetime is isometric to Schwarzschild by Birkhoff’s theorem, and thus is stationary. Inside the matter the metric will be complicated and non-stationary. The spacetime is globally hyperbolic, meaning that it admits a Cauchy surface and thus a well-posed initial value description.\footnote{A Cauchy surface is one such that every causal curve (without an endpoint) intersects it exactly once. Therefore, heuristically, a Cauchy surface registers some information about every point in spacetime, and a globally hyperbolic spacetime is causally well-behaved. See also Geroch (1970).} One such Cauchy surface is denoted $\Sigma$ in figure 1. Given any foliation into Cauchy surfaces, once one surface has intersected the event horizon all subsequent surfaces also will. Thus, these models describe black holes which exist forever after their formation. The final version of a Schwarzschild black hole to consider is evaporation-Schwarzschild. However, first we need Hawking radiation. ### 2.2 Quantum Fields in Black Hole Spacetimes We now turn to how particles are defined in quantum field theory, and how non-stationary spacetimes lead to particle creation. This is the core of Hawking radiation according to the mainstream view. The idea is to quantise a classical field theory by defining a Hilbert space, $\mathcal{H}$, with respect to a time translation symmetry, giving a particle interpretation of the field. Because time translation symmetries are generally local in curved spacetimes, the particle interpretation is generally different in different regions, and this leads to particle creation. In more detail, one begins by modelling a massless complex-valued scalar field, $\Phi$, obeying the covariant wave-equation: $g_{ab} \nabla^a \nabla^b \Phi = 0$. We can now take any of a variety of paths to define a quantum field theory, but roughly one defines a Hilbert space by selecting a subset of the solutions to the covariant wave-equation to represent physical solutions.\footnote{For example, following Wald (1995, p.38), first define a state space for a quantum theory called a “one-particle structure”. The covariant wave equation admits a symplectic vector space of complex-valued solutions, $(\mathcal{S}^C, \Omega)$, where $\Omega$ is the symplectic structure on the space of solutions $\mathcal{S}^C$. Define the Hilbert space, $\mathcal{H}$, representing physical solutions by selecting a subspace of solutions such that: (i) The “inner product” (scare quotes because it is not positive definite on $\mathcal{S}^C$) $\langle v_1, v_2 \rangle = -i \Omega(\bar{v}_1, v_2)$ is positive definite on $\mathcal{H}$, (ii) span$(\mathcal{H}, \mathcal{H}) = \mathcal{S}^C$, and (iii) for all $z_1 \in \mathcal{H}$ and $z_2 \in \mathcal{H}$, $\langle z_1, z_2 \rangle = 0$. Importantly, there will many choices of $\mathcal{H}$ that satisfy these conditions. The Hilbert space of the full QFT will then be $F_S(\mathcal{H})$, the symmetrised Fock space constructed from $\mathcal{H}$.} In stationary spacetimes there is a preferred non-arbitrary way to select this subspace. Namely, we can define a global time coordinate associated with a Killing vector field that characterises time translation symmetry, and choose $\mathcal{H}$ to be the space of positive frequency solutions with respect to this time coordinate (exactly as in Minkowski spacetime for an inertial time coordinate). By non-arbitrarily fixing $\mathcal{H}$, we non-arbitrarily fix a particle interpretation for our QFT.\footnote{Given a positive frequency subspace with respect to a Killing vector field, time translating any state along the Killing vector field will recover a positive frequency state. Thus, the energy of the particle will always be positive when transformed by a time translation symmetry, as we desire for a particle interpretation. See Halvorson and Clifton (2002, pp. 3-4) for a brief discussion.} Thus, there is a preferred, global definition of a particle in stationary spacetimes. However, in general curved spacetimes there will not be a time translation symmetry which we can exploit to define positive frequency solutions. Therefore, there will not exist a non-arbitrary way to define $\mathcal{H}$; so there will be no unique, global definition of a particle. This applies to collapse-Schwarzschild, which is non-stationary. The central idea of Hawking radiation is that the failure of a spacetime to yield a global preferred particle interpretation leads to particle creation. There are local Killing vector fields at past infinity and future infinity but these differ due to non-stationarity in the bulk region of the spacetime. Therefore, given a vacuum state in the past one has particle content in the future. This is the basis of the Hawking (1975) derivation of Hawking radiation. The details are saved for section 3.1, but in summary: we define $\mathcal{H}^\pm$ on $\mathcal{I}^\pm$ and then choose $\Phi$ such that the state is vacuum on $\mathcal{I}^-$; by calculating the unitary operator $U : \mathcal{H}^- \rightarrow \mathcal{H}^+$, we can determine the particle number for $\Phi$ on $\mathcal{I}^+$ with respect to $\mathcal{H}^+$, and one finds that there is particle creation. Specifically, a thermal spectrum of particles is found at $\mathcal{I}^+$. This leads to our next topic, evaporation.\footnote{See Arageorgis, Earman, and Ruetsche (2002) for a challenge to the possibility of formulating unitarily implementable dynamics for quantum field theories on generic, curved spacetimes.} Black hole evaporation cannot be directly inferred from the claim that black holes radiate, because they do not radiate like normal black bodies: no part of Hawking radiation lies in the causal future of the black hole. Instead, evaporation is thought to occur due to the backreaction of the radiation and, without a full theory of quantum gravity, this interaction between the spacetime and the radiation can’t be accounted for in full rigour. One can approximate the backreaction in two ways: either by modelling the radiation as a flux going out to infinity and using a conservation law to infer a flux down over the horizon, or using the semi-classical Einstein equation, $G_{ab} = 8\pi \langle T_{ab} \rangle$. The consensus view is that Hawking radiation implies a black hole loses mass-energy on pain of a “drastic violation of energy conservation” (Fredenhagen & Haag, 1990, pp. 282).\footnote{See Wald (1995, sec. 7.3) for a treatment of the energy flux approach, and Wallace (2018) for a general overview of results in the semi-classical Einstein equation approach.} The above approaches imply the black hole will radiate away all of its mass in finite time. The semi-classical approximation is expected to break down at late times, when the radius of the black hole is of the order of the Planck length. Beyond this point there much disagreement about the description of Hawking radiation and evaporation. However, the consensus has varied very little from Hawking’s original heuristic: “there is not much it can do except disappear altogether.” (Hawking, 1975, pp. 219) Thus, black holes are usually supposed to evaporate entirely, with the spacetime in the region after evaporation isometric to a region of Minkowski spacetime. The conformal diagram for this spacetime, which I call evaporation-Schwarzschild, is depicted in figure 2. Since this spacetime will be central to my discussion, I will highlight a few important features of it. It is very different from collapse-Schwarzschild: the metric in the region exterior to the collapsing matter is not Schwarzschild, it is not globally hyperbolic, it does not admit a timelike Killing vector field and it has a naked singularity, among other things. A consequence of Hawking radiation is that collapse-Schwarzschild is the wrong spacetime to describe the target black hole; it is an idealization. Given Hawking radiation, an uncharged, non-rotating black hole should be described by evaporationSchwarzschild. And yet, Hawking derived the eponymous radiation in collapse-Schwarzschild, in spite of the fact that many properties of collapse-Schwarzschild that are used in Hawking’s derivation do not hold in evaporation-Schwarzschild; this threatens an essential idealization. Thus we arrive at the paradox discussed in the introduction. The rest of this paper defends the claim that throwing away the ladder of collapse-Schwarzschild really leads to inconsistency. 3 Idealization Paradox in Hawking’s Derivation 3.1 Sketch of Hawking’s Derivation To understand exactly what goes wrong for Hawking’s derivation in evaporation-Schwarzschild, we will need a more precise account of it. I give this here, stripped of unnecessary details. In outline, we wish to compare the modes of a quantum field in the distant past with those in the distant future. Consider collapse-Schwarzschild spacetime\textsuperscript{16} containing a massless complex-valued scalar quantum field $\Phi$ (obtained as discussed above). Let $\{f_i\}$ be a complete basis of solutions, so that we may write: $\Phi = \sum_i \{f_i a_i + f_i^\dagger a_i^\dagger\}$, where $a_i$ and $a_i^\dagger$ are the annihilation and creation operators corresponding to the $i$th solution. We choose $\{f_i\}$ to be positive frequency solutions with respect to a time parameter defined by a timelike Killing vector field asymptotically close $\mathcal{I}^-$. We can also describe $\Phi$ as a decomposition into solutions at $\mathcal{I}^+$ and on the event horizon $\mathcal{H}^E$. At $\mathcal{I}^+$ we can again form a Hilbert space generated by positive frequency solutions, $\{p_i\}$, with respect to a time parameter defined by a timelike Killing vector field on $\mathcal{I}^+$. The modes on $\mathcal{H}^E$ play no role in the derivation. $b_i$, $b_i^\dagger$ are the annihilation and creation operators for the $p_i$ modes. Because the spacetime is globally hyperbolic, we can express $\{p_i\}$ and $b_i$ as linear combinations of $\{f_i\}$ and $\{f_i^\dagger\}$ and $a_i$ and $a_i^\dagger$ respectively, $$p_i = \sum_j \{\alpha_{ij} f_j + \beta_{ij} f_j^\dagger\}, \quad b_i = \sum_j \{\bar{\alpha}_{ij} a_j - \bar{\beta}_{ij} a_j^\dagger\}$$ \hspace{1cm} (1) Stipulate that the field is in the state $\|0_-\rangle$ defined as the vacuum state at early times: $a_i \|0_-\rangle = 0$ for all $i$. On $\mathcal{I}^+$, $b_i^\dagger b_i$ has expectation value $$\langle 0_- \| b_i^\dagger b_i \|0_-\rangle = \sum_j \|\beta_{ij}\|^2$$ \hspace{1cm} (2) which will be non-zero because we have different Killing vector fields defining our Hilbert spaces. Thus, to determine the expected number of particles in each mode, one needs to determine the coefficients $\beta_{ij}$. For this calculation, Hawking writes the modes of $\Phi$ in terms of advanced and retarded \textsuperscript{16}Hawking also derived the radiation for charged, rotating black holes, but I focus on the simplest case. Eddington-Finkelstein coordinates: \[ v = t + r_, \quad u = t - r_, \quad r_* = r + 2m \log \left\| \frac{r}{2m} - 1 \right\| \] (3) Hawking considers a mode $ p_i $ on $ \mathcal{I}^+ $ at late retarded time $ u $ of frequency $ \omega $, defined with respect to retarded time, $ p_\omega(u) $. He propagates this mode back along the event horizon through the non-stationary region of the collapsing star onto the $ \mathcal{I}^- $ (see figure 3). The form of the mode on $ \mathcal{I}^- $ is determined by connecting the mode to the event horizon by a null vector normal to the horizon, and parallel transporting this vector onto $ \mathcal{I}^- $.$^{17}$ From the form of the mode on $ \mathcal{I}^- $, one can read off the $ \beta $ coefficients. Thus we arrive at Hawking’s discovery: the expected particle number at frequency $ \omega $ at $ \mathcal{I}^+ $ is that of a black body with temperature, in geometric units, of $ \frac{\kappa}{2\pi} $, where $ \kappa $ is the surface gravity of the black hole. The black hole is seemingly radiating at what is now called the Hawking temperature. Our task now is to investigate why this derivation cannot be carried out in evaporation spacetime. I begin by identifying a globally hyperbolic sub-spacetime of evaporation Schwarzschild that might plausibly admit Hawking’s derivation. I then show that structure used in Hawking’s derivation is not present in evaporation spacetimes and so Hawking’s derivation is evaporation-inconsistent. ### 3.2 Hawking’s Derivation Fails in Evaporation-Schwarzschild Our first task is to find the region of evaporation-Schwarzschild in which to attempt to recover Hawking radiation. In quantum field theory on curved spacetimes, global hyperbolicity is nearly always assumed because this guarantees an initial value problem in the following sense: given an initial data $^{17}$In fact Hawking conducts the calculation on the past horizon of maximally extended Schwarzschild and argues that the conclusions would be the same on $ \mathcal{I}^- $. surface in GR, there exists a unique (up to isometry) spacetime that is the maximal globally hyperbolic development (MGHD) of the data surface. The initial data surface will be a Cauchy surface for this spacetime, and determines the entire spacetime. Moreover, it is clear that derivations of Hawking radiation which map modes in the past to modes in the future (as Hawking’s and Fredenhagen and Haag’s do) will require global hyperbolicity. This is because the state of the field in the past must determine the state of the field in the future. However, evaporation-Schwarzschild is not globally hyperbolic. Therefore, we must find a region of evaporation-Schwarzschild which is globally hyperbolic and has sufficient spacetime structure to admit Hawking’s derivation of Hawking radiation. There are two reasonable sub-spacetime regions of evaporation-Schwarzschild which are globally hyperbolic: a) the causal past of the black hole region ($J^-(B)$), or b) the MGHD of $J^-$, $(D(J^-))$. These two embedded regions are demarcated in figure 4. Which is more suited to deriving Hawking radiation? It is MGHD $J^-$. To see this, consider the causal past of the black hole spacetime. It is a spacetime such that, if a light ray were admitted at a point, it could reach the black hole before it evaporates completely. Near the evaporation event this is a tiny area, so we have deleted most of the spacetime we need for the derivation. More technically, ‘future null infinity’ in the causal past of the black hole will be the boundary of the causal past, $\hat{J}^-(B)$. There will not be a timelike Killing vector field on this boundary; therefore, there will be no preferred time parameter with respect to which we can define a particle interpretation. This is because the boundary bisects the non-stationary exterior region of the spacetime. This also means the boundary won’t be asymptotically flat; instead, it ends at the naked singularity and so it will contain a region of arbitrarily large curvature. Clearly, the causal past of the black hole region is useless for deriving Hawking radiation. MGHD $\mathcal{I}^-$ on the other hand does not suffer these problems, and includes the portion of $\mathcal{J}^+$ where all Hawking radiation will propagate to. Therefore this is the appropriate globally hyperbolic spacetime region to use.\footnote{It can be shown that neither the causal past of the black hole nor MGHD $\mathcal{I}^-$ is conformally equivalent to collapse-Schwarzschild, so proofs of the conformal equivalence of the Hawking temperature (e.g. Jacobson and Kang (1993)) do not help resolve the paradox.} So the question of this section is precisely stated as: which of the necessary assumptions for Hawking’s derivation of Hawking radiation cannot be carried over into MGHD $\mathcal{I}^-$? There are two important differences between MGHD $\mathcal{I}^-$ and collapse-Schwarzschild: the exterior solution is not Schwarzschild and the spacetime is not stationary. How do these changes affect the derivation? Firstly, Hawking’s derivation makes use of ingoing and outgoing Eddington-Finkelstein coordinates, defined in equation (3), which are specified for a particular mass $m$. This constant mass term is unavailable in evaporation spacetimes. Instead, one must analyse how modes defined with respect to coordinates that cover MGHD $\mathcal{I}^-$ behave on an evaporation metric, but nothing like this is carried out for Hawking’s derivation. Next, to calculate the form of modes on $\mathcal{I}^-$, Hawking exploits an isometry with the maximally extended Schwarzschild solution, and analyses modes that propagating onto the past horizon (see footnote 17). When the exterior solution is no longer Schwarzschild this isometry can not be used. Furthermore, and perhaps most strikingly, the failure of stationarity implies that the propagation of the modes back along the horizon will induce an evolution of the modes different to that calculated in collapse-Schwarzschild. Indeed, the normal null vector on the horizon which is used to compute the backwards evolution of the modes will have a different form in MGHD $\mathcal{I}^-$ as compared with collapse-Schwarzschild, precisely because the metric is different and the horizon area is changing. Finally, the non-stationarity will affect the scattering of the modes by the gravitational field. Admittedly, the model of evaporation used here, evaporation-Schwarzschild, is heuristic only and not generally believed to be a realistic model of black hole evaporation. One may wonder whether in more realistic models of black hole evaporation the problems listed here go away. It is in fact the opposite, things are worse in realistic models. For example, in explicitly computed models Schindler, Aguirre, and Kuttner (2020) show that, as well as the above worries still holding true, there is also no event horizon or Killing horizon for an evaporating black hole. Thus there will be no null vector normal to the horizon at all; the very structure Hawking uses to compute the form of the modes on $\mathcal{I}^-$ is non-existent in evaporation spacetimes. So, in realistic evaporation models, more of the spacetime structure exploited by Hawking to derive the radiation is lost. Hawking himself notes that the “negative energy flux will cause the area of the event horizon to decrease and so the black hole will not, in fact, be in a stationary state” (1975, p. 219). He accepts this is a problem, but claims to have a solution, as one can approximate the black hole as “quasi stationary”. In section 6.2 I show that this approximation regime does not, in fact, save Hawking’s derivation because one cannot use the regime to recover the necessary global structure. Therefore, the problems remain. A reader familiar with the vast literature of derivations of Hawking radiation may at this point be thinking of their preferred derivations, and be under the impression that they do not fall victims to the above challenges. I have no objection to such claims. Indeed I will present certain derivations as the best candidates currently available to resolve the paradox for Hawking’s derivation in section 6.3. Nevertheless, this is not a problem for my thesis as I am focused on particular derivations of Hawking radiation, in this case Hawking’s original derivation. Thus, given the amount of structure exploited by Hawking which does not carry over to MGHD $\mathcal{I}^-$, one must accept the conclusion that the derivation of Hawking radiation found in Hawking (1975) falls victim to the idealization paradox. That is to say, remarkably, Hawking’s derivation is evaporation-inconsistent! Fredenhagen and Haag (1990) construct their derivation to avoid a different problematic assumption in Hawking’s derivation, the geometric optics approximation. Thus, Fredenhagen and Haag’s derivation is what most consider to be the watertight derivation. It is to this that I turn next. 4 Idealization Paradox in Fredenhagen and Haag’s derivation Fredenhagen and Haag’s derivation is similar to Hawking’s in that it defines the state outside the black hole at some early time and maps this state to some state at late time. However, it differs in a few important respects. Firstly, the entire calculation is performed on the region of spacetime after the stellar matter has passed the event horizon. Secondly, they use a ‘detector’ at asymptotically late times to model the radiation. Thirdly, they perform the calculation by propagating the detector along the timelike Killing vector field in the exterior region. I sketch this derivation next, and in section 4.2 show that it is also evaporation-inconsistent. 4.1 Sketch of Fredenhagen and Haag’s Derivation This derivation, like Hawking’s, takes place on collapse-Schwarzschild. The region exterior to the event horizon in Schwarzschild can be covered by the coordinates $(t, r, \theta, \phi)$, where we call $t$ Schwarzschild-time, and define $\tau$-time coordinates, $(\tau, r, \theta, \phi)$ where $\tau = t + r^* - r = v - r$, for $v$ and $r^$ defined in (3). $\tau$ is approximately Schwarzschild-time near spacelike infinity, and becomes infinitely negative near the horizon. Let $r = r_s(\tau)$ define the surface of the collapsing star, with $r_s(0) = r_0$ the Schwarzschild radius, such that the star crosses the Schwarzschild radius at $\tau = 0$. As before, let $\Phi$ be a massless complex-valued scalar quantum field which satisfies the covariant wave equation. Fredenhagen and Haag model a detector in a spacetime region $O$ with an observable, $Q^ Q$, which is the counting rate given by $\langle Q^* Q \rangle$, where $Q = \int \Phi(x) h(x) \sqrt{\|g\|} d^4 x$ for a test function $h(x)$ that has support in $O$. They ‘place’ the detector at a large radius at the $\tau$-time for which the collapsing star crosses the horizon (i.e. $h$ has support around $(0, R, \theta_0, \phi_0)$ for $R \gg r_0$). The detector is then translated along the timelike Killing vector field of the Schwarzschild metric. We are interested in the counting rate of the detector at asymptotically late times (given by $Q_T (T \to \infty)$), as displayed in figure 5, in which a collapsing star and the time-translated detector are displayed in $\tau$-time coordinates. ![Figure 5: Set-up of Fredenhagen and Haag’s derivation, in which a detector $Q$ is propagated along the timelike Killing vector field to asymptotically late times, $Q_T (T \to \infty)$.](image) The counting rate is determined by the data on a Cauchy surface in the past of the late time detector. In the asymptotic limit, $T \to \infty$, the contributing data on the Cauchy surface decomposes into a sum of two wave packets, shifting asymptotically to $r \to \infty$ and $r \to r_0$. Wald (1995, pp. 159-162) explains this fact by noting that in maximally extended Schwarzschild, any mode in the region exterior to the black hole will decompose into modes on the past horizon and $\mathcal{I}^-$. Propagating this decomposition along Killing vector fields infinitely far will place modes infinitely close to the future horizon and spatial infinity, as depicted in figure 6. By isometry, we can draw the same conclusion for the exterior of collapse-Schwarzschild.\footnote{This inference is not in fact secure because the MGHD of the spacetime region exterior to the collapsing matter is not maximally extended Schwarzschild, but I ignore this difficulty here as it does not undermine Fredenhagen and Haag’s calculation but only Wald’s explanation of the behaviour of the wave-packet decomposition.} This fact can also be seen as a consequence of the Schwarzschild potential pushing modes onto the horizon and out to infinity. Assuming the state in the distant past is vacuum, the contribution to the counting rate from spatial infinity is zero. The contribution from the wave-packet that accumulates at the horizon is determined by the short-distance behaviour of the quantum field. Without concerning ourselves with the details, the authors assume the leading singularity in the short distance behaviour has a particular form that is implied by the Hadamard condition. Armed with this assumption, Fredenhagen and Haag show that the modes on the horizon contribute a thermal spectrum to the counting rate of the detector at asymptotically late times, with temperature given by the Hawking temperature. 4.2 Fredenhagen and Haag’s Derivation Fails in Evaporation-Schwarzschild We are again interested in whether we can deidealize this derivation: can the assumptions necessary to carry out Fredenhagen and Haag’s derivation be carried over to MGHD $\mathcal{I}^-$? The most prominent difficulty for the derivation is that it relies on the stationarity of the exterior metric. The detector is time translated along the Killing vector field of the Schwarzschild metric. This sends the modes on the Cauchy surface that contribute to the counting rate to spatial infinity and onto the horizon. Moreover, the behaviour of the mode decomposition as the detector is time translated is analysed on the maximally extended Schwarzschild spacetime and, following Wald, arises due to the global Killing field that is timelike in the exterior region. In MGHD $\mathcal{I}^-$ there are no such timelike Killing vector fields and there is no isometry with maximally extended Schwarzschild because the size of the black hole is changing. Indeed, there do not even exist approximate Killing vector fields on the entire spacetime, whatever notion of ‘approximate’ one might try to use. MGHD $\mathcal{I}^-$ contains a large mass black hole at $\tau = 0$, and by the evaporation event it contains a negligible mass black hole. This is clearly a radical change and so the spacetime is in no sense stationary. The behaviour of modes under the time translation symmetry of collapse-Schwarzschild was the core of the derivation, and this is simply not available in evaporation-Schwarzschild. In addition, there is a further difference between the two spacetimes relevant to Fredenhagen and Haag. The global time function on MGHD $\mathcal{I}^-$ does not extend to infinity into the future, whereas it is future infinite on collapse-Schwarzschild. The lack of a future-infinite time coordinate is a problem for Fredenhagen and Haag’s derivation because, whereas Hawking’s asymptotic time assumption was realised by future null infinity, Fredenhagen and Haag translate their detector along a timelike worldline. Every timelike worldline will reach the Cauchy horizon of MGHD $\mathcal{I}^-$ in a finite parameter distance, so one cannot take the asymptotic time limit. This limit was essential to Fredenhagen and Haag’s derivation as it pushed the modes asymptotically close to the horizon, forcing them into the trans-Planckian regime. Fredenhagen and Haag can then describe the modes by their short distance behaviour. Without this limit, we cannot be sure of the derivation. The Fredenhagen and Haag derivation cannot be carried out, in any obvious fashion, in evaporation-Schwarzschild. The idealization paradox thus applies to this approach as well: it too is evaporation-inconsistent. I now turn to algebraic approaches. 5 Idealization Paradox in Algebraic Approaches 5.1 Sketch of the Algebraic Derivation I will not go into any sort of detail in the sketch of algebraic approaches, as they are on the one hand very mathematically heavy, but on the other very conceptually simple. Algebraic approaches function by showing a particular state is the uniquely natural stationary Hadamard vacuum state on the collapse-Schwarzschild spacetime, and that this state is thermal at the Hawking temperature at future null infinity. Algebraic QFT begins with a -algebra of observables $\mathcal{A}$. A state $\omega$ is a completely positive map from $\mathcal{A}$ to $\mathbb{C}$, $\omega : \mathcal{A} \to \mathbb{C}$. For self-adjoint operators the map is real valued. We fix states by demanding they obey certain conditions, such as being vacuum.\footnote{In curved spacetimes, the algebraic approach defines a vacuum as a state that is Gaussian and pure (see Kay and Wald (1991) for details).} Conversely, we can discover facts about states by assessing what conditions they obey, for example a state is thermal with respect to a given Hamiltonian if it obeys the KMS condition.\footnote{See Bratelli and Robinson (1982, p. 13) for a definition.} As usual, we demand that physical states are Hadamard. Finally, one can define the algebra of observables for a scalar field by demanding that the functions used to smear the observables solve the covariant-wave equation. One finds (Dimock and Kay (1987), Dappiaggi et al. (2011)) that the uniquely natural stationary Hadamard vacuum state on the collapse-Schwarzschild spacetime is the Unruh vacuum. The Unruh vacuum has the property of having no particles near $J^-$, but being thermal at the Hawking temperature near $J^+$, with a flux going out to infinity. Thus, one claims that the black hole is emitting Hawking radiation. This sketch is sufficient to analyse the idealization paradox for algebraic approaches, to which I turn now. 5.2 Algebraic Approaches Fail in Evaporation-Schwarzschild Algebraic approaches are the most mathematically rigorous formulation of Hawking radiation. However, they clearly fail to survive the move to evaporation-Schwarzschild, or MGHD $\mathcal{J}^-$. The spacetime we are now interested in is not collapse-Schwarzschild, and not even approximately collapse-Schwarzschild. Therefore, the proof of the unique naturalness of the Unruh vacuum simply does not apply; the Unruh vacuum is uniquely natural on collapse-Schwarzschild, with no implication for the uniquely natural vacuum state on MGHD $\mathcal{J}^-$. Moreover, given that one condition on the Unruh vacuum is that it is stationary, and collapse-Schwarzschild is not stationary, clearly the Unruh vacuum will be the inappropriate vacuum state for MGHD $\mathcal{J}^-$. We can thus conclude that the idealization paradox applies to the algebraic approaches. To conclude, we have three different derivations, each of increasing mathematical rigour, and each with open questions about how they can actually claim to be establishing Hawking radiation in physically realistic models. 6 Paths Toward a Resolution Physics uses idealizations all the time. The idealization used in the derivations here is only particularly striking because it leads to a paradox, rendering the argument each derivation presents for Hawking radiation inconsistent. This paradox is clearly unacceptable, and so we should find a resolution. One aim of this paper, the aim taken up in this section, is to categorise and assess solutions to the idealization paradox. The most natural resolutions are those that deidealize the derivations, to show how they can proceed in evaporation spacetimes. Three sub-categories of deidealization solutions are presented below: - Quantum Gravity (section 6.1) - Approximation Regime (section 6.2) - Essential Structure (section 6.3) The first argues that quantum gravity is needed to describe black hole evaporation and thus resolve the paradox. The second looks to find an approximation regime between collapse-Schwarzschild and evaporation-Schwarzschild. Specifically, I formalise and analyse an approximation regime suggested in Hawking (1975). The third argues that one can weaken the assumptions of the derivations, such that each derivation can derive Hawking radiation whilst assuming only some essential spacetime structure that is present in both evaporation and non-evaporation spacetimes. I find that quantum gravity holds no prospects for resolving the paradox. I find Hawking’s approximation regime achieves varying degrees of success for the different derivations, but even where there are hints of success more work is needed. Finally, I find that essential structure derivations constitute a very fruitful research direction which has already been taken up in Visser (2003); Barcelo, Liberati, Sonego, and Visser (2011b, 2011a). Indeed, this work already points towards deep lessons about the nature of Hawking radiation. The derivations analysed in this paper, as discussed in the introduction, are not exhaustive. So, plausibly, other derivations don’t face the paradox (indeed I will discuss some examples in section 6.3). Given a paradox-free derivation, the consistency conjecture will be true, and so the phenomenon of Hawking radiation will be insulated from the paradox. However, this still leaves us with an idealization paradox for at least some derivations, and this in turn leaves an important and unanswered question: why is such a scientifically revolutionary piece of physics, Hawking (1975), inconsistent? How did it lead to Hawking radiation, when we can’t mesh it with physically realistic black holes? Although, according to many, the context of discovery need not be objective and rational (c.f. Popper (1959, sec. 2)), it still seems highly unlikely that Hawking’s derivation ‘got lucky’ and so an explanation of its success should be sought. I do not consider here resolutions which may be collected under the name deidealization pessimism, examples of such views include: embracing evaporation-inconsistent derivations as essential idealizations (aligning with Batterman (2002, 2005, 2011)), and denying the phenomenon of either black hole evaporation or Hawking radiation. Such approaches would resolve the paradox, but offer a somewhat pyrrhic victory by respectively rejecting either Earman’s principle, or the consensus in black hole physics.\footnote{Wallace (2018, 2019) reviews the arguments in favour of this consensus.} Instead, the categories I propose below (in sections 6.1, 6.2 and 6.3) help to distinguish different ways a derivation may be deidealized to avoid the paradox. ### 6.1 Quantum Gravity It is widely believed that a quantum theory of gravity will resolve the black hole information paradox.\footnote{For taxonomies of such proposals see Belot et al. (1999); Unruh and Wald (2017).} This is because the consensus in the physics community is that a quantum theory of gravity will be required to describe the final stages of black hole evaporation (e.g. Rovelli and Vidotto (2014)). Moreover, it is often claimed that the early stages of black hole evaporation also cannot be fully described without a quantum theory of gravity, as we can’t accurately describe the backreaction of Hawking radiation on the metric. Therefore, a reasonable first suggestion is to expect quantum gravity to resolve the idealization paradox. However, I argue this proposal cannot succeed. The central idea of a quantum gravity resolution to the idealization paradox is that the physics of spacetimes and Hawking radiation occurs in the semi-classical limit, whereas black hole evaporation occurs in a full quantum gravity description. One would argue that this allows one to reject the Inconsistency Claim, premise 3 of the idealization paradox presented in the introduction, which asserts that black hole evaporation leads to the rejection of assumptions required for the derivation of Hawking radiation. In order to reject the Inconsistency Claim, one may argue that because black hole evaporation is a quantum gravity phenomenon, it is not describable in the semi-classical limit and as such tells us nothing about the properties of spacetime in the semi-classical limit. Thus one cannot infer from evaporation to the breakdown of the semi-classical limit spacetime properties required for Hawking radiation. Thus, by acknowledging the need for a quantum theory of gravity to describe black hole evaporation, we can escape the paradox. Unfortunately, quantum gravity does not license us to reject the Inconsistency Claim. To see this, note that any quantum gravity theory of black hole evaporation must be able to represent: i) a black hole of given mass-energy, and ii) the mass-energy of a black hole being reduced in the process of evaporation. If the mass-energy of a black hole is not reducing then one cannot claim to be describing black hole evaporation, it is some other phenomena. This is certainly a possibility, but such a theory would constitute evaporation scepticism by claiming Hawking radiation does not lead black holes to lose mass-energy. Given these minimal representational requirements, the state in our quantum theory of gravity will represent a black hole of mass $m_1$ in the semi-classical limit at some earlier time, and a black hole of mass $m_2$ in the semi-classical limit at some later time, where $m_1 > m_2$. This immediately violates stationarity, one of the properties used in the derivations of Hawking radiation discussed here. Therefore, even a completely quantum gravity model of evaporation implies the breakdown of properties required for derivations of Hawking radiation in the semi-classical limit. Hence we are not licensed to reject the Inconsistency Claim. Why does the black hole information paradox admit a quantum gravity resolution whereas the idealization paradox does not? The difference is that there is no black hole information paradox until the evaporation event\footnote{In the traditional sense, though not in the Page-time paradox sense; see Wallace (2020).} because only then does one have to accept the information has vanished from the universe. Moreover, there is no minimal representational requirement on the evaporation event so we cannot anticipate any aspect of the quantum gravity description. On the other hand, the idealization paradox arises without the need to consider the evaporation event, due to the failure of properties in the entire exterior region such as stationarity. We can then impose our minimal condition on evaporation far before the evaporation event, and this leads to the paradox. ### 6.2 Approximation Regime Perhaps the derivations considered here can be carried out in some appropriate approximation regime: One would find some spacetime region in collapse-Schwarzschild which looks approximately like some corresponding region of evaporation-Schwarzschild. One could then hope to carry out the derivation using this approximating region of collapse-Schwarzschild, then infer the derived radiation back onto evaporation-Schwarzschild. Thus, one would derive the existence of Hawking radiation in the evaporation spacetime. Hawking (1975, p. 219) proposed such a resolution to the paradox: “it is a reasonable approximation to describe the black hole by a sequence of stationary solutions and to calculate the rate of particle emission in each solution.” The regime is justified as follows: The rate of change of the mass of the black hole will (for masses larger than the Planck mass) be much slower than the time taken for light to propagate to a region that can be modelled as approximately flat.\footnote{Initial work on modelling ‘infinity’ at a finite distance so asymptotic flatness can be defined for realistic sub-systems of the Universe can be found in Ellis (2002, sec. 5). Such modelling frameworks will likely be helpful for rigorously analysing approximation regimes such as the one proposed here. However, finite-infinity models are tangential to our current concern because I accept the standard presumption of the literature that astrophysical black holes are well modelled by collapse-Schwarzschild. If one demands finite-infinity models, it is then necessary to show how derivations of Hawking radiation look in these deidealized models, and what the relationship of such finite-infinity models to the models discussed in this paper is.} Thus, one can approximate the variable mass black hole spacetime as a sequence of stationary regions and calculate the rate of particle emission in each solution, avoiding the non-stationarity issues. This is a very intuitive picture if one imagines a black hole as a compact three-dimensional object that evolves in time and produces Hawking radiation via a local mechanism. However, as we have seen, the derivations of Hawking radiation discussed above use global spacetime structure, including the propagation of modes through the collapsing matter region (Hawking, 1975) and timelike Killing vector fields with an future infinite time parameter (Fredenhagen & Haag, 1990). Consequently, we should not assume that the slow rate of evaporation is sufficient to guarantee the derivations are unaffected; indeed, to do so would be negligent of philosophers of physics seeking to understand the derivations, idealizations, and phenomena at hand. To give Hawking a more charitable treatment, let me propose a more promising way to formalise this approximation. We want to identify regions of evaporation-Schwarzschild which are approximated by regions of collapse-Schwarzschild so that the derivations can be carried out using collapse-Schwarzschild. The best candidate regions are the parts of spacetime which Hawking radiation propagates through as it escapes to future null infinity. Figure 7 illustrates the structure of the regions (following Wald (1995, p. 178)). We model photons carrying energy away from the black hole to $\mathcal{J}^+$ and a negative energy flux propagating over the horizon. This is symbolised by two red arrows emerging from a single point, one pointing over the event horizon, the other out to future null infinity. Two such photon emission events are displayed in each conformal diagram in figure 7. The shaded region between the two photon emission events in evaporation-Schwarzschild is quasi-stationary. Thus the corresponding shaded region of the collapse-Schwarzschild spacetime of mass $m$ is approximately isometric to the shaded region of evaporation-Schwarzschild, where $m$ is the mass of the black hole according to the quasi-stationary region.\footnote{The mass of the slices will have to be modelled by the Bondi mass, as this is defined at future null infinity whereas the ADM mass can only be defined at spacelike infinity.} I denote such a quasi-stationary region of evaporation-Schwarzschild as $\mathcal{R}_{QS}$ and a corresponding approximately isometric stationary region of a collapse-Schwarzschild spacetime as $\mathcal{R}_S$. The regime will work by using $\mathcal{R}_S$ instead of $\mathcal{R}_{QS}$ to derive Hawking radiation. One then infers the same result, to some degree of approximation, in the approximately isometric $\mathcal{R}_{QS}$. Repeating this for every $\mathcal{R}_{QS}$ should describe the Hawking effect in evaporation-Schwarzschild. This regime will face two central problems. First, for each of the derivations $\mathcal{R}_S$ will have insufficient structure to derive Hawking radiation because it is a smaller extendable subspacetime. Thus it will be necessary to use the MGHD of $\mathcal{R}_S$. Problematically, although $\mathcal{R}_{QS}$ and $\mathcal{R}_S$ are approximately isometric, the corresponding MGHD for each will be very different. So although it is clear we can use $\mathcal{R}_S$ to draw approximately correct conclusions about $\mathcal{R}_{QS}$, it is less clear that we can use the MGHD of $\mathcal{R}_S$ to draw approximately correct conclusions about $\mathcal{R}_{QS}$. One must therefore justify using the MGHD of $\mathcal{R}_S$ rather than only the approximately isometric region, $\mathcal{R}_S$, despite the different global structure. Second, even if one can justify using the MGHD of $\mathcal{R}_S$ to draw inferences about $\mathcal{R}_{QS}$, neither $\mathcal{R}_{QS}$ nor $\mathcal{R}_S$ contain Cauchy surfaces for evaporation-Schwarzschild or collapse-Schwarzschild respectively. This is obvious in figure 7 where, for example, a massive particle can travel from $i^-$ and reach the singularity and never record data on the quasi-stationary surface. The same is true for $\mathcal{R}_S$ in collapse-Schwarzschild. This means that neither $\mathcal{R}_{QS}$ nor its approximately isometric stationary sibling $\mathcal{R}_S$ determine the entirety of their respective spacetimes. In fact, the past domain of dependence for $\mathcal{R}_{QS}$ does not extend outside of $\mathcal{R}_{QS}$, and so the past is significantly underdetermined. Why can’t we just select a region which does contain a Cauchy surface? Because this region would not be isometric, even approximately, to any region of collapse-Schwarzschild, and so we won’t be able to use approximation to justify performing the derivation on collapse-Schwarzschild and transferring the result of the derivation back over to evaporation-Schwarzschild. Given this, let us see how each of the derivations fair. Consider Hawking’s derivation: it depends on global spacetime structure in the sense of an infinite past prior to collapse that is stationary, and an infinite future after collapse that is stationary, and a non-stationary intervening period. He writes: “To understand how the particle creation can arise from mixing of positive and negative frequencies, it is essential to consider not only the quasi-stationary final state of the black hole but also the time-dependent formation phase.” (Hawking, 1975, p. 204) $ \mathcal{R}_S $ contains none, or at most very little, of this requisite structure. For example, any given $ \mathcal{R}_{QS} $ need not intersect the collapse region; indeed, the majority not, as demonstrated in figure 7. Thus, the approximately isometric region $ \mathcal{R}_S $ will also not intersect the non-stationary collapsing matter and so will have insufficient structure to carry out Hawking’s derivation. In order to recover the necessary structure, one needs to justify moving from $ \mathcal{R}_S $ to a spacetime with the global structure of collapse-Schwarzschild, perform the derivation on the global structure, and then make inferences from the global derivation back to the slice. Even if we assume that the first problem discussed above is solved and so such an inference is permissible, the inference still fails because, as per the second problem above, $ \mathcal{R}_S $ does not contain a Cauchy surface for collapse-Schwarzschild. Therefore, even if one could justify using the very different global structure to draw inferences about $ \mathcal{R}_{QS} $, not enough of the global structure is included in the MGHD of $ \mathcal{R}_S $ to carry out Hawking’s derivation. Therefore, Hawking’s approximation regime fails for Hawking’s derivation. Turning to Fredenhagen and Haag’s derivation: it was designed to not require the propagation of modes through the non-stationary collapse region, so the failure to recover this structure in $ \mathcal{R}_S $ is not problematic. However, the asymptotic time limit is not recovered in $ \mathcal{R}_S $; in fact the time over which the detector can be propagated is even shorter than in MGHD $ \mathcal{I}^- $. Therefore, the modes cannot accumulate arbitrarily close to the horizon, as is needed in Fredenhagen and Haag’s derivation. However, perhaps the result is recovered approximately with this limited time evolution. Moreover, $ \mathcal{R}_S $ does determine the entire future of the spacetime, so if we can overcome the first problem above and justify using the MGHD of $ \mathcal{R}_S $, we can in fact recover the asymptotic time-limit. Hence, the approximation regime holds reasonable promise of succeeding for Fredenhagen and Haag’s derivation. Nonetheless, it needs to be shown that either the use of the MGHD of $ \mathcal{R}_S $ is justified, or $ \mathcal{R}_S $ admits sufficiently long stationary worldlines to allow modes to accumulate sufficiently close to the horizon that the singularity structure of the quantum field will dominate. Neither of these are trivial, but neither seems implausible either. Finally, the algebraic approach: one begins by restricting the algebra of observables on collapse-Schwarzschild of mass $ m $ to an algebra on $ \mathcal{R}_S $. One then infers the vacuum state on this algebra by restricting the Unruh vacuum to $ \mathcal{R}_S $. One then claims that the algebra and vacuum state on $ \mathcal{R}_{QS} $ is approximately that of $ \mathcal{R}_S $. The central challenge for this approach is again the failure of $ \mathcal{R}_S $ to be Cauchy. This means the uniqueness of the state on $ \mathcal{R}_S $ will probably not hold. Without uniqueness, we can’t guarantee the state on $ \mathcal{R}_S $ is the restriction of the state on collapse-Schwarzschild. Moreover, the states on each $ \mathcal{R}_{QS} $ must be smoothly joined together, and therefore one needs to understand how the approximation changes the state, if only slightly.$^{27}$ I do not claim that I have exhausted the possibilities and difficulties for Hawking’s proposal. Nor do I claim that Hawking’s proposal exhausts the possible approximation regimes. I simply claim that, as of yet, this approach hasn’t been completely worked out for any of the derivations. Moreover, if the approximation regime is worked out for one derivation, say Fredenhagen and Haag’s, then the idealization paradox remains for the others, and so interesting open questions remain. The goal of this section has been to emphasise that the inference from that fact that evaporation is slow to the claim that the derivations go through approximately unaffected is non-trivial. In the next section I consider whether we can weaken the premises of the derivations to deidealize them. ### 6.3 Essential Structure The idealization paradox arises because some derivation uses a set of properties $ X $ with which to derive Hawking radiation, and then one finds that evaporation spacetimes don’t instantiate the set of properties $ X $. However, suppose that one could show that the derivation in fact did not require the complete set of properties $ X $ but only some subset of $ X $, call it set $ Y $, the essential structure. Suppose further that evaporation spacetimes could instantiate the essential structure $ Y $. Then the inconsistency would be resolved. Moreover, the essential structure that goes into deriving Hawking radiation would have been identified, and the surplus structure is stripped away. The task of identifying this essential structure is undertaken in Visser (2003); Barcelo et al. (2011a, 2011b). Visser (2003) argues that only three features are required for a derivation of Hawking radiation: an apparent horizon, non-zero surface gravity of the apparent horizon, and slow evolution. Therefore, using these as the set of properties $ Y $ could potentially resolve the idealization paradox. Going further, Barcelo et al. (2011a, 2011b) argue that Hawking-like radiation will occur whenever there is a continuous function mapping an affine parameter on future null infinity to that on past null infinity and the ‘adiabatic condition’ is satisfied.$^{28}$ In Barcelo et al. (2011a) the authors show how these conditions, with added assumptions about the QFT, can be used to derive the Bogoliubov coefficients, making explicit the relationship between their minimal conditions and Hawking’s derivation of Hawking radiation. Deidealization via essential structure derivations is strikingly different to that via the approximation regime. Whereas Hawking sought to find stationary structure within a non-stationary spacetime, $^{27}$Work has begun to formulate algebraic QFT on non-globally hyperbolic spacetimes, e.g. Janssen (2022). $^{28}$This is essentially a slow evolution condition, for details see Barcelo et al. (2011a). these derivations do away with the need for quasi-stationary regions, and instead provide a derivation which would be successful on the global structure of an evaporation spacetime. Both Visser (2003) and Barcelo et al. (2011a, 2011b) require the black hole to evolve slowly, but they do not use this slow evolution to approximate stationarity. By using a different deidealization method, different lessons are drawn. For example, given the very minimal structure used in these derivations, it can be argued that they point towards a kinematic interpretation of Hawking radiation, contra* the dynamical picture given in Hawking (1975). Moreover, these derivations don’t require an event horizon or Killing horizon to form. Similar lessons to those from an approximation regime can be learned here also, for example the spectrum derived is only approximately thermal, and the spectrum can be derived away from the asymptotic future (i.e. before the retarded time coordinate goes to infinity). I do not claim that these derivations face no difficulties, but only that they are very promising candidates for resolving the idealization paradox. A full analysis will be carried out in the future of the project. There is also a semantic issue of what one takes to be the referent of ‘Hawking radiation’ which I ignore here, emphasising only that resolutions to the paradox modify: i) what one takes to be required for something like Hawking radiation to occur, and ii) what is observed at $\mathcal{J}^+$. On a cautious note, it is not clear that one can distinguish between radiation due to the Unruh effect and radiation due to the Hawking effect with these derivations. Although the Unruh effect and Hawking radiation are closely related phenomena, they are not the same (Earman, 2011). If one cannot distinguish between the two a derivation may have insufficient structure. However, this does not seem to me a serious obstacle to these derivations resolving the paradox, but rather an obstacle to the full interpretation of the Hawking effect. The papers discussed here are not the only candidates for essential structure resolutions. Quantum tunnelling approaches, for example Parikh and Wilczek (2000), give a local dynamical account of Hawking radiation. A resolution to the paradox by these derivations would tell a different story. Firstly, they would retain a dynamical ontology for Hawking radiation. Secondly, they would point to lessons about the encoding of local structure by global structure in semi-classical gravity. Unpacking this second point, the definition of a black hole is global and Hawking’s derivation is global, but if a resolution of the paradox along the lines of a local dynamical account is the correct one, we might learn that this global structure is a red herring, and it just encodes local structure that in ways that are, at times, opaque. The goal of this section has been three-fold: 1) To highlight the differences between different deidealization strategies. 2) To emphasise there is an alternative to approximation regimes which make an inference from slow-evaporation to unaffected derivations, 3) To highlight the importance of research programmes such as that undertaken by Visser (2003); Barcelo et al. (2011a, 2011b). I do not claim that the paradox is definitely solved, or even necessarily solved by an essential structure deidealization, but rather that this is a promising option with many lessons to be learnt. Summarising, deidealization can follow multiple different routes and these routes have varying degrees of success. Indeed, the success of a particular deidealization need not be homogeneous across derivations. I only take the quantum gravity route to be completely impotent. Hawking’s approximation regime fails for Hawking’s derivation, but prospects for success are better for Fredenhagen and Haag’s derivation, and other approximation regimes may fare better. The essential structure research programme is very promising, in particular for deidealizing Hawking’s derivation. The lessons we draw from these varying approaches to deidealization depend upon the type of deidealization, and the details of how the deidealization operates. In the final section I consider what options are available if one thinks, against the presumed consensus, that one cannot deidealize the derivations presented here. 7 Conclusion “Paradoxes are just the scar tissue. Time and space heal themselves up around them and people simply remember a version of events which makes as much sense as they require it to make.” – Douglas Adams, Dirk Gently’s Holistic Detective Agency I have argued that Hawking’s derivation of Hawking radiation, along with Fredenhagen and Haag’s and the algebraic approach, are all evaporation-inconsistent. They are carried out on collapse-Schwarzschild but cannot be carried out on evaporation-Schwarzschild. By throwing away the spacetime used to derive the phenomenon, we throw away the very ladder we are standing on, and come tumbling back to inconsistency. There are reasonable (and some unreasonable) paths towards deidealizing the derivations involved, and thus reason to believe that the paradox is just scar tissue from the messy process of scientific development. Presumably, there is a resolution along the lines of an approximation regime or essential structure derivation which will teach us why the inconsistent derivations worked so well, and what they really represent. If so, this will be another victory for the dispensabilists in the idealization literature, and a particularly striking one given that the idealizations used in Hawking radiation derivations were not simply false, but inconsistent. The differences between the different possible deidealizations emphasises the non-triviality of the deidealization project, and the variety of lessons that may be learnt. Most excitingly, deidealizing these derivations may remove the chaff from the conceptual framework of Hawking radiation and give a clear ontological picture of Hawking’s eponymous discovery. Such lessons are the fruits of paying close attention to, and resolving, the idealization paradox; fruits won as reward for not settling with the unjustified inference from slow-evaporation to unaffected derivations. No matter what road we take, we are bound to learn something interesting. References Arageorgis, A., Earman, J., & Ruetsche, L. (2002). Weyling the time away: the non-unitary implementability of quantum field dynamics on curved spacetime. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 33(2), 151–184. Barcelo, C., Liberati, S., Sonego, S., & Visser, M. (2011a). Hawking-like radiation from evolving black holes and compact horizonless objects. Journal of High Energy Physics, 2011(2), 1–30. Barcelo, C., Liberati, S., Sonego, S., & Visser, M. (2011b). Minimal conditions for the existence of a Hawking-like flux. Physical Review D, 83(4), 041501. Batterman, R. (2002). The Devil in the Details: Asymptotic Reasoning in Explanation, Reduction, and Emergence. New York: Oxford University Press. Batterman, R. (2005). Critical phenomena and breaking drops: Infinite idealizations in physics. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 36(2), 225-244. doi: https://doi.org/10.1016/j.shpsb.2004.05.004 Batterman, R. (2011). Emergence, Singularities, and Symmetry Breaking. Foundations of Physics, 41(6), 1031–1050. doi: 10.1007/s10701-010-9493-4 Batterman, R. (2017). Philosophical Implications of Kadanoff’s Work on the Renormalization Group. Journal of Statistical Physics, 167(3-4), 559–574. doi: 10.1007/s10955-016-1659-9 Belot, G., Earman, J., & Ruetsche, L. (1999). The Hawking Information Loss Paradox: The Anatomy of Controversy. The British Journal for the Philosophy of Science, 50(2), 189-229. doi: 10.1093/bjps/50.2.189 Bratelli, O., & Robinson, D. W. (1982). Operator algebras and quantum statistical mechanics. Butterfield, J. (2011). Less is different: Emergence and reduction reconciled. Foundations of physics, 41, 1065–1135. Curiel, E. (2019). The Many Definitions of a Black Hole. Nature Astronomy, 3, 27–34. Curiel, E. (2023). The Hawking Effect, Its Desiderata and Its Discontents. Retrieved from https://www.youtube.com/watch?v=jFZ2HSkMTvY Dappiaggi, C., Moretti, V., & Pinamonti, N. (2011). Rigorous construction and Hadamard property of the Unruh state in Schwarzschild spacetime. Advances in Theoretical and Mathematical Physics, 15(2), 355–447. Dimock, J., & Kay, B. S. (1987). Classical and quantum scattering theory for linear scalar fields on the Schwarzschild metric I. Annals of Physics, 175(2), 366-426. doi: https://doi.org/10.1016/0003-4916(87)90214-4 Dougherty, J., & Callender, C. (2016). Black hole thermodynamics: More than an analogy? Retrieved from https://philsci-archive.pitt.edu/13195/ Earman, J. (2004). Curie’s Principle and spontaneous symmetry breaking. International Studies in the Philosophy of Science, 18(2-3), 173–198. doi: 10.1080/0269859042000311299 Earman, J. (2011). The Unruh Effect for Philosophers. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 42(2), 81–97. doi: 10.1016/j.shpsb.2011.04.001 Ellis, G. F. R. (2002). Cosmology and local physics. New Astronomy Reviews, 46(11), 645-657. doi: 10.1016/S1387-6473(02)00234-8 Fletcher, S. C. (2020). The principle of stability. Philosophers’ Imprint, 20. Fletcher, S. C., Palacios, P., Ruetsche, L., & Shech, E. (2019). Infinite idealizations in science: an introduction. Synthese, 196, 1657–1669. Fredenhagen, K., & Haag, R. (1990). On the Derivation of Hawking Radiation Associated With the Formation of a Black Hole. Commun. Math. Phys., 127, 273. doi: 10.1007/BF02096757 Frigg, R. (2022). Models and theories a philosophical inquiry. Routledge. Frigg, R., & Hartmann, S. (2020). Models in Science. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2020 ed.). Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2020/entries/models-science/. Geroch, R. (1970). Domain of Dependence. Journal of Mathematical Physics, 11(2), 437-449. doi: 10.1063/1.1665157 Gryb, S., Palacios, P., & Thébault, K. (2019). On the Universality of Hawking Radiation. British Journal for the Philosophy of Science, axz025. doi: 10.1093/bjps/axz025 Halvorson, H., & Clifton, R. (2002). No place for particles in relativistic quantum theories? Philosophy of Science, 69(1), 1–28. doi: 10.1086/338939 Hawking, S. W. (1975). Particle Creation by Black Holes. Commun. Math. Phys., 43, 199–220. ([Erratum: Commun. Math. Phys. 46, 206 (1976)]) doi: 10.1007/BF02345020 Hawking, S. W., & Ellis, G. F. (1973). The large scale structure of space-time. Cambridge university press. Jacobson, T., & Kang, G. (1993). Conformal invariance of black hole temperature. Classical and Quantum Gravity, 10(11), L201. Janssen, D. W. (2022). Quantum fields on semi-globally hyperbolic space–times. Communications in Mathematical Physics, 391(2), 669–705. Jones, N. J. (2006). Ineliminable idealizations, phase transitions, and irreversibility (Unpublished doctoral dissertation). The Ohio State University. Kay, B. S., & Wald, R. M. (1991). Theorems on the uniqueness and thermal properties of stationary, nonsingular, quasifree states on spacetimes with a bifurcate Killing horizon. Physics Reports, 207(2), 49-136. doi: https://doi.org/10.1016/0370-1573(91)90015-E Landsman, K. (2013). Spontaneous symmetry breaking in quantum systems: Emergence or reduction? Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 44(4), 379–394. Landsman, K. (2021). Foundations of General Relativity: From Einstein to Black Holes. Radboud University Press. doi: 10.54195/EFVF4478 Manchak, J. B., & Weatherall, J. O. (2018). Paradox Regained? A Brief Comment on Maudlin on Black Hole Information Loss. Foundations of Physics, 48(6), 611–627. doi: 10.1007/s10701-018-0170-3 Maudlin, T. (2017). (Information) Paradox Lost. Menon, T., & Callender, C. (2013). Ch-Ch-Changes Philosophical Questions Raised by Phase Transitions. In R. Batterman (Ed.), The oxford handbook of philosophy of physics (p. 189). OUP. Morrison, M. (2012). Emergent physics and micro-ontology. Philosophy of Science, 79(1), 141–166. doi: 10.1086/663240 Norton, J. D. (2012). Approximation and Idealization: Why the Difference Matters. Philosophy of Science, 79(2), 207–232. doi: 10.1086/664746 Page, D. N. (1994). Black hole information. In Proceedings of the 5th canadian conference on general relativity and relativistic astrophysics (Vol. 1, pp. 1–41). Palacios, P. (2019). Phase Transitions: A Challenge for Intertheoretic Reduction? Philosophy of Science, 86(4), 612–640. doi: 10.1086/704974 Palacios, P. (2022). Emergence and Reduction in Physics. Cambridge University Press. doi: 10.1017/9781108901017 Parikh, M. K., & Wilczek, F. (2000). Hawking Radiation As Tunneling. Physical Review Letters, 85(24), 5042–5045. Retrieved from Popper, K. (1959). The Logic of Scientific Discovery (6th ed.). Routledge. Potochnik, A. (2017). Idealization and the Aims of Science. Chicago: University of Chicago Press. Prunkl, C. E. A., & Timpson, C. G. (2019). Black Hole Entropy is Thermodynamic Entropy. Raju, S. (2022). Lessons from the information paradox. Physics Reports, 943, 1-80. (Lessons from the information paradox) doi: https://doi.org/10.1016/j.physrep.2021.10.001 Rovelli, C., & Vidotto, F. (2014). Planck stars. International Journal of Modern Physics D, 23(12), 1442026. doi: 10.1142/s0218271814420267 Schindler, J. C., Aguirre, A., & Kuttner, A. (2020). Understanding black hole evaporation using explicitly computed Penrose diagrams. Phys. Rev. D, 101, 024010. doi: 10.1103/PhysRevD.101.024010 Sheeh, E. (2018). Infinite Idealizations in Physics. Philosophy Compass, 13(9), e12514. doi: 10.1111/phc3.12514 Sheeh, E. (2023). Idealizations in Physics. Cambridge, UK: Cambridge University Press. Unruh, W. G. (2014). Has Hawking radiation been measured? Foundations of Physics, 44, 532–545. Unruh, W. G., & Wald, R. M. (2017). Information loss. Reports on Progress in Physics, 80. Visser, M. (2003). Essential and inessential features of Hawking radiation. International Journal of Modern Physics D, 12(04), 649–661. Retrieved from Wald, R. M. (1975). On particle creation by black holes. Communications in Mathematical Physics, 45(1), 9–34. Wald, R. M. (1984). General Relativity. Chicago, USA: Chicago Univ. Pr. doi: 10.7208/chicago/9780226870373.001.0001 Wald, R. M. (1995). Quantum Field Theory in Curved Space-Time and Black Hole Thermodynamics. Chicago, IL: University of Chicago Press. Wallace, D. (2018). The case for black hole thermodynamics, Part I: phenomenological thermodynamics. Wallace, D. (2019). The case for black hole thermodynamics part II: Statistical mechanics. Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, 66, 103-117. doi: https://doi.org/10.1016/j.shpsb.2018.10.006 Wallace, D. (2020). Why Black Hole Information Loss Is Paradoxical. In N. Huggett, K. Matsubara, & C. Wüthrich (Eds.), Beyond spacetime: The foundations of quantum gravity (p. 209–236). Cambridge University Press. doi: 10.1017/9781108655705.013 Wüthrich, C. (2019). Are black holes about information? In R. Dardashti, R. Dawid, & K. Thébault (Eds.), Why trust a theory?: Epistemology of fundamental physics (p. 202-224). Cambridge University Press. doi: 10.1017/9781108671224
An advanced scattered moonlight model for Cerro Paranal A. Jones$^1$, S. Noll$^1$, W. Kausch$^1$, C. Szyszka$^1$, and S. Kimeswenger$^{2,1}$ $^1$ Institute for Astro and Particle Physics, Leopold Franzens Universität Innsbruck, Technikerstrasse 25, 6020 Innsbruck, Austria e-mail: firstname.lastname@example.org $^2$ Instituto de Astronomía, Universidad Católica del Norte, 0610 Avenida Angamos, Antofagasta, Chile Received 2 August 2013 / Accepted 23 October 2013 ABSTRACT The largest natural source of light at night is the Moon, and it is the major contributor to the astronomical sky background. Being able to accurately predict the sky background, including scattered moonlight is important for scheduling astronomical observations. We have developed an improved scattered moonlight model, in which the components are computed with a better physical understanding as opposed to the simple empirical fit in the frequently used photometric model of Krisicunas & Schaefer (1991, PASP, 103, 1033). Our spectroscopic model can better trace the spectral trends of scattered moonlight for any position of the Moon and target observation. This is the first scattered moonlight model that we know of which is this physical and versatile. We have incorporated an observed solar spectrum, accurate lunar albedo fit, and elaborate scattering and absorption calculations that include scattering off of molecules and aerosols. It was designed for Cerro Paranal, but can be modified for any location with known atmospheric properties. Throughout the optical range, the uncertainty is less than 20%. This advanced scattered moonlight model can predict the amount of scattered moonlight for any given geometry of the Moon and target, and lunar phase for the entire optical spectrum. Key words. Moon – atmospheric effects – radiative transfer – scattering – methods: data analysis – techniques: spectroscopic 1. Introduction The current trend in astronomy has been to build larger and larger telescopes. The operating costs for running these large telescopes are high and careful planning of observations is needed for telescope time is always in demand. This means that more accurate predictions and estimation of the sky background are needed to understand how long an exposure is necessary for a given observation with a certain signal to noise ratio. The brightest natural source of light in the night sky, and therefore the greatest contributor to the sky background noise, is scattered moonlight. Having a reliable model of the moonlight for sky background estimation is critical. Also, many observers are trying to characterize faint objects spectroscopically, and knowing accurately the spectrum of the background would allow astronomers to predict which spectral features are observable within a given exposure time. By improving the scattered moonlight model within a sky background model, we can increase telescope scheduling efficiency. The long standing scattered moonlight model used by ESO (European Southern Observatory) was by Walker (1987) and provides a table of the magnitudes for five photometric bands of the night sky at five different moon phases. It does not depend on the positions of the Moon or target observation and was measured during solar minimum. This model is limited in producing a scattered moonlight spectrum which is accurate enough for the current and future telescope operations. The current widely used model was developed by Krisciunas & Schaefer (1991), with 52 citations (e.g. Davies et al. 2013; Knoetig et al. 2013; Trinh et al. 2013). It is an empirical fit to data in the $V$-band taken at the 2800 m level of Mauna Kea. Even though the fit was separated into various specific functions, such as initial intensity from the Moon, Rayleigh and Mie scattering, it was still a parametrization based on only 33 observations in one photometric band specifically for Mauna Kea. The accuracy is between 8 and 23% if not near full Moon. In a previous paper, Noll et al. (2012), we presented a spectroscopic extension of the Krisciunas & Schaefer (1991) model. It was optimized for Cerro Paranal and covered the optical regime. The lunar albedo was taken to be constant with respect to wavelength and scaling factors for the different functions were introduced to better fit data from Cerro Paranal. Our advanced scattered moonlight model works from the UV to the near-IR, but has only been fully tested in the optical, due to a current lack of data. It has been calibrated and investigated with 141 optical spectra and has an overall accuracy of $\sigma \lesssim 0.2$ mag. The model has been split into physically based modules which are given by either physical models or the best current fits. The present version is optimized for Cerro Paranal, but can be modified for any location with information about its atmospheric properties. Our model is fully 3D and can predict the amount of scattered moonlight for any configuration of the Moon and target. It includes higher order scattering and is therefore still reliable at high zenith distances (when either object is near the horizon). Because our scattered moonlight model produces a spectrum, it can be used for finding spectral features and trends as well as photometric magnitudes. The original purpose of designing an advanced scattered moonlight model, as part of the full sky background model, was for the Austrian accession to ESO to improve the ETC (Exposure Time Calculator), and thus enhance telescope scheduling and efficiency. Our optical sky background model, but with an older version of the scattered moonlight model, is described in Noll et al. (2012). The full sky background model estimates the amount of background light from 0.3 to 30 micron for Cerro Paranal. It includes all relevant components, such as scattered... Table 1. Properties of the different observing modes in the FORS1 sky background data set. \| Grism \| Filter \| Wav. range (Å) \| Resolution (Å FWHM) \| Dispersion (Å px$^{-1}$) \| Total # \| # with Moon \| # Used \| \|-------\|--------\|----------------\|----------------------\|--------------------------\|---------\|-------------\|--------\| \| 300V \| GG435 \| 4300–8900$^a$ \| 12 \| 2.6 \| 676 \| 188 \| 70 \| \| 300V \| … \| 3615–8900$^b$ \| 12 \| 2.6 \| 163 \| 36 \| 24 \| \| 600B \| OG590 \| 3650–6050 \| 5.3 \| 1.2 \| 143 \| 29 \| 18 \| \| 600R \| GG435 \| 5390–7530 \| 4.5 \| 1.0 \| 207 \| 60 \| 29 \| Notes. Listed here is the total number of sky spectra, the number of sky spectra with moonlight, and the number of sky spectra with moonlight that was used in the analysis (Sect. 2.1). $^a$ Order separation filter, fluxes <4400 Å were not used; $^b$ second-order overlapping, fluxes >6200 Å were not used. moonlight and starlight, zodiacal light, airglow line emission and continuum, scattering and absorption within the Earth’s atmosphere, and thermal emission from the atmosphere and telescope. Each component was designed with the latest knowledge and results in the field and was thoroughly checked with archival ESO data. The new scattered moonlight model is the topic of this paper. The zodiacal light is found using the prescription from Leinert et al. (1998) and the airglow model is based on local observations and semi-empirical modeling (Patat 2008). The scattering and absorption are calculated with two codes, one that is fully 3D for the scattering and a line-by-line radiative transfer code for molecular absorption. The scattering and absorption will also be described in this paper, in the context of the scattered moonlight. The thermal emission is estimated as a gray body. The new sky background model has already been implemented by ESO. In this paper we will discuss the optical portion of the advanced scattered moonlight model. In a subsequent paper, we will discuss the extension of this model into the near-IR using X-Shooter data (Vernet et al. 2011). For the remainder of the paper, the term “moon model” refers to the scattered moonlight model. In Sect. 2, we describe the moon model. In Sect. 3, we compare our scattered moonlight model with the observations and previous models, and finally in Sect. 4 we present our conclusions. 2. The model To accurately calculate the scattered moonlight, there are several components that must be considered. The simplest way to account for the various pieces is to follow the path of light from the source to the instrument. The source of the scattered moonlight is the Sun. Then the light is reflected by the Moon which depends on the lunar albedo. This is mainly a function of the lunar phase. Next, the moonlight enters the Earth’s atmosphere, where it is scattered by molecules (Rayleigh) and aerosols (Mie), and absorbed. The light can be scattered multiple times, including off of the ground, before it reaches the telescope. This depends on the properties of the atmosphere and the positions of the Moon and target observation. In the following subsections, we will discuss each of these steps in more detail. Section 2.1 introduces the calibration data set. In Sect. 2.2 we discuss the solar spectrum, in Sect. 2.3 the lunar albedo, and in Sect. 2.4 a more general discussion of the set-up of the radiative transfer equations. Then, we have a more detailed discussion about Rayleigh scattering in Sect. 2.5, Mie scattering in Sect. 2.6, and absorption in Sect. 2.7. 2.1. Calibration data set We calibrated our full sky background model including the improved scattered moonlight model with a FORS1 (FOcal Reducer/low dispersion Spectrograph) data set from Patat (2008). FORS1 was a low resolution optical spectrograph at Cerro Paranal. The data were taken between 1999 and 2005 with four different observing modes and consist of long-slit spectra, where plain sky could be extracted. For the details about the different observing modes, see Table 1. There are 1186 spectra, but only 26% have moonlight. For the analysis, we considered only 141 spectra with good weather conditions according to ESO’s ambient weather conditions database\footnote{\url{http://archive.eso.org/asm/ambient-server}}. The condition of good weather depended on two criteria, humidity and clouds. First, we only considered spectra taken when the humidity was $\leq 20\%$. Second, there must not be any detection of possible clouds in the night sky, within 5 h of the observation. Since we are measuring scattered moonlight, the entire sky must be free of clouds to ensure stable atmospheric conditions. Figure 1 shows several properties of the data: the distribution of the humidity with the cutoff, moon distance, lunar phase angles with waxing and waning, and angular distance $\rho$. 2.2. Solar spectrum The source of the scattered moonlight is the Sun. We use the spectrum from Colina et al. (1996), which covers wavelengths from 0.12 to 2.5 micron. The UV and optical portion are from measurements from satellites and ground based data. The near-IR spectrum is from data by the NASA CV-990 aircraft with a model spectrum. The optical and near-IR spectrum has an uncertainty of $\leq 5\%$, while the UV has an uncertainty of $\sim 20\%$ due to solar variability, with a resolution of 1 to 2 nm. Figure 2 shows the optical solar spectrum. It resembles the black body spectrum with typical absorption lines for a G2 star. 2.3. Lunar albedo The lunar albedo ($A$) determines the amount of sunlight that is reflected off the lunar surface towards the Earth. It is directly related to the intensity of light that will enter the Earth’s atmosphere, called $I^$. They are related by (Kieffer & Stone 2005), $$I^ = I_{\text{sol}} \frac{\Omega_M}{\pi} A \left( \frac{384\,400}{M_{\text{dis}}} \right)^2,$$ $I_{\text{sol}}$ is the spectrum of the light intensity of the Sun (see Sect. 2.2), and $\Omega_M$ is the solid angle of the Moon, where we take $\Omega_M = 6.4177 \times 10^{-5}$ sr. For our set of sky observations (Patat 2008), we found the moon distances ($M_{\text{dis}}$) with the online tool called JPL Horizons\footnote{\url{http://ssd.jpl.nasa.gov/?horizons}}. The distance of 384\,400 km is the average moon distance. Fig. 1. Properties of the FORS1 data from Patat (2008). The data shown has been preselected to not have any clouds during observation and plots b), c), and d) also have $\leq 20\%$ humidity. a) The distribution of the amount of humidity at the time of observation as given by ESO Ambient Conditions Database. Only observations with $\leq 20\%$ humidity were analyzed and is shown here as the filled in part of the histogram. b) Shows the distribution of the relative moon distance. A moon distance of 1 corresponds to the average moon distance of 384,400 km. c) Here we have plotted the distribution of lunar phase angles $g$. Over-plotted is the number of spectra taken during a waxing and waning Moon, which effects the solar selenographic longitude. d) The distribution of the angular distance between the Moon and target observation $\rho$ is displayed. See Sect. 2.1 for more details. Fig. 2. Optical solar spectrum from Colina et al. (1996) with principal absorption lines and bands labeled (Wallace et al. 2011). This was used as the source for our scattered moonlight model (Sect. 2.2). The lunar albedo $A$ depends on several factors, including lunar phase, solar selenographic longitude, and wavelength. We use an empirical fit based on the ROLO survey (Kieffer & Stone 2005). They used over 100,000 images of the Moon in 32 different photometric bands in the optical and near infrared for certain lunar phases. The fit given by Kieffer & Stone (2005) is, $$\ln A_\lambda = \sum_{i=0}^{3} a_{i,\lambda} g^i + \sum_{j=1}^{3} b_{j,\lambda} (\Phi^{2j-1} + d_{1,\lambda} e^{-g/p_1} + d_{2,\lambda} e^{-g/p_2} + d_{3,\lambda} \cos [(g - p_3)/p_4]. \tag{2}$$ The lunar phase is described by the parameter $g$, where $g = 0^\circ$ corresponds to a full Moon and $g = 180^\circ$ is a new Moon. $\Phi$ is the solar selenographic longitude which describes whether the Moon is waxing or waning. Because the maria are not uniformly distributed on the lunar surface, the overall reflectivity of the Moon varies depending on which portion of the Moon is illuminated. We used the Astronomical Ephemeris by NASA\footnote{\url{http://eclipse.gsfc.nasa.gov/TYPE/ephemeris.html}} to find whether the Moon was waxing or waning. In Eq. (2) the parameters $a_{i,\lambda}$, $b_{j,\lambda}$, and $d_{x,\lambda}$ for the 32 different wavelengths $\lambda$ are provided in Table 4 and the constants $p_n$ are in Eq. (11) of Kieffer & Stone (2005). The last three terms in Eq. (2) are to correct for the opposition effect (exponential functions) and to minimize the residuals between the data and fit (cosine function). The opposition effect is where the Moon has an enhanced brightness near full Moon (Whitaker 1969). We have neglected the terms that depend on the observer selenographic latitude and longitude. They have a small impact, maximum of $\sim 7\%$ over a full Saros (libration) cycle, and are variables that are not easily available in astronomical data. The ROLO fit only covers the phase angles between $g = 1.55^\circ$ and $97^\circ$. We extrapolated this fit to $g = 180^\circ$. For the terms that depend on $\Phi$, when $g > 97^\circ$ we took the values of $\Phi$ at $g = 97^\circ$. This physically means that we are assuming that the ratio of maria to highlands is the same when $g > 97^\circ$ as $g = 97^\circ$. The other terms in Eq. (2) are still well behaved when $g > 97^\circ$ and fit reasonably well with our data (Patat 2008). It is expected from previous studies that the general dependence of $A$ on $g$ is exponential for all values of $g$ (outside the opposition effect) (e.g. Lane & Irvine 1973), which we see in our extrapolation as well. As $g$ increases the overall flux from the Moon decreases, and becomes less important compared with the other components of the sky background. Thus, we believe the extrapolation to higher $g$ is reasonable. In Fig. 3, we show the dependence of the lunar albedo on moon phase including the extrapolation when $g > 97^\circ$. Also shown is the effect of $\Phi$ which is a bit larger at redder wavelengths and is over-plotted as a band of values in yellow. At $g = 97^\circ$, $\Phi$ can cause deviations of $\sim 6.8\%$. The fit from the ROLO data was only done photometrically, however we require a full spectrum of $I^$ for the scattered moonlight model. To attain a fit for the entire optical spectrum, we performed a simple linear interpolation between the fit values provided. This interpolation is shown in Fig. 4 for several different moon phases. In general the wavelength dependence is fairly smooth and becomes less significant towards larger $g$. There are some small variations where the photometric values are clustered. According to Velikodsky et al. (2011), there are some discrepancies in the overall flux calibration amongst the various Moon observations. They find that the ROLO data at 603 nm is $13\%$ too faint. They used an interpolation of the ROLO fit to find the albedo at 603 nm to compare it with their observations. Since there is no information about the wavelength dependence for this correction, we divide the albedo at all wavelengths by 0.87. We currently neglect the effects of polarization. We assume that the change in the resulting scattered moonlight due to polarization is smaller than our current errors. In general the polarization of the lunar surface is small, less than 10% at maximum. However, the maria can have a decent amount of polarization ($\sim 30\%$) (Dollfus & Bowell 1971). ### 2.4. Radiative transfer The light from the Moon can be scattered many times or absorbed in the Earth’s atmosphere before reaching the telescope. We have developed a fully 3D single scattering code with estimates for double and higher order scattering. At the various scattering points, an effective airmass is calculated for absorption. The transmission of light $t$ can be directly related to the optical depth at zenith $\tau_0$ or the extinction coefficient $k$ and airmass $X$ by the following, $$t(\lambda) = e^{-\tau_0(\lambda)X} = 10^{-0.4k(\lambda)X}. \quad (3)$$ We will use this formalism in the following sections to describe the effects of scattering. #### 2.4.1. Single scattering The single scattering can be done fully in 3D (Wolstencroft & van Breda 1967; Staude 1975; Bernstein et al. 2002). We wish to obtain the integrated scattered light towards the azimuth $A_0$ and zenith distance $z_0$. We take the Moon to be a point source. We consider the scattering path elements $S$ of density $n(\sigma)$, with $\sigma$ being the radius vector from the center of Earth $C$ to $S$. $S$ is located between the top of atmosphere $T$ to the observer $O$, at height $H_0$ above the surface (see Fig. 5). The distance between $O$ and $C$ is $\sigma_0 = H_0 + R$, where $R$ is the radius of Earth (6371 km for the mean radius). For each path element $S$ at distance $s_2$ from $O$, the contributions of radiation from $M$, where the moonlight enters the atmosphere, to the intensity at $S$ along $s_1$ are considered. This intensity includes the scattering of light out of the path (and possible absorption), which depends on the effective column density of the scattering/absorbing particles. This is given by, $$B_1(z, \sigma) = \int_0^{s_1(z, \sigma)} n(\sigma') \, d\sigma'. \quad (4)$$ The scattering intensity also depends on the wavelength-dependent extinction cross section $C_{\text{ext}}(\lambda)$ of the various processes, Rayleigh, Mie and absorption. They are related to the optical depth $\tau$ (see Eq. (3)) and Eq. (4) at the zenith by, $$C_{\text{ext}}(\lambda) = \frac{\tau_0(\lambda)}{B_0(0, \sigma_0)}. \quad (5)$$ After scattering at $S$ the light travels along path $s_2$ and is extincted by $B_2(z_0, \sigma)$. The total scattering intensity per solid angle $I_{\text{scat}}$ at $(A_0, z_0)$ for the flux of the Moon $F^$, related to $I^$ (Eq. (1)), at $(A_M, z_M)$ as seen by $O$, for Rayleigh and Mie scattering separately, is given by, $$I_{\text{scat}}(A_0, z_0) = \frac{C_{\text{scat}}(\lambda)}{4\pi} \int_0^{r_2(z_0, \sigma_0)} n(\sigma) P(\theta)$$ $$\times F^(A_M, z_M) \exp^{-\tau} \, ds,$$ where $$\tau = (C_{\text{exl,R}} + C_{\text{exl,A}}) \times [B_{1,R}(z, \sigma) + B_{2,R}(z_0, \sigma_0)]$$ $$+ C_{\text{exl,M}} \times [B_{1,M}(z, \sigma) + B_{2,M}(z_0, \sigma_0)].$$ Equation (7) is for Rayleigh (R), Mie (M), and absorption (A). In Eq. (6), $C_{\text{scat}}$ is the wavelength-dependent scattering cross section, which will deviate from $C_{\text{exl}}$ if absorption occurs. The scattering phase function $P$ depends on the scattering angle $\theta$, the angle between the paths $s_1$ and $s_2$, and is related to the zenith distances and azimuths at $S$ by, $$\cos \theta = \cos \tilde{z}_0 \cos \tilde{z}_M + \sin \tilde{z}_0 \sin \tilde{z}_M \cos (\tilde{A}_0 - \tilde{A}_M).$$ Here, $\tilde{z}_0$, $\tilde{A}_0$, $\tilde{z}_M$, and $\tilde{A}_M$ are the zenith distances and azimuths of the target and Moon at $S$. In Eq. (7), the various $C_{\text{exl}}$ are calculated by Eq. (4) for their respective densities. $C_{\text{exl,A}}$ is calculated assuming the same particle distribution as Rayleigh scattering. This is optimal for molecular oxygen absorption. We neglect the effect of polarization on the scattering phase function. Scattered moonlight does have some degree of polarization, similar to scattered sunlight, and has large areas of the sky with very little polarization (Horváth et al. 1998; Gál et al. 2001). The distribution of polarized light is not fully understood and would be difficult to implement. For the vertical distribution of the scattering molecules, we use the standard barometric formula, $$n(h) = n_0 \exp(-h/h_0).$$ Here, $h = \sigma - R$, the sea level density $n_0 = 2.67 \times 10^{19}$ cm$^{-3}$, and the scale height $h_0 = 7.99$ km above the Earth’s surface (Staude 1975; Bernstein et al. 2002). For the troposphere and the lower stratosphere, where most of the scattering occurs, this is a good approximation. Cerro Paranal is at an altitude of $H_0 = 2.64$ km. For the thickness of the atmosphere, we take $H_{\text{max}} = 200$ km. For Rayleigh scattering, $C_{\text{scat}} = C_{\text{exl}}$, i.e. no absorption is involved. For the height distribution of aerosols, we also use Eq. (9) with $n_0 = 1.11 \times 10^4$ cm$^{-3}$ and $h_0 = 1.2$ km. This is the tropospherical distribution of Elterman (1966). Dust, in particular soot, also absorbs radiation, thus $C_{\text{scat}}$ is lower than $C_{\text{exl}}$. We used the OMI (Ozone Monitoring Instrument) satellite data to find the median ratio of 0.97 (range between 0.90 and 0.99) for the area around Cerro Paranal (Levelt et al. 2006). The model is not very sensitive to changes in the molecular or aerosol distribution, because we scale them with the extinction curve. ### 2.4.2. Double scattering The above scattering equations only deal with single scattering. We also considered double scattering. For this, we divided the path $s_1$ into two different possible paths $s_{1,1}$ and $s_{1,2}$ where scattering and absorption can occur at the path elements $D$ (for double scattering), in between the two paths (see Fig. 5). At $D$ the scattering is treated in a similar way as in Eq. (6), but over the additional path elements. To simplify this integral and speed up the computing time, we made an approximation that allowed the integration to become an analytical expression with percent level accuracy (Eq. (4)). We assumed that the Moon coordinates where absolute coordinates with respect to the observer, such that the path $s_2 \ll R$ and the Moon is sufficiently far enough away to disregard slight angular shifts to the Moon coordinates for the different scattering points $S$ and $D$. Near the horizon and when scattering occurs at a far distance from the telescope, this assumption breaks down. However, most of the scattering occurs in the lower troposphere and a few km from the observer (with mean distances of $\sim 10$ km). The calculation was performed along a grid, which exponentially grows with distance. Where $\theta$ is small and therefore the contribution from the forward scattering peak of aerosols is high, we increased the resolution of the grid. For the higher order scattering, including double scattering, absorption by molecules via an effective airmass was no longer considered. In the case of double scattering, scattering can also occur at the ground, labeled as point $G$ in Fig. 5. Since Cerro Paranal is located at 2635 m in a mountainous region, we took $G$ to be at a height of 2 km, although the exact height has little influence. To estimate the ground reflection, we used the A1 profile from Sutter et al. (2007) based on soil samples taken in the Chilean desert. We then scaled it to the values from OMI (Levelt et al. 2006) at the three provided wavelengths for the region around Cerro Paranal. We followed the procedure of Kleipool et al. (2008) to obtain a proper average ground reflection, since there are many errors associated with this calculation. This gave us a wavelength dependent ground reflection at our desired location, as shown in Fig. 6. Double scattering is the sum of two different single scattering events, which can be off of molecules, aerosols, or the ground. We sum the contributions from the three different types of scattering to arrive at a final double scattering intensity, called $I_{DS}$. Because the moon model was intended for the ETC, it needed to be computationally quick and the exact position of the Moon is irrelevant. For this purpose, we made additional approximations for $I_{DS}$. We computed weighted averages over both the zenith distance to the target $z_0$ and to the Moon $z_M$. The weighting for $z_0$ was based on all the sky spectra from Patat (2008) and for $z_M$ from theoretical modeling of the positions of the Moon. $I_{DS}$ is then only a function of the angular separation between the Moon and target $\rho$ and the optical depth $\tau$. The average uncertainty for this simplification is on the order of 5%, and is higher at large zenith distances but quickly decreases towards smaller zenith distances. ### 2.4.3. Multiple scattering We now consider the intensity for multiple scattering, anything higher than two, called $I_{MS}$. The intensity for single scattering will be labeled as $I_{SS}$. For $I_{MS}$, we compared the contributions for $I_{DS}$ with $I_{SS}$. We assumed that the ratio between $I_{DS}$ and $I_{SS}$ would be the same for each consecutive order of scattering. We then summed over the geometric series and multiplied it by $I_{DS}$, $$I_{MS} = I_{DS} \left( \frac{1}{1 - I_{DS}/I_{SS}} - 1 \right). \quad (10)$$ We imposed an upper limit for $I_{DS}/I_{SS}$ of 0.9, so $I_{MS}$ would not diverge. Similar to the weighted averages for $I_{DS}$, we calculated a weighted average over the difference between the azimuths for the target and Moon $A_0 - A_M$, which were weighted evenly. With Eq. (8), this gave a weighted average over the angular separation $\rho$. In the end, $I_{MS}$ is only a function of optical depth $\tau$, which can be related to wavelength. The total scattering intensity for the Moon $I_{tot}$ is simply the sum of the contributions from $I_{SS}$, $I_{DS}$, and $I_{MS}$. With the weighting over the various parameters, we created a table of correction factors $f$ for the added amount of scattering due to $I_{DS}(\tau, \rho)$ and $I_{MS}(\tau)$, $$I_{tot}(\tau, \rho, z_0, z_M) = I_{SS}(\tau, \rho, z_0, z_M)f(\tau, \rho). \quad (11)$$ The higher order scattering, $I_{DS}$ and $I_{MS}$, typically contribute only a few percent and is strongest at high optical depth. Some typical scattering curves are shown in Fig. 7 for three different wavelengths. At the redder wavelengths, Mie scattering is dominating and the scattering function approaches an exponential. At bluer wavelengths, the characteristic Rayleigh scattering phase function is the most influential. We have compared our scattering intensities $I_{tot}$ with a radiative transfer code called libRadtran (Mayer & Kylling 2005) for a grid of zenith distances of the Moon and target, and the angular separation $\rho$ in $10^{-5}$ steps for many different optical depths (corresponding to the full optical wavelength range). We have made this comparison to check the validity and accuracy of our scattering code. libRadtran is a widely used third party code for calculating radiative transfer in the atmosphere. It has several different numerical solvers, including the default which uses the plane parallel approximation. With this approximation, libRadtran is not accurate when either the source (Moon) or target are near the horizon. For all of the cases, 85% of $I_{tot}$ from our code and libRadtran agreed within a relative error of 20%. For $I_{tot}$ with zenith angles less than 70°, 75% agreed to within 10% relative error. The comparison between our code and libRadtran is shown in Fig. 8. With zenith angles ≤50°, there is very good agreement. As expected, at higher zenith angles where libRadtran is no longer accurate, the two codes diverge. In the next three subsections, we will describe how Rayleigh, Mie and absorption are treated within the context of the scattering and absorption equations. ### 2.5. Rayleigh scattering The scattering off of molecules in the Earth’s atmosphere can be well described by Rayleigh scattering, which assumes the particles are much smaller than the wavelength. Rayleigh scattering is characterized with having a steep dependence on wavelength $\sim \lambda^{-4}$ and at Cerro Paranal it is quite stable (Noll et al. 2012). For the extinction due to Rayleigh scattering we use the following parametrization from Liou (2002), $$\tau_R(\lambda) = \frac{P}{1013}(8.6 \times 10^{-3} + 6.5 \times 10^{-6}H)\lambda^{-(3.9+0.0741+0.050/\lambda)}. \quad (12)$$ Here, $\lambda$ is in $\mu m$, and $P$ is the pressure at a height $H$ of the observer, which we take to be 744 ± 1.5 hPa and 2.64 km, respectively (Noll et al. 2012). The phase function for Rayleigh scattering is also well defined and is given by, \[ P(\theta) = \frac{3}{4}(1 + \cos^2 \theta), \] where $ \theta $ is the scattering angle. The Rayleigh scattering is taken at each path element $ S $ and $ D $, and is a well described and stable component of the model. ### 2.6. Mie scattering For the aerosol scattering, we have gathered information from several different sources and tried to build the most physical description. We take the Ångström Law fit derived by Patat et al. (2011) for Cerro Paranal in the optical wavelengths. We then decomposed this fit into several aerosol types given by Warnecke & Williams (2012), and produced the phase function using a Mie scattering code for log normal distributions based on Bohren & Huffman (1983). A typical way to describe the optical depths $ \tau $ and wavelength dependence $ \lambda $ of aerosols in the optical is with the Ångström Law, $ \tau = \beta \lambda^{-\alpha} $, where $ \alpha $ and $ \beta $ are the fitting parameters (Ångström 1929). The fit by Patat et al. (2011) is based on 600 observations of 8 different spectrophotometric standard stars taken with FORS1. The data were taken within a short time span of 6 months in 2009, so there could be some unknown, long-time variations. They compared the leftover aerosol signature with what was predicted by the radiative transfer codeLBLRTM (Line-By-Line Radiative Transfer Model; Clough et al. 2005). The tropospheric aerosols (height $<10$ km) needed to be scaled down by 25% in LBLRTM to match the observations. The optical depth at Cerro Paranal is very low and the aerosols are almost purely background aerosols, with no urban component (Patat et al. 2011). In the optical, their fit is given by, \[ k_{\text{aer}} = k_0 \lambda^a, \] where $ k_0 = 0.013 \pm 0.002 $ and $ a = -1.38 \pm 0.06 $ mag airmass$^{-1}$, valid from 0.4 to 0.8 $ \mu $m. We use the best fit $ k_0 $ value quoted in Patat et al. (2011) of 0.014 mag airmass$^{-1}$. The aerosol component can fluctuate significantly, as can be seen in data taken at Mauna Kea (Buton et al. 2013). They have the Ångström exponent of $ a = -1.3 \pm 1.4 $ and state that at 330 nm it can vary as much as 0.4 mag airmass$^{-1}$ during photometric nights. We are interested in extending the scattered moonlight model into the infrared, as well as wanting a more physical basis for the aerosol extinction to more accurately calculate the phase function. Therefore, we decomposed the Ångström Law. We used the size distributions given in Warnecke & Williams (2012) for the remote continental tropospheric and stratospheric aerosols. The tropospheric aerosols are split into three types: nucleation, accumulation, and coarse modes. Nucleation consists of newly produced particles from either direct emissions of combustion products or gas-phase condensation reactions. Accumulation is from coagulation of nucleation particles, condensation of products from the gas-phase chemical reactions onto particles, and chemical reactions in the aqueous phase of clouds. The mass of the coarse mode comes from mineral dust, sea salt, and biogenic material. The stratospheric aerosols are in a layer around 25 km above sea level and consist mostly of sulfuric acid particles with an admixture of nitrosyl sulfates and solid granules containing silicates (Warnecke & Williams 2012). We use a log normal distribution given by, \[ \frac{\mathrm{d}N(r)}{\mathrm{d}\log r} = \frac{1}{\sqrt{2\pi} \log s} n \exp \left[ -\frac{(\log r/R)^2}{2(\log s)^2} \right]. \] $ N $ is the cumulative number density distribution in particles cm$^{-3}$, and $ r $ is the particle radius in $ \mu $m. Table 2 provides the values for the parameters $ n $, $ R $, and $ \log s $ of the various aerosol modes, which represent the number of particles, average radius, and a distribution parameter, respectively. To calculate the extinction curve for each aerosol type we used a Mie scattering IDL code for log normal distributions written by G. Thomas$^4$ based on Bohren & Huffman (1983) for single particles (Grainier et al. 2004). With the extinction curves for the various aerosol types along with the fit for Cerro Paranal, we determined how much each curve needed to be scaled to match the overall fit. However, there are degeneracies. The refractive index $ N $ is not known for these background aerosols, so we varied it between 1.3 and 1.5. Varying $ N $ changes the scaling of the different aerosol components needed to match the fit. There is also a degeneracy between two of the aerosol types, remote continental tropospheric coarse mode and stratospheric, because their optical extinction curves are similar. In Fig. 9a we have plotted the various Mie extinction curves. The extinction from the different aerosol types with a range in $ N $ between 1.3 and 1.5 are shown, along with the total curve that is consistent with the observed one. It can easily be seen that the tropospheric nucleation mode has little to no --- Table 2. Aerosol modes. \| Type \| $ n $ cm$^{-3} $ \| $ R $ 10$^{-1} $ $ \mu $m \| $ \log s $ 10$^{-1} $ \| \|-----------------------\|------------------------\|-----------------------------------\|-----------------------------\| \| Trop nucleation \| $ 3.20 \times 10^3 $ \| 0.10 \| 1.61 \| \| Trop accumulation \| $ 2.90 \times 10^3 $ \| 0.58 \| 2.17 \| \| Trop coarse modes \| $ 3.00 \times 10^{-1} $ \| 9.00 \| 3.80 \| \| Stratospheric \| $ 4.49 \times 10^0 $ \| 2.17 \| 2.48 \| Notes. The values used for Mie scattering of remote continental aerosols from Warnecke & Williams (2012) (Sect. 2.6). --- $^4$ [http://www.atm.ox.ac.uk/code/mie/mie_lognormal.html](http://www.atm.ox.ac.uk/code/mie/mie_lognormal.html) effect, whereas the tropospheric accumulation has the greatest influence. With the current FORS1 data set (Patat 2008), there is little to no difference in the overall results between $N = 1.3$ or 1.5 and the various amounts of the different aerosol components. We chose $N = 1.5$, with 45% tropospheric accumulation, 5% tropospheric coarse modes, and 100% stratospheric aerosols, where the total matches the Patat et al. (2011) fit. The tropospheric nucleation contributed extremely little to the overall aerosol scattering, and so was left at 100%. Since Cerro Paranal is located at 2635 m, we lowered the amount of tropospheric aerosols and kept all the stratospheric particles. With the log normal Mie scattering code we also calculated the Mie phase function of the different aerosol components and the overall Mie phase function. This method of determining the phase function gives a realistic forward scattering peak, but can poorly approximate the back scatter. This is caused by the fact that the aerosol particles are not spherically symmetric, and therefore the scattering phase function tends to flatten out at higher angles, rather than dip to a minimum as in the Mie approximation (e.g. Horvath et al. 2006). Since the forward scattering dominates the back scatter by a few orders of magnitude, we prefer the Mie scattering phase function over the Henyey-Greenstein (Henyey & Greenstein 1941) approximation, which depends on the average asymmetry parameter of the aerosols. Additionally, the majority of the stratospheric aerosols in this region are sulfates, which are in droplets, and so are fairly round. The amount of coarse mode aerosol is low, which is the least round type of aerosol. This will also provide another method to break the degeneracies amongst the various aerosol modes. Finally, in astronomical observations it is rare to have the target and Moon at angular distances > 120°, where the back scatter peak becomes important. In Fig. 9b, the phase functions for the different aerosol types are plotted along with the total Mie phase functions. Once again each mode is shown with a range of $N$ between 1.3 and 1.5. ### 2.7. Absorption The absorption is calculated using the radiative transfer code LBLRTM (Clough et al. 2005) with a merged atmospheric profile for Cerro Paranal and the effective airmass $X_{\text{eff}}$ at point $O$ calculated from the single scattering at $S$. LBLRTM is a widely used radiative transfer code in atmospheric sciences and it uses the atomic line database HITRAN (High-Resolution TRANsmission; Rothman et al. 2009). This gives the amount of absorption from the various atmospheric molecules. It requires an atmospheric profile, and we use a merged one, which combines the temperature, pressure, and chemical composition as a function of height from three different sources. We combined MIPAS (Michelson Interferometer for Passive Atmospheric Sounding) equatorial profile (see Seifahrt et al. 2010) with one from GDAS (Global Data Assimilation System)$^5$, and finally scale the quantities of the lower atmosphere to the local meteorological data taken at Cerro Paranal. The MIPAS profile contains height information on 30 molecular species. The GDAS database, which is based on measurements, provides modeled profiles of temperature, pressure, and the relative humidity up to a height of about 26 km on a 3 h basis with a spatial grid of $1^\circ \times 1^\circ$. The local meteorological data from the ESO MeteoMonitor$^6$ gives the local temperature, relative humidity, and pressure. We currently use bimonthly mean, merged profiles for all the nights. For each single scattering point $S$, an effective airmass $X_{\text{eff}}$ relative to the direct path from the zenith to observer is calculated to scale the amount of absorption due to the given geometry of the Moon and target. We do not calculate $X_{\text{eff}}$ from the higher order scattering, since this contribution is quite small. $X_{\text{eff}}$ then modifies the transmission from the moonlight as calculated by LBLRTM. For determining $X_{\text{eff}}$, we compare the intensities with and without extinction, $I^{(+)}$ and $I^{(-)}$, respectively. We first look at a specific path element for a given $S$. We find the column density of that path element for both Rayleigh $B_{n,R}$ and Mie scattering $B_{n,M}$ (similar to Eq. (4)). Then with $C_{\text{scat}}$ and phase function $P(\theta)$, we calculate $I^{(-)}$ by, $$I^{(-)} = \sum_{n}^{n} I_{n}^{(-)},$$ where, $$I_{n}^{(-)} = C_{\text{scat},R} B_{n,R} P_{R}(\theta_{n}) + C_{\text{scat},M} B_{n,M} P_{M}(\theta_{n}).$$ $^5$ [http://ready.arl.noaa.gov/gdas1.php](http://ready.arl.noaa.gov/gdas1.php) $^6$ [http://archive.eso.org/asm/ambient-server](http://archive.eso.org/asm/ambient-server) To calculate $I^{(+)}$, we use $I^{(-)}$ and $\tau$ from Eq. (7), and is given by, $$I^{(+)} = \sum_{n=1}^{n} I_n^{(-)} \exp(-\tau_n).$$ Here, $\tau_n$ is given at a reference wavelength, where the absorption is strong and the scattering is weak. Within this constraint, we arbitrarily choose a wavelength of 1 $\mu$m for the scattering at the zenith, which gives $\tau_{R+M} = 0.02$. Then with $I^{(+)}$ and $I^{(-)}$ along with the optical depth at zenith $\tau_0$, $X_{\text{eff}}$ is, $$X_{\text{eff}} = \frac{\log(I^{(-)}/I^{(+)})}{\tau_0}.$$ Here, $\tau_0$ is the reference $\tau_{R+M} = 0.02$ plus $\tau_A$. Finally, the scaled scattering intensity $I_{\text{scat},A}$ for absorption is given by, $$I_{\text{scat},A} = I_{\text{scat}} \exp(-\tau_A X_{\text{eff}}).$$ The three main absorbing molecules in the optical are O$_2$, H$_2$O, and O$_3$. LBLRTM (Clough et al. 2005) with the merged atmospheric profile produces a transmission curve for the absorption. We use the particle distribution from Rayleigh scattering as an approximation for the absorbing molecules, which is optimal for O$_2$. The water vapor, however, is distributed differently from the main air molecules. The absorption from H$_2$O mostly occurs at low altitudes. Since the distribution of particles is scaled with the extinction curve, the amount of H$_2$O is fairly accurate. Changing the profile did not significantly change the resulting modeled spectrum. The ozone is treated separately using the van Rhijn formula with a height of 25 km, which is given by, $$X_{\text{o2}} = \frac{1}{\sqrt{1 - 0.992 \sin^2 z_M}}.$$ Here $X_{\text{o2}}$ is the effective airmass for ozone and $z_M$ is the zenith distance to the Moon. There are two assumptions made. The first one is that we treat the ozone only at a height of 25 km and not at a distribution of heights. The thickness can vary with season and geographic location. This assumption allows for a much simpler calculation of $X_{\text{o2}}$. The second is that we are only considering the light that is coming along the direct path from the Moon, by using the Moon’s zenith distance. We assume that the majority of the light has not been scattered prior to passing through the ozone layer. Since most of the scattering takes place in the lower troposphere, which is distinctly below 25 km, this appears to be a safe assumption. Following Patat et al. (2011), we multiply the amount of ozone given in the standard atmospheric profile by 1.08. ### 3. Comparison with FORS1 data and previous models We have compared our model with the FORS1 data (see Sect. 2.1 for details) (Patat 2008) and with the previous moon model from Noll et al. (2012) based on Krisicunas & Schaefer (1991). We found that overall, for the complete sky background model, an uncertainty of $\sigma \sim 0.2$ mag with the FORS1 data set. For the moon model, the uncertainty is $\sim 0.15$ mag. Figure 10 shows two examples of how well the model fits the observed data. The spectrum in Fig. 10a has moderate moonlight and is around 1st quarter and (b) has significant moonlight and is near full Moon. The full sky model is consistent with the observed spectrum in both scenarios. Also shown is the scattered moonlight model which is a large portion of the overall sky background flux. For (b), the model tends to slightly overestimate the amount of background light. To better understand the errors of the moon model, we made a mean and $\sigma$ spectrum of the difference between the FORS1 observations and the full sky model for all the spectra, those with moonlight, and those without moonlight (Fig. 11). The mean spectrum is centered around zero, with increased fluctuations at the redder wavelengths due to airglow emission. The $\sigma$ spectrum without moonlight is slightly lower, but all three groups tend to lie around 0.2 mag, and increase slightly at the red end. Loosely comparing the spectrum with moonlight to those without, gives a rough estimate of the quality of our advanced moon model. To better evaluate only the scattered moonlight model, we subtracted the other components of the sky background, using the sky background model, from the observations. This, in principle, should provide only an observed scattered moonlight spectrum. We then compare this observed moon-only spectrum with the scattered moonlight model. Such a comparison contains all the errors associated with the full sky background model as well as the errors from the moon model. For an estimate of the uncertainty in the other sky background components, we looked at the full sky model without moonlight. We then computed a weighted $\sigma$ to represent the uncertainty from the other sky background components, by comparing the relative flux from the moonlight with full sky background for the Moon observations used in this analysis. If there is less moonlight, then $\sigma$ from the other components should be larger, versus an observation dominated by moonlight where $\sigma$ should mostly come from the moon model. We did this comparison using all the data with a significant amount of moonlight ($> 100 \text{ phot s}^{-1} \text{ m}^{-2} \text{ \mu m}^{-1} \text{ arcsec}^{-2}$) and decent weather conditions. Also, we ignored one of the observing modes, Grism 600B, which had only 8 data points and the flux calibrations appeared to greatly deviate from the other observing modes. This left us with a total of 82 spectra. The results can be seen in Fig. 12. There is a jump in the mean around 500 nm where the majority of the observations changes from one observing mode to another. We believe this comes from errors in the flux calibration of the Patat (2008) data. For the mean of these spectra to be centered at zero, we needed to multiply the moon model by a factor of 1.2 to correct for the flux calibration. The error bars shown here are the uncertainties from the moon model squared minus the square of the weighted $\sigma$ from the other components. At blue wavelengths $\sigma$ is small, and it gets larger towards the red. The estimated weighted $\sigma$ of the other sky background components is inflated in the UV and red wavelengths. The airglow/residual continuum model probably overestimates the flux, and so the residuals, especially in the red, are large. This causes the Moon contribution to be underestimated and $\sigma$ to be inflated, so the last three continuum nodes are not shown for this reason. In Fig. 11a, this effect due to the airglow model can be seen from the downturn in the mean difference spectra. Also plotted in Fig. 12 with the same analysis is the previous moon model discussed in Noll et al. (2012), which is an extension of Krischiunas & Schaefer (1991) moon model and scaled to Patat (2008) data, labeled as KS91. The mean and $\sigma$ for this previous model are worse than our new advanced moon model. KS91 has a larger $\sigma$ at blue wavelengths and the mean is more off centered at redder wavelengths compared with this work. The mean and $\sigma$ for our new advanced moon model and the previous one, before being corrected for the errors from the other sky background components are shown in Table 3, along with the weighted $\sigma$ from the rest of the sky background. ![Fig. 11.](image) Fig. 11. a) Mean and b) $\sigma$ of the difference between the observed and modeled spectra for the full sky background of all the spectra, those with moonlight, and only those without moonlight. See Sect. 3 for more details. ![Fig. 12.](image) Fig. 12. Means and uncertainties for the scattered moonlight versus observed data at several 4 nm wide continuum bands. The $y$-axis is the average of the observed minus modeled fluxes for data with good weather conditions and a significant amount of moonlight. Over-plotted is the same analysis with the previous model from Noll et al. (2012) based on Krischiunas & Schaefer (1991), labeled as KS91. The numbers below each point are the number of spectra considered. See Sect. 3 for more details. Table 3. Narrow band filter statistics for our new moon model (new), Noll et al. (2012) moon model (old), and the weighted $\sigma$ of the other sky background components (SBC). \| Filter (nm) \| # \| $\langle \text{new} \rangle$ \| $\sigma_{\text{new}}$ \| $\langle \text{old} \rangle$ \| $\sigma_{\text{old}}$ \| $\sigma_{\text{SBC}}$ \| \|------------\|---\|-----------------\|-----------------\|-----------------\|-----------------\|-----------------\| \| 369 \| 16\| 0.259 \| 0.222 \| 0.102 \| 0.390 \| 0.190 \| \| 387 \| 16\| 0.100 \| 0.213 \| −0.038 \| 0.357 \| 0.180 \| \| 420 \| 16\| 0.137 \| 0.191 \| −0.081 \| 0.331 \| 0.107 \| \| 450 \| 62\| −0.026 \| 0.185 \| −0.155 \| 0.242 \| 0.091 \| \| 480 \| 62\| −0.025 \| 0.192 \| −0.126 \| 0.248 \| 0.101 \| \| 510 \| 62\| −0.060 \| 0.208 \| −0.136 \| 0.264 \| 0.117 \| \| 543 \| 82\| −0.076 \| 0.220 \| −0.077 \| 0.287 \| 0.170 \| \| 575 \| 82\| −0.080 \| 0.299 \| −0.021 \| 0.372 \| 0.265 \| \| 608 \| 82\| −0.106 \| 0.330 \| 0.048 \| 0.410 \| 0.290 \| \| 642 \| 66\| −0.150 \| 0.314 \| 0.130 \| 0.374 \| 0.272 \| \| 675 \| 66\| −0.143 \| 0.326 \| 0.193 \| 0.384 \| 0.283 \| \| 720 \| 66\| −0.084 \| 0.364 \| 0.269 \| 0.413 \| 0.433 \| \| 820 \| 46\| −0.152 \| 0.676 \| 0.298 \| 0.721 \| 0.791 \| \| 872 \| 45\| 0.016 \| 0.761 \| 0.483 \| 0.818 \| 1.029 \| Notes. The listed $\sigma_{\text{new}}$ and $\sigma_{\text{old}}$ are before subtracting the uncertainties from the other sky background components. Calculates fully 3D double scattering calculations, including ground reflection, then simplifies for faster computing. Approximates multiple scattering by comparing double and single scattering. Calculates the absorption and single scattering simultaneously. Decomposes the Mie scattering Ångström Law into typical aerosol distributions for Cerro Paranal. Calculates the Mie phase function based on the aerosols found by decomposing the Ångström Law. Provides a scattered moonlight spectrum, not just photometric magnitudes (Walker 1987; Krisciunas & Schaefer 1991). The uncertainties across the entire optical range are much lower than previous (Krisciunas & Schaefer 1991; Noll et al. 2012). Acknowledgements. This study is carried out in the framework of the Austrian ESO In-kind program funded by BM:wf under contracts BMWF-10.490/0009-II/10/2009 and BMWF-10.490/0008-II/3/2011. References Ångström, A. 1929, Geografiska Annaler, 11, 156 Bernstein, R. A., Freedman, W. L., & Madore, B. F. 2002, ApJ, 571, 56 Bohren, C. F., & Huffman, D. R. 1983, Absorption and scattering of light by small particles (New York: Wiley) Buton, C., Copin, Y., Aldering, G., et al. 2013, A&A, 549, A8 Clough, S. A., Shephard, M. W., Mlawer, E. J., et al. 2005, J. Quant. Spec. Radiat. Transf., 91, 235 Colina, L., Böhlke, R. C., & Castelli, F. 1996, AJ, 112, 307 Davies, T. W., Bennie, J., Inger, R., & Gaston, K. J. 2013, Nature Sci. Rep., 3, 22 Dollfus, A., & Bowell, E. 1971, A&A, 10, 29 Elterman, L. 1966, Appl. Opt., 5, 1769 Gál, J., Horváth, G., Barra, A., & Wehner, R. 2001, J. Geophys. Res., 106, 22647 Grainger, R. G., Lucas, J., Thomas, G. E., & Ewen, G. B. L. 2004, Appl. Opt., 43, 5386 Henyey, L. G., & Greenstein, J. L. 1941, ApJ, 93, 70 Horváth, G., Gál, J., Pomozzi, I., & Wehner, R. 1998, Naturwissenschaften, 85, 333 Horvath, H., Kasahara, M., Tohno, S., & Kocifaj, M. 2006, Aerosol Science, 37, 1287 Kieffer, H. H., & Stone, T. C. 2005, AJ, 129, 2887 Kleipool, Q. L., Dobber, M. R., de Haan, J. F., & Levelt, P. F. 2008, J. Geophys. Res. (Atmospheres), 113, 18308 Knoetig, M. L., Biland, A., Bretz, T., et al. 2013 [arXiv:1307.6116] Krisciunas, K., & Schaefer, B. E. 1991, PASP, 103, 1033 Lane, A. P., & Irvine, W. M. 1973, AJ, 78, 267 Leinert, C., Bowyer, S., Haikala, L. K., et al. 1998, A&AS, 127, 1 Levelt, P. F., van den Oord, G. H. J., Dobber, M. R., et al. 2006, IEEE Trans. Geosc. Remote Sens., 44, 1093 Liou, K. N. 2002, An introduction to atmospheric radiation, 2nd edn. (Academic Press) Mayer, R., & Kylling, A. 2005, Atm. Chem. Phys., 5, 1885 Noll, S., Kausch, W., Bartlen, M., et al. 2012, A&A, 543, A92 Patat, F. 2008, A&A, 471, 575 Patat, F., Moehler, S., O’Brien, K., et al. 2011, A&A, 527, A91 Rothman, L. S., Gordon, I. E., Barbe, A., et al. 2009, J. Quant. Spec. Radiat. Transf., 110, 533 Seifahrt, A., Käuß, H. U., Zängl, G., et al. 2010, A&A, 524, A11 Staude, H. J. 1975, A&A, 39, 325 Sutter, B., Dalton, J. B., Ewing, S. A., Amundson, R., & McKay, C. P. 2007, J. Geophys. Res. (Biogeosciences), 112, 4 Trinh, C. Q., Ellis, S. C., Bland-Hawthorn, J., et al. 2013, MNRAS, 432, 3262 Velikodvskiy, Y. I., Opanasenko, N. V., Akimov, L. A., et al. 2011, Icarus, 214, 30 Vernet, J., Dekker, H., D’Odorico, S., et al. 2011, A&A, 536, A105 Walker, A. 1987, NOAO Newsletter, 10, 16 Wallace, L., Hinkle, K. H., Livingston, W. C., & Davis, S. P. 2011, ApJS, 195, 6 Warneck, P., & Williams, J. 2012, The Atmospheric Chemist’s Companion., (Springer) Whitaker, E. A. 1969, NASA Sp. Publ., 201, 38 Wolstencroft, R. D., & van Breda, I. G. 1967, ApJ, 147, 255
Upstream sediment input effects on experimental dune trough scour in sediment mixtures M. G. Kleinhans Faculty of Geosciences, Department of Physical Geography, Utrecht University, Utrecht, Netherlands Received 25 May 2004; revised 30 September 2004; accepted 7 March 2005; published 19 October 2005. [1] Understanding causes of dune irregularity, especially dune trough scour, is important for the modeling of vertical sorting of sediment mixtures in morphological models of rivers with sediment mixtures. Sediment in dunes is generally sorted in a fining-upward manner, which affects the sediment transport composition depending on the scour depth distribution of the dunes. Why dunes become more irregular and develop deep scour holes in some conditions is only partially understood. Moreover, there is a feedback between vertical sorting and dune irregularity. In gravelly sands, erosion-resistant coarse layers may form that decrease or inhibit dune trough scour. The causes of dune irregularity and the feedback by coarse sediment layers are explored in experiments and are demonstrated to be related partly to the upstream sediment boundary condition determined by the experimental setup: sediment feeding or recirculating. Sediment recirculation flumes promote fining of the transported sediment and formation of a less or immobile gravel lag layer in the dune troughs. Sediment feed flumes may force the transported sediment at the flume entry to be equal to the bed sediment (equal mobility condition) and hence allow no lag layer formation. Experiments show that the dune trough scour in recirculation is less deep than in feed flumes, and the vertical sorting and transport sediment composition are different. The experiments also indicate that dune irregularity in feed flumes is related to dimensionless shear stress similarly as in uniform sediments. Interpretation of experiments of dune dynamics in sediment mixtures should therefore account for differences in the upstream sediment supply condition. More in general, it is hypothesized that the upstream sediment supply condition (resembling either recirculation or feed flumes) in rivers may affect dune and vertical sorting dynamics. Citation: Kleinhans, M. G. (2005), Upstream sediment input effects on experimental dune trough scour in sediment mixtures, J. Geophys. Res., 110, F04S06, doi:10.1029/2004JF000169. 1. Introduction [2] Morphological modeling of channel beds with sediment mixtures has made strong progress in the past decade. To model the sediment transport process and the exchange between transported and bed sediment mixtures in rivers with dunes correctly, the bed level variations due to migrating dunes must be accounted for [Ribberink, 1987; Piola and Borgman, 1991; Parker et al., 2000; Leclair and Bridge, 2001] as well as the vertical sediment sorting in the dunes [Parker et al., 2000; Kleinhans, 2001, 2002] (see http://www.geog.uu.nl/fg/mkleinhans). Bed level variation is caused by dune migration and hence related to dune irregularity. Following the definition of Costello and Southard [1981], regular dunes are called two-dimensional (2-D) and the irregular 3-D. This can be expressed quantitatively in the vertical direction as a probability distribution of trough depths or a probability distribution of bed level variations [Parker et al., 2000], and in the horizontal direction as the sinuosity of the dune crest lines and connectivity (or length) between crest lines. Sediment is generally sorted in dunes in a fining-upward manner, which in turn affects the sediment transport composition depending on the scour depth distribution of the dunes at a later point in time. Both the bed level variation and sediment sorting must therefore be implemented in model concepts [Parker et al., 2000; Blom, 2003], but are not yet well predictable for various flow conditions and sediments [e.g., Leclair and Blom, 2005]. [3] Dunes in uniform sediment become more irregular in horizontal and vertical directions with increasing shear stress up to the transition to upper stage plane bed where the dunes flatten again [Reineck and Singh, 1973; Boothroyd and Hubbard, 1975; Costello and Southard, 1981; Terwindt and Brouwer, 1986]. Southard and Boguchwal [1990] presented bed form stability diagrams in which the dune stability field is divided into 2-D and 3-D regimes based on experiments. Although the exact division is arbitrary, it indicates a transition from 2-D to 3-D dunes with increasing shear stress (or flow velocity at a given water depth). This diagram is also valid for gravels [Carling, 1999]. When dunes have a highly variable trough scour depth, the set thickness of the dune deposits is larger and highly... Table 1. Experimental Parameters of the Experiments \| Exp \| u, m/s \| h, m \| S, $10^{-3}$ \| D$_{50}$, $10^{-3}$ m \| D, $10^{-3}$ m \| $\tau'$, Pa \| $\theta'$ \| Temperature, °C \| Description \| \|-----\|--------\|------\|-------------\|-----------------\|----------------\|------------\|---------\|----------------\|-------------\| \| Equ1 \| 0.51 \| 0.048\| 4.1 \| 1.26 \| 28 \| 5.27 \| 1.89 \| 0.093 \| 12.2 \| Armoring, "plane" bed, repeated as Non1 \| \| Equ2 \| 0.51 \| 0.12 \| 3.6 \| 1.26 \| 28 \| 5.27 \| 3.09 \| 0.147 \| 12.2 \| 2-D dunes \| \| Equ3 \| 1.08 \| 0.209\| 3.6 \| 1.40 \| 31 \| 5.63 \| 5.03 \| 0.222 \| 11.5 \| 3-D dunes \| \| T10 \| 0.54 \| 0.16 \| 1.11 \| 1.28 \| 29 \| 9.29 \| 1.88 \| 0.091 \| 14 \| Barchans \| \| T5 \| 0.70 \| 0.23 \| 1.47 \| 1.06 \| 24 \| 9.49 \| 2.50 \| 0.146 \| 14 \| Transitional barchans to 2-D dunes \| \| T7 \| 0.80 \| 0.35 \| 1.52 \| 1.21 \| 27 \| 10.25 \| 2.89 \| 0.148 \| 14 \| 2-D dunes \| \| T9 \| 0.70 \| 0.26 \| 1.69 \| 1.39 \| 31 \| 9.44 \| 2.41 \| 0.107 \| 14 \| 2-D dunes \| Spatially variable in thickness [Harms and Farnesstock, 1965; Reineck and Singh, 1973; Paola and Borgen, 1991; Leclair and Bridge, 2001]. Since vertical sediment sorting is generated within a set [Kleinhaus, 2001], set thickness variability decreases the net vertical sorting in the bed averaged over several dune lengths and set lengths. On reentrainment of these fining-upward sets by passing dunes, the depth and variability of the trough scour determines the entrainment of the coarser, deeper sediments. [4] For dunes in sediment mixtures, two complicating factors are known. First, lower mobility or immobility of the coarser sediment leads to armoring, which in turn leads to sediment supply-limited conditions. This means that the amount of mobile sediment above the armor layer is insufficient to build dunes up to their equilibrium dimensions for the given flow and characteristics of the mobile sediment. Instead, flow parallel sand ribbons or barchan dunes develop. Barchan dunes are highly 3-D in the horizontal direction but trough scour is inhibited by the armor layer [Kleinhaus et al., 2002]. Second, during high discharge when large dunes form, coarse sediment concentrates in the troughs of the dunes. If this coarse lag layer is coarse or thick enough, a situation comparable (but not equal) to armoring might develop [Kleinhaus, 2001]. The lag layer just below the active dunes hinders dune growth and sediment entrainment from the bed below the active dunes [Klaassen, 1991], and inhibits the formation of deep scour holes that are characteristic for irregular dunes [Hooke, 1968; van der Zwaard, 1973]. Thus the trough scour depth variation, which determines lag layer formation and thus vertical sorting, is itself strongly coupled to vertical sediment sorting. This multivariate interaction has already been explored by Leclair and Blom [2005] and will be further explored herein. [5] A gravel lag layer forms when the gravel is slightly less mobile than the sand. The sediment mobility differences between grain size fractions are commonly related to a hiding exposure effect: the hiding of small grains in the lee of large grains and the exposure of large grains to the flow. A well-known mobility condition in unimodal sediments is equal mobility, in which the sediment composition of bed load (temporarily “stored” in dunes) and sediment underlying the surface sediment are equal. This implies that the critical shear stress is equal for all grain size fractions, which is attained by mobile armoring in the (nearly) plane bed case [Parker and Klingeman, 1982]. Bimodal or very widely distributed sediments, on the other hand, commonly exhibit smaller mobility of the coarser sediment [Wilcock, 1993; Kühnle, 1993]. A basic question for the plane bed or bed load sheet case is whether the composition of the bed surface armor layer changes with changing flow or with a change in composition of the upstream sediment supply. Wilcock [2001] hypothesized that the nature of the sediment supplied from upstream controls the mobility differences between grain size fractions. [6] A new hypothesis is developed here to extend Wilcock’s [2001] hypothesis to duned beds. The above mentioned mobility differences between grain size fractions strongly affect the lag layer development, which in turn determine the dune irregularity due to trough scour. In short: the nature of the sediment supplied from upstream controls dune irregularity. This is highly relevant for experimental studies of dune migration in sediment mixtures, because two fundamentally different methods for upstream sediment supply have been used: by feeding a constant sediment mixture or by recirculating the transported sediment from the tail end of the flume to the entrance (periodic boundary condition). In both flumes, the discharge, water depth and the initial slope are specified (and uniform flow can be maintained). In the sediment feed flume, the rate and composition of the sediment entering the flume is specified, while in the recirculating flume, it is determined by the (selective) transport process; thus the transport rate and composition is a dependent parameter [Parker and Wilcock, 1993]. [7] The aims of this paper are to investigate experimentally whether (1) dune trough scour irregularity in sediment mixtures without lag layers is similarly related to the shear stress magnitude as in uniform sediment and (2) the upstream sediment supply has an effect on lag layer formation and therefore on dune trough scour. 2. Methods 2.1. Methodology [8] The difference between the two types of experimental setups is employed to contrast two cases: one with equal mobility (feed flume) and one with higher mobility of finer sediment (recirculation flume). Low and high shear stress equilibrium experiments were done in a feed flume to test whether the 2-D versus 3-D bed form irregularity is related to shear stress as it is in uniform sediment. The use of the feed flume with unimodal sediment ensures that no gravel lag is formed, so the effect of shear stress on dune irregularity can be isolated from the effect of gravel lags on dune irregularity. A second set of experiments was done in a recirculating flume to demonstrate the effect of a gravel lag (in addition to shear stress) on dune irregularity and to compare this condition to one without a gravel lag. [9] Summary parameters are given in Table 1. The depth-averaged flow velocity ($u$) was determined by calibrated flow discharge measurement, the width of the flume ($w$) and the average water depth ($h$) along the measurement section as $u = Q/(wh)$. Subcritical flow ($Fr < 0.84$) was maintained, with $ \text{Fr} = u/(gh)^{0.5} $ ($ g $ = gravitational acceleration). From the slope (S) and the hydraulic radius ($ R_c $) the total shear stress ($ \tau $) was determined with $ \tau = \rho gR_c S $ ($ \rho $ = density of water). The hydraulic radius was corrected for sidewall roughness with the method of Vanoni-Brooks. The dimensionless shear stress on grains can be computed as $ \sigma_{50} = \tau'/(p_s - \rho)gD_{50} $, in which $ p_s $ = density (of sediment), $ g $ = gravitational acceleration and $ D $ = sediment diameter (50th percentile), $ \tau' $ = shear stress on the grains, computed as $ \tau' = \rho g[u/C']^2 $ in which $ u $ is the depth-averaged flow velocity, $ C' $ is the grain-related Chézy coefficient: $ C' = 180\log[12R_c/k'_s] $, with $ k'_s $ the grain roughness, assumed to be equal to the $ D_{50} $ of the sediment mixture as installed in the flumes. By assuming grain shear stress, the effect of bed forms on roughness and shear stress is removed. The median grain size is made dimensionless with $ D^ = D_{50}/[(p_s - \rho)g/(\nu \rho)]^{1/3} $ with $ \nu $ the kinematic viscosity of water ($ \sim 1.2 \times 10^{-6} \text{ m}^2 \text{ s}^{-1} $). ### 2.2. Setup of Experiments in a Sediment Feed Flume [10] Three feed flume experiments were done in the Tilting Bed Flume at St. Anthony Falls Laboratory, which has a length of 14 m and a width of 0.9 m [Kleinhaus, 2002] (Figure 1). Equ1 was a mobile armoring experiment for reference to the plane bed condition and Equ2 and Equ3 had dunes (Figure 2). Uniform flow was maintained by adjusting the downstream weir. The test section was between 4 and 13 m from the sediment feeder. Time series of the bed level at a fixed point (11 m from flume entrance) and water and bed surface profiles were collected in the middle of the flume with an ultrasonic device. The flow was maintained until the system was in equilibrium, which was defined as the condition at which the variations of bed form dimensions, sediment transport and average bed level change became smaller than the measurement accuracy and variability. The bed was remixed after each experiment. It should be noted that the flume was too short to attain equilibrium of the dune height for the largest flow depth; the dune height increased toward the downstream end of the flume as in most experiments with dunes reported in literature. The sediment was a log-normally distributed unimodal sediment (Figure 3), installed at a bed slope equal to that of the flume. Sediment of the same composition as the initial bed sediment was fed into the upstream end of the flume by a rising platform of 0.9 m wide and long, which rose at a constant and adjustable speed. Helley Smith measurements (ratio of nozzle exit area to entrance area 1:1.0, bag of 100 $ \mu \text{m} $ mesh size) were done at 13 m to check whether the transport rate and composition were equal to that of the feeder. In equilibrium conditions, the Helley Smith gave a mean transport rate within a few percent of the feed rate. The composition of transported sediment as measured with the Helley Smith is reported herein. Suswhile T7 and T9 had dunes. The sediment was recirculated with sediment pumps. Suspended load transport was negligible. Bed and water surface profiles along the flume and bed load transport were automatically collected. The experiments were started with a mixed bed of slightly bimodal sediment, installed at a bed slope that is equal to the expected water surface slope of the experiments. The flow was maintained until the system was in equilibrium (T10 and T5). The next step on the bed of T5 was to generate a flow with a higher bed shear stress until a new equilibrium was reached (T7), and then again lower (T9) without remixing the bed. Samples have been taken from the transported sediment that was measured automatically in the recirculation system. 3. Results 3.1. Sediment Feed Flume [12] Near the end of the Equ2 and Equ3 experiments (Figure 2), the composition of the transported sediment is almost equal to that of the original bed sediment and feeder sediment (Figures 3a and 3b). This shows that all the grain sizes were in motion in almost the same abundance as they occur in the bed sediment, approximating equal mobility with increasing discharge as expected. Small differences in fine grain sizes are due to suspended sediment not sampled in the Helley Smith. [13] The dunes in Equ2 (lower shear stress) were much more regular than in Equ3 (high shear stress), in the sense that the latter had more curved crest lines and deep scour pits every now and then, while the regular ones were more straight crested and did not have pronounced scour pits (Figure 2). The dunes in Equ3 are about twice as high as in Equ2. From the time series taken at a single position, probability distributions of the bed level were computed (Figure 4a). These distributions indicate the variations of bed level due to passing dunes, normalized with respect to the average bed level at that position in that time period, and made dimensionless with water depth (Figure 4c). The Equ3 experiment has a skewed distribution with much deeper dune troughs than dune tops, while the distribution for Equ2 is about symmetrical. The narrow, symmetrical distribution of Equ1 is solely due to grain dynamics. 3.2. Sediment Recirculation Flume [14] Experiments T10 and T5 had barchan dunes whereas T7 and T9 had slightly irregular 2-D dunes. There were no trough scour pits as was found after making trenches along the flume; the gravel lag was continuous and nearly plane. The bed load composition in all experiments is finer than the original bed sediment (Figures 3c and 3d), but the finest in T9 because most gravel was worked downward in T7 and was no longer entrained in the shallower dune troughs in T9. Consequently, the bed load compositions of T5 and T9 are different even though the shear stresses were approximately equal. The bed level distributions were computed from the downstream half of a number of profiles along the flume (in equilibrium conditions) after linear detrending of the profile. Comparison of T5, T7 and T9 reveals a close resemblance of T7 and T9 (Figure 4b) even though the shear stress in T7 is larger than in T9. So, relative to T5, the trough scour depth in T9 is much more... Figure 3. Comparison between grain size distributions of the bed sediment (equal to the feed sediment) and the bed load sediment at the end of the feed flume experiments and between the bed sediment and recirculated sediment of the recirculation flume experiments. The bed load sediments of (a, b) Equ2 and Equ3 approximate the bed sediments, demonstrating (near) equal mobility, whereas the bed load sediment of (c, d) T5, T7, T9 and T10 is much finer than the bed sediment, demonstrating higher mobility of the finer sediment. pronounced, although still fairly regular. The reason is that in T5 and T7 there was a gravel layer beneath the troughs which hindered the trough scour, whereas in T9 the gravel had been worked down too deep for the dunes to be affected. See Kleinhans et al. [2002] and Blom et al. [2003] for more extensive descriptions and vertical sorting data. 4. Discussion 4.1. Causes of Dune Irregularity [18] From the feed flume experiments, in which lag layer formation is not allowed, it can be concluded that the transition from 2-D to 3-D dunes in nonuniform sediments occurs with increasing shear stress similarly to uniform sediments. The probability distributions of bed levels for 3-D dunes (Equ3) at high shear stress have a much longer tail down into the bed than for 2-D dunes (Equ2) in lower shear stress (Figure 4). The choice for a grain size parameter in the dimensionless shear stress parameter is discussed by Kleinhans et al. [2002]. The effect of choosing grain size percentiles from bed sediment, transported sediment or lag layer sediment has a large effect on the magnitude of the shear stress, but the trends between experiments are the same except for T9. The (dimensionless) shear stress for T9 given in Table 1 is probably not correct as the chosen sediment diameter refers to the original sediment rather than the finer sediment activated in T9 by the dunes over time. [16] The dunes in the Equ experiments become irregular at high shear stress as expected from literature. To further demonstrate the relation between three-dimensionality of the gravelly sand dunes and flow conditions, the available experiments reported in this paper and from literature are plotted in the bed form stability diagram of Southard and Boguchwal [1990, Figure 5] (Figure 5). This diagram is based on flow velocity rather than shear stress, but the limited depth range for which it is given means that a diagram with (dimensionless) grain shear stress gives similar results. The result for the diagram of van den Berg and van Gelder [1993] (based on dimensionless grain shear stress and $D^$) is indeed similar, but is not shown here because van den Berg and van Gelder [1993] never plotted the transition from 2-D to 3-D dunes in their diagram. The data follow the predicted increase of three-dimensionality in the diagram to some extent, even though the dunes have been classified somewhat arbitrarily in two- and three-dimensional and larger dune height relative to the flume width may obscure three-dimensionality in some experiments. However, the 29SAFL experiment deviates from the results and the reason for this is unclear and probably related to parameters not given in the diagram. Unfortunately no data are available for 2-D dunes in sediment mixtures with grain sizes of 0.4—0.7 mm. 4.2. Mixture Feedbacks on Dune Irregularity [17] While the results in the feed flume indicated that dune irregularity depends on shear stress in the same way as for uniform sediment when the formation of coarse layers is inhibited, the recirculation flume experiments demonstrated the two ways in which such a coarse layer affects dune irregularity. [18] First, in extreme cases (very wide mixtures, low shear stresses and fine sediment supply), isolated barchan dunes migrate over a stable armor layer [Kleinhans et al., 2002], and the trough-scouring flow is ineffective (experiment T10). Thus the exchange of sediment between the bed load and substrate sediment through the armor layer is insignificant. The dunes are “irregular” in planform (barchanoid) but do not have widely varying scour depths. In this case the mixture feedback on trough scour-related irregularity is strong and outweighs the shear stress control on (vertical) irregularity. Moreover, the irregularity is completely controlled by either the bimodality of the sediment and/or the upstream sediment supply (recirculation) which is much finer than the bed sediment, both resulting in the low mobility of coarse sediment. [19] Second, for the same initial combination of slope, discharge and water depth and sediment mixture properties the feed and recirculation flumes give different results. Starting with a fully mixed bed and a relatively low shear stress, in a recirculating flume only the finer sediment is entrained and migrates over the immobile coarse sediment. The sediment entering the flume is this same fine sediment. The result is therefore fine mobile sediment (possibly barchan dunes) migrating over immobile coarse gravel (lag deposit), which is a strong deviation from equal mobility. In a feed flume the coarse fractions in the feed sediment cannot be transported and therefore are deposited in the upstream part of the flume. This leads to an increase in bed slope (and, maintaining uniform flow, also water surface slope), and consequently to an increase of the bed shear stress, until the coarser sediment is transported as well. The feed system is eventually forced (in equilibrium) to transport all the sediment that is fed in. At moderate discharge, this equal mobility condition is only attained when a mobile armor layer is formed [Parker and Klingeman, 1982]. So, in a recirculation flume or with a fine upstream sediment supply or without sediment supply at all, the armor layer becomes stable whereas in a feed flume an armor layer is mobile. A mobile armor layer is able to adapt to changing conditions, whereas the stable armor layer usually needs a shear stress above a high critical value before it is broken up. [20] Equivalently for the duned bed, the coarse layer formed at the base of dunes can be characterized as mobile or stable. In stable coarse layer-forming conditions when the sediment supply is limited and/or finer than the substrate (recirculating flume), the immobility of the lag layer will outweigh the tendency of dunes to scour their troughs, and Figure 5. Bed form stability diagram of Southard and Boguchwal [1990] from their Figures 10.3 and 10.8 for water depths of $0.1–0.4$ m. Open symbols represent 3-D dunes, and closed symbols represent 2-D dunes. The transition from dunes to antidunes occurs above Froude numbers of 0.84, which is given for two water depths in the graph. The experiments plotted are Equ1–3 (feed flume) of Kleinhans [2002], the barchan dunes of A1 and B1, 2-D dunes of A2 and 3-D dunes of B2 (recirculation flume) from Blom et al. [2003], lower plane bed of T0, barchan dunes of T5 and T10 and 2-D dunes of T7 and T9 (recirculation flume) from Kleinhans et al. [2002] and Blom et al. [2003], 3-D dunes of BU14 and 29SAFL (recirculation flume) from Leclair and Blom [2005], 3-D dunes in uniform sand runs UUA and UUB (recirculating flume) from J. H. van den Berg and I. van Enckevort (personal communication, 2003), and 3-D river dunes (runs 1–3) of Nordin [1971] with uniform sediment. All these data were collected in water depths of $0.15–0.35$ m. In extreme cases the dunes remain isolated barchanoids. In “mobile lag layers” (feed flume) dunes will be able to scour their troughs according to the tendency to become more irregular, and there barely is a feedback of a gravel lag on dune trough scour. In other words, the mobility of the coarse sediment in the trough zone is not only a function of the shear stress and the trough depth variation, but also of the composition of upstream supplied sediment relative to the sediment in the (active) bed as hypothesized. This extends the findings of Wilcock [2001] for plane beds and Leclair and Blom [2005] for duned beds, and must be taken into account when experimental bed level probability distributions from sediment feed and recirculation setups are compared and interpreted with the aim to apply the resulting model to rivers. For instance, Leclair and Blom [2005] compared the probability distributions of the bed surface levels for the experiments A1 and A2 of Blom et al. [2003] and 29SAFL of Leclair and Bridge [2001], both in a recirculation flume, of which the former were done with a wide sand-gravel mixture while the latter was done with a narrow sand mixture. The wide mixture had a more or less symmetric probability distribution, while the narrow mixture had a long tail from the deep irregular scour. Leclair and Blom [2005] attributed the difference between the 3-D dunes in the narrow mixture (29SAFL) and the 2-D dunes in the wide mixture (A1, A2) to the presence of coarse sediment in the trough zone which is less mobile due to the low dimensionless shear stress. In the light of the analysis above, the cause may also have been a combination of both low sediment mobility and lag formation, where the lag formation is due to the recirculating flumes of A1,A2 rather than the sediment mobility. This suggests that an experiment similar to A2 in a feed flume would give deeper dune trough scours in the absence of a lag layer. The differences between upstream sediment boundary conditions may also be relevant for rivers and modeling in general. The feed flume likely represents field conditions with equilibrium sediment transport and sediment composition best because the gravel is not worked down irreversibly below the active layer (also see Parker and Wilcock [1993] for discussion). The recirculating flume, on the other hand, may better represent bed sediment dynamics during floods, where hysteresis in armor layer and dune development cause the vertical sorting and composition of sediment in transport to be different before and after the flood peak [Klaassen, 1991; Kleinhans, 2001]. Bimodal sediments commonly show nonequal mobility [e.g., Wilcock, 1993], which, as an upstream sediment input condition similar to a recirculating flume, might cause lag layer formation or downstream fining [Paola et al., 1992] more often than unimodal sediments. 5. Conclusions The two common experimental setups of sediment feed and recirculation represent different, extreme cases of upstream sediment supply conditions compared to natural streams. Sediment recirculation flumes promote the formation of lag layers, contrary to feed flumes. The dune irregularity (expressed as a probability distribution of bed levels) in the absence of coarse sediment layer formation as in feed flumes is related to dimensionless shear stress similarly as uniform sediment. A negative feedback by vertical sorting (coarse layer formation) is likely to occur in mixtures with nonequal mobility, as in conditions where the sediment supplied upstream is finer than the substrate as in recirculation flumes. This must be taken into account in the use of experimental data of both vertical sorting and dune irregularity for future models. Notation* - $C$ Chézy roughness coefficient ($\text{m}^{0.5} \text{s}^{-1}$). - $D$ sediment diameter (m). - $D^$ dimensionless sediment diameter. - $g$ gravitational acceleration (9.81 m s$^{-2}$). - $h$ water depth. - $k_a$ Nikuradse equivalent sand roughness (m). - $R_e$ hydraulic radius corrected for sidewall roughness (m). - $S$ slope. - $u$ depth-averaged flow velocity (m s$^{-1}$). - $\tau$ dimensionless shear stress (Shields parameter). - $\rho$ density of water (kg m$^{-3}$). - $\rho_s$ density of sediment (kg m$^{-3}$). - $\tau_t$ total shear stress (including form and grain drag) (N m$^{-2}$). - $\nu$ kinematic viscosity (m$^2$ s$^{-1}$). Subscripts and superscripts* - $f$ referring to skin friction, grain roughness. - 50 50% percentile. - 90 90% percentile. --- [24] Acknowledgments. Astrid Blom, Suzanne Leclair, Gary Parker, Chris Paola, John Bridge, Pim van Santen, and Roy Frings are thanked for ongoing discussion on the complex interaction between dunes and sorting. Gary Parker and Jeff Marr and coworkers (SAFL) are thanked for their help with the field experiments and discussions about field and recirculation flames. Suzanne Hulscher and coworkers are acknowledged for organizing the MARID conference in Enschede at which this work was presented. The investigations in the feed flume were in part supported by the Netherlands Earth and Life Sciences Foundation (ALW) with funds from the Netherlands Organization for Scientific Research (NWO), and by the St Anthony Falls Laboratory, Minneapolis MN, USA. The constructive reviews by Suzanne Leclair and Enrico Foti were helpful and well appreciated. References Blom, A. (2003), A continuum vertical sorting model for rivers with non-uniform sediment and dunes, Ph.D. thesis, Univ. of Twente, Enschede, Netherlands. Blom, A., J. S. Ribberink, and H. J. de Vriend (2003), Vertical sorting in bed forms: Flume experiments with a natural and a trimodal sediment mixture, J. Geophys. Res., 108(C9), 4079–4092 [2002WR001088]. Boothroyd, J. C., and L. K. Habibar (1975), Genesis of low-symmetry mesotidal estuaries, in Estuarine Research, vol. 2, Geology and Engineering, edited by L. E. Cronin, pp. 217–234, Elsevier, New York. Carling, P. A. (1999), Subaqueous gravel dunes, J. Sediment. Res., 69, 533–543. Costello, W. R., and J. B. Southard (1981), Flume experiments on lower-flow-regime bed forms in coarse sand, J. Sediment. Petrol., 51, 849–864. Harms, J. C., and R. K. Fahnestock (1965), Stratification, bed forms, and flow phenomena (with an example from the Rio Grande), in Primary Sedimentary Structures and Their Hydrodynamic Interpretation, Spec. Publ. 12, edited by G. V. Middleton, pp. 84–115, Soc. of Econ. Paleontol. and Mineral., Tulsa, OK. Hutchinson, J. N. (1948), Laboratory study of the influence of granules on flow over a sand bed, Giol. Soc. Am. Bull., 59, 495–500. Klaassen, G. J. (1991), Experiments on the effect of gradation and vertical sorting on sediment transport phenomena in the dune phase, paper presented at the 1st Great Sand Dunes Seminar, Int. Assoc. for Hydraul. Res., Ascona, Switzerland, 21–25 Oct. Kleinhans, M. G. (2001), The key role of fluidic flows in transport and deposition of sand-gravel mixtures, a preliminary note, Sediment. Geol., 143, 7–13. Kleinhans, M. G. (2002), Sorting out sand and gravel: Sediment transport and deposition in sand-gravel bed rivers, Netherlands Geogr. Stud., 293, 317 pp. Kleinhans, M. G., and L. C. van Rijn (2002), Stochastic prediction of sediment transport in sand-gravel bed rivers, J. Hydraul. Eng., 128, 412–425. Kleinhans, M. G., A. W. E. Wilbers, A. De Swaaf, and J. H. van den Berg (2002), Sediment-only-limited bedforms in sand-gravel bed rivers, J. Sediment. Res., 72, 100–108. Kuhle, M. (1993), Incipient motion of sand-gravel sediment mixtures, J. Hydraul. Eng., 119, 1400–1415. Leclair, S. F., and A. Blom (2005), A qualitative analysis of the distribution of bed surface elevation and the characteristics of associated deposit for subaqueous dunes, in Sedimentology VII, Spec. Publ. Int. Assoc. Sedimentol., vol. 35, edited by M. D. Blum, S. Marriott, and S. Leclair, pp. 121–134, Blackwell, Malden, Mass. Leclair, S. F., and J. S. Bridge (2001), Quantitative interpretation of sedimentary structures formed by river dunes, J. Sediment. Res., 71, 713–716. Nordin, C. E. (1971), Statistical properties of dune profiles, U.S. Geol. Surv. Prof. Pap., 562-F, 41 pp. Paola, C., and P. C. Bormgardt (1991), Reconstructing topography from preserved stratification, Sedimentology, 38, 553–565. Paola, C., G. Parker, R. Yeat, S. K. Sinha, J. B. Southard, and P. R. Wilcock (1992), Downstream fining by selective deposition in a laboratory flume, Science, 258, 1757–1760. Parker, G., and P. C. Kitchenson (1982), On why gravel bed streams are patchy, Water Resources Res., 18, 1409–1423. Parker, G., and P. R. Wilcock (1993), Sediment feed and recirculating flumes: Fundamental difference, J. Hydraul. Eng., 119, 1192–1204. Parker, G., C. Paola, and S. Leclair (2000), Proabilistic theory of sediment transport for mixtures with no active layer, J. Hydraul. Eng., 126, 818–825. Reineck, H.-E., and I. B. Singh (1973), Depositional Sedimentary Environments, 439 pp., Springer, New York. Ribberink, J. S. (1976), Mathematical modelling of one-dimensional morphological changes in rivers with non-uniform sediment, Ph.D. thesis, Delft Univ., Netherlands. Southard, J. B., and A. Boguchwal (1990), Bed configurations in steady unidirectional open flows, part 2. Synthesis of flume data, J. Sediment. Petrol., 60, 658–670. Terwindt, J. H. J., and M. J. N. Brouwer (1986), The behaviour of intertidal sandwaves during neap-spring tide cycles and the relevance for palaeo-flow reconstructions, Sedimentology, 33, 29–31. van den Berg, J. H., and A. van Groesen (1993), A new bedform stability diagram with emphasis on the transition of ripples to plane bed in flows over fine sand and silt, Spec. Publ. 17, Int. Assoc. of Sedimentol., Santa Barbara, Calif. van der Zwet, J. J. (1973), Roughness aspects of sand transport over a fixed bed, Publ. J18, Delft Hydraulics, Delft, Netherlands. Wilcock, P. R. (1993), Critical shear stress of natural sediments, J. Hydraul. Eng., 119, 491–505. Wilcock, P. R. (2001), The flow, the bed and the transport: Interaction in the flume and field, in Gravel-Bed Rivers, vol. 5, edited by M. P. Mosley, pp. 212–214, N. Z. Hydrol. Soc., Wellington. M. G. Kleinhans, Faculty of Geosciences, Department of Physical Geography, Utrecht University, P.O. Box 80115, 3508 TC Utrecht, Netherlands (firstname.lastname@example.org)
Be Strong in the Lord! When a ship goes to sea, there are three essential questions its captain must address: 1. Why are we out here? Essential ethics 2. How do we keep from sinking? Individual ethics 3. How do we keep from bumping into other ships? Social ethics --- Tactics of Terror Isaiah 14:12 5 – The Five “I Wills” of Satan 1. “I will ascend into heaven” - Satan wanted to take God’s abode away from Him - SELF-WILL 2. “I will raise my throne above the stars of God” - Satan wanted to rule over the angelic beings - SELF-EXALTATION 3. “I will sit on the mount of assembly” - Satan wanted to rule over the Earth - SELF-ENTHRONEMENT 4. “I will ascend above the heights of the clouds” - Satan wanted to displace God’s rule - SELF-ASCENSION 5. “I will make myself like the Most High” - (Satan wanted to be God!) - SELF-DEIFICATION --- Tactics of Terror “The worst is always very near the best; there is something much worse than Atheism which is Satanism; otherwise known as Being God.” (G.K. Chesterton) --- Tactics of Terror 1. What is the general strategy of the devil? - Our first clue comes from a few of his names. - Two names are the most popular names given him in the scriptures: 1. Satan (adversary) (Job 1:6-12; Job 2:1-7; Zechariah 3:1-2; Matthew 4:10; Matthew 12:26; Matthew 16:23; Mark 1:13; Mark 3:23, 26; Mark 4:15; Mark 8:31; Luke 4:18; Luke 11:18; Luke 13:16; Luke 22:31; John 13:27; Acts 5:3); Acts 1:18; Acts 19:10; Acts 16:20, 1 Corinthians 5:5; 2 Corinthians 11:3; 2 Thessalonians 2:11; 2 Corinthians 11:14; 2 Corinthians 12:7; 1 Thessalonians 2:18; 2 Thessalonians 2:8; Hebrews 2:14; 1 Timothy 2:15; 1 Timothy 2:15; Revelation 2:9, 13, 24; Revelation 3:9; Revelation 12:9; Revelation 20:2, 7) 2. the devil (accuser) (Matthew 4:1, 5, 8, 11; Matthew 13:39; Matthew 25:41; Luke 4:2-3, 5, 9, 13; Luke 8:12; John 8:44; John 13:2; John 14:30; Acts 5:3; Acts 13:10; Acts 13:11; Acts 13:11; 1 Timothy 3:8-7; 2 Timothy 2:26; Hebrews 2:14; James 3:15; James 4:7; 1 Peter 5:8; 1 John 3:8, 10; Jude 9; Revelation 2:10; Revelation 12:9, 12; Revelation 20:2, 19) --- Tactics of Terror 1. What is the general strategy of the devil? - Our first clue comes from a few of his names. - Some of his descriptive names give insight into his power: 13. angel of the Abyss (Revelation 9:11) 14. the god of this age (2 Corinthians 4:4) 15. the prince of demons (Matthew 9:34; Matthew 12:24; Mark 3:22; Luke 11:15) 16. the prince of this world (John 12:31; John 14:30; John 16:11) 17. the ruler of the kingdom of the air (Ephesians 2:2) Tactics of Terror 1. What is the general strategy of the devil? - Our first clue comes from a few of his names. - Other descriptive names give insight into his character: 18. the accuser of our brothers (Revelation 12:10) 19. the enemy (Matthew 13:39; Luke 10:19; 1 Timothy 5:14) 20. the tempter (1 Peter 5:8) 21. a liar (John 8:44) 22. the father of lies (John 8:44) 23. the lawless one (2 Thessalonians 2:8-9) 24. murderer (John 8:44) 25. the tempter (Matthew 4:3; 1 Thessalonians 3:5) 26. the thief (John 10:10) 27. the evil one (Matthew 5:37; Matthew 6:13; Matthew 13:19, 38; John 17:15; Ephesians 6:16; 2 Thessalonians 3:3; 1 John 2:13-14; 1 John 3:12; 1 John 5:16-19) Tactics of Terror 1. What is the general strategy of the devil? - Our first clue comes from a few of his names. - He is also known by these “proper” names: 28. Beelzebub (Matthew 10:25; Matthew 12:24, 27; Mark 3:22; Luke 11:15) – “the lord of flies,” “the lord of dung,” or “the dung-god.” 29. Belial (2 Corinthians 6:15) – “worthlessness” 30. Abaddon (Revelation 9:11) – Hebrew for “destruction.” In the OT, the dwelling place of the dead. In the NT, the Hebrew name for the ruler of Sheol. 31. Apollyon (Revelation 9:11) – the Greek translation of the Hebrew word Abaddon, meaning “the destroyer.” Tactics of Terror 1. What is the general strategy of the devil? - Our first clue comes from a few of his names. - ‘the Destroyer’ - What does it mean to “destroy?” - To kill - To spoil - To tear down - To break up - To demolish - To carry away with - To kill - To subdue - To defeat - To crush - To render useless or ineffective Tactics of Terror 2. How does he plan to do this? - The Bible makes clear that the tactics of the devil fall into two major divisions. 1. He attacks the human race directly: - Demon possession (caution!) 1. Man of Capernaum, healed by Christ in the synagogue on the Sabbath (Mark 1:24; Luke 4:33) 2. Maniac of Gergesa, possessed by and healed of a legion of demons (Matt. 8:28–32; Mark 5:1–20; Luke 8:26–39) 3. A mute man healed by Christ, causing the multitude to rejoice (Matt. 9:32–33) 4. A demon-possessed girl, healed by Christ at the home of her parents (Mark 9:17–29; Matt. 17:16; Mark 1:22; Luke 9:43) 5. An epileptic son of Mount Hermon, healed at the request of his heartbroken father (Matt. 17:16; Mark 1:22; Luke 9:43) 6. A demon-possessed man, healed of paralysis and accused of healing by the power of Beelzebub (Matt. 12:22; Luke 11:14) 7. Woman possessed by seven demons, healed by Christ in a synagogue on the Sabbath (Luke 13:10–17) 8. Man possessed by seven demons, healed by Christ of seven demons (Matt. 16:8; Luke 8:2) 9. Judas lecariot, possessed by Satan himself (Luke 22:3; John 6:70; 13:27) 10. A slave girl with powers of divination, healed by Paul at Philippi (Acts 16:18–19) 11. Servo’s sons, renegade Jews at Ephesus (Acts 19:11) Tactics of Terror 2. How does he plan to do this? - The Bible makes clear that the tactics of the devil fall into two major divisions. 1. He attacks the human race directly: - Direct spiritual battle in the unseen realm - 2 Kings 6:17 - Daniel 10:10-14 2. By far, the majority of the attacks of the devil against Christians are indirect. 1. This is why they are called the “wiles,” “schemes,” “tricks” of the devil. - Wiles means deviousness, circuity, something not obvious, difficult to detect. 2. The two channels through which Satan tricks us are the world and the flesh. - Some have said that the Satanic Trilogy is Satan, the world, and the flesh. - This is not the case, as Satan is responsible for the two “…it is through the channels of the world and the flesh that the Devil makes his indirect and most insidious attack upon human life.” (Stedman, Spiritual Warfare) Tactics of Terror 2. How does he plan to do this? By far the majority of the attacks of the devil against Christians are inward. (cont.) 1. The two channels through which Satan tricks us are the world and the flesh. 2. Satan’s attacks us through our soul: mind, will, emotions. 3. The world: (1) the prince of this age (2 Corinthians 4:4; 1 Cor. 2:6-8) 1. the god of this age (2 Corinthians 4:4; Matthew 12:24; Mark 3:22; Luke 11:15) 2. the prince of demons (Matthew 9:34; Matthew 12:24; Mark 3:22; Luke 11:15) 3. the ruler of this world (John 12:31; John 14:30; John 16:11) 4. the ruler of the kingdom of the air (Ephesians 2:2) 5. Anointed Cherub (Ezekiel 28:14) 6. Angel of Light (2 Corinthians 11:14) 4. Conformity! Romans 12:2 It is easier to keep up than not to be different. Those who do swim up stream are ostracized! Tactics of Terror “The ‘world’ is human society, blindly and universally accepting false values, shallow concepts and insights and deluded ideas of reality, as well as almost desperately insisting upon conformity to those standards and insights.” (Stedman, Spiritual Warfare) Tactics of Terror 2. How does he plan to do this? “The flesh is that inward urge within us toward total independence, toward being our own little gods and running our worlds to suit ourselves. It is that continual drift within us toward self-centeredness and selfishness.” (Stedman, Spiritual Warfare) Tactics of Terror “It is part of the devil’s business to keep the Christian’s spirit imprisoned. He knows that the believing and justified Christian has been raised up out of the grave of his sin and trespasses. From that point on, Satan works that much harder to keep us bound and gagged, actually imprisoned in our own grave clothes. He knows that if we continue in this kind of bondage we will never be able to claim our rightful spiritual heritage. He knows also that while we continue bound in this kind of enslavement we are not much better off than when we were spiritually dead.” (A.W. Tozer, Renewed Day by Day) Tactics of Terror “...[Satan] has interposed himself between God and humanity and perverted the life (or love) of God and twists and distorts it so that it becomes outward directed, as it came from God, but it becomes inward directed; that is, it no longer is loving, but it becomes selfish. That life is called ‘the flesh.’ This, then, is the primary characteristic of the flesh: it is self-serving. It is God’s life, misused. It can have all the outward appearance of the life of God–loving, working, forgiving, creating, serving, etc., but it is all self-serving, aimed always and solely at the advancement of self. It thus becomes the rival of God – another god!” (Stedman, Authentic Christianity, page 67)
THE PILGRIMS. And who are they? That humble band, And is each one a pilgrim? Who journey in a weary land, And worship as they go? With steadfast faith, and upward eye, Unmoved by present shame; They wait a glory from on high, A Prince of heavenly fame. They look, and yet no cloud appears, And watch still undismayed; Nor seem to heed the doubts and fears, That other hearts pervade. Weary, and yet pursuing still The path of their love; Affirming that a shadow will Appear in clouds above. While multitudes their hope revile, And dare the threatened day, They patiently the hours beguile, Content to watch and pray. Then who are they? Their garments seem All suddenly to change! And all their way-worn faces gleam With radiance sweet and strange. He comes! He comes! Their faithful King, Embraced on either side of light; And rising with angels they sing, And triumph in his sight. EXPOSITION OF DANIEL 7th CHAPTER; OR, VISION OF THE FOUR BEASTS—BY GEO. STORRS. (Concluded.) We are now prepared for the inquiry—who, or what is the little horn here spoken of? We will inquire. 1st. What is the character of this horn? 1. It makes war with the saints. 2. It speaks great words against the Most High. Let us see if we can find a description of the same character elsewhere in the Bible. See Rev. xiii, 6, 7: "And he opened his mouth in blasphemy against God, to blaspheme his name and his tabernacle, and them that dwell in heaven. And it was given unto him to make war with the saints and to overcome them;" Daniel says, "he prevailed against them." Now see 2 Thess. ii, 3, 4: "That man of sin be revealed, who opposeth and exalteth himself above all that is called God, &c. Can there be any doubt of the identity of the character? Daniel's "little horn," Paul's "man of sin," and John's "blasphemous beast," are clearly identified. 2d. Let us inquire, has a power of this description arisen? It must be admitted that there has; and that power is Papacy. The titles the Popes have assumed, of "Most Holy Lord," and their attempts to pardon sin, even before its commission, if we had nothing else, sufficiently establishes the blasphemous character of that power. Pope Innocent III, writes—"He [Christ] hath set one man over the world, him whom he hath appointed his vicar on earth; and as to Christ I bend every knee in heaven, in earth, and under the earth; so shall obedience and service be paid to his vicar by all, that there may be one fold and one shepherd.—Again, Pope Gregory VII, says, "The Roman Pontiff alone is by right universal. In him alone is the right of making laws. Let all kings kiss the feet of the Pope. His name alone shall be heard in the churches. He is the only name in the world. It is his right to depose kings. His word is not to be repealed by any one. It is to be repealed by himself alone. He is to be judged by none. The church of Rome has never erred; and the Scriptures testify it never shall err." Surely here is a power diverse from all others, and proud and blasphemous enough to answer the character of the little horn. 3d. Let us now inquire, When this little horn arose? Or, which is the same thing, When did Papacy arise? There has been a difference of opinion on that question. But it appears to me the question is not one so difficult to settle now as in former years. First, therefore, it did not arise before the ten horns. Hence it did not arise before A. D. 438, when the tenth horn came up. It did not arise until three of the first horns fell, or were plucked up. It did not come up after that, because it came up among the ten horns, and three of those horns fell before it. It must then have come up soon after establishing the position where the third horn fell. If that point can be settled, it seems to me there can be no reasonable doubt as to the time Papacy arose. In the year of our Lord 493, the Heruli in Rome and Italy were conquered by the Ostrogoths. In 534, the Vandals, who were under Arian influence, were conquered by the Greeks, for the purpose of establishing the supremacy of the Catholics. The Ostrogoths, who held possession of Rome, were under an Arian monarch, who was an enemy to the supremacy of the Bishop of Rome; hence, before the decree of Justinian, (a Greek emperor at Constantinople,) could be carried into effect, by which he had constituted the Bishop of Rome head of all the churches, the Ostrogoths must be plucked up. This conquest was effected by Justinian's army in the month of March, 538; at which time, the Ostrogoths, who had retired without the city, and besieged it in their turn, raised the siege and retired, leaving the Greeks in possession of the city; thus the third horn was plucked before Papacy, and for the express purpose too of establishing that power. [See Gibbon's Decline and Fall of the Roman Empire.] How exactly do the facts answer to the prophecy. I will here introduce the letter of Justinian to the Bishop of Rome, of A. D. 533: "Justinian, pious, fortunate, renowned, triumphant, conqueror, &c., &c., to John, the most holy Archbishop of our city of Rome and patriarch: "Rendering honor to the apostolic see, and to your holiness, (as always was and is our desire,) and, as it becomes the honor of your blessedness, we have laid, within these few days, before the notice of your holiness all things pertaining to the state of the church. Since it has always been our earnest study to preserve the unity of the holy sees, and the state of the holy churches, of God with us, we have obtained, what we will remain, without any interfering opposition; therefore we hasten to subject, and to unite to your holiness all the priests of the whole East. As to the matters which are presently agitated, although clear and unobjectionable, and according to the decree of your holiness, held assuredly resolved and decided by all priests, we have yet deemed it necessary to lay them before your holiness. Nor do we suffer anything which belongs to the state of the church, however manifest and undeniable, that is agitated, to pass without the knowledge of your holiness, who are the head of all the holy churches. For in all things (as had been said or resolved) we are prompt to increase the honor and authority of your see." "The authenticity of the title," says Mr. Croley, "receives unanswerable proof from the edicts of the Novella's of the Justinian code. The preamble of the 9th, states, 'that as the elder Rome was the founder of the laws; so was it not to be questioned, that in her was the supremacy of the pontificate.' The 131st, on the Ecclesiastical Titles and privileges, chapter ii., states: 'We therefore decree that the most holy Pope of the elder Rome is the first of all the priesthood, and that the most blessed archbishop of Constantinople, the new Rome, shall hold the second rank, after the holy apostolic chair of the elder Rome.'"—Croley, pp. 114, 115. Some suppose that Phocas, A. D. 606, by applying the title "Universal Bishop" to the Pope, first gave him his supremacy; but this cannot be, for it does not agree with the prophecy that three of the first horns were to be plucked up before it, as it came up, and this happened more than half a century before. Again, Mr. Croley, a writer of the Church of England, says—"The highest authorities among the civilians and annalists of Rome spurn the idea that Phocas was the founder of the supremacy of Rome; they ascend to Justinian, as the only legitimate source, and rightly date the title from the memorable year 533." Imperial Rome fell about A. D. 476, and was in the hands of the barbarians. Thus it continued till the conquest of Rome by Belisarius, Justinian's general, 536 to 538, when the Ostrogoths left it in possession of the Greek emperor, March, 538. Thus the way was open for the dragon to give the beast his power, and his seat, and great authority. Rev. xii, 2.—This fact, from Revelation, also, settles the point that the Pope could not receive his power from the King of France. It was the dragon that gave him his seat, Rome, his power, as head of the churches. The Roman emperors had stood at the head of the churches with power to make important decisions for the church—it is now transferred to the Pope; and he has given him, also, great authority, under the Justinian code of laws, to judge and punish heretics. 4th. The next point we want to settle is, the length of time this power was to continue. Daniel says, "a time, times, and the dividing of time." The Revealatory says, [Chap. xiii, 5, "Power was given unto him to continue 42 months."] He was to make war upon the saints—the church; and in Rev. xii, 6, we are told, the woman, the church, fled into the wilderness 1260 days; and at the 14th verse, that it was for "a time, and times, and half a time." Here then we have the period of the continuance of this power given us in three forms of expression, which settles the point that the time, times, and dividing of time is 42 months, or 1260 prophetic days or years. 5th. Did the continuance of papal dominion, as a horn of the beast, cease at the end of that period?—1260 years from 538 would extend to 1798. Did anything transpire that year to justify the belief that the dominion of Papacy was taken away that year? It is a historical fact, that, on Feb. 10th, 1798, Berthier, a French general, entered the city of Rome and took it. On the 15th of the same month the Pope was taken prisoner and shut up in the Vatican. The Papal governments, which had continued from the time of Justinian, was abolished, and a republican form of government given to Rome. The Pope was carried captive to France, where he died in 1799. Thus, he that led others into captivity, went into captivity; and he who killed with the sword, those he was pleased to call heretics, was himself killed [subdued] with the sword; i.e., his dominion was taken away by war. See Rev. xiii. 10. Verse 26: "but the judgment shall sit, and they shall take away his dominion, [he is cut off from being a horn on the beast, or deprived of his civil power, so that he can no longer wield the sword against dissenters] to consume and destroy it unto the end." See 2 Thess. ii, 8: "When the Lord shall consume with the spirit of his mouth, and shall destroy with the brightness of his coming to judge the world in the burning day, when the beast and false prophet will be cast alive into the lake of fire, burning with brimstone; then the little horn will be destroyed." Some tell us the civil power of Papacy is not taken away. That the Pope was restored, or a new one chosen, is admitted, and that he may have some civil power in Italy is not denied. But that he has power to depose kings and put to death the saints now, is denied. When he was a horn on the beast, he depopulated kings at pleasure, for centuries, and silenced heretics by the flame, the rack, prison, and the sword. Can he do it now? No. Nor has he been able to do it since 1798—since that time the church is out of the wilderness; and Papacy is compelled to tolerate Protestantism. Hear the Pope himself on that subject. Here is his letter, dated Sept., 1840, at Rome. Read it, and see if you think Papacy is now a horn on the beast, or is possessed of power to war against the saints—unto death, as formerly. "Encyclical Letter of Our Most Holy Lord Gregory XVI, by Divine Providence Pope, to all Patriarchs, Primates, Archbishops, and Bishops. Gregory XVI. Pope. Venerable Brethren,—Health and the apostolic Benediction. "You will know, Venerable Brothers, how great are the calamities with which the Catholic Church is beset on all sides in this most sorrowful age, and how pitifully she is afflicted. You know by what a deluge of errors of every kind, and with what unbridled audacity of the erroneous, our Holy Religion is attacked; and how cunningly and by what frauds, heretics and infidels are endeavoring to pervert the hearts and minds of the faithful. In a word, you know there is almost no kind of effort or machination which is not employed, to overthrow, from its deepest foundations, if it were possible, the invincible edifice of the Holy City. "Indeed, are we not, (Oh, how shameful!) compelled to see the most crafty enemies of the truth, ranging far and wide with impunity; not only attacking religion with ridicule, the church with contumely, and Catholics with insults and slander, but even entering into cities and towns, establishing schools of error and impiety, publishing in print the poison of their doctrines, skilfully concealed under the deceitful veil of the natural sciences and new discoveries, and even penetrating into the cottages of the poor, traveling through rural districts, and insinuating themselves into familiar acquaintance with the lowest of the people and with the farmers! Thus they leave no means unattempted, whether by corrupt Bibles in the language of the people, or pestiferous news papers and other little publications, or caviling conversation, or pretended charity, or, finally, by the gift of money, to allure ignorant people, and especially youth, into their nets, and induce them to desert the Catholic faith. "We refer to facts, Venerable Brethren, which not only are known to you, but of which you are witnesses; even you, who, though you mourn, and as your pastoral duty requires, are by no means silent, are yet compelled to tolerate in your dioceses these aforesaid propagators of heresy and infidelity; those shameless preachers, who, while they walk in sheep's clothing, but inwardly are ravening wolves, cease not to lay in wait for the flock and tear it in pieces. Why should we say more? There is now scarcely a barbarous region in the universal world, to which the well known Central Boards of the heretics and unbelievers have not, regardless of expenses, sent out their explorers and emissaries, who either insidiously, or openly and in concert, making war upon the Catholic religion, its pastors and its ministers, bear the faithful out of the bosom of the church, and intercept her approach to the infidels. "Hence it is easy to conceive the state of anguish unto which our soul is plunged day and night, as we, being charged with the superintendence of the whole field of Jesus Christ, and the care of all the churches, must give account for his sheep to the Divine Priest of Pastors. And we have thought fit, Venerable Brethren, to recall to your minds by our present letter the causes of those troubles which are common to us and you, that you may more attentively consider how important it is to the church, that all holy priests should endeavor, with redoubled zeal, and with united labors, and with every kind of efforts, to repel the attacks of the raging foes of religion, to turn back their weapons, and to forewarn and fortify the subtle ambushments which they often use. This, as you know, we have been careful to do at every opportunity; nor shall we cease to do it; as we also are not ignorant that you have always done it hitherto, and confidently trust that you will do hereafter with still more earnest zeal. * * * * * * * "Given at Rome, at St. Mary the Greater, on the 18th of the Kalends of September, of the year 1840, the tenth of our pontificate. Gregory XVI. Pope." You see what is to come next after the fall of the little horn. Now let us see whereabouts we are in the prophetic chain. Have we passed the Lion—Babylon? Yes. Have we gone by the Bear with three ribs in his mouth? Yes. Has the sign of the Leopard with four wings of a fowl and four heads been passed? It has. The dreadful and terrible beast, with ten horns,—has he been seen? Yes. Have we got past the little horn having eyes like the eyes of a man? That is among the things numbered with the past.—How far beyond it are we? Forty five years, nearly. What comes next? The judgment, followed by the everlasting kingdom of God. How far off is that? That question I shall answer, hereafter, definitely. But one thing is certain; it cannot be at a great distance. It is the next prophetic event. Awake, ye slumbering virgins! Behold the Bridegroom cometh, go ye out to meet him. No time to sleep now. The second trumpet is preparing the sound. "Awake, ye dead!" will soon thunder through the skies.—Happy day to those that are waiting. "Awful day to those who are saying—"My Lord delayeth his coming." Awake—Awake!! The 70 Weeks, or part of the 2300 days. [Two lectures were preached on this subject by Rev. Andrews. We conclude. We give the following extract from the exposition of Dan. viii, by Geo. Storr, published in 1843. It shows the position then taken by the whole Advent body.] "We will now give you the 13th and 14th verses of the eighth chapter of Daniel, leaving out what our translators have supplied:—"Then I heard one saint speaking, and another saint said unto that certain saint which spake, How long the vision, the daily and the transgression of desolations, to give both the sanctuary and the host to be trodden under foot? And he said unto me, Unto two thousand and three hundred days; then shall the sanctuary be cleansed." The inquiry, 'How long the vision,' clearly related to the 'Ram' and 'Goat,' as well as 'little horn'—and, the 2300 days are given in answer to the question,—"How long the vision?" It was the meaning of the vision Daniel sought—verse 15: 'It came to pass, when I had seen the vision, and sought for the meaning,' &c. It was to make Daniel understand the vision, Gabriel was sent—verse 16: 'I heard a man's voice between the banks of Uzai, which called and said, Gabriel, make this man to understand the vision.' It was to make Daniel understand the vision, Gabriel came—verse 17: 'So he came near where I stood, and said unto me, Understand, O son of man.' The first thing Gabriel would have Daniel understand, was, that the vision was down to the end,—verse 17: 'At the time of the end shall be the vision.' He would have Daniel understand that the end intended was the last end of indignation,—verse 19: 'Behold, I will make thee know what shall be in the last end of the indignation;' and he would have Daniel know that, 'at the time appointed, the end shall be.' The 2300 days is the only time appointed. That time cannot be applied to a particular agent or event, without violence to the whole subject. Now let us inquire what Daniel did understand, and what not. The angel explained every thing to him about the ram, the goat, and little horn. But Daniel tells us that he 'understood' it. Was he satisfied with the vision, but none understood it? What did Daniel understand? There were, evidently, three things he did not understand: 1st. What 'sanctuary' was intimated in verse 13; 2d. He did not understand how to reckon the days; and 3d. Where to commence his reckoning. As Gabriel is not to be charged with disobedience to the command to make Daniel understand the vision, and as he has not fulfilled that command in this chapter, we must look elsewhere to see if he ever did what he was directed to do, and what he promised Daniel he would do. Let us now look into the 9th chapter. Daniel there informs us that he found out 'by books, the number of years whereof the word of the Lord came to Jeremiah the prophet, that he would accomplish seventy years in the desolations of Jerusalem.' We might here inquire, Why did not Daniel 'find that out before?' It was written in the book, but he did not discover it until now. Our opponents seem to think it is a conclusive argument, that we are wrong, because the time of the end of the world has not been found out before now. But is it any more marvelous than that Daniel did not learn that the captivity of the Jews in Babylon was to be seventy years, till those years were accomplished? When Daniel discovered this fact, fifteen years had passed since the vision of the eighth chapter, and he had all that time been in uncertainty about the points that were not explained to him in that vision. He now seems to catch at the thought, that it must be the sanctuary at Jerusalem, to which the vision related, and he at once commences praying accordingly. He, at the 17th verse, prays especially about the sanctuary. 'Now, therefore, O our God, hear the prayer of thy servant, and his supplications, and cause thy face to shine upon thy sanctuary that is desolate for the Lord's sake.' Daniel's mind is entirely on the vision, and he seems to suppose he has got the clue to the sanctuary that is to be cleansed; but Gabriel comes flying swiftly, to stop Daniel in the midst of his prayer. See verse 21: 'Yea, while I was speaking in prayer, even the man Gabriel, whom I had seen in the vision at the beginning, being caused to fly swiftly, touched me about the time of the evening oblation.' Gabriel, why this haste? Why, I see Daniel is wrong—he don't understand the matter—he thinks the vision related to the sanctuary at Jerusalem, and I must stop him, for he is going astray. 'Gabriel, whom I had seen in the vision,' says Daniel. What vision? Where had Daniel seen Gabriel in vision? Evidently nowhere but in the vision of the 2300 days. Well, says Gabriel, 'I am now come forth to give thee rest, and to comfort—therefore understand the matter, and consider the vision.' How is it possible the anything can be plainer than that both Daniel and Gabriel have the vision in mind, that 'none understood,' at the close of the eighth chapter? Now, says Gabriel, 'understand the matter, as by your prayer I see you did not, and consider the vision—direct your attention to what I have now to say of it.' Seventy weeks are determined [cut off, so the word signifies] upon thy people.' Cut off from what? Surely not from the indefinite space, but from some that previously given. What time had Daniel given him before? Only except the 2300 days. The natural inference, then, is that the 70 weeks were cut off from those days: there is nothing else to cut them off from. For what are they cut off? Several objects are specified; but one especially, viz., 'to seal up [or as the word signifies, see Dan. vi. 17, make sure] the vision.'" THE LAST DAYS. AN EXTRACT FROM AN OLD SERMON. Time grows old! The Dread Majesty of the heavens arises to shake terribly the earth! The day is near that shall burn as an oven, and all the proud and them that do wickedly shall be shaken! The prophet Daniel, speaking of this very period of time, says, that many shall be punished at once while and together, probably in times of great persecution; but the wicked shall do wickedly—go from bad to worse with long and rapid strides—enjoy glory in their shame and proud ignorance, and, therefore, none of the wicked shall understand; but the wise shall understand. Before the evening of the great and dreadful day of the Lord, there shall be, according to St. Paul, a departure from the faith. Those also that retain the form, but deny the power of godliness that they may gratify their lusts and clothe their carnality, will heap to themselves teachers, complimentary and waxen images of men having itching ears that cannot bear the truth. In these perilous times, men shall be lovers of their own selves: covetous, hoarders, proud, blasphemers, disobedient to parents, unthankful, unholy, without natural affection, truce breakers, false accusers, incontinent, fierce, despisers of those that are good, traitors, heady, high-minded, lovers of pleasure more than lovers of God. And St. Peter says, that in the last days shall come scoffers, walking after their own lusts, and saying, Where is the promise of his coming? for since the fathers fell asleep all things continue as they were from the beginning of the creation. For this they are willingly ignorant of, the destruction of the antediluvian world, and deny and despise the future certainty of the conflagration of the present heavens and earth, which we know by the word of God, continues to be left in store, reserved unto fire against the day of judgment and perdition of ungodly men. As infidelity increases, the times will be more strongly marked, and voluptuousness and security among the wicked, make an additional sign of the coming of the Son of man: For as it was in the days of Noah, so shall it be also in the days of the son of man. They did eat, they drank, they married wives, they were given in marriage, until the day that Noah entered into the ark, and the flood came and took them all away. Likewise also as it was in the days of Lot. They did eat, they drank, they bought, they sold, they planted, they builded, but the same day that Lot went out of Sodom, it rained fire and brimstone from heaven and destroyed them all. And a cautionary direction is given to the believers that in this general security they should watch, lest at any time their hearts should be overcharged with intemperance and worldly cares, and so that day which shall come as a snare on all them that dwell on the face of the whole earth, should overtake them also by surprise. At some period not long before the great day, there shall be signs in the sun, and in the moon, and in the stars—and upon the earth distresses [revolutions] of nations with perplexity, the sea and waves roaring—either literally or symbolically, to express tumultuous combinations of people mad to throw off every restraint, civil or religious—men's hearts failing them for fear, and for looking after those things which are coming on the earth—for the powers of heaven shall be shaken. And then shall appear the sign of the Son of man—one somberious display understood as we said before by faithful believers. And when these things begin to come to pass, says the evangelist, then look up and lift up your heads, for your redemption draweth nigh. Lo! the hour is come! the mystery of God is finished! It is the determined counsel of the Eternal. What voice is that I hear from the dread throne, as the sound of many waters? "I am Alpha and Omega, the beginning and the end, the first and the last, saith the Lord." Behold, I come quickly, and my reward is with me, to give every man according as his work shall be! I see the Son approaching to the throne of majesty in the heavens. I saw, says the prophet Daniel, in the night visions, and behold one like the Son of Man came with the clouds of heaven, and came to the Ancient of days, and they brought him near before him. And thus he moves his Father: O Father! the year of my redeemed is come—the day of vengeance is in mine heart. Long has my Spirit strove with men. All that love and wisdom infinite can give, has been bestowed to effect their salvation. And now the ingrates vainly imagine that they shall efface thy remembrance, a sense of thine existence, and their obligations to love and serve thee, and to worship thee, the only true God, sole earth. Come, therefore, let me get down, for the harvest is ripe, the press is full, the fats overflow! For their wickedness is great—that I may wipe away the reproach of my people, and perform my faithful promise, which their steadfast faith has long expected, in bringing them home to these fair realms, to behold the glory which I had with thee before all worlds, and with me to enjoy thy smiles and love eternal—and that I may condemn and banish, the impetuous rebels of mankind and the disobedient angels to their own place, as vessels of wrath fitted for destruction. And there was given him, continues the prophet, dominion and glory and a kingdom, that all nations and languages should serve him: his dominion is an everlasting dominion which shall not pass away, and his kingdom that which shall not be destroyed. And now he ascends the great white throne, raised on "the chariot of paternal deity." Above, beneath and round about him pour the marshaled hierarchies of angels and the innumerable company of the souls of just men made perfect; for them also who sleep in Jesus, will God bring with him. The brightest military spirit of heaven leads on the angelic hosts: down through the regions of space, by unnumbered worlds they mark their rapid way, till they decay far off this poor terrestrial, sin-polluted orb, at once for their impudence and rejection of the Saviour, the disgrace and scorn of all created worlds. The mighty angel whose name John saw standing upon the sea and upon the earth, at the signal given, lifts up his hand to heaven and swears by him that liveth for ever and ever, who created the heavens, and the earth, and the sea, and the things which are therein, that time should be no longer. What shout is that I hear? It is no other than the bursting joy of angels attendant on the Lord from heaven—from ten thousand thousand strong hanged cherubs breaks the glad peal: "The joy, the shout, the harmony, "Piles o'er the mountains while, "Low from four hundred numbers number, "Sweet as from blest voices uttering praise." The jubilee begins! Hark! hark! do you hear that awful sound? It is the trump of God. Louder and louder the great archangel swells the tremendous roar! The awful peal reverberates through continents, thro seas and islands! The steadfast earth reels to and fro like a tempest-beset cottage! See, the opening grave! the cairn teems with resurrection! Behold those glorious forms! how sweetly they accord—What love and joy in their resplendent countenances! These are the saints of God—these are they that slept in Jesus! For the dead in Christ shall rise first—Then we which are alive and remain, shall drop our dull mortality. For we shall all be changed in a moment, in the twinkling of an eye at the last trump, says the Apostle.—This corruption shall put on incorruption, and this mortal shall put on immortality. By the ministry of glorious angels, both the living and resuscitated saints ascend to meet the Lord in the air: for, Behold he cometh! "Far off his coming shines!" Nearer and nearer the bright pomp approaches! Behold, he cometh! and every eye shall see him, and they also that pierced him, and all kindreds of the earth shall wail because of him. The angelic hosts dispart on either hand, and the King immortal appears in all the glory of his Father, with the holy angels. The judgment is set, and the books are opened. Lift up your eyes, ye saints of the living God, and behold on the seat of judgment enthroned your Judge and Saviour! Around him through the vast concourse are spread innumerable animated forms, beaming augucially bright. Is this the babe of Bethlehem? Is this the manger's weeping care? Is this the man arraigned at Pilate's bar? Is this the bleeding astonishment of Calvary, whose death the blushing heavens refused to behold? Yes, this is he—that object of grief. But O, how changed! And I saw a great white throne, says St. John, and him that sat on it, from whose face the earth and the heavens fled away and there was found no place for them. To the Saints scattered abroad. Dear Brethren and Sisters:—Do we believe with all the heart that Christ is soon coming? And that we are now having the last message of mercy that is ever to be given to a guilty world? Is our example what it should be? And do we show to those around us, by our lives and holy conversation, that we are looking for the glorious appearing of our Lord and Saviour Jesus Christ, to change these vile bodies and fashion them like his most glorious body? I fear that we do not believe, and realize these things as we should. Those who believe the important truths that we profess to believe, should act out their faith, in the immediate coming of Christ. There is too much seeking of amusements, and things to take up the mind here in this world; the mind is left too much to run upon pride of dress; and the tongue is engaged too often in light and trifling conversation, which gives the lie to our profession, for the conversation is not in heaven from whence we look for the Saviour. And we are watching over us, to guard us; and we often give those angels by indulging in trifling conversation; jesting and joking, and also by sinking down in a careless, stupid state. And although we may make an effort now and then for the victory, and obtain it, yet if we do not keep it, but sink down in the same careless, indifferent state, unable to endure temptations, and to resist the enemy, it is not enduring the trial of our faith, that is more precious than gold. It is not suffering for Christ's sake, and glorying in tribulation. There is a great lack of christian fortitude, and serving God from principle. We should not seek to please and gratify self; but to honor and glorify God, and in all we do and say, have a single eye to his glory.—If we would let our hearts be impressed with the following important words, and ever bear them in mind, we should not so easily fall into temptation; but our words would be few, and well chosen. "He was wounded for our transgressions, he was bruised for our iniquities; the chastisement of our peace was upon him; and with his stripes we are healed." "Every idle word that men shall speak, they shall give account thereof in the day of judgment."—"Thou God seest me." We could not think of these important words, and call to mind the sufferings of Jesus for our sinner, that we might receive pardon from our sins, and be redeemed unto God by his most precious blood, without feeling a holy restraint upon us, and an earnest desire to suffer for him, who suffered and endured so much for us. If we dwell on these things, dear self, with its dignity, will be humbled; a child-like simplicity will take its place, which will bear reproof from others, and will not be easily provoked, and suffer a self-willed spirit. to come in and rule the soul. The true christian's joys, amusements and consolation, must and will be in heaven. "Upward God be the heart's adoration, Where ever is flowing pure streams of salvation." The longing soul of those who have tasted of the powers of the world to come, and have feasted on heavenly joys, will not be satisfied, or amused, with things of earth. Such will find enough to do in their leisure moments. Their souls will be drawn out after God. Where the treasure is, there will be their heart holding sweet communion with the God they love and worship. Their amusements will be in contemplating their treasures—the holy city—the earth made new—their eternal home. And while they dwell upon these things, which are lofty, pure and holy, heaven will be brought near, and they will feel the power of the Holy Spirit, which will tend to wean them from the world more and more, and cause their consolation and chief joy to be in the things of heaven, their sweet home.—The power of attraction to God and heaven will be so great, that nothing can draw their mind from the great object of securing their soul's salvation, and honoring and glorifying God. "Brighter joys than earth can give, win me away, Pleasures that for ever live—I cannot stay." As I realize how much has been done for us, to keep us right. I am led to exclaim, O, what love! What wondrous love hath the Son of God for us poor sinners! Should we be stupid and careless, while every thing is being done for our salvation that can be done? All heaven is interested for us. We should be alive and awake, to honor, glorify and adore the High and Lofty One. Our hearts should flow out in love and gratitude to him who has been so full of love and compassion to us. With our lives we should honor him, and with pure and holy conversation show that we are born from above; that this world is not our home, but that we are pilgrims and strangers here, traveling to a better country. Many who profess the name of Christ, and profess to be looking for his speedy coming, know not what it is to suffer for Christ's sake. Their hearts are not subdued by grace, and they are not dead to self; but it often appears in various ways; and, at the same time, they are talking of having trials. But the principal cause of their trials, is an unsubdued heart, which makes self so sensitive, that it is often crossed. If such could realize what it is to be an humble follower of Christ, a true christian, they would begin to work in good earnest, and begin right. They would first die to self, then be instant in prayer, and check every passion of the heart. Give up your self-confidence, and self-sufficiency, and follow the meek pattern.—Ever keep Jesus in your mind, that he is your example, and you must tread in his footsteps. Looking unto Jesus, the author and finisher of our faith; who, for the joy that was set before him, endured the cross, despised the shame. He endured the contradiction of sinners against himself. Is not the reward, at the end of the race, great and rich enough? What greater inducements could be held up before us, than has been held up to encourage us to be bold and valiant soldiers, to overcome the world, the flesh, and the devil? Eternal life is ours, if we endure the trial of our faith. Is it not enough? Will any complain of the roughness of the way?—Would you enter heaven if you could without suffering, and dwell in the presence of that Jesus, who suffered so much for us, whose loveliness and glory is unspeakable? He for your sins, was once the meek slain lamb, wounded, bled, smitten and afflicted. O, it would be no place for you. Any other place would be far preferable. You would feel that you had no right there. Let us, then, cheerfully suffer something for Jesus' sake, crucify self daily, be a partaker of Christ's sufferings here, that we may be made partakers with him of his glory, and be crowned with glory, honor, immortality and eternal life. Ellen G. White. THE Immediate Coming of Christ. In the last Review, we briefly examined a portion of the evidences on which the advent faith rested, and showed that the passing of a few years beyond the period of confident expectation for the coming of the Lord, does not affect these evidences for the immediate coming of Christ. We noticed the metallic image of Dan. ii., the prophetic chain of Dan. vii., the special signs in the sun, moon and stars, also, the state of the moral and physical, and the spiritual wonders of the present day, performed by acts of devils. We will now notice briefly Daniel chapter viii. This chapter presents the ram, which represents the Medo-Persian kingdom, the lamb as the symbol of chapter vii., and the goat, which represents Greece, the same as the leopard of chapter vii., and the bolly and sides of brass of the image of chapter ii.; and the little horn that waxed exceeding great, which represents the Roman kingdom, the same as the ten horned beast of chapter vii., and the legs and feet of the image of chapter ii. The question is then asked, how long this vision, to give both the Summary and Detail to the Lord's Word. The answer is, "Unto 2300 days, when shall the Sanctuary be cleansed." Dan. viii., 13, 14. Those of us that looked for the Lord with such confidence in 1844, understood the text as though it read, Unto 2300 days, then Christ will come and raise the dead and change the living saints, &c. Here is the secret of our disappointment. We did not then understand the Sanctuary, and its cleansing, as we now may. The time was then, "The time of Antichrist," as it is perfectly plain in No. 16, of the Review. The cleansing the Sanctuary, we did not understand; hence, looking for the second coming of Christ at the end of the days, and were disappointed. We acknowledge that we were disappointed, and did not then understand the events to occur at the end of the days, but we do confess that we were in the least affected the evidence of the immediate coming of Christ. We now see more clearly than ever, the event that marks the commencement of the 2300 days, n. c. 457, also, we see clearly what marks their termination in 1844. This subject has been clearly presented in the late numbers of the Review, by Miss Andrews. We hope that others are as well informed as the doctrine of Christ's soon coming will be a thorough investigation. The period of time we now occupy is called by Apostle Paul, the "little while." It is also called "the patience of the saints." It is emphatically the waiting, watching, trying time, in which the last work of mercy is to be accomplished, to prepare the world for the day of truth, and to stand when the Son of man shall appear. Referring to the present persecution, and trials of those looking for Christ, the Apostle says: "Cast not away therefore your confidence, which has great recompense of reward. For ye have need of patience, that, after ye have done the will of God, ye might receive the promise." For a yet little while, and he that shall come, will come, and will not tarry. Now ye must still live by faith; but if ye will draw near, ye must first learn that hereafter ye shall. But we are not of them who draw back unto perdition; but of them that believe to the saving of the soul." Heb. x., 35-29. The phrase, "For yet a little while, and he that shall come will come, and will not tarry," shows that the Apostle's words apply to the period just prior to the Second Advent, and that they were addressed especially to those who should be looking for the Lord's coming. Then the expression, "Cast not away therefore your confidence," refers to those who have had great confidence in his immediate coming. And the words, "For ye have need of patience, that, after ye have done the will of God, ye might receive the promise," certainly apply to the faithful, who looked for the Lord with such confidence, and were greatly disappointed, fearful and tempted, at the time of the most distressing persecution. But did the Advent people to the will of God in looking for Christ, and proclaiming the messages of the first and second angels, prior to the Autumn of 1844, where they were sadly disappointed? Let the Apostle answer: "For ye have need of patience, that, after ye have done the will of God," &c. We repeat it, that this language applies to no other people, and to no other period of time, only to those who are looking for the Lord, and prior to his coming, or when he is coming in "a little while." Says Paul to his Thessalonian brethren: "Let no man deceive you by any means; for that day shall not come, except there comes a falling away first, and that man of sin be revealed, the son of perdition; who opposeth and exalteth himself above all that is called God, or that is worshipped." 2 Thess. ii., 3, 4. Here is a warning against looking for the Lord prior to the period when he should be expected. And this warning reaches over the predicted 1250 years of cruel, blasphemous rule of the man of sin, or Papacy, down to about 1788. Then it is impossible to give the above words of Paul a general application, or to apply them to the church, prior to 1788. We therefore, feel bound to apply them to the Advent movement, and the trials which have followed. "Now the just shall live by faith. Faith in what? Answer, that he did the will of God in proclaiming the Advent, and that Christ is coming in "a little while." "But if any man draw back, my soul shall have no pleasure in him." Hundreds have drawn back, and have incurred the Lord's displeasure, and are now, far from God, in a "lukewarm" state. Some are counseled to buy "gold," "white raiment" and "eye-salve," and are, by the True Witness, the "little flock," exhorted to be "zealous" and "repent." While in their endeavours. "But we are not of them who draw back unto perdition; but of them that believe to the saving of the soul." There are but two sides to this question. We may doubt, draw back, and reject the light now shining forth from the sure word, which has come to the past Advent movement, and close our ears to the Shepherd's voice, and the voice of mercy, heard in this message of Rev. xiv., 9-12, and finally sink in apostatization. Or, we may believe in the saving of the soul. Brethren, in the fullness of our soul, we say, believe. And we are, from a conviction of God's truth, compelled to take the position, that saving faith lies here; instead of casting away our confidence in the past Advent movement, to believe that God's will was done that moment, and that in "a little while, and he that shall come will come." The third angel makes all plain, and shows that we are still on the track of prophecy. "Here is the patience of the saints; here are they that keep the commandments of God, [the Father,] and the faith of Jesus," [the Son.] "Blessed are they that do his [the Father's] commandments, that they may have right to the tree of life, and may enter in through the gates into the City." Amen. THOUGHTS OF THE PAST WORK OF WILLIAM MILLER, AND HIS ADHERENTS, RESPECTING THE TRUE STARTING POINT OF DAN. VIII., 14; IX., 24, 25. By Joseph Bates. On a recent visit to Low Hampton, N. Y., I visited the grave of this noble martyr. The inscription on the monument erected over his head, shows to the passing traveler, a book chiseled on the front of the marble slab, representing a Bible. On the first page, in large black letters, are the following appropriate words: "And he said unto me, Unto two thousand and three hundred days, shall the sanctuary be cleansed." Dan. viii., 14. Here, thought I, lies the venerable man that so successfully repeated these words of holy writ. With this mighty weapon of the Lord of hosts, accompanied by his Spirit, in a few years did he make up and move out a company of faithful believers who associated with him, and awakened and aroused "them that slept in the dust of death." Rev. iii., 6, 7. Where, thought I, is the advent believer that does not know that the world was aroused by a calculation on time, from the 2300 days of Dan. viii., 14, the vision of days that had been seen up from the church of God for nearly 2400 years. [For example, see Dan. viii., 17; Dan. viii., 4, 9, 12.] The opening of this book [Rev. x.] in "the time of the end," and the presentation of this mighty truth, was what aroused the people to prepare for the cleansing of the sanctuary, and the coming of the Lord Jesus. No advent believer can doubt but this was the main cause that drew out of the world, and stirred up the starting point of the 2300 days. All that have read his works, or heard him preach, know that he commenced the days from the 7th year of the reign of Artaxerxes, n. c. 457, in harmony with Dan. viii., 24, 25. That he was right in starting the vision with this date, the church of God was in the past advent history, and, when many in the churches and outside raised their objections, and were all silenced by the united efforts of himself and his associates, before the ending of the days in the Fall of 1844. Then the thought of the strange scenes of distraction that followed his work, even by his associates and professed friends, since the passing of the time. From the Albany, N. Y. Conference of April, 1845, to the Spring of 1850, the work of moving this starting point went forward. March 2d, 1850, the Advent Herald gave forth an article, showing that the plainly stated year had brought us more to the original starting point of Daniel's vision of the 2300 days, from the same press, by S. Bliss, entitled, Analysis of Sacred Chronology, covering the whole ground of unanswered arguments. The Herald speaks as follows: "It is by the Council of Ptolemy that the great prophetic period is fixed every weekday at 12 o'clock. This council places the seventh year of Artaxerxes in the year A. D. 457; and the accuracy of this date is demonstrated by the concurrent agreement of more than twenty scholars. The majority breaks off from the going forth of a decree respecting the restoration of Jerusalem. The council, however, cast themselves back to the seventh year of Artaxerxes. This data we cannot change from 457, without first demonstrating the incapacity of Ptolemy's Canon. To do this, it would be necessary to show that the large number of eclipses, in which his inscriptions have been repeatedly demonstrated, have not been correctly computed; and such a result would unsettle every chronological date." LETTERS. [Some of the brethren and sisters, who have not the privilege of meeting with those of like precious faith, will write that the paper published in the Review is all the same among things they have. This is the case with quite a large portion of our readers. This is the reason why we have published so many letters. This paper contains a large number of short letters, which will be cheering to at least a portion of the readers of the Review. Those who write letters for publication, should guard against a cold formal style. Write as you would speak, in a warm, clear, lively style; then the more your views and feelings. Then look over what you have written, and strike out useless words, and then copy with care on another sheet.] From Sister Huntley. Dear Bro. White:—Since last we met and parted with you, I have had a desire to write to you. I often think of the wonderful display of God's power and goodness, in raising me up at the time of the Washington Conference. Truly, I was sickly then, but for the truth's sake he did it. I can never call those days by remembrance, without feeling a kindling in my soul of the sacred flame. The Lord truly stood by us after we returned home to protect and deliver either from taking cold, and from the power of the enemy. I was taken the second night about midnight, with a severe attack of the pleurisy, in the side that was lame. We cried unto the Lord. We thought of no other remedy, and dare not look to any other source for help. I never was so sensible of the power of the enemy before. It seemed his hold was as an iron grasp about me. But the Lord was near to deliver in this time of need, and we soon retired again to rest, praising the Lord for victory. We found the promise sure. James iv, 14, 15. Again we saw our faith must be tried. And more precious was it to us, than gold seven times purified. I am very sure that had it not been for the prayer of faith, I should have been confined there for weeks. I feel to praise the Lord that I know his promise is sure. When we fulfill our promise, it is you and among us. We have been sadly disappointed, and sore afflicted of late. Death has entered our dwelling. Our little son, two years old, has fallen asleep under its power. We weep and mourn, yet not as those who have no hope. We know he sleeps in Jesus, and such, God will soon bring from the land of the enemy.—The circumstances are so distressing I can hardly relate them. When I reflect that one forethought of my own might have prevented it, I feel that I can never forgive myself. Friday, Dec. 17th, I took a large kettle with a pailful of hot water from the stove, sat it down in the adjoining room, stepped back to prepare the fire for another kettle, when I heard the noise of the kettle and sprang to the door, and behold my son upon the floor, and the hot water pouring upon him; you can have a little idea of my feelings as I took him up, strangling for breath; the scene was dreadfully painful. * * * * We applied what we had, that we at first thought of, and plead with the Lord for his life. After being dressed, he fell asleep. When he awoke he appeared better than was expected. It was about noon when the accident happened. It was immediately after a storm, the roads were bad, and the distance being such, it seemed almost impossible then to follow the direction of James v, 14. The next day being the Sabbath, and our meeting being only three miles distant, we thought we should have an opportunity to send for the brethren, and we did not expect that Bro. Wheeler was at home. But ere the morrow came, another scene was before us. Death had claimed the child for its own, and we were left in sorrow, no more to hear his prattling voice, or share his sweet kiss. It seemed at first that we could not be reconciled; but the blessed hope of Jesus' soon coming, and the promise of God to bring all that sleep in Jesus with him, cheers our bereaved hearts, and we are admonished to weep not for him. Through the providence of God, Bro. Wheeler was kept at home, and was here to attend the funeral.—He spoke from 1 Thess. iv, 13, 14. The Holy Spirit so accompanied the word, that for a season our mourning was turned to rejoicing. I do not wonder that the Apostle exhorted to "comfort one another with these words." Some of our dear brethren and sisters from Washington, Newport, and Unity, were present to sympathize with us, and assist in singing, this was a source of consolation also. Hubbard was taken down with a cold, which fastened upon his lungs, about a week after our little boy was buried. He was confined to the house a fortnight, when his cough became so alarming, we felt something must be done. According to James v, 14, sent for the Elders, Bro. Wheeler was gone.—Bro. J. Stowell, N. Mead, T. B. Mead, and S. Smith, came and followed the directions of the word. The Lord was true to his promise. His cough was almost entirely gone, his appetite restored, insomuch that it was ordered as formerly that something should be given him to eat. He has been out about his business every day since. As I look upon the past, and see how wonderfully the Lord has delivered us, through the prayer of faith, I feel that we had not followed the same direction in the case of little son. You would like to know how the cause is prospering with us, I rejoice to say that it is rising. Our meetings are interesting, and profitable. When we commenced holding meetings in this vicinity, our number was small, now there is quite a room full.—Then we had meetings only once a month, the rest of the time we went to Washington. Now we have them nearly every Sabbath at different places, in this vicinity. A young man in Newport has recently come out. There are others that seem interested. I think they will decide to be on the Lord's side. Bro. S. Smith, with a number of others, have held meetings the two First-days past at Claremont, and Charleston. The brethren have a mind to work, are all of one heart and mind. L. G. HUNTLEY. Lampeter, N. H., Jan. 18th, 1853. From Bro. Dyer. Dear Bro. White:—I feel anxious to add my testimony to the truth, and tell what the Lord has done for my soul. About a year ago last November, I was deeply convicted of my guilt as a sinner. I strove against the Holy Spirit until I was subdued by its power, and was willing to have Christ rule and reign over me. The place where I trust I was made happy in a Saviour's love, was in the city of Brooklyn. I there joined the Methodist class. In February I removed to the city of New York, where I was still under Methodist influence. I there commenced attending Baptist meetings, and became interested in the subject of Baptism, and was baptised as a member of the Shiloh Baptist Church. There I remained until last October, when I received a letter of dismissal to unite with the Second Baptist Church in Rochester. The brethren of this church, most of them, seemed to be in a lukewarm state. I soon heard of a place where the power of religion was felt. I commenced attending meetings of those who observe the seventh-day Sabbath, and very unexpectedly I was convinced of having broken the fourth commandment. I now endeavored to get rid of this; but my conscience, and the evidence I could obtain from the Bible, led me to lift the cross, and confess the truth. Previous to this, I had heard very little of so speedy a close of time, or the third angel's message. But I do thank God for this glorious truth, and that he has now kept the seventh day as the Sabbath of the Lord our God. Our Savior suffered without the gates of the city, and I now go without the camp, bearing the reproach. There is something here that plainly shows a mark, or a seal, that God uses to set apart and sanctify a people for himself. Here is something that separates us from the love of the world, from its fashions and customs, and unites us in closer communion with our heavenly Father. "Here is the patience of the saints; here are they that keep the commandments of God, and the faith of Jesus." I am willing to suffer for one. Christ strengthening me I can do all things, though thorns and thorns may tear me here, yet my eye is fixed upon the Morning Star. "For yet a little while and he that shall come will come, and will not tarry." O may God give me sufficient grace to endure, until I meet him in peace. We who embrace the third angel's message, and refuse the mark of the beast, in a little while shall see the King in his beauty, and shout our sufferings over. The pearly gates of the New Jerusalem will swing back upon their glittering hinges, and Jesus will bid us enter in. Is not such a hope as this enough to give us courage? A home in glory is enough to cause us to leave this world behind, and seek a holier, happier clime, where we shall gaze on the unveiled glory of King Jesus for ever. Your brother striving for eternal life. ISRAEL DEVOL. Rochester, N. Y. From Bro. Flower. Dear Bro. White:—The Conference in this place was one of much interest, and we trust will prove a blessing to the scattered remnant in this region. The third angel's message was clearly presented by Bro. Wheeler, who came to us richly laden with the blessings of the gospel. The Holy Spirit was poured in the Spirit, and the power of God attended to the close. The brethren from South Hawley met with us, and a few from other places. While the word was preached, a solemnity seemed to rest upon the people; and notwithstanding that some scoffed, and the enemy marshalled his army against the truth, yet here and there could be seen the falling tear among the crowd. Some arose and requested prayers, others, since the close of the meeting, have become serious, and some we trust, have found peace in believing. O, I bless the name of the Lord that he has power on earth to forgive sins. The Lord is still to work among the children of the believing parents in this place, and to him be all the praise. There is a power in the third angel's message that honest souls cannot, and will not resist. I feel like giving all for Christ, laying all upon the altar.—An entire regeneration is needed, in order that we may be enabled to stand in the great day of the Lord's anger. It is indispensable. Brethren, let us seek for it, until it is obtained. Yours, looking for him who is our Life. H. FLOWER. Ashfield, Mass., Jan. 30th, 1853. From Bro. Case. Dear Bro. White:—I write a few lines to you, although tired and worn down by constant labor and hard journeys. I left Grand Rapids this morning with Bro. M. G. Kellogg for Bedford. The Lord has, we trust, some jewels in Grand Rapids. I tarried with them over one week, and gave them ten lectures. The first that was ever given in this place. The interest increased all the time, but I could not stay any longer, as my appointment called me to Bedford.—Much labor is wanted here at this time, by those who are experienced. I have more calls than I can possibly attend, yet by the help of the Lord I will give myself anew to the glorious work, as long as life and strength is granted me. I see the truth is taking hold of the people, and there is an increase of interest to hear among many that belong to the nominal churches. They appear to have been fairly starved out for spiritual food. How cruel are those shepherds that have fed themselves, and not the flock. I feel to praise the Lord that he is at work for the remnant, and the honest ones are beginning to see. My heart is cheered to hear from the brethren through the Review. May the Lord bless them in their arduous labors. But do not forget the West. We do want help, but the will of God be done. Yours in love, H. S. CASE. Hendaye, Mich., Jan. 12th, 1853. From Bro. Hart. Dear Bro. White:—I rejoice that the third angel's message is gaining ground, and that the people are beginning to wake up to the subject. Our meetings are becoming more interesting, and the Lord is calling to our numbers, such as I hope, will be saved. The light is increasing, and the saints are growing stronger. The Lord is with his people that keep his commandments. We have every reason to believe that this is the Lord's truth; and surely it is time for those who believe in the message, to be waking up. Satan is rallying all his forces, and seems to choose those who have had great light to work through. May we be found with our loins girt with truth, and we like men that wait for the Lord, when he shall return from the wedding. JOSIAH HART. Northfield, Vt., Feb. 2d, 1853. From Bro. Rockwell. Dear Bro. White:—I wish to say to the remnant scattered abroad, Hold fast whereto you have attained, that no man take thy crown. We know that we have the truth, and the truth makes us free, so long as we obey it by doing all His commandments; Christ says, He that heareth these sayings of mine, and doeth them, I will liken him unto a wise man who built his house upon a rock. But he that heareth them, and doeth them not, is likened unto a foolish man that built his house upon the sand. We must not be forgetful hearers, but doers of the word.—Some say they are not under the law, but under grace. Now the law was made for the lawless and disobedient. And when one person breaks, or disregards any one or all of the ten precepts of the immutable laws of God, he is no longer under grace; because he is a transgressor of the law, and must repent and forsake his sins, and have that godly sorrow which worketh repentance unto life, before he can say that he is under grace. Grace implies favor, or the love of God. This is the love of God, that ye keep his commandments. Truth makes us free and hold to rebuke sin, and contend earnestly for the faith once delivered to the saints. Now, dear brethren, walk in the love and simplicity of the gospel, keeping the unity of the Spirit in the bonds of peace. Speaking the same things, being of one mind, loving one another with a pure heart fervently. And may the God of peace sanctify you wholly. Our Saviour's language was, Father, sanctify them through thy truth, thy word is truth. We have the word of God to read, and by it we shall be condemned, or sanctified by faith on the Son of God, who hath given himself for us, that he might redeem us from under the curse of the broken law. For, verily, if we keep the whole law, both in the spirit and in the letter too, we shall only be saved by faith on the Son of God. For we have all broken the law, which Paul says is holy, just and good. Oh, brethren, pray in faith that our Great High Priest will make a full atonement for us, that all our sins and errors may be confessed over the head of the scape-goat, and blotted out. We having confessed and forsaken them, will find mercy, and the author of all sin shall bear them to a land not inhabited, or wilderness. Otherwise, our sins will be upon our own heads, and we shall be the subjects of divine wrath. Oh, try to make sure work, while he yet maketh intercession for us. From your unworthy brother, in patient waiting for the coming kingdom, N. W. ROCKWELL, Parsham, C. B., Feb 2d, 1853. From Bro. Jackson. Dear Bro. White:—I write a few lines to let you know that we are trying to keep the commandments of God, and the faith of Jesus. We are striving to be ready to meet our blessed Lord when he shall come from heaven, in like manner as he ascended. O what a glorious scene that will be for those that are ready! We are striving to have our hearts in order, that we may share largely of the refreshing when it shall come from the presence of the Lord. The Review is a welcome messenger to us; we know not what we should do without it. We have no Sabbath School, for there is no children here except our two little ones, one eight, the other four. They are much interested with their little paper, the Instructor. Joseph Jackson. Corunna, Mich., Jan. 23d, 1853. From Bro. Barr. Dear Bro. White:—I wish to say that our Conference of two days at this place, has been one of thrilling interest to the saints who came together from a large number of towns in this vicinity. Our Heavenly Father presided over it, from its commencement to its close. The Spirit of the Lord filled the hearts of his waiting people, causing them to rejoice with joy unspeakable and full of glory. God's power, to preserve soul, body and spirit blameless unto the coming of the Lord Jesus, was made known. The power of the enemy is mighty; but the power of our God is almighty. Many never saw it on this wise before.—It was truly a time of the Lord's power. A number for the first time confessed the truth, with a determination to be purified by obeying it, and to go with the remnant that keep the commandments of God. The third angel's message will soon become a loud cry. The saints begin to realize it, while indications of the latter rain at hand, are thickening on every side. My soul shall rejoice in God for what he has done and is still doing for his people. It seems as though the remnant in Northern Vermont had heard the command, Move forward, and were obedient. My cry is, Lord speed the remnant on to Mount Zion. Yours expecting the crown after the trial, E. L. Barr. Johnson, Vt., Jan. 24th, 1853. From Sister Paine. Dear Bro. White:—In looking over the letter I wrote to you Dec. 12th, I see there is a mistake. I wrote that there was no meetings within five miles. Eagle Harbor, and Albion are each five miles from us; but there are meetings nearer. Far be it from me to wish to state what is not true, or do any thing to injure the cause of Christ. (Yet I do not consider it as such in the sight of God.) But there are some in this place who feel disposed to call it a lie. May God forgive them the wrong they have done in trying to circulate it. And O, may the Lord enable me so to watch every word and act, that the world may have no reason to gainsay. It is my earnest prayer that my daily walk may be such as shall adorn my profession by a well ordered life and godly conversation. Since I wrote to you, Brn. Chapin and Lindsey have been here and have been studying. There are six in this place, who are endeavoring to keep all of the commandments of God. We meet three times in a week, and the Lord does truly meet with us and bless us. O, I do rejoice in the glorious hope of soon seeing Jesus. Then if we are his children, we shall be made like him, and shall reign with him for ever. Yours in hope of eternal life, E. J. Paine, Barre, N. Y., Feb. 27th, 1853. From Sister Prior. Dear Bro. White:—I am thankful to God that I have the privilege of reading the Letters in the Review, from the children of God, that are scattered abroad, for I feel that I am a stranger, in a strange land, in every sense of the word. I have recently moved into this State, and know no one that keeps the Sabbath of the Lord, although they think that they are keepers of the commandments. I have lived in this place three months, and am a stranger to all but my family. I am looked upon as a Millerite; and with some persons, almost everything that pertains to the second coming of our Saviour, is called Millerism. I have not the privilege of meeting with any one on the Sabbath. I do wish that my husband could hear the truth preached; he would if he had a chance. I want to reflect the image of Jesus, and in heart I want to be meek and lowly. I do wish that a preacher of the truth would come this way. Shall I ever hear them speak again? O, I pray that I may! I live thirty-seven miles south of Cleveland. Yours in hope, Lorette H. Prior. Seville, Medina Co., Ohio, Jan. 23d, 1853. From Bro. Lawrence. Dear Bro. White:—The Lord is adding to our number here in Bangor. The cause of God is rising in a number of places in this part of the State. Some in St. Lawrence Co. have embraced the present truth since my return, and they are more favorable, and begin to see a harmony in the truths connected with the third angel's message, and are willing to investigate them. We are not without our trials in common with God's people elsewhere; but find it our privilege to overcome and maintain our freedom, by obeying the truth. I am more than ever satisfied with our present position, and feel that it is our duty to give the trumpet a certain sound, and thus give the warning of coming wrath. Horace W. Lawrence. Bangor, N. Y., Jan. 23d, 1853. From Bro. Hastings. Dear Bro. White:—The Review comes to us laden with the truths of God's word, and is truly meat in due season. I love the Spirit manifested, and the cheering letters of the brethren from abroad. I rejoice for the increase of numbers, and for the prospect that the truth will soon go with a loud "voice." I desire to live very humble, and be willing to suffer with Jesus the little time which is allotted us in this state of trial, that I may reign with him in his coming kingdom. I think the cause in New Hampshire is rising. Our meetings are becoming more spiritual, and the brethren and sisters are more interested in the present truth. Last Sabbath, the 15th, we held our meeting at Sister Smith's, Wilton. Bro. and Sr. Mead were present. A good degree of the Spirit pervaded the place. Bro. M. brings a good report of the brethren in Washington. Yours in hope, L. Hastings. New Ipswich, N. H., Jan. 24th, 1853. From Bro. Davis. Dear Bro. White:—For the first time I take the liberty of writing a few lines, to let you know that I am a Sabbath-keeper. I have been taking your paper for a few months, and am much pleased with the gems of truth which it contains, and am desirous of having it spread through our country as widely as possible. W. C. Davis. Sullivan, Ind., Jan. 10th, 1853. Providence of God in Minute Events. It is an erroneous view to think of God as governing the grand phenomena of nature, and leaving those which are minute to the operation of a set of laws which he does not uphold at every moment in all the fullness of their application. "We cannot," says Chalmers, "disjoin God from one particle of the universe of God." We may despise what is small as beneath the notice of our pride, but nothing is too microscopic for Him to know, while "he measures the waters in the hollow of his hand, and draweth out heaven with a span, and comprehends the depth of the earth in measure, and weigheth the mountains in scales, and the hills in a balance;" yet numbers the very hairs of our head, and knows of every sparrow that falls to the ground. The minuteness with which God provides for all wants is well brought out in the sixty-fifth Psalm, where David speaks of him as attending to the very setting of the furrows of the field, and the watering of the ridges. "Thou makest the out-goings of the morning and evening to rejoice; Thou visitest the earth and waterest it; Thou greatly enrichest it with the river of God, which is full of water; Thou preparest them corn, when Thou hast so provided food; Thou wastest the ridges thereof abundantly; Thou settest the furrows thereof; Thou makest it soft with showers; Thou blessest the springing thereof; Thou crownest the year with thy goodness, and thy paths drop fatness."—Balfour. Riches of the Old Testament. There is scarcely a surer evidence of low attainments in religion, than undervaluing the Old Testament. The mature Christian finds it rich in the same gospel as the New, and the New is the key to the Old. "If the Psalms," says Irving, in his brilliant introduction to Horne, "contain not the argument of the simple doctrines, and the detail of the issues of the gospel, to reveal which the Word of God became flesh and died at Calvary, yet now is the key given, and the door of spiritual life is opened, where we may find such spiritual treasures as the book of Psalms, wherein are revealed the depths of the soul's sinfulness, the stoutness of her rebellion against God, the horrors of spiritual desolation, the agonies of contrition, the blessedness of pardon, the joys of restoration, the constancy of faith, and every other variety of Christian experience? And if they contain not the narrative of Messiah's birth, life and death; or the labors of his apostolic servants, and the struggles of his infant church, as these are written in the books of the New Testament, where in the whole Scriptures, can we find such declarations of the work of Christ, in his humiliation and its glory, the spiritu
Unsupervised Modeling of Topical Relevance in L2 Learner Text Ronan Cummins ALTA Institute Computer Lab University of Cambridge UK, CB3 0FD email@example.com Helen Yannakoudakis ALTA Institute Computer Lab University of Cambridge UK, CB3 0FD firstname.lastname@example.org Ted Briscoe ALTA Institute Computer Lab University of Cambridge UK, CB3 0FD email@example.com Abstract The automated scoring of second-language (L2) learner text along various writing dimensions is an increasingly active research area. In this paper, we focus on determining the topical relevance of an essay to the prompt that elicited it. Given the burden involved in manually assigning scores for use in training supervised prompt-relevance models, we develop unsupervised models and show that they correlate well with human judgements. We show that expanding prompts using topically-related words, via pseudo-relevance modelling, is beneficial and outperforms other distributional techniques. Finally, we incorporate our prompt-relevance models into a supervised essay scoring system that predicts a holistic score and show that it improves its performance. 1 Introduction Given the increase in demand for educational tools and aids for L2 learners of English, the automated scoring of learner texts according to a number of predetermined dimensions (e.g., grammaticality and lexical variety) is an increasingly important research area. While a number of early approaches (Page, 1966; Page, 1994) and recent competitions$^1$ (Shermis and Hammer, 2012) have sought to assign a holistic score to an entire essay, it is more informative to give detailed feedback to learners by assigning individual scores across each such writing dimension. This more specific feedback facilitates reflection both on learners’ strengths and weaknesses, and focuses attention on the aspects of writing that need improvement. Recent work outlines a number of broad competencies that systems should assess (Kakkonen and Sutinen, 2008). These include morphology, syntax, semantics, discourse, and stylistics, noting that the specific assessment tasks that might aim to measure these areas of competency may vary. One dimension against which a piece of text is often scored is that of topical relevance. That is, determining if a learner has understood and responded adequately to the prompt. This aspect of automated writing assessment has received considerably less attention than holistic scoring.$^2$ Topical relevance is not so much concerned with whether an L2 learner has constructed grammatically correct and well-organised sentences, as it is concerned with whether the learner has understood the prompt and attempted a response with appropriate vocabulary. Other reasons for measuring the topical relevance of a text include the detection of malicious submissions, that is, detecting submissions that have been rote-learned or memorised specifically for assessment situations (Higgins et al., 2006). In this paper, we employ techniques from the area of distributional semantics and information retrieval (IR) to develop unsupervised prompt-relevance models, and demonstrate that they correlate well with human judgements. In particu- $^1$https://www.kaggle.com/c/asap-aes $^2$We note that a recent paper (Persing and Ng, 2014) has referred to this task as prompt-adequacy, while we use the terms prompt-relevance and topical-relevance interchangeably throughout this paper. lar, we study four different methods of expanding a prompt with topically-related words and show that some are more beneficial than others at overcoming the ‘vocabulary mismatch’ problem which is typically present in free-text learner writing. To the best of our knowledge, there have been no attempts at a comparative study investigating the effectiveness of such techniques on the automatic prediction of a topical-relevance score in the noisy domain of learner texts, where grammatical errors are common. In addition, we perform an external evaluation to measure the extent to which prompt-relevance informs (Rotaru and Litman, 2009) the holistic score. The remainder of the paper is outlined as follows: Section 2 discusses related work and outlines our contribution. Section 3 presents our framework and four unsupervised approaches to measuring semantic similarity. Section 4 presents both quantitative and qualitative evaluations for all of the methods employed in this paper. Section 5 performs an external evaluation by incorporating the best prompt-relevance model as features into a supervised preference ranking approach. Finally, Section 6 concludes with a discussion and outline of future work. 2 Related Research There are a number of existing automated text-scoring systems (sometimes referred to as essay scoring systems). For an overview, the interested reader is directed to reviews and advances in the area (Shermis and Burstein, 2003; Landauer, 2003; Valenti et al., 2003; Diki, 2006; Phillips, 2007; Briscoe et al., 2010; Shermis and Burstein, 2013). In this section, we review related research on topical-relevance detection for automated writing assessment, and outline the key differences between our approach and that of existing work. A wide variety of computational approaches (Miller, 2003; Landauer et al., 2003; Higgins et al., 2004; Higgins and Burstein, 2007; Chen et al., 2010) have been used to automatically assess L2 texts. Early work on topical relevance (Higgins et al., 2006) posed the problem as one of binary classification and aimed to identify whether a text was either on or off-topic. The main motivation of the research was to detect off-topic text, text submitted mistakenly (within an online assessment setting), or text submitted in bad faith (i.e., possibly memorised on an unrelated topic). They adopted an unsupervised approach to the problem, where they matched each text to its corresponding prompt using tf-idf weighted content vectors and a similarity function. One of the heuristic approaches employed in that work was to calculate the similarity of an essay to a number of unrelated prompts. If the essay was closer to an unrelated prompt than the relevant one, the essay was deemed to be off-topic. Briscoe et al. (2010) tackle the problem of off-topic detection using more complex distributional semantic models that tend to overcome the problem of vocabulary mismatch. However, they frame the task as binary classification and evaluate their approach by determining if it can associate a learner text with the correct prompt. The work which is closest in spirit to that of our own is by Louis and Higgins (2010), who expand prompts using morphological variations, synonyms, and words that are distributionally similar to those that appear in the prompt. Their work builds on the earlier work by Higgins et al. (2006), and again pose the problem as one of binary classification. The most recent work of Persing and Ng (2014) involves scoring L2 learner texts for relevance on a seven-point scale using a feature-rich linear regression approach. While they demonstrate that learning one linear regression model per prompt is a useful supervised approach, it means that substantial training data is needed for each prompt in order to build the models. For the task of determining topical relevance, this places a substantial burden on manually annotating texts for each individual prompt.\footnote{In fact, it is often the case that there are multiple prompts per exam, which change for every exam sitting.} As a result, supervised prompt-specific approaches are impractical and less flexible in an operational setting; if, for example, a new previously-unseen prompt is required for an upcoming assessment, the model cannot be applied until a sizeable amount of manually-annotated response texts are collected and annotated for that prompt. A dataset developed from the international corpus of learner data (ICLE) (Granger et al., 2009) consisting of 830 essays measured for relevance against one of 13 prompts on a seven-point scale was released as part of that work (Persing and Ng, 2014). We make use of this new resource in our work as it is the only such public dataset.\footnote{www.hlt.utdallas.edu/~persingq/ICLE/paDataset.html} We make the following contributions to the automated assessment of topical relevance: - We perform the first systematic comparison of several unsupervised methods for assessing topical relevance in L2 learner text on a publicly available dataset. - We adopt a new unsupervised pseudo-relevance feedback language-modelling approach and show that it correlates well with human judgements and outperforms a number of other distributional approaches. - We perform an external evaluation of our best prompt-relevance models by incorporating them into the feature set of a supervised prompt-independent text-scoring system, and show that they improve its performance. 3 Semantic Prompt Relevance Previous research (Higgins et al., 2006) has shown that representing a prompt $p$ and an essay $s$ as tf-idf weighted vectors\footnote{We use bold lower-case letters throughout to denote vectors, including probability vectors.} $p$ and $s$ in the term space $\mathbb{R}^v$ (where $v$ is the vocabulary of the system) yields useful representations for exact matching using cosine similarity as follows: $$cos(p, s) = \frac{\sum_{t \in v} p_t \cdot s_t}{\sqrt{\sum_{t \in v} p_t^2 \cdot \sum_{t \in v} s_t^2}} \quad (1)$$ However, it is likely that many L2-learner texts will use words that are related to the prompt, but which do not have an exact match to any words contained in the prompt. Therefore, we extend this approach by aiming to expand the prompt $p$ with a set of topically related expansion terms $e$ using one of a number of distributional similarity techniques. 3.1 Prompt Expansion As a general method of prompt expansion, we represent the prompt $p$ and each candidate expansion word $w$ as vectors $p$ and $w$ in an $n$-dimensional space $\mathbb{R}^n$, and then use some measure of similarity between the two vectors (e.g. cosine similarity) to rank the candidate expansion words according to how close they are to the original prompt. We then select the top $\|e\|$ most similar expansion terms to add to the original prompt. Once the $\|e\|$ closest terms are selected and added to the original prompt $p$, we create a tf-idf weighted expanded prompt vector $p_{p+e}$ and compare it to the tf-idf essay vector $s$ using cosine similarity in the original space $\mathbb{R}^v$ as per Equation (1). In our approach, we conduct the essay matching in the term space $\mathbb{R}^v$ as it allows us to analyse the quality of the expansion terms, and subsequently to understand the merits and demerits of the various approaches. We now outline four methods of selecting candidate prompt expansion terms. 3.2 Traditional Distributional Semantics Our first approach involves building traditional distributional vectors by constructing a matrix of co-occurrence frequencies. For a specific word $w$, its vector is constructed by counting the words (its context words $c$) that it co-occurs with in a specified context (usually a window of a few words). The row for a specific word $w$ then represents the vector for that word. We weight the vector elements using the PPMI (positive pointwise mutual information) weighting scheme (Turney et al., 2010). We build word vectors using a lemmatised version of Wikipedia from 2013. We removed from the corpus all words that appeared less than 200 times and used the 96,811 remaining words as both potential expansion words $w$ and as contexts $c$. We used a 5 word context window (2 words either side of the target word) and reduced the size of the resultant vectors by only storing dimensions that had a PPMI greater than 2.0 (Turney et al., 2010). The resultant vectors are competitive with the best reported results for traditional word vectors on a word–word similarity task (Spearman-$\rho = 0.732$ on 3000 word-pairs from the MEN dataset) (Levy et al., 2015). We create a vector representation for the prompt $p$ in $\mathbb{R}^{\|v\|}$ by summing the PPMI word-vectors of the words occurring in the prompt. Finally, the $\|e\|$ closest words to the prompt vector $p$, as measured by cosine similarity, can then be selected as expansion terms. 3.3 Random Indexing Random Indexing (RI) (Kanerva et al., 2000) is an approach which incrementally builds word vectors in a dimensionally-reduced space. Words are initially assigned a unique random index vector in a space $\mathbb{Z}^n$, where $n$ is user-defined. These near-orthogonal vectors are updated by iterating over a corpus of text. In particular, the word vector for a specific word $w$ is altered by adding to it the vectors of the words in its contexts. The process proceeds incrementally and therefore only requires one pass over the data. In this way, words that occur in similar contexts will be pushed towards similar points in the space $\mathbb{Z}^n$. We use Random Indexing to build word vectors using the S-Space package\footnote{https://github.com/tozzetthebeat/S-Space} using the same preprocessed Wikipedia corpus as outlined in the previous section. We used a dimensionality of 400 with window sizes up to 5 words (finding a window of 5 words to create better vectors for the word-word similarity task). The resultant vectors are not as competitive as those built using the traditional approach on a word-word similarity task (Spearman-$\rho = 0.432$ on 3000 word-pairs from the MEN dataset). Again, we create a vector representation for the prompt $p$ by summing the RI vectors, and find the closest words vectors $w$ to the prompt. 3.4 Word Embeddings The continuous bag-of-words architecture (cbow) and the skip-gram architectures (skip) in word2vec have been shown to be particularly well-suited to learning word-embeddings (i.e. low-dimensional vector representations of words) (Mikolov et al., 2013). The word2vec package\footnote{https://code.google.com/p/word2vec/} from Mikolov is the original implementation of these models. We use word2vec to learn distributed representations for prompts in a similar manner to that just outlined (in Section 3.2 and Section 3.3). In particular, we learn distributed vectors using both cbow and skip and the same preprocessed version of Wikipedia as used previously. We used word vectors of length 400 for both architectures with a window of 5 for cbow and 10 for skip-gram as recommended in the original documentation. For both approaches we use negative sampling. The performance of these approaches on the word-word MEN dataset are $\rho = 0.737$ and $\rho = 0.764$ for cbow and skip respectively. As with previous approaches, we create a vector representation for the prompt $p$ by summing the vectors of the words in the prompt. 3.5 Pseudo-Relevance Feedback Pseudo-relevance feedback (PRF) is a technique in IR for expanding queries with topically related words. In PRF, the top $\|F\|$ ranked documents for a query are deemed relevant and candidate terms occurring in these documents are analysed and selected according to a term-selection function. Each candidate word can be viewed as being described by a vector of contexts of dimensionality $\|F\|$ (i.e. where the entire document $d \in F$ is the context). We use this approach by using a prompt analogously to a query. In the popular relevance modelling (RM) framework (Lv and Zhai, 2009), the term-selection value can be viewed as the dot-product of a prompt vector $p$ (the vector of similarities between the initial prompt $p$ and the document contexts $d \in F$) and the candidate word vector $w$ (the vector of weights for word $w$ in its contexts $d \in F$) as follows: $$PRF(p, w) = \sum_{d \in F} f(w, d) \cdot Pr(p\|d)$$ \hspace{1cm} (2) where $f(w, d)$ is the weight of the candidate word $w$ in document $d$ and $Pr(p\|d)$ is the probability that $d$ generated $p$, i.e. the query-likelihood (Ponte and Croft, 1998). Furthermore, by selecting only the most important dimensions (i.e. top $\|F\|$ documents), dimensional reduction is automatically incorporated in an operationally efficient manner. PRF can be viewed as a dimensionally-reduced probabilistic version of Explicit Semantic Analysis (ESA) (Gabrilovich and Markovitch, 2007). The typical dimensionality used for PRF is usually of around $\|F\| = 20$. In the language modelling framework, documents are assumed to have been generated by a mixture of a topical model $\alpha_c$ and a background model $\alpha_b$, such that $d \sim (1 - \omega) \cdot \alpha_c + \omega \cdot \alpha_b$ where $\omega$ is the mixture parameter. Given a candidate term $w$ appearing in $d$, the probability that it was generated by the topical model is as follows: $$f(w, d) = p(\alpha_r \| w) = \frac{(1 - \omega) \cdot \alpha_r}{(1 - \omega) \cdot \alpha_r + \omega \cdot \alpha_c}$$ \hspace{1cm} (3) and therefore, we use this probability of topicality $f(w, d)$ as the vector weights for $w$. Assuming that documents have been generated by a multivariate Pólya distribution (Cummins et al., 2015), $f(w, d)$ is as follows: $$f(w, d) = \frac{tf_{w,d}}{tf_{w,d} + \frac{\omega m_d df_w}{(1-\omega)\sum_{w'}df_{w'}} + \frac{\|d\|}{m_d}}$$ \hspace{1cm} (4) where $tf_{w,d}$ is the term-frequency, $df_w$ is the document frequency of $w$ in the collection being searched, $m_d$ is the number of unique terms in the document, $m_c$ is the background mass (Cummins et al., 2015), and $\omega = 0.8$ is a stable hyper-parameter that controls the belief in the background model. Essentially, this approach (denoted $PRF$) selects terms that occur more frequently in the top $\|F\|$ documents than they should by chance. As our documents, we use the same preprocessed Wikipedia corpus as outlined previously. ### 4 Evaluation of Expansion Methods In this section, we present results on the effectiveness of the unsupervised approaches for the task of assessing the prompt relevance of an essay. #### 4.1 Data and Experimental Setup For the first set of experiments, we use 830 L2 learner essays from the ICLE dataset that are assessed for prompt relevance across 13 prompts. This corpus consists of essays written by higher intermediate to advanced learners of English, which corresponds to approximately B2 level, or above, of the CEFR (Common European Framework of Reference for Languages). The scores assigned to the essays range from 1.0 to 4.0 in increments of 0.5 (although all essays received a score of 2.0 or more in the dataset as seen in Table 1). The essays were double-marked and the linear correlation$^8$ between the assessors was 0.243 (a weak correlation). The distribution of essays per prompt is included in Table 2. We lemmatised all prompts and essays using RASP (Briscoe et al., 2006). A point worth noting is that there are minimal essay-length effects in operation on this dataset. The Spearman correlation between the length of the essay and the human-assigned prompt-relevance score across all 830 essays is $\rho = 0.007$. As a baseline approach, we use the cosine similarity between the original prompt (unexpanded) and the essay $cos(p, s)$. For all expansion approaches, we set the number of expansion terms $\|e\| = 200$ and use the weight of association between the prompt and the expansion term as the expansion term’s frequency $tf$ value in the expanded prompt. We evaluate the approaches by calculating Spearman’s rank ($\rho$) correlation coefficient between each method’s predicted similarity score and the scores assigned by the assessors. #### 4.2 Results for Prompt Relevance Table 2 (Top) shows the performance of the approaches over 11 prompts.$^9$ On average, all approaches increase over the baseline. We can see that the most consistent approach is the PRF approach as it improves over the baseline in 10 out of 11 prompts. The RI approach also performs well and is the best approach on many of the prompts. However, to measure the topical quality of the expansion words selected by each approach in isolation, we removed the original prompt words from the expanded prompts and again calculated the performance of the different approaches. This more rigorous evaluation in Table 2 (Bottom) shows that the topical quality of the expansion words from the PRF approach tends to be better than the other approaches. We next look at the actual expansion words selected for two prompts. --- $^8$While this seems to suggest that the upper-bound on this dataset is quite low, the original work notes that 89% of the time, assessors graded within a point of each other. Furthermore, correlation is affected by scale (Yannakoudakis and Cummins, 2015). $^9$The two remaining prompts have only three essays associated with them. \| Prompt \| 1 \| 2 \| 3 \| 4 \| 5 \| 6 \| 7 \| 8 \| 9 \| 10 \| 11 \| Mean \| \|--------\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|------\| \| # of essays \| 237 \| 53 \| 64 \| 58 \| 131 \| 43 \| 80 \| 28 \| 49 \| 71 \| 13 \| \| \| lemmatized cos(p, s) \| 0.113 \| 0.026 \| 0.063 \| 0.211 \| 0.023 \| -0.111 \| 0.103 \| -0.115 \| 0.056 \| 0.171 \| 0.520 \| 0.045 \| \| ds\textsubscript{p+c} \| 0.328 \| 0.141 \| 0.182 \| 0.114 \| 0.208 \| -0.011 \| 0.340 \| 0.519 \| 0.280 \| 0.082 \| 0.637 \| 0.256 \| \| RI\textsubscript{p+c} \| \textbf{0.372} \| 0.098 \| 0.103 \| \textbf{0.214} \| 0.192 \| \textbf{0.093} \| \textbf{0.398} \| \textbf{0.720} \| 0.259 \| 0.116 \| 0.449 \| 0.274 \| \| cbow\textsubscript{p+c} \| 0.345 \| 0.125 \| 0.131 \| 0.114 \| 0.209 \| 0.068 \| 0.328 \| 0.581 \| 0.265 \| -0.024 \| 0.637 \| 0.253 \| \| skip\textsubscript{p+c} \| 0.359 \| 0.160 \| 0.183 \| 0.139 \| 0.245 \| 0.026 \| 0.363 \| 0.571 \| 0.278 \| -0.064 \| 0.677 \| 0.267 \| \| PRF\textsubscript{p+c} \| 0.348 \| \textbf{0.188} \| 0.126 \| 0.145 \| \textbf{0.260} \| 0.034 \| 0.340 \| 0.598 \| \textbf{0.335} \| 0.078 \| \textbf{0.679} \| \textbf{0.285} \| \| Prompt \| 1 \| 2 \| 3 \| 4 \| 5 \| 6 \| 7 \| 8 \| 9 \| 10 \| 11 \| Mean \| \|--------\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|-----\|------\| \| # of essays \| 237 \| 53 \| 64 \| 58 \| 131 \| 43 \| 80 \| 28 \| 49 \| 71 \| 13 \| \| \| ds\textsubscript{v\setminus p} \| 0.008 \| 0.043 \| -0.098 \| -0.073 \| -0.017 \| -0.092 \| 0.126 \| 0.619 \| \textbf{0.202} \| 0.029 \| 0.375 \| 0.102 \| \| RI\textsubscript{v\setminus p} \| \textbf{0.097} \| 0.016 \| -0.195 \| 0.326 \| 0.061 \| 0.091 \| \textbf{0.206} \| 0.572 \| 0.030 \| \textbf{0.185} \| -0.082 \| 0.119 \| \| cbow\textsubscript{v\setminus p} \| 0.080 \| 0.025 \| -0.209 \| 0.165 \| 0.071 \| \textbf{0.266} \| 0.088 \| \textbf{0.677} \| -0.079 \| -0.118 \| 0.230 \| 0.110 \| \| skip\textsubscript{v\setminus p} \| 0.087 \| 0.133 \| \textbf{-0.052} \| 0.167 \| 0.149 \| 0.188 \| 0.173 \| 0.592 \| 0.000 \| -0.171 \| 0.222 \| 0.135 \| \| PRF\textsubscript{v\setminus p} \| 0.079 \| \textbf{0.184} \| -0.055 \| \textbf{0.363} \| \textbf{0.151} \| 0.155 \| 0.157 \| 0.612 \| 0.161 \| 0.125 \| \textbf{0.455} \| \textbf{0.217} \| Table 2: Correlation (Spearman’s $\rho$) between essay–similarity scores and human annotations for each prompt (higher values indicate a better approach) for expansion methods when including original prompt terms (Top – denoted $p+c$) and when removing original prompt terms from the expanded prompt (Bottom – denoted $v\setminus p$). Best result in bold. # 2 – Most University degrees are theoretical and do not prepare us for the real life. Do you agree or disagree? \| ds \| RI \| cbow \| skip \| PRF \| \|----\|----\|------\|------\|-----\| \| faculty \| 0.222 \| accept \| 0.948 \| accept \| 0.598 \| however \| 0.077 \| theory \| 0.544 \| \| graduate \| 0.214 \| wish \| 0.919 \| psychology \| 0.558 \| nevertheless \| 0.677 \| study \| 0.444 \| \| professor \| 0.21 \| experience \| 0.918 \| understand \| 0.553 \| indeed \| 0.675 \| science \| 0.414 \| \| PhD \| 0.204 \| idea \| 0.915 \| study \| 0.551 \| insist \| 0.672 \| differ \| 0.396 \| \| mathematics \| 0.199 \| from \| 0.913 \| teach \| 0.55 \| accept \| 0.671 \| student \| 0.396 \| \| philosophy \| 0.195 \| work \| 0.912 \| philosophy \| 0.549 \| fact \| 0.67 \| philosophy \| 0.394 \| \| theory \| 0.194 \| acknowledge \| 0.911 \| knowledge \| 0.545 \| s \| 0.67 \| topic \| 0.392 \| \| sociology \| 0.189 \| nevertheless \| 0.911 \| argument \| 0.538 \| would \| 0.664 \| educate \| 0.372 \| \| science \| 0.185 \| notice \| 0.91 \| discuss \| 0.538 \| while \| 0.66 \| academy \| 0.361 \| \| study \| 0.182 \| nonetheless \| 0.909 \| theory \| 0.528 \| nonetheless \| 0.656 \| argue \| 0.354 \| # 9 – Feminists have done more harm to the cause of women than good. feminist harm cause women \| ds \| RI \| cbow \| skip \| PRF \| \|----\|----\|------\|------\|-----\| \| symptom \| 0.310 \| likewise \| 0.883 \| feminism \| 0.612 \| feminism \| 0.671 \| feminism \| 0.910 \| \| disease \| 0.275 \| consequence \| 0.882 \| sexual \| 0.583 \| landdyke \| 0.632 \| sex \| 0.896 \| \| illness \| 0.270 \| furthermore \| 0.879 \| violence \| 0.577 \| woman \| 0.617 \| sexual \| 0.896 \| \| chronic \| 0.266 \| effect \| 0.875 \| stigma \| 0.573 \| affect \| 0.594 \| oppress \| 0.883 \| \| treatment \| 0.260 \| response \| 0.873 \| perceive \| 0.573 \| twwa \| 0.580 \| argument \| 0.875 \| \| infect \| 0.256 \| moreover \| 0.871 \| affect \| 0.564 \| argue \| 0.580 \| rape \| 0.800 \| \| diagnosis \| 0.255 \| hinder \| 0.871 \| detriment \| 0.553 \| provoke \| 0.578 \| men \| 0.787 \| \| patient \| 0.255 \| excuse \| 0.869 \| homosexual \| 0.547 \| believe \| 0.574 \| gender \| 0.762 \| \| illness \| 0.253 \| lady \| 0.866 \| consequence \| 0.545 \| consequence \| 0.573 \| anti \| 0.749 \| \| disorder \| 0.247 \| perceptive \| 0.863 \| oppres \| 0.545 \| sexism \| 0.573 \| right \| 0.740 \| Table 3: The top 10 non-prompt words and their similarity to the prompt in a lemmatised Wikipedia corpus of 4.4M documents. ### 4.3 Qualitative Evaluation of Expansion Terms Table 3 shows the expansion words selected by each approach for two prompts (prompts # 2 and # 9). For prompt # 2 we can see the top words selected for RI and skip do not seem topically similar to the prompt. The top words for $ds$, $cbow$, and PRF seem on-topic and might be part of useful feedback to a learner writing for this prompt. For prompt # 9, $ds$ and RI do not tend to promote topically related words. The words for the $ds$ approach seem to be related to topic of diseases as it may have been mislead by some of the prompt words. In fact, the top terms promoted by the RI approach are not particularly on-topic for any of the 11 prompts, despite the empirical evaluation in the previous section. This could be because some topical words appear further down the ranking for RI. We believe the main reason that the PRF approach outperforms the others is that topicality is a quality that spans larger segments of text (e.g. documents). For the other approaches, the words that are promoted are very close in proximity to the prompt words (due to the smaller context sizes), and this is more likely to capture local aspects of word usage. Furthermore, in the PRF approach the most important contexts are those in which all prompt words appear together, and this aids automatic disambiguation. Regardless, due to the empirical results in the previous section and the perceived topical quality of the terms from the PRF approach, we make use of the PRF approach as a feature in the next experiment. 5 Prompt-Relevance for Holistic Scoring We now evaluate the effectiveness of a supervised essay scoring system that incorporates tf-idf similarity features and the PRF approach for the task of predicting an overall essay quality score. 5.1 Data and Experimental Setup For this experiment, we used a dataset consisting of 2,316 essays written for the IELTS (International English Language Testing System) English examination from 2005 to 2010 (Nicholls, 2003). The examination is designed to measure a broad proficiency continuum ranging from an intermediate to a proficient level of English (A2 to C2 in the CEFR levels). The essays are associated with 22 prompts that are similar in style (i.e. essay style) to those in the ICLE dataset. Candidates are assigned an overall score on a scale from 1 to 9. Prompt relevance is an aspect that is present in the marking criteria, and it is identified as a determinant of the overall score. We therefore hypothesise that adding prompt-relevance measures to the feature set of a prompt-independent essay scoring system (i.e. that is designed to assess linguistic competence only) would better reflect the evaluation performed by examiners and improve system performance. The baseline system is a linear preference ranking model (Yannakoudakis et al., 2011; Yannakoudakis and Briscoe, 2012) and is trained to predict an overall essay score based on the following set of features: - word unigrams, bigrams, and trigrams - POS (part-of-speech) counts - grammatical relations - essay length (# of unique words) - counts of cohesive devices - max-word length and min-sentence length - number of errors based on a presence/absence trigram language model We divided the dataset into 5-folds in two separate ways. First, we created prompt-dependent folds, where essays associated with all 22 prompts appear in both the training and test data in the appropriate proportions. This scenario allows the system to learn from essays that were written in response to the prompt. Second, we created prompt-independent folds, where all essays associated with a specific prompt appear in only one fold. This second dataset is a more realistic real-world scenario (see Section 2) whereby the system learns on one set of prompts (possibly from previous years) and aims to predict the score for essays associated with different prompts. For both of these supervised experiments, we measured system performance using Spearman’s and Pearson’s correlation between the output of the system and the gold essay scores (human judgements). In order to examine the effect of prompt relevance on these datasets, we added to our baseline system two sets of features. The first set of features labelled PR includes the cosine similarity between the essay and the prompt $ \cos(p, s) $, the fraction of essays words that appear in the prompt $ \cov(p, s) $, and the fraction of prompt words that appear in the essay $ \cov(s, p) $. The second set of features labelled semPR is the same as the first set except that the prompt is expanded using the PRF method from earlier. 5.2 Results for Overall Scoring The results of the experiment are outlined in Table 4. Firstly, we observe that the effectiveness of the baseline system is higher on the prompt-dependent folds ($ \rho = 0.661 $) than on the prompt-independent folds ($ \rho = 0.637 $). This confirms expectations as the prompt-dependent folds allow the baseline model to learn useful features from essays written specifically for those prompts. When adding the exact matching prompt-relevance features – referred to as PR in Table 4 – we observe an increase in performance on the prompt-independent folds. When we add the semantic prompt-relevance models – referred to as semPR in Table 4 – we again observe a modest increase in \| System \| Spearman-$\rho$ \| Pearson-$r$ \| \|------------\|-----------------\|-------------\| \| Baseline \| 0.661 \| 0.686 \| \| + PR \| 0.659 \| 0.685 \| \| + semPR \| 0.662 \| 0.691 \| Table 4: Performance of systems using 5-fold cross-validation on prompt-dependent folds (top) and prompt-independent folds (bottom) when adding unsupervised prompt-relevance (PR) features and semantic prompt-relevance features (semPR) on a set of 2316 essays. † means statistically significant compared to the baseline using Steiger’s test (1980). performance on the prompt-independent folds. We can see that both Spearman and Pearson correlations approach the performance of the baseline system on the prompt-dependent folds. On the other hand, there is little or no increase in performance when adding the PR and semPR features on the prompt-dependent folds. One suspected reason for this is that it is likely that the lexical features in the prompt-dependent folds are performing prompt-relevance modelling (by learning appropriate weights for lexical features in essays written for that prompt). Overall, this is an interesting result as it shows that the features developed in this paper are useful and contribute to the holistic score in real-world examinations. 6 Discussion Firstly, the results from Section 4 are not directly comparable with previous research using the ICLE dataset, as that work (Persing and Ng, 2014) reported metrics averaged over all essays where each prompt was not isolated individually. Ignoring prompt effects may lead to favouring systems that perform well only on a few prompts, and that are not robust across the types of prompt that may be used operationally. Table 5 shows the results of the approaches outlined in this paper against those from the original research using the ICLE dataset that used supervised models. Importantly, we achieve these correlations without any training data. \| System \| Baseline* \| tf-idf \| PRF \| Persing* \| \|------------\|-----------\|--------\|-----\|----------\| \| Pearson’s-$r$ \| 0.233 \| 0.261 \| 0.277 \| 0.360 \| Table 5: Pearson correlation of systems over all 830 essays. * means from original paper. Interestingly, we have shown that the PRF prompt expansion is effective and is easily analysable. In an operational setting, prompt expansion is likely to be a highly important feature. Observing non-prompt words that are related to the prompt, in a learner text is likely to be indicative of a learner who has a good understanding of the vocabulary of the topic. The expansion step issues the entire prompt to a Wikipedia index to gather candidate expansion terms. While this has been shown to be a useful approach on average, there may be cases when aspects of the prompt are not adequately reflected by the candidate expansion terms. In such cases it may be better to partition the prompt into useful phrases that can be expanded in isolation, or to manually rephrase the prompt before expanding it with related terms. 6.1 Conclusion and Future Work We have shown that using an unsupervised pseudo-relevance language modelling approach to measuring relevance in learner texts is beneficial as it correlates with human annotators. The expansion terms in isolation have been shown to be useful and we argue that they are an important feature for overcoming vocabulary mismatch in learner text. The estimation of an L2 learner’s language model from lexemes produced by the learner is an intuitive and theoretically-motivated way to assess many lexical aspects of writing. However, compositionally-motivated language modelling approaches exist (Mitchell and Lapata, 2009), and it would be interesting to investigate these across different areas in assessment. The approaches developed herein may also be useful for providing feedback and/or suggestions to learners during the process of writing. Future work will look at supplying feedback in pedagogically sound ways. Acknowledgements We would like to thank Cambridge English Language Assessment for supporting this research, and the anonymous reviewers for their useful feedback. References Ted Briscoe, John Carroll, and Rebecca Watson. 2006. The second release of the rasp system. In Proceedings of the COLING/ACL on Interactive presentation sessions, pages 77–80. ACL. Ted Briscoe, Ben Medlock, and Øistein Andersen. 2010. Automated assessment of esol free text examinations. University of Cambridge Computer Laboratory Technical Reports. UCAM-CL-TR-790. Y. Y. Chen, C. L. Liu, T. H. Chang, and C. H. Lee. 2010. An Unsupervised Automated Essay Scoring System. IEEE Intelligent Systems, 25(5):61–67. Ronan Cummins, Jiaul H. Paik, and Yuanhua Lv. 2015. A Pólya urn document language model for improved information retrieval. ACM Transactions of Information Systems, 33(4):21. Semire Dikli. 2006. An overview of automated scoring of essays. Journal of Technology, Learning, and Assessment, 5(1). Evgeniy Gabrilovich and Shaul Markovitch. 2007. Computing semantic relatedness using wikipedia-based explicit semantic analysis. In Proceedings of the 20th International Joint Conference on Artificial Intelligence. IJCAI’07, pages 1606–1611, San Francisco, CA, USA. Morgan Kaufmann Publishers Inc. Sylviane Granger, Estelle Dagneaux, Fanny Meunier, Magali Paquot, et al. 2009. The international corpus of learner english. version 2. handbook and cd-rom. Louvain-la-Neuve:Presses universitaires de Louvain. Derrick Higgins and Jill Burstein. 2007. Sentence similarity measures for essay coherence. In Proceedings of the 7th International Workshop on Computational Semantics, pages 1–12. Derrick Higgins, Jill Burstein, Daniel Marcu, and Claudia Gentile. 2004. Evaluating multiple aspects of coherence in student essays. In Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics, pages 185–192. ACL. D. Higgins, J. Burstein, and Y. Attali. 2006. Identifying off-topic student essays without topic-specific training data. Natural Language Engineering, 12(2):145–159. Tuomo Kakkonen and Erkki Sutinen. 2008. Evaluation criteria for automatic essay assessment systems—there is much more to it than just the correlation. In International Conference on Computers in Education (ICCE), pages 111–116. Pentti Kanerva, Jan Kristoferson, and Anders Holst. 2000. Random indexing of text samples for latent semantic analysis. In In Proceedings of the 22nd Annual Conference of the Cognitive Science Society, pages 103–6. Erlbaum. Thomas K Landauer, Darrell Laham, and Peter W Foltz. 2003. Automated scoring and annotation of essays with the intelligent essay assessor. Automated essay scoring: A cross-disciplinary perspective, pages 87–112. Thomas K Landauer. 2003. Automatic essay assessment. Assessment in education: Principles, policy & practice, 10(3):295–308. Omer Levy, Yoav Goldberg, and Ido Dagan. 2015. Improving distributional similarity with lessons learned from word embeddings. Transactions of the Association for Computational Linguistics, 3:211–225. Annie Louis and Derrick Higgins. 2010. Off-topic essay detection using short prompt texts. In Proceedings of the NAACL HLT 2010: Fifth Workshop on Innovative Use of NLP for Building Educational Applications, pages 92–95. ACL. Yuanhua Lv and ChengXiang Zhai. 2009. A comparative study of methods for estimating query language models with pseudo feedback. In Proceedings of the 18th ACM Conference on Information and Knowledge Management, CIKM ’09, pages 1895–1898, New York, NY, USA. ACM. Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean. 2013. Distributed representations of words and phrases and their compositionality. In Advances in neural information processing systems, pages 3111–3119. Tristan Miller. 2003. Essay assessment with latent semantic analysis. Journal of Educational Computing Research, 29(4):495 – 512. Jeff Mitchell and Mirella Lapata. 2009. Language models based on semantic composition. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 430–439. ACL. Diane Nicholls. 2003. The cambridge learner corpus: Error coding and analysis for lexicography and elt. In In Proceedings of the Corpus Linguistics 2003 conference, pages 572–581. Ellis B Page. 1966. The imminence of grading essays by computer. Phi Delta Kappan, 47:238–243. Ellis Batten Page. 1994. Computer grading of student prose, using modern concepts and software. The Journal of experimental education, 62(2):127–142. Isaac Persing and Vincent Ng. 2014. Modeling prompt adherence in student essays. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, pages 1534–1543, Baltimore, Maryland, June. ACL. Susan Phillips. 2007. Automated essay scoring: A literature review, volume 30. Society for the Advancement of Excellence in Education (SAEE). Jay M. Ponte and W. Bruce Croft. 1998. A language modeling approach to information retrieval. In Proceedings of the 21st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’98, pages 275–281, New York, NY, USA. ACM. Mihai Rotaru and Diane J Litman. 2009. Discourse structure and performance analysis: Beyond the correlation. In Proceedings of the SIGDIAL 2009 Conference: The 10th annual meeting of the special interest group on discourse and dialogue, pages 178–187. Association for Computational Linguistics. Mark D Shermis and Jill C Burstein. 2003. Automated essay scoring: A cross-disciplinary perspective. Routledge. Mark D Shermis and Jill Burstein. 2013. Handbook of automated essay evaluation: Current applications and new directions. Routledge. M Shermis and B Hammer. 2012. Contrasting state-of-the-art automated scoring of essays: analysis. Technical report, The University of Akron and Kaggle. James H Steiger. 1980. Tests for comparing elements of a correlation matrix. Psychological bulletin, 87(2):245. Peter D Turney, Patrick Pantel, et al. 2010. From frequency to meaning: Vector space models of semantics. Journal of artificial intelligence research, 37(1):141–188. Salvatore Valenti, Francesca Neri, and Alessandro Cucchiarelli. 2003. An overview of current research on automated essay grading. Journal of Information Technology Education, 2:3–118. Helen Yannakoudakis and Ted Briscoe. 2012. Modeling coherence in esol learner texts. In Proceedings of the Seventh Workshop on Building Educational Applications Using NLP, pages 33–43. Association for Computational Linguistics. Helen Yannakoudakis and Ronan Cummins. 2015. Evaluating the performance of automated text scoring systems. In Proceedings of the 10th Workshop on Innovative Use of NLP for Building Educational Applications. Helen Yannakoudakis, Ted Briscoe, and Ben Medlock. 2011. A New Dataset and Method for Automatically Grading ESOL Texts. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies.
TOWARDS INTRODUCING LONG-TERM STATISTICS IN MUSE FOR ROBUST SPEECH RECOGNITION Christopher Kermorvant and Chaïc Mokbel IDIAP P.O.Box 592, Martigny, Switzerland. kermorva,email@example.com ABSTRACT In this paper, we propose new developments of the Multi-path Stochastic Equalization techniques (MUSE). The MUSE technique is based on an enriched model of speech, composed of both a classical model of clean speech with HMM and equalization functions. This technique is able to reduce the recognition error rate due to a mismatch between the training and testing conditions. In order to track long-term variation of this mismatch, the introduction of a priori statistics on the equalization function is studied. In the case of Bias Removal, this approach has been implemented in HTK and tested on the Numbers95 database. Experiments show that the convergence of the bias computation is fast enough and limits the effect of the a priori values. However, both the fast convergence property and the proposed framework open research directions towards more complex equalization functions. 1. INTRODUCTION Recent advances made it possible to integrate speech recognition techniques in commercial products and applications. However, the performance of the recognisers still highly rely on the conditions in which they are used. If the recognition takes place in an environment which is close to the environment the recogniser has been designed for, high recognition scores can be achieved. But as soon as there is a mismatch between the training and testing environments, performance drops rapidly. Many equalization scheme have been developed to reduce this mismatch both in the spectral [2][6] or in the cepstral [1][7] domain. However, the equalization process has always been separated from the recognition process. The first attempt to combine an equalization scheme with HMM modeling during recognition was proposed by Ephraim [4]. Then, the Stochastic Matching technique [9] has been proposed: this technique uses a Maximum Likelihood approach to compute the parameters of a mapping function in order to reduce the mismatch between the observed utterance and the speech models during recognition. In this case, both the mapping function and the state sequence are optimized using the EM (Expectation-Maximisation) algorithm. Recently, a Multi-path Stochastic Equalization (MUSE) technique has been developed [8]. MUSE provides an enriched model of speech signals. By combining usual HMM models and equalization functions, MUSE can model both the variations of the speech signals and the variations of the environment. In the case of bias removal, MUSE has already shown its ability to track local variation of the bias. In this paper, we introduce a method to learn and integrate long-term characteristics of the bias. The implementation of MUSE and of the proposed extension into a classical decoder, namely HTK[10], is also presented. This article is organized as follow. The theoretical framework behind MUSE is recalled in Section 2. A particular application of MUSE to bias removal is developed in Section 3. In this section, the introduction of a priori statistics on the bias is also presented. Recognition experiments designed to assess this approach are presented and analyzed in Section 4. Finally, conclusions is given in Section 5. 2. THEORETICAL FRAMEWORK The basic idea behind MUSE is to associate an equalization function to every possible state sequence hypothesized during the decoding. The parameters of the equalization function are computed using either a Maximum Likelihood or a Maximum A Posteriori criterion, as developed in [8]. In this section, we recall the theoretical development of the technique using a Maximum A Posteriori (MAP) criterion. We denote by $Y = Y_1, Y_2, \ldots, Y_T$ a sequence of T speech frames observed at the output of a transmission channel and by $\lambda$ the parameter set of the HMM modeling clean speech. We suppose that we can obtain an estimate of the clean speech $\hat{X}$ from the observed sequence $Y$ using an equalizing function $T_\theta(.)$: $$\hat{X}_t = T_\theta(Y_t)$$ \hspace{1cm} (1) For a given path $s_t$ in the HMM model, we can derive the optimal values for the equalizing function parameter with a Maximum A Posteriori criterion: \[ \hat{\theta}(s_t) = \arg \max_{\theta} [p_Y(Y_1, \ldots, Y_t \| \theta(s_t), s_t, \lambda) \\ \cdot p(\theta(s_t) \| s_t, \lambda)] \] (2) \[ \hat{\theta}(s_t) = \arg \max_{\theta} \left[ \frac{p_X(T_{\theta(s_t)}(Y_1), \ldots, T_{\theta(s_t)}(Y_t))\|s_t, \lambda}{\prod_{\tau=t_0}^{t} \\|J(Y_\tau)\\|} \right. \\ \cdot p(\theta(s_t) \| s_t, \lambda) \] (3) where $\\|J(Y_\tau)\\|$ is the absolute value of the Jacobian associated to the change of variable. With certain types of function $T_\theta(.)$, Equation 3 can be solved analytically to find the values of the equalization parameters $\theta$. If we assume, as it is generally the case in HMM, that the distribution of the state $i$ is Gaussian with a mean vector $\vec{\mu}_i$ and a covariance matrix $\Sigma_i$, then $\theta(s_t)$ is the solution of the following equation: \[ \sum_{\tau=t_0}^{t} \left[ T_{\theta_t}(Y_\tau) - \vec{\mu}_{s_\tau} \right]^# \cdot \Sigma_{s_\tau}^{-1} \cdot \left[ \frac{\partial}{\partial \theta_t} T_{\theta_t}(Y_\tau) \right] \] \[+ \left[ \frac{\partial}{\partial \theta_t} \log \\|J(Y_\tau)\\| \right] \right\} - \frac{\partial}{\partial \theta_t} \log p_\theta(\theta_t \| s_t, \lambda) = 0_{1,q} \] (4) Once $\theta$ is computed for each path, the most likely state sequence can be found by: \[ \hat{s}_t = \arg \max_{s_t} [p(s_t \| Y_1, \ldots, Y_t, \theta(s_t), \lambda)] \] \[= \arg \max_{s_t} \left[ \frac{p_X(T_{\theta(s_t)}(Y_1), \ldots, T_{\theta(s_t)}(Y_t))\|s_t, \theta(s_t), \lambda}{\prod_{\tau=t_0}^{t} \\|J(Y_\tau)\\|} \right. \\ \cdot p(s_t\|\lambda) \] (5) ### 3. INTRODUCTION OF LONG-TERM STATISTICS FOR BIAS REMOVAL The approach presented in the previous section can be applied to a simple equalization function like a bias removal function in the form $T_\theta(Y_t) = Y_t - \vec{b}$. In this case, in Equation 4, we have: \[ \frac{\partial}{\partial \theta_t} T_{\theta_t}(Y_\tau) = -I_p \] \[ \log \\|J(Y_\tau)\\| = 0 \] where $I_p$ is the p-dimensional identity matrix. In order to introduce a priori statistics on bias, we suppose that having a bias $\vec{b}_t$ at time $t$ is equivalent to having $\vec{b}_1 = \ldots = \vec{b}_{t-1} = \vec{b}_t$. We also suppose that the a priori distribution of the bias is Gaussian with mean $\vec{\mu}_{apriori}$ and variance $\frac{\Sigma_{apriori}}{\alpha^t}$. The form of this distribution is based on the fact that the importance of the a priori distribution of the bias is high at the beginning of the decoding, and decreases with the time. The probability of a bias $\vec{b}_t$ at time $t$ given its a priori distribution is expressed as: \[ p(\vec{b}_t\|s_t, \lambda) = \prod_{\tau=t_0}^{t} N(\vec{b}_t, \vec{\mu}_{apriori}, \frac{\Sigma_{apriori}}{\alpha^t}) \] (6) Then, In Equation 4, we have \[ \frac{\partial}{\partial \theta_t} \log p_\theta(\theta_t \| s_t, \lambda) = \sum_{\tau=t_0}^{t} \alpha^\tau (\vec{b}_t - \vec{\mu}_{apriori})^# \Sigma_{apriori}^{-1} \] (7) By solving Equation 4, we can derive an analytical formula to compute the bias: \[ \vec{b} = \frac{\left[ \sum_{\tau=t_0}^{t} [\vec{Y}_\tau - \vec{\mu}_{s_\tau}]^# \cdot \Sigma_{s_\tau}^{-1} + \alpha^\tau \vec{\mu}_{apriori}^# \Sigma_{apriori}^{-1} \right]}{\left[ \sum_{\tau=t_0}^{t} (\Sigma_{s_\tau}^{-1} + \alpha^\tau \Sigma_{apriori}^{-1}) \right]} \] (8) With this bias value, the log likelihood of a sequence $Y$ of observation vectors $Y_\tau$ (of dimension $n$) given the states $S_t$ is now given by: \[ \log(p_X(T_{\theta_t}(Y)\|S_t, \vec{b}, \lambda)) = -\frac{nt \log(2\pi)}{2} \] \[-\frac{1}{2} \sum_{\tau=t_0}^{t} \left\{ \log(\\|\Sigma_{s_\tau}\\|) + (Y_\tau - \vec{\mu}_{s_\tau})^# \Sigma_{s_\tau}^{-1} (Y_\tau - \vec{\mu}_{s_\tau}) \right\} \] \[+ \left[ \sum_{\tau=t_0}^{t} [\vec{Y}_\tau - \vec{\mu}_{s_\tau}]^# \cdot \Sigma_{s_\tau}^{-1} \right] \vec{b} - \frac{1}{2} \vec{b}^# \left[ \sum_{\tau=t_0}^{t} \Sigma_{s_\tau}^{-1} \right]^{-1} \vec{b} \] (9) In this equation, one can recognize the classical formula for the likelihood computation (first and second line) plus two terms associated with the bias (third line). ### 4. EXPERIMENTS #### 4.1. Experimental Setups The speech database chosen for the experiments is the Numbers’95 database [3] from the Center for Spoken Language Understanding (CSLU). This database contains digits sequences continuously spoken over the telephone. We used the 3590 sentences of the training set for the training of our models and we tested them on the 1206 sentences of the development-test set. We used the front-end developed at IDIAP to extract the feature vectors from the speech files. We computed 26 mel-scaled filter bank coefficients, over a 32 ms hamming window, with a 10 ms shift. Then 13 mel-cepstral coefficients were derived together with their first and second order derivatives (for a total of 39 coefficients). The recognition system was based on Gaussian mixture HMMs. It was trained with HTK [10]. The system was composed of 81 triphones modeled by 3 states HMMs; each state had a 10 Gaussian mixture pdf and a diagonal covariance matrix. No language model was used. The recognition has been done with a modified version of HTK in which has been implemented the MUSE technique. 4.2. Implementation issues 4.2.1. Bias computation with a priori statistics In order to compute the bias, we used two accumulators $\tilde{A}_t$ and $\tilde{B}_t$ defined as: $$\tilde{A}_t = \tilde{A}_{t-1} + (\vec{Y}_t - \vec{\mu}_{s_t})^T \Sigma_{s_t}^{-1} + \alpha^t \tilde{A}_{apriori} \cdot \tilde{B}_{a_priori}$$ $$\tilde{B}_t = \tilde{B}_{t-1} + \Sigma_{s_t}^{-1} + \alpha^t \tilde{B}_{upriori}$$ These two accumulators are updated with each new frame and with a priori values of the bias. We have computed $\tilde{A}_{apriori}$ and $\tilde{B}_{apriori}$, using the first and second order statistics of the bias, on the training database. The multiplicative factor $\alpha^t$ exponentially decreases with time ($0 < \alpha < 1$), and allows to give more importance to the a priori estimation of the bias at the beginning of the utterance. Several values have been tested for the parameter $\alpha$. Results are shown in the Section 4.3. For each frame, once the two accumulators are updated, the bias is then computed as: $b = \tilde{A}_t \cdot \tilde{B}_t^{-1}$. 4.2.2. On-line estimation of a priori statistics We have also introduced an on-line process to estimate the a priori estimation of the bias. At the end of each utterance, the a priori means and variances of the bias are updated recursively with the current value of the bias with parameter $\beta$ and $1 - \beta$. This allows to track the long-term variation of the bias, and therefore to adapt on-line the a priori statistics of the bias. 4.3. Results We first present the results of an adaptation to a mismatch between training and testing conditions using MUSE. The best recognition results we obtained so far on the Numbers95 database was obtained by pre-processing the data with a Cepstral Mean Subtraction (CMS) scheme [5]. However, this preprocessing is not frame synchronous, and thus can not be used in real systems. Using classical MFCC feature vectors with the models trained on data pre-processed with CMS yields a decrease in recognition results. As shown on table 1, the recognition word error rate increase from 5.33% to 11.16% when using MFCC feature vectors with models trained with CMS. This bias can be significantly reduce by using MUSE. Indeed, using MUSE reduces the WER from 11.16% to 6.10% and almost recovers the WER obtained in matching conditions. It is also important to note that using MUSE in mismatch conditions (MFCC features with CMS based models) outperforms the baseline system, based on both MFCC features and models. \| WER \| baseline \| CMS/CMS \| MFCC/CMS \| MUSE \| \|-----\|----------\|---------\|----------\|------\| \| \| 7.07 \| 5.33 \| 11.16 \| 6.10 \| Table 1: Recognition error rate on the Numbers95 database in matching conditions (CMS/CMS), mismatch conditions (MFCC/CMS) and mismatch condition with MUSE (MUSE). The second set of results concerns the introduction of a priori statistics on the equalization function parameters in the MUSE technique. These a priori statistics have been computed on the train set and consist in the mean bias and its variance for each cepstral coefficient. The data are then used in the MUSE decoding process as described in the previous section. Table 2 presents the variation of the recognition Word Error Rate according to the value of the factor $\alpha$. First, when $\alpha$ increases, the recognition score decreases. This is due to the fact that the a priori statistics should be used only at the beginning of the recognition. Therefore, $\alpha$ should be small enough not to hinder the convergence of the estimated bias. Second, even with a small value for $\alpha$, we get no improvement of the recognition results. Two reason might explain why the introduction of a priori statistics on the bias does not improve the recognition scores. The first one is illustrated in Figure 1. This figure shows the convergence of the estimated bias compared to the real bias computed by a forced Viterbi alignment. We can see that the convergence of the estimated bias is quite fast. This means that, in the case of bias removal, the estimation of the parameters of the equalization function is simple enough not to need many frames to converge. The introduction of the a priori values for the bias is therefore of no help for the convergence. Secondly, when looking at the a priori values of the bias computed on the training set, we see that these values present a very large variance. The effect of these values in the computation of the bias is thus very limited. \| WER \| 0.5 \| 0.7 \| 0.8 \| 0.9 \| \|-----\|-----\|-----\|-----\|-----\| \| \| 6.17\| 6.17\| 6.21\| 6.47\| Table 2: Recognition error rate on the Numbers95 database in mismatch condition with MUSE for different values of the factor $\alpha$. However, these results give us some perspective for the future work. Firstly, the fast converge of the MUSE algorithm allows to consider more complex equalization functions. Secondly, the MAP approach to MUSE allows to introduce equalization functions which depend on phonemes. In these two cases, the introduction of a priori statistics on the parameter of the equalization functions, as proposed in this paper, will be necessary. 5. CONCLUSION In this paper, we have proposed a development of the MUSE technique towards the introduction of a priori statistics on equalization functions, in the case of Bias Removal. The experiments on the Numbers95 database have shown that in the case of Bias Removal, the convergence of the MUSE algorithm is quite fast. Therefore, the introduction of the a priori statistics does not improve the recognition score. However, the proposed framework for the introduction of adaptative a priori statistics and the good convergence properties of MUSE open research directions towards more complex equalization functions. Acknowledgments: This work is supported by the Swiss Federal Office for Education and Science (FOES) through the European TMR SP-HEAR and RESPITE projects. 6. REFERENCES [1] B. S. Atal. Effectiveness of linear prediction characteristics of the speech wave for automatic speaker identification and verification. JASA, 55(6):1304–1312, June 1974. [2] Steven F. Boll. Suppression of acoustic noise in speech using spectral subtraction. IEEE Trans. on Acoustics, Speech and Signal Processing, 27(2), April 1979. [3] R.A. Cole, M. Noel, T. Lander, and T. Durham. New telephone speech corpora at cslu. Proc. European Conf. on Speech Communication and Technology, 1:821–824, 1995. [4] Yariv Ephraim. Statistical-model-based speech enhancement systems. Proceedings of the IEEE, 80(10):1555, October 1992. [5] C. Kermorvant. A comparison of noise reduction techniques for robust speech recognition. IDIAP-RR 10, 1999. [6] L. Mauuary. Blind equalization for robust telephone based speech recognition. In Proc. European Signal Processing Conference, 1996. [7] L. Mauuary. Blind equalization in the cepstral domain for robust telephone based speech recognition. In Proc. European Signal Processing Conference, 1998. [8] C. Mokbel. Muse : Multipath stochastic equalization, a theoretical framework to combine equalization and stochastic modeling. In ESCA-NATO Tutorial and Research Workshop on Robust Speech Recognition for Unknown Communication Channel, Pont-à-Mousson, April 1997. [9] A. Sankar and C.H. Lee. A maximum-likelihood approach to stochastic matching for robust speech recognition. IEEE Trans. on Acoustics, Speech and Signal Processing, 4(3):190–202, May 1996. [10] Steve Young. The HTK Book. Cambridge University, March 1997.
50 Methods and Principles in Medicinal Chemistry Edited by Dennis A. Smith, Charlotte Allerton, Amit S. Kalgutkar, Han van de Waterbeemd, and Don K. Walker Pharmacokinetics and Metabolism in Drug Design Methods and Principles in Medicinal Chemistry Edited by R. Mannhold, H. Kubinyi, G. Folkers Editorial Board H. Buschmann, H. Timmerman, H. van de Waterbeemd, T. Wieland De Clercq, Erik (Ed.) Antiviral Drug Strategies 2011 ISBN: 978-3-527-32696-9 Vol. 50 Klebl, Bert / Müller, Gerhard / Hamacher, Michael (Eds.) Protein Kinases as Drug Targets 2011 ISBN: 978-3-527-31790-5 Vol. 49 Sotriffer, Christoph (Ed.) Virtual Screening Principles, Challenges, and Practical Guidelines 2011 ISBN: 978-3-527-32636-5 Vol. 48 Rautio, Jarkko (Ed.) Prodrugs and Targeted Delivery Towards Better ADME Properties 2011 ISBN: 978-3-527-32603-7 Vol. 47 Smit, Martine J. / Lira, Sergio A. / Leurs, Rob (Eds.) Chemokine Receptors as Drug Targets 2011 ISBN: 978-3-527-32118-6 Vol. 46 Ghosh, Arun K. (Ed.) Aspartic Acid Proteases as Therapeutic Targets 2010 ISBN: 978-3-527-31811-7 Vol. 45 Ecker, Gerhard F. / Chiba, Peter (Eds.) Transporters as Drug Carriers Structure, Function, Substrates 2009 ISBN: 978-3-527-31661-8 Vol. 44 Faller, Bernhard / Urban, Laszlo (Eds.) Hit and Lead Profiling Identification and Optimization of Drug-like Molecules 2009 ISBN: 978-3-527-32331-9 Vol. 43 Sippl, Wolfgang / Jung, Manfred (Eds.) Epigenetic Targets in Drug Discovery 2009 ISBN: 978-3-527-32355-5 Vol. 42 Todeschini, Roberto / Consonni, Viviana Molecular Descriptors for Chemoinformatics Volume I: Alphabetical Listing / Volume II: Appendices, References 2009 ISBN: 978-3-527-31852-0 Vol. 41 Edited by Dennis A. Smith, Charlotte Allerton, Amit S. Kalgutkar, Han van de Waterbeemd, and Don K. Walker Pharmacokinetics and Metabolism in Drug Design Third, Revised and Updated Edition WILEY-VCH Verlag GmbH & Co. KGaA Series Editors Prof. Dr. Raimund Mannhold Molecular Drug Research Group Heinrich-Heine-Universität Universitätsstrasse 1 40225 Düsseldorf Germany firstname.lastname@example.org Prof. Dr. Hugo Kubinyi Donnersbergstrasse 9 67256 Weisenheim am Sand Germany email@example.com Prof. Dr. Gerd Folkers Collegium Helveticum STW/ETH Zurich 8092 Zurich Switzerland firstname.lastname@example.org Volume Authors Dr. Dennis A. Smith 4, The Maltings Walmer, Kent CT14 7AR United Kingdom email@example.com Dr. Charlotte Allerton School House, School Lane Ickham, Canterbury CT3 1QU United Kingdom Dr. Amit S. Kalgutkar Pfizer Global R&D Pharmacokinetics, Metabolism MS 8220-3529 Groton CT 06340 USA Dr. Han van de Waterbeemd 14 rue de la Rasclose 66690 Saint Andre France Dr. Don K. Walker 8 Whitecross Green Broadstairs, CT10 3JS United Kingdom firstname.lastname@example.org All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de. © 2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photostripping, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Composition Thomson Digital, Noida, India Printing and Binding Markomo Print Media Pte Ltd, Singapore Cover Design Schulz Grafik-Design, Fußgönheim Printed in Singapore Printed on acid-free paper Print ISBN: 978-3-527-12954-0 ePDF ISBN: 978-3-527-64530-6 eBook ISBN: 978-3-527-64576-4 Epub ISBN: 978-3-527-64529-2 Mobi ISBN: 978-3-527-64531-4 ## Contents A Personal Foreword IX 1 Physicochemistry 1 1.1 Physicochemistry and Pharmacokinetics 2 1.2 Partition and Distribution Coefficients as Measures of Lipophilicity 2 1.3 Limitations on the Use of 1-Octanol 5 1.4 Further Understanding of log $P$ 6 1.4.1 Unraveling the Principal Contributions to log $P$ 6 1.4.2 Hydrogen Bonding 7 1.4.3 Polar Surface Area 8 1.4.4 Molecular Size and Shape 9 1.5 Alternative Lipophilicity Scales 10 1.6 Computational Systems to Determine Lipophilicity 10 1.7 Membrane Systems to Study Drug Behavior 10 1.8 Dissolution and Solubility 12 1.9 The BCS Classification and Central Role of Permeability 13 References 15 2 Pharmacokinetics 19 2.1 Setting the Scene 20 2.2 Intravenous Administration: Volume of Distribution 21 2.3 Intravenous Administration: Clearance 22 2.4 Intravenous Administration: Clearance and Half-life 23 2.5 Intravenous Administration: Infusion 24 2.6 Oral Administration 26 2.7 Repeated Doses 27 2.8 Development of the Unbound (Free) Drug Model 29 2.9 Unbound Drug and Drug Action 29 2.10 Unbound Drug Model and Barriers to Equilibrium 32 2.11 Pharmacodynamic Models 34 2.12 Slow Offset Compounds 35 2.13 Factors Governing Unbound Drug Concentration 38 References 40 3 Absorption 41 3.1 The Absorption Process 42 3.2 Dissolution 42 3.3 Membrane Transfer 44 3.4 Barriers to Membrane Transfer 49 3.5 Prodrugs to Increase Oral Absorption 51 3.6 Active Transport 55 3.7 Models for Absorption Estimation 56 3.8 Estimation of Absorption Potential and other Computational Approaches 56 References 57 4 Distribution 61 4.1 Membrane Transfer Access to the Target 62 4.2 Brain Penetration 63 4.2.1 Accumulation of Lower Permeability Compounds into the Brain 67 4.2.2 Distribution into Tumors 68 4.2.3 Volume of Distribution and Duration 70 4.2.4 Distribution and $T_{\text{max}}$ 77 References 78 5 Clearance 81 5.1 The Clearance Processes 82 5.2 Role of Transport Proteins in Drug Clearance 83 5.3 Interplay Between Metabolic and Renal Clearance 87 5.4 Role of Lipophilicity in Drug Clearance 87 5.5 Active Metabolites 88 5.6 Balancing the Rate of Renal and Metabolic clearance and Potency 91 References 101 6 Renal Clearance 103 6.1 Kidney Anatomy and Function 103 6.2 Lipophilicity and Reabsorption by the Kidney 105 6.3 Effect of Charge on Renal Clearance 106 6.4 Plasma Protein Binding and Renal Clearance 106 6.5 Balancing Renal Clearance and Absorption 108 6.6 Renal Clearance and Drug Design 109 References 110 7 Metabolic (Hepatic) Clearance 111 7.1 Symbols 111 7.2 Function of Metabolism (Biotransformation) 112 7.3 Cytochrome P450 112 7.3.1 Catalytic Selectivity of CYP2D6 115 7.3.2 Catalytic Selectivity of CYP2C9 117 7.3.3 Catalytic Selectivity of CYP3A4 119 7.4 Other Oxidative Metabolism Processes 126 7.4.1 Aldehyde Oxidase 126 7.4.2 Flavin-Containing Monoxygenases 130 7.4.3 Monoamine Oxidases 133 7.5 Oxidative Metabolism and Drug Design 138 7.6 Nonspecific Esterases 138 7.6.1 Function of Esterases 138 7.6.2 Ester Drugs as Intravenous and Topical Agents 140 7.7 Prodrugs to Aid Membrane Transfer 142 7.8 Enzymes Catalyzing Drug Conjugation 144 7.8.1 Glucuronosyl- and Sulfotransferases 144 7.8.2 Methyl Transferases 147 7.8.3 Glutathione-S-Transferases 148 7.9 Stability to Conjugation Processes 149 7.10 Pharmacodynamics and Conjugation 152 References 153 8 Toxicity 159 8.1 Toxicity Findings 160 8.1.1 Pharmacologic Mechanism-Based Toxicity 160 8.1.2 Chemotype-Dependent Toxicity 161 8.1.3 Metabolism-Induced Toxicity 164 8.2 Structure–Toxicity Analyses 167 8.3 Reactive Metabolite Screening in Drug Discovery 171 8.4 Structural Alerts/Toxicophores in Drug Design 173 8.5 Dealing with Reactive Metabolite Positives in Drug Discovery: Risk Assessment Strategies – Effect of Daily Dose 173 8.6 Dealing with Reactive Metabolite Positives in Drug Discovery: Risk Assessment Strategies – Competing Detoxication Pathways 182 8.7 Stratification of Toxicity 183 8.8 Toxicity Prediction: Computational Toxicology 183 8.9 Toxicogenomics 184 8.10 Pharmacogenomics 185 8.11 Enzyme Induction and Drug Design 186 8.12 Enzyme Inhibition and Drug Design 191 8.12.1 Quasi-Irreversible Inhibition 191 8.12.2 Irreversible CYP Inactivation via Apoprotein and/or Heme Covalent Modification 193 8.12.3 CYP Inhibition by Nitrogen-Containing Heterocycles 195 References 202 9 Predicting Human Pharmacokinetics 209 9.1 Objectives of Predicting Human Pharmacokinetics 210 9.2 Allometric Scaling of Preclinical In Vivo PK Parameters 211 9.2.1 Volume of Distribution 211 9.2.2 Clearance 214 9.3 Prediction of Human PK Parameters Using In Vitro Data 220 9.3.1 Predicting Human Volume of Distribution from In vitro Data 220 9.3.2 Predicting Human Clearance from Human In Vitro Data 222 9.3.3 Species Scaling: Incorporating Differences in Metabolic Clearance 223 9.4 Elimination Half-Life 224 9.5 Moving Forward 224 References 225 10 ADME Screening 229 10.1 The High-Throughput Synthesis and Screening Trend 230 10.2 The Concept of ADME Space 231 10.3 Drug Metabolism and Discovery Screening Sequences 233 10.4 Physicochemistry 234 10.4.1 Solubility 235 10.4.2 Ionization 236 10.4.3 Lipophilicity 236 10.4.4 Polar Surface Area 237 10.5 Absorption/Permeability 238 10.6 Metabolism, Induction, and Inhibition 239 10.7 Transporters 240 10.8 Protein Binding 242 10.9 Pharmacokinetics 243 10.10 In silico Approaches to ADME 243 10.10.1 QSAR Approaches to ADME 244 10.10.2 Theoretical Models for Predicting Metabolism 244 10.10.3 Physiologically-Based Pharmacokinetic Modeling 245 References 246 Index 251 A Personal Foreword Time seems to have passed so quickly since we updated the second edition, but as we began to assemble the list of changes and incorporations we needed to make for the third edition, we realized just how much the science has moved on. It really was time to renew the book and keep it at the forefront of the application of pharmacokinetics and metabolism in drug design. Not only the science had moved on, though, but also the people and finally the actual workplace. In this edition, more attention is paid to the role of drug permeability in determining ADME fate. Much more emphasis is also placed on drug transport (in parallel to permeability). In addition, the role of free drug is described more fully. The latest concepts and knowledge in drug distribution, drug clearance, and drug toxicity are incorporated along with more descriptive sections on human prediction and screening methods. What has not changed is the layout of chapters, trying to condense the huge breadth of topics covered under a readable size. Each chapter can be read alone or in sequence. We hope readers who attempt the sequence find it not a sprint, not a marathon, but a satisfying middle distance: more magic mile than magic methyl. Han and Don took retirement shortly before the third edition project began and Amit and Charlotte have stepped into their place as authors. Both have very strong chemistry backgrounds, with Charlotte a practicing and distinguished medicinal chemist before stepping across into the land of uncertainty and promise (ADME), and Amit is recognized worldwide for his experience in enzymology and metabolism chemistry. Don and Han have not disappeared, they have worked meticulously on the various chapter proofs, ensuring what we thought we had clearly written could actually be comprehended by the reader. Finally the workplace. What brought us together was Pfizer Drug Metabolism (or Pharmacokinetics, Dynamics and Metabolism as it became known later). The original trio of authors of the first and second editions worked together at Sandwich Laboratories. The year 2011 marked the closure of Drug Discovery at Sandwich. The laboratories that buzzed with the electrical energy of drug discovery creating doxazosin, fluconazole, amlodipine, voriconazole, eletriptan, sildenafil, maraviroc, and more now are almost deserted. I feel, as they do, that working at Sandwich during its almost magical productive period was a privilege that left us with a profound sense of achievement. Charlotte too was Sandwich based, and I would like to dedicate the third edition to the people who made the Sandwich site so special in terms of drug discovery and drug metabolism. It will remain in the memory as a crucible of many of the thoughts and ideas in this book. Dennis A. Smith 1 Physicochemistry Abbreviations \| Abbreviation \| Description \| \|--------------\|-------------\| \| CPC \| Centrifugal partition chromatography \| \| CoMFA \| Comparative field analysis \| \| CsA \| Cyclosporine A \| \| 3D-QSAR \| Three-dimensional quantitative structure–activity relationships \| \| HDM \| Hexadecane membrane \| \| IUPAC \| International Union of Pure and Applied Chemistry \| \| MLP \| Molecular lipophilicity potential \| \| PAMPA \| Parallel artificial membrane permeability assay \| \| PGDP \| Propylene glycol dipelargonate \| \| PSA \| Polar surface area \| \| RP-HPLC \| Reversed-phase high-performance liquid chromatography \| \| SF \| Shake flask, referring to traditional method of measuring log $P$ or log $D$ \| \| TPSA \| Topological polar surface area \| Symbols \| Symbol \| Description \| \|--------\|-------------\| \| AP$_{SUV}$ \| Absorption potential measured in small unilamellar vesicles (SUV) \| \| $\Delta \log D$ \| Difference between $\log D$ in octanol/water and $\log D$ in alkane/water \| \| $\Delta \log P$ \| Difference between $\log P$ in octanol/water and $\log P$ in alkane/water \| \| $f'$ \| Rekker or Leo/Hansch fragmental constant for $\log P$ contribution \| \| $K_a$ \| Ionization constant \| \| $\Lambda$ \| Polarity term, mainly related to hydrogen bonding capability of a solute \| \| $\log P$ \| Logarithm of the partition coefficient (P) of neutral species \| \| $\log D$ \| Logarithm of the distribution coefficient ($D$) at a selected pH, usually assumed to be measured in octanol/water \| \| $\log D_{\text{oct}}$ \| Logarithm of the distribution coefficient ($D$) at a selected pH, measured in octanol/water \| \| $\log D_{\text{chex}}$ \| Logarithm of the distribution coefficient ($D$) at a selected pH, measured in cyclohexane/water \| \[\log D_{7.4}\] Logarithm of the distribution coefficient (D) at pH 7.4 MW Molecular weight $\pi$ Hansch constant; contribution of a substituent to $\log P$ $\text{p}K_a$ Negative logarithm of the ionization constant $K_a$ ### 1.1 Physicochemistry and Pharmacokinetics The body can be viewed as primarily composed of a series of membrane barriers dividing aqueous filled compartments. These membrane barriers are principally comprised of the phospholipid bilayers that surround cells and form intracellular barriers around the organelles present in cells (mitochondria, nucleus, etc.). These are formed with the polar ionized head groups of the phospholipid facing toward the aqueous phases and the lipid chains providing a highly hydrophobic inner core. To cross the hydrophobic inner core, a molecule must also be hydrophobic and able to shed its hydration sphere. Many of the processes of drug disposition depend on the ability or inability to cross membranes and hence there is a high correlation with measures of lipophilicity. Moreover, many of the proteins involved in drug disposition have hydrophobic binding sites further adding to the importance of the measures of lipophilicity [1]. At this point, it is appropriate to define the terms hydrophobicity and lipophilicity. According to published IUPAC recommendations, both terms are best described as follows [2]: Hydrophobicity is the association of nonpolar groups or molecules in an aqueous environment that arises from the tendency of water to exclude nonpolar molecules. Lipophilicity represents the affinity of a molecule or a moiety for a lipophilic environment. It is commonly measured by its distribution behavior in a biphasic system, either liquid–liquid (e.g., partition coefficient in 1-octanol/water) or solid–liquid (retention on reversed-phase high-performance liquid chromatography or thin-layer chromatography (TLC) system). Key physicochemical properties that are associated with hydrophobicity and lipophilicity include solubility, hydrogen bonding capacity, and the ionization state [3]. All these properties have a strong influence on membrane permeability that affects absorption [4], distribution, and balance of elimination by transporter-mediated processes and metabolism [5]. ### 1.2 Partition and Distribution Coefficients as Measures of Lipophilicity The inner hydrophobic core of a membrane can be modeled by using an organic solvent. Similarly, a water or aqueous buffer mimics the aqueous media surrounding cells or present within cells. If the organic solvent is not miscible with water, then a two-phase system can be used to study the relative preference of a compound for the aqueous (hydrophilic) or organic (hydrophobic and lipophilic) phase. For an organic compound, lipophilicity can be described in terms of its partition coefficient $P$ (or $\log P$ as it is generally expressed). This is defined as the ratio of concentrations of the compound at equilibrium between the organic and the aqueous phases: $$P = \frac{[\text{drug}]_{\text{organic}}}{[\text{drug}]_{\text{aqueous}}} \quad (1.1)$$ The partition coefficient ($\log P$) describes the intrinsic lipophilicity of the collection of functional groups and carbon skeleton, which combine, to make up the structure of the compound, in the absence of dissociation or ionization. Methods to measure partition and distribution coefficients have been described [6, 7]. Every component of an organic compound has a defined lipophilicity, and calculation of partition coefficient can be performed from a designated structure. Likewise, the effect on $\log P$ of the introduction of a substituent group into a compound can be predicted by a number of methods as pioneered by Hansch [8–11] ($\pi$-values), Rekker [12, 13] ($f$-values), and Leo and Hansch [8–10, 14, 15] ($f'$-values). These values break molecules down into fragments allowing the total lipophilicity to be calculated. Partitioning of a compound between aqueous and lipid (organic) phases is an equilibrium process. When in addition the compound is partly ionized in the aqueous phase, a further (ionization) equilibrium is set up since it is assumed that under normal conditions only the unionized form of the drug penetrates the organic phase [16]. This traditional view is shown schematically in Figure 1.1. This model is consistent with many observations, but partitioning of some compounds into octanol has been shown to occur as an ion pair [17]. Such ion pairs include chloride with basic compounds and sodium with acidic compounds. Whether such behavior can occur with a biological membrane is still not clear, some evidence exists for this with the strongly acidic drug proxicromil. The lipophilicity of 1. Is a function of acid/base strength $\text{pK}_a$ 2. Is a function of $P$ ($\log P$) Figure 1.1 Schematic depicting the relationship between $\log P$ and $\log D$ and $\text{pK}_a$. the drug above pH 6 in octanol–buffer partition experiments depends on ion pair formation and largely unaffected by change in pH in contrast to Eq. (1.5). Similar trends were demonstrated for the in vitro partition of the compound into rat gastrointestinal rings. Furthermore, the absorption of the compound from the perfused gastrointestinal tract of anesthetized rats in vivo was not consistent with classical nonionized drug absorption theories and supported ion pair formation as a mechanism of proxicromil absorption. While phenomena such as these are occasionally reported [18], it is probable that for 99% of examples the theory of unionized drug being the “lipophilic” species is sound. The intrinsic lipophilicity ($P$) of a compound refers only to the equilibrium of the unionized drug between the aqueous phase and the organic phase. It follows that the remaining part of the overall equilibrium, that is, the concentration of ionized drug in the aqueous phase, is also of great importance in the overall observed partition ratio. This in turn depends on the pH of the aqueous phase and the acidity or basicity ($pK_a$) of the charged function. The overall ratio of drug, ionized and unionized, between the phases has been described as the distribution coefficient ($D$), to distinguish it from the intrinsic lipophilicity ($P$). The term has become widely used in recent years to describe, in a single term, the effective (or net) lipophilicity of a compound at a given pH taking into account both its intrinsic lipophilicity and its degree of ionization. The distribution coefficient ($D$) for a monoprotic acid (HA) is defined as $$D = \frac{[\text{HA}]_{\text{organic}}}{([\text{HA}]_{\text{aqueous}} + [\text{A}^-]_{\text{aqueous}})} \quad (1.2)$$ where $[\text{HA}]$ and $[\text{A}^-]$ represent the concentrations of the acid in its unionized and dissociated (ionized) states, respectively. The ionization of the compound in water is defined by its dissociation constant ($K_a$) as $$K_a = \frac{[\text{H}^+][\text{A}^-]}{[\text{HA}]} \quad (1.3)$$ sometimes referred to as the Henderson–Hasselbalch relationship. The combination of Eqs. (1.1)–(1.3) gives the pH distribution (or “pH partition”) relationship: $$D = P/(1 + \{K_a/[\text{H}^+]\}) \quad (1.4)$$ more commonly expressed for monoprotic organic acids in the form of Eqs. (1.5) and (1.6) as follows: $$\log(\{P/D\} - 1) = \text{pH} - pK_a \quad (1.5)$$ or $$\log D = \log P - \log(1 + 10^{\text{pH} - pK_a}) \quad (1.6)$$ For monoprotic organic bases (BH$^+$ dissociating to B), the corresponding relationships are $$\log(\{P/D\} - 1) = pK_a - \text{pH} \quad (1.7)$$ or \[ \log D = \log P - \log(1 + 10^{pH-pK_a}) \] (1.8) From these equations, it is possible to predict the effective lipophilicity ($\log D$) of an acidic or basic compound at any pH value. The data required in order to use the relationship in this way are the intrinsic lipophilicity ($\log P$), the dissociation constant (p$K_a$), and the pH of the aqueous phase. The overall effect of these relationships is the effective lipophilicity of a compound, at physiological pH, is the $\log P$ value minus one unit of lipophilicity, for every unit of pH the p$K_a$ value is below (for acids) and above (for bases) pH$_{7.4}$. Understanding the role of p$K_a$ and intrinsic lipophilicity ($\log P$) in the lipoidal permeability of molecules is extremely important. Drugs can be rendered with more effective lipophilicity by i. increasing intrinsic lipophilicity ($\log P$) by addition of lipophilic fragments or the substitution of polar fragments for lipophilic ones; ii. increasing for an acidic compound the p$K_a$ value of the acidic function, thus increasing the proportion of unionized drug available to distribute into the lipoidal medium; iii. decreasing for a basic compound the p$K_a$ value of the basic function, thus increasing the proportion of unionized drug available to distribute into the lipoidal medium. It is important to cross validate p$K_a$ and $\log D$ and $\log P$ values using Eqs. (6) and (8) or the rule of thumb referred to above. This may illustrate badly measured values. For an ionizable compound, p$K_a$ values can be determined by measuring $\log D$ across a pH range. Obviously for compounds with multifunctional ionizable groups, the relationship between $\log P$ and $\log D$, as well as $\log D$ as function of pH, becomes more complex [19]. Zwitterionic compounds are compounds with an acidic and basic function with p$K_a$ values separated to allow both to be ionized at a given pH. Such compounds are permanently ionized and are at their most lipophilic when both functions are ionized ($\log D_{\text{max}}$) see Figure 1.2. The permanent ionized state is explained by the separation of p$K_a$ values such that even when one function is suppressed the other is ionized. The “lipophilic” charge neutral species occurs when both functions are ionized. ### 1.3 Limitations on the Use of 1-Octanol Octanol is the most widely used model of a biological membrane [20, 21] and $\log D_{7.4}$ values above 0 normally correlate with effective transfer across the lipid core of the membrane, while values below 0 suggest an inability to traverse the hydrophobic barrier. Figure 1.2 Example of a zwitterionic compound with basic and acidic functionality both ionized at a given pH. Octanol, however, supports H bonding. Besides the free hydroxyl group, octanol also contains 4% v/v water at equilibrium. This obviously comes in conflict with the exclusion of water and H bonding functionality at the inner hydrocarbon core of the membrane. For compounds that contain functionality capable of forming H bonds, therefore, the octanol value can overrepresent the actual membrane crossing ability. These compounds can be thought of as having a high hydration potential and difficulty in shedding their water sphere. The use of a hydrocarbon solvent such as cyclohexane can discriminate these compounds either as the only measured value or as a value to be subtracted from the octanol value (Δlog P) [22–24]. Unfortunately, cyclohexane is a poor solvent for many compounds and does not have the utility of octanol. Groups that bond with hydrogen and attenuate actual membrane crossing compared to their predicted ability based on octanol are listed in Figure 1.3. Later, various measures will be used to describe H bonding groups, but as a simple rule the presence of two or more amide functions in a molecule will significantly impact the membrane crossing ability and will need substantial intrinsic lipophilicity in other functions elsewhere in the molecule, to provide sufficient hydrophobicity and to penetrate the lipid core of the membrane. 1.4 Further Understanding of log P 1.4.1 Unraveling the Principal Contributions to log P The concept that log P or log D is composed of two components [25], that of size and polarity, is a useful one. This can be written as Eq. (1.9), \[ \log P \text{ or } \log D = aV - \Lambda \] where $V$ is the molar volume of the compound, $\Lambda$ is a general polarity descriptor, and $a$ is a regression coefficient. Thus, the size component will largely reflect the carbon skeleton of the molecule (lipophilicity), while the polarity will reflect the hydrogen bonding capacity (normally oxygens and nitrogens). The positioning of these properties to the right and left of Figure 1.3 reflects their influence on the overall physicochemical characteristics of a molecule. ### 1.4.2 Hydrogen Bonding Hydrogen bonding is recognized as an important property related to membrane permeation. Various scales expressing H bonding have been developed [26]. Some of these scales describe total hydrogen bonding capability of a compound, while others discriminate between donors and acceptors [27]. In this book, various measures of H bonding will be exemplified and referenced since similar correlations are obtained regardless of methods [28]. Of particular importance is the need to separate hydrogen bonds that form with solvent and internal H bonds. These are not exclusive and compounds can show a change of conformation that hides an “external H bond” by internal H-bonding within the interior of a membrane. Cyclic peptides have poor intrinsic membrane permeability, but they can be designed to have favorable properties. When the differing membrane permeabilities of a series of cyclic peptides were examined, these were found to be related to their ability to form internal hydrogen bonds [29]. Simply, the ability to form internal hydrogen bonds can promote passive membrane permeability by reducing the free energy cost of desolvating the compounds as they insert into the membrane. Cyclic peptides have poor membrane permeability, but there are exceptions, including cyclosporine A (CsA), a cyclic undecapeptide that is used as an orally active immunosuppressive drug. When the partition coefficient of cyclosporine A was measured in octanol/water and heptane/water, it was deduced that the hydrogen-bonding capacity of CsA changed dramatically from an apolar solvent (where it is internally bonded to H) to a polar solvent (where it exposes its H-bonding groups to the solvent). The log $P$ values were 2.9 in octanol and 1.4 in heptane, which is a remarkably small difference between the solvents. In comparison, cyclo (PHe-Phe) had values of 1.6 and $< -3.0$, respectively. Molecular dynamics simulations in water and CCl$_4$ showed that CsA underwent a solvent-dependent conformational change [30]. This interconversion process is slow on the molecular dynamics timescale. The larger the molecule, the process of separating inter- and intra-H bonds becomes more complex, and apart from natural products it is an area not well researched in terms of new drug design. ### 1.4.3 Polar Surface Area Polar surface area is a concept that has gained popularity in recent years. PSA is a simple measure of total hydrogen bonding capacity [31]. The PSA of a molecule is defined as the area of its van der Waals surface that arises from all oxygen and nitrogen atoms plus the hydrogen atoms attached to them. Thus, PSA is also a measure of the ability of a compound to form hydrogen bonds and accounts for three-dimensional (3D) structural features such as shielding or burial of polar groups by other parts of the molecule. PSA has been used to predict passage across the blood–brain barrier [32, 33], flux across a Caco-2 monolayer [34], and human intestinal absorption [35, 36]. The physical explanation is that polar groups are involved in desolvation when they move from an aqueous extracellular environment to the more lipophilic interior of membranes. PSA thus represents, at least in part, the energy involved in membrane transport. PSA depends on the conformation, and the original method of calculation [31] is based on a single minimum energy conformation. Others [32] have taken into account conformational flexibility and coined a dynamic PSA, in which a Boltzmann-weighted average PSA is computed. However, it was demonstrated that PSA calculated for a single minimum energy conformation is in most cases sufficient to produce a sigmoidal relationship to intestinal absorption (see Figure 3.14), differing very little from the dynamic PSA described above [36]. A fast calculation of PSA as a sum of fragment-based contributions, called topological polar surface area (TPSA), has been published [37], which no longer uses a 3D representation of the molecule. This technique brings TPSA to a similar footing to clog $P$ and the two descriptors can be readily calculated for all molecules, even for large data sets such as combinatorial or virtual libraries. Poorly absorbed compounds have been identified as those with a PSA > 140 Å$^2$. Considering more compounds, more scatter was found around the sigmoidal curve observed for a smaller set of compounds [33]. This is partly due to the fact that compounds that are not of high permeability do not usually show simple passive diffusion as they are affected by active uptake carriers and efflux mechanisms involving P-glycoprotein (P-gp) and other transporter proteins. Figure 1.4 ADME space bounded by the interconnected physicochemical properties of molecular weight, polar surface area, and lipophilicity. Drugs with desirable pharmacokinetic properties such as absorption are much more likely to occupy the space. MW is not important per se but reflects that molecules of 500 MW or more are likely to exceed desirable PSA or clog $P$ limits. 1.4.4 Molecular Size and Shape The role of molecular size in oral absorption has been the subject of debate. The Lipinski rule-of-5 proposes an upper limit of MW 500 as acceptable for orally absorbed compounds [38]. Another view is that of Smith [40] who proposed that Eq. (1.9) is a virtual formula relating $\log P$ to MW–PSA and that MW is incidental to the two key terms $\log P$ and PSA [39]. Such a concept allows ADME (absorption, distribution, metabolism, and excretion) to be viewed in terms of a physicochemical space, which includes the properties most likely to be associated with drug-like properties. The dimensions of such a space and its boundaries are illustrated in Figure 1.4. The formula explains the interconnectivity of the physicochemical properties. Lipophilicity will increase membrane permeability but, at its upper limits, solubility will be so low that adequate dissolution will not be achieved at clinical doses. PSA as stated above is the energy cost in desolvation for the molecule to enter the membrane; so, as the higher values of PSA are reached, membrane permeability is energetically unfavorable. The interconnection with molecular weight has spawned a belief in that particular property being important per se. It is the belief of Smith and coworkers that drugs are largely products of carbon (lipophilicity), oxygen, and nitrogen (PSA) and that as molecular weight approaches 500, the chances of too high a lipophilicity or too great a PSA increases. In fact, it is difficult not to exceed a $\log P$ of 5 or a PSA of 140 Å$^2$ when synthesizing molecules of greater than 500 MW. 1.5 Alternative Lipophilicity Scales Since 1-octanol has certain limitations (see Section 1.3), many alternative lipophilicity scales have been proposed [40] (see Figure 1.3). A critical quartet of four solvent systems of octanol (amphiprotic), alkane (inert), chloroform (proton donor), and propylene glycol dipelargonate (PGDP) has been advocated [41, 42]. By measuring distribution in all four, a full understanding of partitioning properties should be obtained. Experience has indicated that the added value of more phases has not been sufficient and octanol has become the “gold standard” now universally adopted for both measurement and computational reference. Various solvents have been used in membrane systems such as PAMPA (see Chapter 10) including the standard $n$-hexadecane, 2% w/v dioleoylphosphatidylcholine in $n$-dodecane, and 20% w/v lecithin in $n$-dodecane [43]. These systems have also been compared with dodecane partitioning per se without really showing advancement on octanol-based systems. 1.6 Computational Systems to Determine Lipophilicity In the design of new compounds as well as the design of experimental procedures, an \textit{a priori} calculation of log $P$ or log $D$ values is useful. Methods may be based on the summation of fragmental [44–46] or atomic contributions [47–49], or a combination of these [50, 51]. Many other methods continue to be advanced, but commercial software usually uses either fragment-based (clog $P$) or atom-based (alog $P$) methods and these have become standardized with clog $P$ usually favored. 1.7 Membrane Systems to Study Drug Behavior The standardization on octanol does not allow all aspects of compound behavior to be understood. Often acidic, neutral, and basic molecules with similar lipophilicities will show different behaviors in a biological system. To study some of these behaviors, different \textit{in vitro} models have been utilized. For instance, the distribution of molecules has been studied between unilamellar vesicles of dimyristoylphosphatidylcholine and aqueous buffers. These systems allow the interaction of molecules to be studied with the whole membrane that includes the charged polar head group area (hydrated) and the highly lipophilic carbon chain region. Such studies indicate that for amine compounds ionized at physiological pH, partitioning into the membrane is highly favored and independent of the degree of ionization. This is believed to be due to electrostatic interactions with the charged phospholipid head group. This property is not shared with acidic compounds even for the “electronically neutral” phosphatidylcholine [52]. Such ionic interactions between basic drugs are even more favored for membranes containing “acidic” phospholipids such as phosphatidylserine [53]. The structures of these two phospholipids are shown in Figure 1.5 together with the structure of the basic drug chlorphentermine. Table 1.1 shows the binding of chlorphentermine to phosphatidylcholine and phosphatidylserine containing membranes. The extent of binding of chlorphentermine and various amphiphilic drugs occurred in increasing order with liposomes prepared from sphingomyelin (SM), phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS). The higher extent of binding to phosphatidylserine does not result from higher affinities for phosphatidylcholine but rather from higher capacity of phosphatidylserine liposomes compared to PC or SM liposomes. The divalent cationic drug \| Phospholipid \| $k \times 10^{-4}$ M \| $n_{\text{max}}$ \| \|-----------------------\|----------------------\|------------------\| \| Phosphatidylserine \| 2.17 \| 0.67 \| \| Phosphatidylcholine \| 1.26 \| 0.05 \| chloroquine showed especially high binding to phosphatidylserine compared to monovalent drugs. These systems potentially predict the actual affinity for the membrane, rather than the ability to transfer across a membrane. Membrane affinity, and hence tissue affinity, is particularly important in the persistence of drugs within the body and relates in part to these interactions. This topic will be covered in Section 4.4. 1.8 Dissolution and Solubility Each cellular membrane can be considered a combination of a physicochemical and biological barrier to drug transport. Poor physicochemical properties may sometimes be overcome by an active transport mechanism. Before any absorption can take place at all, the first important properties to consider are dissolution and solubility. Many cases of solubility-limited absorption have been reported and therefore solubility is now seen as a property to be addressed at the early stages of drug discovery. A compound only in solution is available for permeation across the gastrointestinal membrane. Solubility has long been recognized as a limiting factor in the absorption process leading to the implementation of solubility screens in early stages of drug design [54]. Estimates of desired solubility for good oral absorption depend on the permeability of the compound and the required dose, as illustrated in Table 1.2 [54]. The incorporation of an ionizable center, such as an amine or similar function, into a template can bring a number of benefits including water solubility (see Chapter 3). In the drug design and discovery phase, simple solubility testing is the only practical method and this is often limited to kinetic solubility. This practice examines the solubility of a drug after mixing a small aliquot of a DMSO solution of the drug into aqueous media. Apart from solvent effects, the measurement does not measure the solubility of crystal forms. As the key, first step toward oral absorption, considerable effort went into the development of computational solubility prediction [55–59]. However, partly due to a lack of large sets of experimental data measured under identical conditions, today’s methods are not robust enough for reliable predictions [60]. Further fine-tuning of the models can be expected as now high-throughput data have become available to construct such models. Table 1.2 Desired solubility correlated with expected doses [54]. \| Dose (mg/kg) \| Permeability high \| Medium \| Low \| \|-------------\|-------------------\|--------\|-----\| \| 0.1 \| 1* \| 5 \| 21 \| \| 1 \| 10 \| 52 \| 207 \| \| 10 \| 100 \| 520 \| 2100\| * µg/mL. 1.9 The BCS Classification and Central Role of Permeability Despite the apparent complexity of the drug absorption process, which combines physicochemical properties of the drug, physiological factors of the gastrointestinal tract and the physics of the dosage form the work of Amidon and coworkers [61] who revealed that the fundamental events controlling oral drug absorption are the permeability of the drug through the GI membrane and the solubility/dissolution of the drug dose in the GI milieu. These key parameters have been characterized in the Biopharmaceutics Classification System (BCS) by three dimensionless numbers: absorption number ($A_n$), dissolution number ($D_n$), and dose number ($D_0$). Importantly in this, the solubility is related to the dose size so drugs with high potencies that have adequate solubility for their low dose size may be classed differently to lower potency drugs with better solubility but still inadequate relative to dose size. These numbers take into account both physicochemical and physiological parameters and are fundamental to the oral absorption process. On the basis of their solubility and intestinal membrane permeability characteristics, drug substances have been classified into one of the four categories according to the BCS (Figure 1.6). Wuand and Benet [62] extended the influence of the BCS classification much further when they correlated the high-permeability characteristics of BCS class I and II with metabolic fate. Highly permeable drugs will not be subjected to major transporter influences and will have ready access to metabolizing enzymes within the hepatocytes; moreover, these drugs will not be eliminated renally due to tubular absorption. This analysis showed that if the drug was highly permeable, then the drug will be cleared by metabolism, whereas if the drug is of low permeability the major route of clearance will be renal and/or hepatobiliary. Such is the nature of the ![Figure 1.6](image) Figure 1.6 BCS classification of drugs. The classification not only allows to understand absorption behavior but also, particularly around permeability, provides a framework for all drug dispositions to be referenced. relationship with permeability that it can be used to characterize the clearance fate of the drug, be it via passive permeability and metabolism or via carrier-mediated transport. These properties can also be related to ADME space illustrated in Figure 1.5. The relationship between PSA/log $P$ can rationally be grouped with the categories of drugs in the following classification: \| Class \| PSA/log $P$ \| Category \| \|-------\|-------------\|----------\| \| 1 \| \| Medium \| \| 2 \| \| Low \| \| 3 \| \| High \| \| 4 \| PSA and log $P$ \| High \| This can be exemplified by \| Class \| Drug \| PSA/log $P$ \| \|-------\|------------\|-------------\| \| 1 \| Propranolol\| 41/3.1 \| \| 2 \| Phenytoin \| 58/1.4 \| \| 3 \| Aliskiren \| 146/2.7 \| \| 4 \| Nelfinavir \| 127/7.3 \| Propranolol is a basic compound with nM potency and a dose size around 100 mg. In contrast, phenytoin is a neutral compound with $\mu$M potency and a dose size up to 1 g, explaining the characterization of solubility. Both are cleared exclusively by metabolism. Because the BCS classification includes solubility (which is relevant only in absorption), an altered form can be made to understand disposition and include the physicochemical descriptor PSA/log $P$ ratio and its Table 1.3 Relationship of the disposition fate of a compound with its permeability across a biological membrane. Permeability of a biomembrane is favored by lipophilicity and attenuated by polar functionality (PSA). \| Lipoidal permeability \| Low \| Medium \| High \| \|-----------------------\|-----\|--------\|------\| \| PSA/log $P$ \| \| \| \| \| Absorption \| \| \| \| \| Bioavailability \| \| \| \| \| Clearance \| \| \| \| Low (e.g., aliskiren) unless MWt less than 250 Da and absorbed by paracellular route (e.g., atenolol) As for absorption Renal or biliary (possible transporter involvement) Variable. Influenced by permeability and transporters (e.g., nelfinavir) As for absorption and metabolism Transporters and metabolism High via transcellular route (e.g., propranolol) Variable. Influenced by metabolism Metabolism correlation with permeability. In this, low-permeability compounds cross membranes paracellularly (aqueous pore) and medium- and high-permeability compounds cross membranes by the transmembrane lipoidal route. This classification is exemplified in Table 1.3 [39]. References 1 Pliska, V., Testa, B., and Van de Waterbeemd, H. (eds) (1996) Lipophilicity in Drug Action and Toxicology, Wiley-VCH Verlag GmbH, Weinheim. 2 Van de Waterbeemd, H., Carter, R.E., Grassy, G., Kubinyi, H., Martin, Y.C., Tute, M.S., and Willett, P. (1998) Annual Reports in Medicinal Chemistry, 33, 397–409. 3 Van de Waterbeemd, H. (2002) Physicochemical properties, in Medicinal Chemistry: Principles and Practice, 2nd edn (ed. F.D. King), RSC, London. 4 Van de Waterbeemd, H., Lennernas, H., and Artursson, P. (2003) Drug Bioavailability, Wiley-VCH Verlag GmbH, Weinheim. 5 Van de Waterbeemd, H., Smith, D.A., Beaumont, K., and Walker, D.K. (2001) Journal of Medicinal Chemistry, 44, 1313–1333. 6 Deirdren, J.C. and Bresnien, G.M. (1988) Quantitative Structure–Activity Relationships, 7, 133–144. 7 Hersey, A., Hill, A.P., Hyde, R.M., and Livingstone, D.J. (1989) Quantitative Structure–Activity Relationships, 8, 288–296. 8 Hansch, C. and Leo, A. (1979) Substituent Constants for Correlation Analysis in Chemistry and Biology, Wiley-Interscience, New York. 9 Hansch, C., Leo, A., and Hoekman, D. (1995) Exploring QSAR. Hydrophobic, Electronic and Steric Constants, ACS, Washington. 10 Hansch, C. and Leo, A. (1995) Exploring QSAR. Fundamentals and Applications in Chemistry and Biology, ACS, Washington. 11 Fujita, T., Iwasa, J., and Hansch, C. (1964) Journal of the American Chemical Society, 86, 5175–5180. 12 Rekker, R.F. and De Kort, H.M. (1979) European Journal of Medicinal Chemistry, 14, 479–488. 13 Rekker, R.F. and Mannhold, R. (1992) Calculation of Drug Lipophilicity, Wiley-VCH Verlag GmbH, Weinheim. 14 Leo, A. and Abraham, D.J. (1988) Proteins: Structure, Function and Genetics, 2, 130–152. 15 Leo, A., Hansch, C., and Elkins, D. (1971) Chemical Reviews, 71, 525–616. 16 Manners, C.N., Payling, D.W., and Smith, D.A. (1988) Xenobiotica: The Fate of Foreign Compounds in Biological Systems, 18, 331–350. 17 Davis, M.G., Manners, C.N., Payling, D.W., Smith, D.A., and Wilson, C.A.J. (1984) Journal of Pharmaceutical Sciences, 73, 949–953. 18 Reynold, F., Carrupt, P.A., Testa, B., and Girault, H.H. (1999) Chemistry: A European Journal, 5, 39–47. 19 Smith, R.N., Hansch, C., and Ames, M.M. (1975) Journal of Pharmaceutical Sciences, 64, 599–605. 20 Avdeef, A. (1996) Assessment of distribution-pH profiles, in Lipophilicity in Drug Action and Toxicology (eds V. Pliska, B., Testa, and H. Van de Waterbeemd), Wiley-VCH Verlag GmbH, Weinheim, pp. 109–139. 21 Avdeef, A. (2003) Absorption and Drug Development, Wiley-Interscience, Hobroken, NJ. 22 Young, R.C., Mitchell, R.C., Brown, T.H., Ganellin, C.R., Griffiths, R., Jones, M., Rana, K.K., Saunders, D., Smith, I.R., Sore, N.E., and Wilks, T.J. (1988) Journal of Medicinal Chemistry, 31, 656–671. 23 El Tayar, N., Tsai, R.S., Testa, B., Carrupt, P.A., and Leo, A. (1991) Journal of Pharmaceutical Sciences, 80, 590–598.
A. CALL MEETING TO ORDER, FLAG SALUTE, OPENING STATEMENT: 7:00PM Adequate notice of this regular meeting of the Lopatcong Township Board of Education has been provided in accordance with the Open Public Meetings Act by publication of the annual notice in the Express-Times. A copy of the agenda has been filed with the Township Clerk and posted at both school locations. B. ROLL CALL by Donna Tolley, Business Administrator Present: Steve Ruane, Terry Glennon, Ernie Gallant, Bill Taggart, Debbie Hille, Pam Thomas, and Gena James Absent: Yvette Day, and Bob McFarlane C. READING OF MISSION STATEMENT: Mrs. Roncoroni, Superintendent The Lopatcong School District is committed to provide a safe, secure and supportive learning environment that challenges all students to meet or exceed developmentally appropriate expectations based on New Jersey Core Curriculum Content and Common Core State Standards through interactions with highly qualified and dedicated staff. With community collaboration, the district will promote a positive atmosphere that acknowledges diversity and encourages students to be self-reflective and civically responsible. By providing a comprehensive educational program using varied instructional strategies, recognizing individual differences, incorporating technology and digital literacy and encouraging critical thinking, the district will prepare students for college and careers in the 21st century. D. PRESENTATION: NONE E. PUBLIC COMMENTS ON AGENDA ITEMS: NONE F. REVIEW OF OFFICIAL CORRESPONDENCE: NONE G. ADOPTION OF MINUTES FROM PREVIOUS MEETINGS (Attachment) June 30, 2016 - Regular Meeting Minutes June 30, 2016 - Executive Session Minutes Voice Vote: Motion by Mrs Thomas, second by Mr. Gallant, unanimous. Vote: 7-0 H. COMMITTEE REPORTS a. Ad-hoc - None b. Education - None c. Facilities and Technology: Mr. Glennon stated that the Committee did meet regarding the Facility Use Forms that were tabled at a previous meeting for the LAA/LWA. The Committee has another meeting scheduled for July 18th to discuss all four activities schedules. d. Finance - None e. Negotiations - None f. Policy - None I. FINANCE Motion by , second by , Resolutions R-001-17 will be moved in one roll call R-001-17 Report of Awarded Contracts: BE IT RESOLVED, Pursuant to PL 2015, Chapter 47 the Lopatcong Township Board of Education intends to renew, award, or permit to expire the following contracts previously awarded by the board of education. These contracts are, have been, and will continue to be in full compliance with all state and federal statutes and regulations; in particular, New Jersey Title 18A:18.et.seq, NJAC Chapter 23, and Federal Uniform Administrative Requirements 2CFR, Part200. Barnes & Nobel Book Fair Midland School Camp Warren Education and Recreation Ctr. MRESC Natural Gas Contract with the Alliance for Competitive Energy Services (“ACES”) CDK Systems New Road School Cisco NJCIE for Consultative Services Comegno Law Group, P.C Commission for the Blind De Lange Landen Penn Tele Data Dalrymple Transportation Contract for 2016-17 Eastern Data Communications Phillipsburg School Distict Greenwich Township School District ESC of Morris County Pitney Bowes Educators, Inc. for Educational Services Pohatcong Township School District Extel, Inc RICOH USA Genesis Educational Services R & L Payroll Systems HCESC Honeywell Building Solutions School Alliance Insurance Fund Brown and Brown HP Financial Services Shoretel IEP Direct Transfinder Integra One Settebrino Architects Maschio’s Food Service Warren County Health Department Warren County Technical School WCSSD Centris Group Strauss Esmay Assoc. Elliott Lewis Integra Business Center Eccotrol Inc. Crowder Jr. Company Aqua of New Jersey JCP&L Matheny Medical and Education Center Meals Plus Arrow Elevator Kistler O’Brien Security Services Company RAMM Environmental Services Roll Call: by Donna Tolley, Business Administrator Motion by Mr. Gallant, second by Mrs. Hillie, unanimous, motion carried Vote:7-0 J. EDUCATION Resolutions R-002-17 through R-014-17 and R-024-17 R-002-17 Approve HIB Self Assessment Grade Reports: (Attached) BE IT RESOLVED, that the Board of Education, upon recommendation by the Superintendent, approve the year end HIB Self-Assessment Grade Reports with an overall score of 68/78 for the Elementary School and 70/78 for the Middle School for the reporting period July 1, 2015-June 30, 2016. R-003-17 Permission to Hire Staff BE IT RESOLVED, that the Board of Education hereby authorizes the Superintendent to take appropriate action to effectuate any necessary emergent hires in order to address any unfilled summer positions, unanticipated resignations and/or retirements that may occur between today and the date of the next Board meeting so that the District’s schools are appropriately staffed to provide support and instruction during the 2016-2017 school year and the 2016 summer programs. Any such emergent hires are subject to ratification and approval by the Board at its next regularly scheduled meeting. R-004-17 Vacation Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the vacation payout to Kevin Osifchin as former Vice Principal, for 18 vacation days at a rate of $275.43/day for a total payout of $4,957.74. R-005-17 Vacation Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the vacation payout to Michele McCann as former Business Administrator, for 10 vacation days at a rate of $350/day for a total payout of $3,500.00. R-006-17 Retirement Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the retirement payout to Rosemary Azzilina, retired BSI Teacher, for 26 years at a rate of $450/year for a total payout of $11,700.00. R-007-17 Retirement Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the retirement payout to Teri Fogt, retired HR Assistant, for 26 years at a rate of $450/year for a total payout of $11,700.00. R-008-17 Retirement Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the retirement payout to Dave Bommieno, retired Custodian, for 32 years at a rate of $450/year for a total payout of $14,400.00. R-009-17 Retirement Payout: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the retirement payout to Marianne Burd, retired Guidance Counselor, for 36 years at a rate of $450/year for a maximum payout of $15,000.00. R -010-17 Approve Ancillary Ed. Services Contract with ESC of Sussex County 2016-2017 BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent, approve the Ancillary Educational Services contract with ESC of Sussex County for Student State ID#2143437538 on an as needed basis for the 2016-2017 school year. R -011-17 Approve Contract with Commission for the Blind 2016-2017 BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent, approve the contract with the New Jersey Commission for the Blind and Visually Impaired as per individual student needs for Student State ID#8519869570 and Student State ID#8944858641 for the 2016-2017 school year. R -012-17 Approve Contract Clinical Affiliates 2016-2017 BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent, approve the contract with Clinical Affiliates as per individual student needs for Student State ID#8519869570 for the 2016-2017 school year. R -013-17 Approve Contract with Jennifer K. Rueber Consulting Services 2016-2017 BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent, approve the contract with Jennifer K. Rueber Consulting Services as per individual student needs for Student State ID#8519869570 for the 2016-2017 school year. R-014-17 Acknowledge The Data Submission for the Electronic Violence and Vandalism Reporting System (EVVRS): (Attachment) BE IT RESOLVED, that the Board of Education, upon recommendation by the Superintendent, acknowledge the data submitted for the Electronic Violence and Vandalism Reporting System (EVVRS) for the Report Period Two, January 1, 2016 through June 30, 2016 and Harassment and Bullying – Investigations, Trainings and Programs (HIB-ITP) for Report Period Two, January 1 – June 30, 2016. Roll Call: by Donna Tolley, Business Administrator Motion by Mrs. James, second by Mrs. Thomas, motion carried Vote: 7-0 (Mr. Ruane and Mrs. Thomas said No to R-024-17) R-024-17 Approve Part Time Guidance Counselor: BE IT RESOLVED, that the Board of Education, upon recommendation by the Superintendent, approve the hiring of Ms. Angela Sodtalbers, as Elementary School Part Time (0.5) Guidance Counselor effective August 30, 2016, at Step MA/Z as per the Collective Bargaining Agreement.. Roll Call: by Donna Tolley, Business Administrator Motion by Mrs. James, second by Mrs. Thomas, motion carried Vote: 5-2 (Mr. Ruane and Mrs. Thomas said No to R-024-17) K. FACILITIES & TECHNOLOGY Resolutions R-015-17 through R-020-17 R-015-17 Approve Contract with Kistler O’Brien: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with Kistler O’Brien for $1,730 for the annual fire system inspections at the elementary school. R-016-17 Approve Contract with Kistler O’Brien: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with Kistler O’Brien for $3,708.40 for the annual fire system inspections at the middle school. R-017-17 Approve Contract with LMR Clean Out and Disposal: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with LMR Clean Out and Disposal for $7,160 for annual regular trash and recycling pick up at both schools R-018-17 Approve Contract with Honeywell Building Solutions: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with Honeywell Building Solutions in the amount of $6,430 for 5 visits in the 2016-2017 school year. R-019-17 Approve Contract with Arrow Elevator: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with Arrow Elevator for $2,256.00 for annual elevator preventative maintenance for the middle school. R-020-17 Approve Contract with Security Services Company: BE IT RESOLVED, that the board of Education, upon recommendation by the Superintendent, approve the contract with Security Services Company for $1,642.32 for annual monitoring of fire and burglar alarms at both schools. Roll Call: by Donna Tolley, Business Administrator Motion: by Mr. Gallant, second by Mrs. James, unanimous motion carried Vote: 7-0 L. POLICY Resolutions R-021-17 through R-023-17 R -021-17 FIRST Reading of Revised Policies and Regulations: BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent and the Policy Committee, approve FIRST Reading of the following revisions and adoption of the policies and regulations. \| Code \| Description \| \|--------\|--------------------------------------------------\| \| 1140 \| Affirmative Action Program \| \| 1220 \| Employment of Chief School Administrator \| \| 1310 \| Employment of School Business Administrator/Board Secretary \| \| 1523 \| Comprehensive Equity plan \| \| 1530 \| Equal Employment Opportunities \| \| R1530 \| Equal Employment Opportunities Complaint Procedure \| \| 1550 \| Affirmative Action Program for Employment and Contract Practices \| \| 2200/R2200 \| Curriculum Content \| \| 2260 \| Affirmative Action Program for School and Classroom Practices \| \| 2411/R2411 \| Guidance Counseling \| \| 2423/R2423 \| Bilingual and ESL Education \| \| 2610 \| Educational Program Evaluation \| \| 2622 \| Student Assessment \| 3111 Creating Positions 3124 Employment Contract 3125 Employment of Teaching Staff members 3125.2 Employment of Substitute Teachers 3126/R3126 District Mentoring Program 3141 Resignation 3144/R3144 Certification of Tenure Charges 3159 Teaching Staff member/School District Reporting Responsibilities 3231 Outside Employment as Athletic Coach 3240 Professional Development for Teachers and School Leaders R3240 Professional Development 4159 Support Staff Member/School District Reporting Responsibilities 5305 Health Services Personnel R5330 Administration of Medication 5339 Screening for Dyslexia 5350 Student Suicide Prevention R5350 Student Suicide 5514 Student Use of Vehicles on School Grounds 5750 Equal Educational Opportunities 5755 Equity in Educational Programs and Services 7510 Use of School Facilities 8441/R8441 Care of Injured and ill Persons 8630 Bus Driver/Bus Aide Responsibilities R8630 Emergency School Bus Procedures 9541 Student Teachers/Interns R -022-17 FIRST Reading of New Policy: BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent and the Policy Committee, approve FIRST Reading of the following new policy: 7481 Unmanned Aircraft Systems (UAS also known as Drones) R -023-17 Abolition of Board Policy and Regulation: BE IT RESOLVED, that the Board of Education, by recommendation by the Superintendent and the Policy Committee, approve the abolition of the following Policy and Regulation. 3244/R3244 In-Service Training Roll Call: by Donna Tolley, Business Administrator Motion: by Mr. Taggart, second by Mrs. Hillie unanimous, motion carried Vote: 7-0 M NEGOTIATIONS - None N. ADMINISTRATOR’S REPORT- NONE O. LEGISLATIVE UPDATES - NONE P. REPORT FROM PHILLIPSBURG BOARD OF EDUCATION REPRESENTATIVE Mr. Glennon stated that the Committee has not met recently. He did notice that all the temporary trailers at the high school have wheels under them and are ready to be towed away. The intersection work has been stopped until the utility company moves the light poles to their new location. Q. ITEMS OF EMERGENT NATURE: Mrs. Roncoroni reviewed with the board members the availability dates for the board retreat. The options were Aug 16th, Aug 17th, or Aug 24th. After some discussion, the board decided to schedule Aug 16th from 5:00 – 7:00 pm. Comments by Board of Education members Mrs. Thomas was concerned about budget shortages for the 2016-2017 school year and wanted to make sure that we plan for shortages in the 2017-2018 budget process. R. PUBLIC COMMENTS Pursuant to the Open Public Meetings Act, the Board has set aside two portions of this meeting for public comment. Specifically, during both the “Public Comment on Agenda Items” and the “Other Public Comments” sections noted on the agenda, this meeting will be open to members of the public who wish to speak or make comment on agenda items or a school district issue that may be of concern to the residents of the Township of Lopatcong. In that respect, please limit your comments or questions during the “Public Comment on Agenda Items” to agenda items only, and save any other questions or comments that you may have for the “Other Public Comments” portion of the meeting. During both portions of the meeting, the Board requests that the following procedures be observed: 1. Any person who wishes to speak must wait until they have been recognized by the presiding Board Officer; 2. Before beginning, each speaker must state their name and address and, if speaking on behalf of an organization, the name of that organization; 3. Each speaker is limited to one (1) opportunity to speak during each of the two (2) portions that have been set aside; 4. Each speaker is limited to three (3) minutes in length so that other members of the public who wish to speak may have an opportunity to do so. Additional time may be afforded to a speaker at the sole discretion of the presiding Board Officer; 5. The presiding Board Officer will advise each speaker when the three (3) minute period has expired; 6. Out of respect for other members of the public that may wish to speak, the Board requests that each speaker cede the floor to the next member of the public as soon as they finished making their respective comment(s) and/or when their allotted time has expired; and 7. If your questions or comments pertain to litigation, student or personnel matters, the Board asks that you see the Superintendent after the meeting since the Board does not, pursuant to the Open Public Meetings Act, discuss or respond to these items in public. Mrs. Lance has a question about the retirement payouts. She wanted to know if there was anything in the CBA that allowed for delayed payment of these types of payouts. Mrs. Tolley stated that they have a 90 day payout clause from the date of retirement. Some people we paid by June 30th and the balance we put into 2016-2017 since we wanted to make sure we had the maximum funds to place in surplus. S. EXECUTIVE SESSION: None T. ADJOURNMENT: 7:34 PM Motion by, Mr. Gallant, and second by Mr. Taggart to adjourn the meeting at 7:34 p.m., unanimous, Vote: 7-0. Donna Tolley, Board Secretary
Throughput Maximization for Full-Duplex UAV Aided Small Cell Wireless Systems Meng Hua, Student Member, IEEE, Luxi Yang, Senior Member, IEEE, Cunhua Pan, and Arumugam Nallanathan, Fellow, IEEE Abstract—This paper investigates full-duplex unmanned aerial vehicle (UAV) aided small cell wireless systems, where the UAV serving as the base station (BS) is designed to transmit data to the downlink users and receive data from the uplink users simultaneously. To maximize the total system capacity, including uplink and downlink capacities, the UAV trajectory, downlink/uplink user scheduling, and uplink/user transmit power are alternately optimized. The resulting optimization problem is mixed-integer and non-convex, which is challenging to solve. To address it, the block coordinate descent method and successive convex approximation techniques are leveraged. Simulation results demonstrate the significant capacity gain can be achieved by our design compared with the other designs. Index Terms—Unmanned aerial vehicle (UAV), trajectory optimization, full-duplex. I. INTRODUCTION Recently, unmanned aerial vehicles (UAVs) have received significant research interests both from academia and industry as a promising technique for various applications such as data collection, wireless power transfer, hot-spot offloading, data transmission, etc. [1]–[4]. A typical functionality of UAV is acted as a mobile base station (BS). The authors in [1] studied UAV-aided data collection problem with the objective of minimizing the maximum energy consumption of all sensors by jointly optimizing the communication access strategy and the UAV trajectory. The authors in [2] studied the single UAV-enabled multiuser wireless power transfer system that targets at maximizing the amount of energy transferred to the total users by optimizing the UAV trajectory. The hotspot problem was addressed by [3], where the authors used the UAV to cover cell-edge users and offload the data traffic from the overloaded BS. A multi-UAV enabled system for serving multiple users was presented in [4] to improve throughput by carefully designing the UAV trajectories and their transmit power. A sustainable UAV communication was investigated in [5], where the authors proposed a solar-powered UAV to serve users with energy harvested from sun by adjusting its altitude and horizontal trajectory. In addition, the UAV can also act as a relay. For example, work [6] studied the UAV-aided relay system, where the user communicated with BS with the help of UAV to minimize the system outage by optimizing the UAV trajectory and transmit power. The full-duplex technique allows the downlink and uplink transmission operating at the same time and frequency, and thus can double the system capacity compared with the half-duplex technique [7]. At present, there have been some work on the research of full-duplex UAV [8], [9]. In [8], the authors considered the time-sensitive scenario, where the full-duplex UAV acts as a relay to minimize the relaying system outage probability. The authors in [9] further considered a more complicated scenario, where the full-duplex UAV serving in a device-to-device underlaying cellular system was studied. However, both of them focus on studying the UAV relaying system, the full-duplex UAV acting as mobile BS is still not investigated. In this paper, we deploy a full-duplex UAV-BS to serve the targeted small cell users, including uplink and downlink users. In the uplink transmission phase, multiple uplink users transmit their data to the UAV with TDMA manner. Meanwhile, the full-duplex UAV-BS transmits the data to multiple downlink users in the downlink transmission phase still with TDMA manner. However, the downlink users will receive strong interference from the uplink users. Therefore, a fundamental question for the proposed full-duplex UAV-BS enabled systems is how to jointly optimize the UAV trajectory, uplink user transmit power, downlink and uplink user scheduling so as to maximize the system uplink and downlink capacity. To tackle this challenge, we divide the resulting problem into four sub-problems and optimize one subset of variables while keeping other variables fixed, and then alternately optimize the four sub-problems in an iterative way by using the block coordinate descent method and successive convex optimization techniques. The numerical results demonstrate that the proposed design significantly outperforms the benchmarks. II. SYSTEM MODEL AND PROBLEM FORMULATION We consider a UAV-enabled communication system where the UAV serves as a full-duplex BS that can communicate with $K_D$ single-antenna downlink users and $K_U$ single-antenna uplink users using the same and frequency resource as shown in Fig. 1. The UAV is equipped with two antennas, in which one is used for data transmission in the downlink and the other is used for data reception in the uplink. Define $ K_D = \{1, 2, \ldots, K_D\} $ and $ K_U = \{1, 2, \ldots, K_U\} $ as the sets of downlink and uplink users, respectively. We denote the horizontal coordinate of user $ k $ as $ w_k $, $ k \in K_D \cup K_U $. The UAV altitude is fixed at $ H $. The given time period $ T $ is equally divided into $ N $ time slots with duration $ \delta = T/N $. Then, the horizontal location of UAV at any slot $ n $ is denoted as $ q[n] $. As pointed out by the 3GPP, the UAV-ground channel model depends on the environmental scenarios, such as the suburban with less scattering and the macro urban with rich scattering. Especially when the UAV flies above 40m in the rural area, the UAV offers a nearly 100% LoS probability for UAV-ground channel as shown in the 3GPP specification [10]. As a consequence, the downlink channel gain of UAV to the $ j $-th downlink user at time slot $ n $ is given by [9], [11]–[13] \[ h_{b,j}[n] = \beta_0 \left( \\|q[n] - w_j\\|^2 + H^2 \right)^{-1}, \quad j \in K_D, \] where $ \beta_0 $ represents the reference channel gain at $ d = 1 \text{m} $. Similarly, the uplink channel gain from the $ i $-th uplink user to the UAV at time slot $ n $ is denoted as $ h_{u,i}[n] $, $ i \in K_U $. The channel model from the $ i $-th uplink user to the $ j $-th downlink user follows Rayleigh fading with channel power gain denoted by $ g_{i,j}[n] = \beta_0 d_{i,j}^{-\alpha} \xi $, where $ d_{i,j} $ is the distance between the $ i $-th uplink user and the $ j $-th downlink user, $ \alpha $ denotes the path loss exponent, and $ \xi $ is a random variable following the exponential distribution with unit mean. We adopt a TDMA manner for both downlink and uplink users, and assume that the UAV can only communicate with at most one user at one time slot in the uplink/downlink [4], [11], which yields the following user scheduling constraints \[ \sum_{j=1}^{K_D} x^d_j[n] \leq 1, \forall n, \] \[ x^d_j[n] \in \{0, 1\}, \forall j, n, \] \[ \sum_{i=1}^{K_U} x^u_i[n] \leq 1, \forall n, \] \[ x^u_i[n] \in \{0, 1\}, \forall i, n, \] where (2)–(3) denote the downlink scheduling constraints, and (4)–(5) denote the uplink scheduling constraints. The lower bound for the downlink ergodic capacity from the UAV to the $ j $-th downlink user at time slot $ n $ is given by \[ R^d_j[n] = \log_2 \left( 1 + \frac{p_b h_{b,j}[n]}{\sum_{i=1}^{K_D} x^u_i[n] \beta_0 d_{i,j}^{-\alpha} p_i[n] + \sigma^2} \right), \] where $ \sigma^2 $ denotes the noise power, and $ p_b $ represents the UAV transmit power. Similarly, the uplink capacity from the $ i $-th uplink user to the UAV is given by \[ R^u_i[n] = \log_2 \left( 1 + \frac{p_i[n] f_{b,i}[n]}{f_b[n] + \sigma^2} \right), \] where $ f_b[n] $ denotes the self-interference at time slot $ n $ from the transmit antenna to the receive antenna at the full-duplex UAV (Here, we assume that $ f_b[n] $ is a constant to represent the maximal self-interference on the UAV [9]), and $ p_i[n] $ denotes the $ i $-th uplink user transmit power at time slot $ n $. Let $ A_D = \{x^d_j[n], \forall j, n\} $, $ A_U = \{x^u_i[n], \forall i, n\} $, $ P = \{p_i[n], \forall i, n\} $, $ Q = \{q[n], \forall n\} $. We aim at maximizing the total system capacity, including uplink and downlink links capacity, which is formulated as follows \[ \begin{align} (P) \max_{P, Q, A_D, A_U} & \sum_{n=1}^{N} \sum_{j=1}^{K_D} x^d_j[n] R^d_j[n] + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x^u_i[n] R^u_i[n] \\ \text{s.t.} & \quad 0 \leq p_i[n] \leq P_{\text{max}}, \forall i, n, \\ & \quad \\|q[n] - q[n-1]\\| \leq V_{\text{max}} \delta, \forall n, \\ & \quad q_I = q[0], q_F = q[N], \end{align} \] (2), (3), (4), (5), where $ P_{\text{max}} $ and $ V_{\text{max}} $ respectively denote the maximum uplink user transmit power and UAV speed, $ q_I $ and $ q_F $ represent UAV’s initial and final location, respectively. ### III. Proposed Algorithm The objective function is a function of $ Q, P, A_D, $ and $ A_U $, which is not jointly convex with these variables. In addition, the binary constraints of (3) and (5) make the optimization more difficult to solve. The optimal solution is hard to obtain even using exhaustive search. First, the search space is $ O \left( (K_D K_U)^N \right) $ for solving downlink/uplink user scheduling, which means the complexity is exponentially increasing with the number of time slots $ N $. Second, even with the fixed downlink and uplink user scheduling, the sub-problem is still non-convex with respective to uplink user transmit power and UAV trajectory, which indicates that the optimal solution still cannot be obtained. To deal with these issues, we first relax the binary variables $ A_D $ and $ A_U $ into continuous ones, and transform the binary constraints (3) and (5) into the linear constraints, which are respectively given by \[ 0 \leq x^d_j[n] \leq 1, \forall j, n, \] \[ 0 \leq x^u_i[n] \leq 1, \forall i, n. \] Then, we propose a four-stage iterative optimization algorithm for solving problem (P), and a local solution is obtained. A. Stage 1: Downlink user scheduling design First, we consider the downlink user scheduling problem with the given uplink user scheduling $A_U$, transmit power $P$, and UAV trajectory $Q$. Then, Problem P becomes the following optimization problem \[ (P1) \max_{A_D} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] R_j^d[n] \] s.t. (11). Obviously, problem (P1) is a standard linear programming problem, and can be efficiently solved by using standard optimization packages such as CVX. B. Stage 2: Uplink user scheduling design Second, we study the uplink user scheduling problem with the given transmit power $P$, downlink user scheduling $A_D$, and UAV trajectory $Q$, which can be formulated as \[ (P2) \max_{A_U} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] R_j^d[n] + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u[n] R_i^u[n] \] s.t. (12). The term $R_j^d[n]$ in objective function of problem (P2) is a strictly convex with respect to (w.r.t.) $x_i^u[n]$, and hence not concave. Hence, problem (P2) is a non-convex optimization problem. To deal with this issue, we approximate the convex function as its lower bound, which is linear function of the optimization variables that is much easier to solve. By taking the first-order Taylor expansion at any feasible point $\{x_i^{u,r}\}$, we have \[ R_j^d[n] \geq R_j^{d,lb}[n] \triangleq \log_2 \left( 1 + \frac{p_0 h_{b,j}[n]}{I_j^r[n]} \right) - \sum_{i=1}^{K_U} \beta_0 d_{i,j}^{-\alpha} p_i[n] \frac{\log_2^5}{p_0 h_{b,j}[n]} (x_i^{u,r}[n] - x_i^{u,r}[n]), \] where $I_j^r[n] = \sum_{i=1}^{K_U} x_i^{u,r}[n] \beta_0 d_{i,j}^{-\alpha} p_i[n] + \sigma^2$. As a result, for any given local point $\{x_i^{u,r}\}$, define the following problem \[ (P2') \max_{A_U} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] R_j^{d,lb}[n] + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u[n] R_i^u[n] \] s.t. (12). Now, problem (P2) can be readily shown to be a convex optimization problem that can be efficiently solved. Then, (P2) can be approximately solved by successively updating the uplink user scheduling $A_U$ obtained from (P2'). C. Stage 3: UAV trajectory design Third, we study the UAV trajectory optimization problem with the given transmit power $P$, downlink user scheduling $A_D$, and uplink user scheduling $A_U$, which is given by \[ (P3) \max_{Q} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] R_j^d[n] + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u[n] R_i^u[n] \] s.t. (9), (10). Problem (P3) is a non-convex optimization problem since the objective function is non-convex. To handle the non-convex objective function, the successive convex approximation technique is applied. It can be observed that $R_j^d[n]$ is convex w.r.t. $\\|q[n] - w_j\\|^2$, but it is not convex w.r.t. $q[n]$. By taking the first-order Taylor expansion at any given local point $\\|q^[n] - w_i\\|^2$, we can obtain its convex lower bound as follows \[ R_j^d[n] \geq \log_2 \left( 1 + \frac{C_j[n]}{\\|q^[n] - w_j\\|^2 + H^2} \right) = \Sigma_j[n] \times \left( \\|q[n] - w_j\\|^2 - \\|q^[n] - w_j\\|^2 \right) \triangleq \varphi^{lb}(R_j^d[n]), \] where $\Sigma_j[n] = \frac{C_j[n] \log_2}{(C_j[n] + \\|q^[n] - w_j\\|^2 + H^2)(\\|q^[n] - w_j\\|^2 + H^2)}$ and $C_j[n] = \frac{\kappa_P}{\sum_{i=1}^{K_U} x_i^u[n] \beta_0 d_{i,j}^{-\alpha} p_i[n] + \sigma^2}$. Similarly, to tackle the non-convexity of $R_i^u[n]$, for any given local point $\\|q^[n] - w_i\\|^2$, we have \[ R_i^u[n] \geq \log_2 \left( 1 + \frac{E_i[n]}{\\|q^[n] - w_i\\|^2 + H^2} \right) = F_i[n] \times \left( \\|q[n] - w_i\\|^2 - \\|q^[n] - w_i\\|^2 \right) \triangleq \varphi^{lb}(R_i^u[n]), \] where $F_i[n] = \frac{\log_2 E_i[n]}{(E_i[n] + \\|q^[n] - w_i\\|^2 + H^2)(\\|q^[n] - w_i\\|^2 + H^2)}$ and $E_i[n] = \frac{p_i[n] \beta_0}{f_i[n] + \sigma^2}$. As a result, with (14) and (15), problem (P3) can be simplified as \[ (P3') \max_{Q} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] \psi^{lb}(R_j^d[n]) + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u[n] \varphi^{lb}(R_i^u[n]) \] s.t. (9), (10). Problem (P3') is now a convex optimization problem. Then, we can obtain the locally optimal solution of (P3') by iteratively solving problem (P3') to update the UAV trajectory. D. Stage 4: Uplink user transmit power control Finally, we study the uplink user transmit power optimization problem with the given UAV trajectory $Q$, downlink user scheduling $A_D$, and uplink user scheduling $A_U$, which is given by \[ (P4) \max_{P} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d[n] R_j^d[n] + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u[n] R_i^u[n] \] s.t. (8). The term $R_j^d[n]$ is convex w.r.t. $p_i[n]$ which makes problem (P4) be a non-convex optimization problem. Similar to that of (P2'), by taking the first-order Taylor expansion at local point $\{p_i^[n]\}$, we can approximate it as its lower bound as follows \[ R_j^d[n] \geq \varphi^{lb}(R_j^d[n]) \triangleq \log_2 \left( 1 + \frac{p_0 h_{b,j}[n]}{G_j^r[n]} \right) - \sum_{i=1}^{K_U} \beta_0 x_i^u[n] d_{i,j}^{-\alpha} p_i h_{b,j}[n] \log_2^5 (p_i[n] - p_i^[n]), \] where $ G_j^r [n] = \sum_{i=1}^{K_U} x_i^u [n] \beta_0 d_{i,j}^{-\alpha} p_i^r [n] + \sigma^2 $. Then, with (16), problem (P4) is simplified as \[ (P4) \max_{x^d, p^r} \sum_{n=1}^{N} \sum_{j=1}^{K_D} x_j^d [n] \varphi^{16} \left( R_j^d [n] \right) + \sum_{n=1}^{N} \sum_{i=1}^{K_U} x_i^u [n] R_i^r [n] \] s.t. (8). Problem (P4) is a convex optimization problem. Then, (P4) can be approximately solved by successively updating the uplink user transmit power obtained from problem (P4). ### E. Overall algorithm Based on the above four-stage sub-problems, we optimize the four-stage sub-problems in an iterative way, which is summarized in Algorithm 1. Note that Algorithm 1 is guaranteed to converge a local solution, which can be found in [4], [9]. At last, the continuous user scheduling variable is reconstructed into binary one by adopting the following simple criteria [11]: $ x = \begin{cases} 1, & \text{if } x \geq 0.5, \\ 0, & \text{if } x < 0.5, \end{cases} $ where $ x \in \{x_i^u [n], x_j^d [n], \forall i, j, n\} $. Algorithm 1 Alternating optimization for problem (P) 1: Initialize $ q^r [n], x_i^{u,r} [n], p_i^r [n] $, and set $ r \leftarrow 0 $ as well as tolerance $ \epsilon > 0 $. 2: repeat. 3: Solve (P1) for given $ \{q^r [n], x_i^{u,r} [n], p_i^r [n]\} $, and denote the optimal solution as $ \{x_i^{u,d,r+1} [n]\} $. 4: Solve (P2) for given $ \{q^r [n], x_i^{u,r} [n], p_i^r [n]\} $, and denote the optimal solution as $ \{x_i^{d,r+1} [n]\} $. 5: Solve (P3) for given $ \{x_i^{u,r+1} [n], x_j^{d,r+1} [n], p_i^r [n]\} $, and denote the optimal solution as $ \{q^{r+1} [n]\} $. 6: Solve (P4) for given $ \{x_i^{u,r+1} [n], x_j^{d,r+1} [n], q^{r+1} [n]\} $, and denote the optimal solution as $ \{p_i^{r+1} [n]\} $. 7: $ r \leftarrow r + 1 $. 8: until the fractional increase of the objective value of (P) is less than tolerance $ \epsilon $. It is worth pointing out that all the sub-problems (P1), (P2), (P3), and (P4) are convex, thus the computational complexity of Algorithm is $ O \left( L_4 \left( (K_D N)^{3.5} + L_1 (K_U N)^{3.5} + L_2 (2N)^{3.5} + L_3 (K_U N)^{3.5} \right) \right) $ with $ L_1, L_2, L_3, $ and $ L_4 $ being the iterative numbers. ### IV. Simulation Results In this section, we evaluate the performance of full-duplex UAV system by our proposed algorithm. In our example, we consider four downlink users and four uplink users, i.e., $ K_D = 4 $ and $ K_U = 4 $. The channel gain of the system and noise power are respectively set as $ \beta_0 = -60 $ dB and $ \sigma^2 = -110 $ dBm [4]. The system bandwidth is assumed to be $ B = 1 $ MHz. The UAV maximum transmit power and speed are respectively assumed to be $ p_0 = 0.1 $ W and $ V_{\text{max}} = 50 $ m/s [12]. We set $ \alpha = 3, \delta = 0.5 $ s, $ F_{\text{max}} = 0.1 $ W. Unless otherwise specified, we set $ H = 100 $ m and $ f_b [n] = -130 $ dB, $ \epsilon = 10^{-3} $. Fig. 2 shows the obtained UAV trajectories by our proposed Algorithm for three different periods, i.e., $ T = 30 $ s, $ T = 50 $ s, and $ T = 150 $ s. The UAV’s initial and final location are $ q_I = (0, 500 \text{ m})^T $ and $ q_F = (1000 \text{ m}, 500 \text{ m})^T $, respectively. The circle and square represent locations of uplink and downlink users, respectively. $ D_i $ and $ U_i $ denotes the $ i $-th downlink and uplink user, respectively. It is observed that the UAV prefers moving closer to the downlink users rather than the uplink users for all the three different periods. The reason is that the downlink users are exposed to the strong interference from the uplink users, and the UAV moving closer to the downlink users can enlarge the downlink capacity. To illustrate it clearly, the uplink user transmit power for $ T = 50 $ s is plotted in Fig. 3. We can observe from this figure that for the proximity of two uplink users, one transmits with the maximal power and the other transmits with a lower power in order to reduce the interference to the nearby downlink users. In Fig. 4, we investigate the impact of self-interference on the sum of system capacity for period $ T = 70 $ s. It is observed that the sum of capacity is monotonically non-increasing with the self-interference. Especially, when the power of self-interference $ f_b [n] $ below $-150$ dB, the sum of capacity is not be changed. This is because the power level of self-interference is much smaller than the noise power, i.e., $ \sigma^2 = -110 $ dBm, thus the impact of self-interference on the system can be neglected. However, when the power of self-interference $ f_b [n] $ exceeds $-150$ dB, the power of interference cannot be ignored compared to the noise power, which results in a poor system performance. In Fig. 5, the effect of UAV altitude on the system performance is studied. It is observed that the system performance is significantly decreasing with the UAV altitude. This is expected since the higher UAV altitude resulting in a lower signal-to-noise (SNR) for uplink and downlink. This result can also be directly seen in expressions (6) and (7). In Fig. 6, we compare our proposed design with the folfollowing benchmarks: 1) Ideal no interference: in this scheme when calculating the objective function value, we ignore the interference from the uplink users to the downlink users. Hence, this scheme can serve as the performance upper bound. 2) Proposed design: we jointly optimize the UAV trajectory, downlink/uplink user scheduling, and uplink transmit power. 3) No power control scheme: we assume that the uplink users transmit with the maximal power, and other variables are still optimized. 4) Straight flight scheme: the UAV flies in a straight line from $q_I$ to $q_F$ with constant speed $\frac{\|q_I - q_F\|}{T}$. 5) Static scheme: UAV stays at a position with a fixed UAV altitude $H$ that minimizes the sum of distance to all the users, i.e., the horizontal UAV location is calculated from $q_{static}^opt = \min_{K_S \cup K_D} \sum_{i=1}^{N} \\|q_{static} - w_i\\|^2$. 6) Half-duplex scheme (HD): UAV operates in a half-duplex mode, the interference imposed on the downlink users is disappeared. Note that each time slot is further divided into two time sub-slots with the same duration $\delta/2$, and at each time slot, one time sub-slot is assigned to uplink users and the other is assigned to downlink users. In addition, for straight flight and static benchmarks, the downlink/uplink user scheduling and uplink user transmit power are still optimized. Several insights can be made from Fig. 6. First, we can observe that the interference from the uplink users to the downlink users severely deteriorates the system capacity. Second, the system capacity can be prominently improved by designing the UAV trajectory. Third, the half-duplex UAV performs worst over other benchmarks. This result shows that with the help of full-duplex technique, it provides much system performance gain compared to the half-duplex technique. At last, we find that the system capacity gain of our proposed algorithm over the power control scheme is marginal. This is because that the downlink capacity can be improved by reducing the uplink user transmit power while the uplink capacity will be decreased. Therefore, the sum of downlink and uplink capacity will not improve too much. However, if we consider a fairness metric over the users, we will show that the uplink user transmit power will significantly impact on the system performance in the next figure. In Fig. 7, the benchmarks are the same for that of Fig. 6, but the goal of this design is to maximize the minimum average capacity over both uplink users and downlink users for a fair consideration. Similar results can be obtained from that of Fig. 6 except for the last insight. It is observed from Fig. 7 that the proposed scheme with uplink user power control significantly outperforms no power control scheme in terms of achievable rate. This is because for achieving a fairness of system performance over the uplink and downlink users, the uplink link transmit should be carefully optimized. V. CONCLUSION This paper studied the UAV acted as a full-duplex base station to serve the ground users. We formulated a sum of uplink and downlink throughput maximization problem by alternately optimizing the UAV trajectory, downlink/uplink user scheduling, and uplink user transmit power. To address the resulting optimization problem, we developed an efficient iterative algorithm to solve it. The simulation results showed that a significant performance gain is improved compared to that of half-duplex UAV. In addition, the results also showed that the system performance was prominently improved by optimizing the uplink user transmit power as well as UAV trajectory in terms of average capacity. REFERENCES [1] C. Zhan, Y. Zeng, and R. Zhang, “Energy-efficient data collection in UAV enabled wireless sensor network,” IEEE Wireless Communications Letters, vol. 7, no. 3, pp. 328–331, 2018. [2] J. Xu, Y. Zeng, and R. Zhang, “UAV-enabled wireless power transfer: Trajectory design and energy optimization,” IEEE Transactions on Wireless Communications, vol. 17, no. 9, pp. 5092–5106, 2018. [3] J. Lyu, Y. Zeng, and R. Zhang, “UAV-offloaded for cellular hotspot,” IEEE Transactions on Wireless Communications, vol. 17, no. 6, pp. 3988–4001, 2018. [4] Q. Wu, Y. Zeng, and R. Zhang, “Joint trajectory and communication design for multi-UAV enabled wireless networks,” IEEE Transactions on Wireless Communications, vol. 17, no. 3, pp. 2109–2123, 2018. [5] Y. Sun, D. Xu, D. W. K. Ng, L. Dai, and R. Schober, “Optimal 3D-trajectory design and resource allocation for solar-powered UAV communication systems,” IEEE Transactions on Communications, vol. 67, no. 6, pp. 4281–4296, 2019. [6] S. Zhuang, H. Zhang, Q. He, X. Bian, and L. Song, “Joint trajectory and power optimization for UAV relay networks,” IEEE Communications Letters, vol. 22, no. 1, pp. 161–164, 2018. [7] J. I. Choi, M. Jain, K. Srinivasan, P. Levis, and S. Katti, “Achieving single channel, full duplex wireless communication,” in Proceedings of the sixteenth annual international conference on Mobile computing and networking. ACM, 2010, pp. 1–12. [8] M. Hua, Y. Wang, Z. Zhang, C. Li, Y. Huang, and L. Yang, “Outage probability minimization for low-altitude UAV-enabled full-duplex mobile relaying systems,” China Communications, vol. 15, no. 5, pp. 9–24, 2018. [9] Y. Wang, J. Wang, G. Ding, J. Chen, Y. Li, and Z. Han, “Spectrum sharing planning for full-duplex UAV relaying systems with underlaid D2D communications,” IEEE Journal on Selected Areas in Communications, vol. 36, no. 9, pp. 1986–1999, 2018. [10] 3GPP, “Enhanced LTE support for aerial vehicles,” accessed on Nov. 1, 2019. [Online]. Available: http://portal.3gpp.org/ngppapi/CreateTdoc.aspx?mode=view&contributionUid=RP-1727779 [11] M. Hua, Y. Wang, Z. Zhang, C. Li, Y. Huang, and L. Yang, “Power-efficient communication in UAV-enabled wireless sensor networks,” IEEE Communication Letters, vol. 22, no. 6, pp. 1264–1267, 2018. [12] J. Zeng and R. Zhang, “Energy-efficient UAV communication with trajectory optimization,” IEEE Transactions on Wireless Communications, vol. 16, no. 6, pp. 3747–3760, 2017. [13] B. Duo, Q. Wu, X. Yuan, and R. Zhang, “Energy efficiency maximization for full-duplex UAV secrecy communication,” arXiv preprint arXiv:1906.07346, 2019.
Colin Campbell CV EXHIBITIONS (selected) 2001 Saw Gallery, Ottawa, Ontario (Solo) 2000 'Magnetic North-Experimental Canadian Video', Tour, Walker Art Centre, Minneapolis, Minnesota. Catalogue. Art Gallery Of Ontario, Toronto 'Between Geography and Imagination', Trinity Square Video, Toronto, Screening/Panel. 'Borderstasis', University of Toronto 'Disheveled Destiny World Premier', Owen's Art Gallery, Sackville, N.B. 1999 'Appropriate Behaviour', British/Canadian Video Exchange, London, England. Catalogue. 'Tranz-Tech', Toronto International Video Art Biennial. Catalogue. 'Fragile Electrons' National Gallery of Canada: Toured; Art Gallery of Nova Scotia, Mendel Art Gallery. Catalogue. 'Possible Maps' Vitape, Toronto. Catalogue. 1998 'La Quizaine de la Video', Videograph, Montreal. 'New Toronto Works', Pleasure Dome, Toronto 'Picturing Toronto-The Queen Street Years', Power Plant, Toronto. Catalogue. C Magazine 12 page insert., issue #59, Sept.-Nov. 'Toronto Life magazine. 'Close Encounters/Contacts Time', Ottawa Art Gallery Art Gallery of York University, Toronto. 1997 'Synthetica', Walter Phillips Gallery, Banff 'Flaming Creatures', Agnes Etherington Arts Centre, Kingston. Catalogue. Concordia University, Montreal Jury Member, Exhibition Committee for: Alex Wilson Community Project Design Competition, The Design Exchange, Toronto Premier of 'Rendez-Vous' at 'So High I Could Almost See Eternity' Symposium on Performance Art, Mount Allison University 1996 Vancouver Art Gallery, Vancouver. Love Gasoline, Mercer Union, Toronto. Video Art Plastique, Centre d'Art Contemporain de Basse Normandie, France. 1995 'Pressing The Flesh', Toronto Photographers Workshop, Toronto Solo: Colin Campbell: Early Tapes, Art Gallery of Ontario, Toronto 1994 Art Video Art, TVO Broadcast Television, Toronto Wexner Centre for the Arts, Columbus, Ohio World Conference on AIDS, Berlin Rutgers University, New Brunswick, N. J. New York University, New York EM Media, Calgary Budapest Gay and Lesbian Festival Pyramid Arts Centre, Rochester, New York. Owen's Art Gallery, Sackville, N.B. Pacific Cinematheque, Vancouver, B.C. Brock University, St. Catherines Forest City Gallery, London, Ont. Cinematheque, Toronto Videographe, Montreal 1993 Centre Georges Pompidou, Paris, France Dress Codes, Institute of Contemporary Art, Boston. Video Catalogue. Vision Television, Toronto Artist's Television Launch, Squeeze Club, Toronto Oakville Galleries Owen's Art Gallery, Sackville, N.B. Walter Phillips Gallery, Banff Centre for Art Tapes, Halifax Solo: Retrospective, Colin Campbell Media Works 1972-90, Art Gallery of Nova Scotia. Catalogue. Invention, Video Compilation Launch, Squeeze Club, Toronto. Catalogue. 1992 3rd International Istanbul Biennale. Catalogue. Saw Gallery, Ottawa Video Inn, Vancouver Concordia University, Montreal Aids Committee of Ontario, Ottawa First Night, Toronto Centre for Art Tapes, Halifax Walter Phillips Gallery, Banff 1991 Whitney Museum, New York British Film Institute Gay and Lesbian Film Festival, London, England Melbourne Film Festival, (in competition), Australia 1991 Solo: Retrospective, Colin Campbell Media Works 1972-90, National Gallery of Canada, Ottawa. Catalogue. Retrospective, Colin Campbell Media Works 1972-90, The Power Plant, Toronto. Catalogue. 1990 Festival of Festivals, Toronto Images 90, Toronto Chicago International Film Festival Cinema du Cinq Jour Film Festival, Montreal Women and AIDS Benefit, The Euclid, Toronto Yorkton Short Film and Video Festival, Yorkton Los Angeles Gay and Lesbian Festival, Los Angeles Solo: Retrospective, Colin Campbell Media Works 1972-90, Winnipeg Art Gallery, Winnipeg. Catalogue. Retrospective, Image and Nation Film and Video Festival, Colin Campbell: Video Work, Montreal. Catalogue. 1989 Simon Fraser University, Vancouver Trinity Square Video, Toronto Images Festival, Toronto 'Trade Initiative', Vancouver Art Metropole, Toronto Hamilton Artist's Inc., Hamilton, Ontario Buddies In Bad Times Theatre, Toronto 1988 YYZ, Toronto National Gallery of Canada, Inaugural Exhibition Museum of Modern Art, New York, 'Video: New Canadian Narrative' 1987 Retrospective, Video In, Vancouver 'Sea To Shining Sea', The Power Plant, Toronto. Catalogue 'A Play of History', The Power Plant, Toronto. Catalogue. 4TH Piccadilly Film and Video Fest, London, England. American Film Institute Film and Video Festival Los Angeles Video Culture Canada, Toronto California Institute of the Arts, Valencia, California Concordia University, Montreal University of Saskatchewan, Regina 1986 Channel Four Television, 'Ghosts In The Machine', London, England. Video Series 86/87 49TH Parallel, New York Habits, YYZ Gallery, Toronto Vintage Video, A.R.C., Toronto Trinity Video Archives IV, A Space, Toronto Saw Gallery International Fest. of Video Art Toured: Toronto, Ottawa, Halifax, London, Winnipeg Calgary. Catalogue Montreal-Toronto Video Exchange, Videograph, A Space. 1986 New Work Show, Theatre Passe Muraille, Toronto 1985 Videograph, Montreal Montevideo, Amsterdam Time Based Arts, Amsterdam Day of Resistance, Shaw Fest. Plug-in Winnipeg Ed Media, Guelph, ON. Hallwalls, Buffalo Braffet Gallery, Houston, Texas A.R.C., Toronto Signal Approach, The Funnel, Toronto. Catalogue. Video: A Retrospective, Long Beach Museum, Long Beach, California. Catalogue Gallery 101, Ottawa Concordia University, Montreal University of Victoria, Victoria, B.C. 1984 Recent Acquisitions, Museum of Modern Art, New York The New Work Show, Toronto Sex In Canada, The Kitchen, New York British Video Exchange, London Video Arts, London Eng. Toronto Video, Rio de Janeiro, Brazil American Film Institute, Los Angeles 13th Festival Du Nouveau Cinema, Montreal 1983 New Narrative Video, IMAGE, Video Film Centre, Atlanta Canada Video, Künstlerhaus, Stuttgart, West Germany Gay Lesbian Video Festival, U.C.L.A., Los Angeles Maison de la Culture de St. Etienne des images et des Sons, St. Etienne, France. Toured: Lyon, Paris, Grenoble, Strasbourg. Video Art; A History, Part 2, Museum of Modern Art, New York Carlton University, Ottawa 1982 MANNERism, Vancouver Art Gallery, Vancouver, B.C. VIDEO, Vehicule, Montreal Gay Film and Video Festival, Montreal Flaunting It, Gay Film and Video, Toronto Androgyny in Art, Emily Lowe Gallery, Hofstra University, Hempstead, New York. Catalogu New Narratives For Living Room Viewing, Art Gallery Of Ontario, Toronto O KANADA, Kunstmuseum, Berlin West Germany. Catalogue 1981 VIDEO VIDEO, Festival of Festivals, Toronto Salon Theatre, Toronto Theatre Festival, Toronto Persona, The New Museum, New York. Catalogue Sydney Biennale, Sidney, Australia Catalogue A Space Gallery, Toronto Concordia University, Montreal 1980 Venice Biennale, Venice. Catalogue Canadian Video, PSI New York Canadian Video Open, Cineplex Toronto Installations, Performances, Video, Artist Publications, Harbourfront. Toronto. Catalogue Cabana Room, Toronto Musee Nacional de Arte Contemporanea, Lisbon, Portugal Prospect 80's Museum van Hedendaagse Kunst, Ghent, Belgium Catalogue Le Plan K, Brussels, Belgium Libra Gallery, Los Angeles Projects XXXII, Museum of Modern Art, New York Pumps Gallery, Vancouver, B.C. 'Meet The Artist', Donnell Library, New York 1979 Art Gallery of Hamilton, Hamilton, Ontario Diciannni di Videotape, Rome, Italy. Catalogue. L.A.T.C.A., Los Angeles Art Metropole, Toronto Canada House, London, England Cabana Room, Toronto Arthur Strong Gallery, Winnipeg Foundation For Art Resources, Los Angeles University Art Museum, The University of California, Berkeley, California Ontario College of Art, Toronto 1978 Bonnefantenmuseum, Amsterdam IN VIDEO, Dalhousie University, Halifax, Canada. Catalogue Art Metropole, Toronto Kanadische Künstler, Kunsthalle, Basel, Switzerland Museum Of Modern Art, New York 1977 DOCUMENTA., West Germany Another Dimension, National Gallery of Canada, Ottawa XIV Bienal Internacional de Sao Paolo, Brazil Southland Video Anthology, Long Beach Museum, Long Beach, California Catalogue Thomaslewallen Gallery, Los Angeles 1976 Video International, Aarhus Museum of Art, Aarhus, Denmark Artspace, Peterborough, ON Saw Gallery, Ottawa 1975 Videoart, Institute of Contemporary Art, Philadelphia. Catalogue. Art Metropole, Toronto 1974 Video Art Confrantin 74, Festival d'Automme, Musée d'art Moderne de la Ville de Paris. Catalogue A Space Gallery, Toronto Memorial Art Gallery, Rochester, N.Y. Present Company, Art Gallery of Ontario, Toronto. Catalogue. Project 74, Kunsthalle Bliet Kunst, Wallraf-Richartz Museum, Cologne, West Germany. Videoscape, Art Gallery of Ontario. Toronto. Catalogue. 1973 A Space Gallery, Toronto Toronto Video Artists, Everson Museum of Art, Syracuse, New York. Video Circuits, University of Guelph, Guelph, Ontario. Catalogue. 1972 The Mezzanine Gallery, Nova Scotia College of Art and Design, Halifax, Nova Scotia. REVIEWS (selected) Bronson, A.A. 'Truth and Beauty', File, Toronto, Vol. 3, No. 1 Fall 1975. P. 44 Burstyn, Varda, Reviews, Canadian Art, Vol. 3 No.3 Fall-Sept. 1986 Cameron, Eric, 'Videotape and The University Art Programme', Studio International, VOL 187, No 967, June 1974 P. 289-91 Cameron, Eric, 'Colin Campbell; The Story of Art Star', Vie Des Arts (Montreal), Vol XX No. 78, Spring 1978 p. 46-7 Ditta, Sue, 'A Work In Progress-An Interview With Colin Campbell", Colin Campbell Media Works Catalogue, Winnipeg Art Gallery, 1991 Durand, Doug, Fleming, Martha "Video Performance Art", Body Politic, Toronto, Vol 6 Fall I(&$P.37 Fleming, Martha, "Reviews: Toronto" Artforum, Vol 19, No. 5 pp 78-79 Ferguson, Bruce, Canada Video, Ottawa: The National Gallery of Canada, 1980 Published in conjunction with the Venice Biennale. Ferguson, Bruce, 'Otherwise Worldly' in conjunction with the Winnipeg Art Gallery Publication COLIN CAMPBELL, MEDIA WORKS, 1991 Gale, Peggy, "Video Art In Canada; Four Worlds". Studio International, Vol 191, No. 981, May/June 1976 pp224-29 Gale, Peggy, "Colour Video; Vulgar Potential", Parachute(Montreal) No 20 Vol 1 Dec. 1982 Gale, Peggy, ed. Video By Artists, Toronto: Art Metropole Gale, Peggy, Bronson A.A. eds. Performance By Artists, Toronto: Art Metropole. 1979 Gale, Peggy, ed. Invention: Colin Campbell Video Compilation, Toronto, Art Metropole and Video Tape. Catalogue Gildiner, Alinia, 'Poetic Politics From Danceworks', The Globe and Mail, (Toronto), Sept. 1, 1984 Guest, Tim, 'Modern Love', Centrefold (Toronto) Vol 3 No 4 May 1979, pp196-97 Hanna, Dierdre, 'Retrospective tracks career of video visionary Campbell'. Now Magazine (Toronto) Dec. 12-18, 1991 Hume, Christopher, "Making Privateness In Stunning Art", The Toronto Star April 4, 1986 James, Geoffry, 'An official guide to the periphery', Macleans (Toronto) August 3, 1987 Johnson, Brian, 'Hot art in a cool medium', Macleans (Toronto), March 3, 1986 King, Helen, Images 89, Now Magazine Cover Story (Toronto), May 4, 1989 Kwinter, Kerri, 'Modern Love', Fuse (Toronto) VOL 4 No 2 January 1980 pp8-9 Lamereaux, Johanne, 'On coverage: Performance, Seduction, Flatness' Artscanada Issue 240-241, March-April, pp 25-7 Maitland-Carter, Kathleen, 'Colin Campbell Interviewed', Cineaction! (Toronto), Vol No 9 1987 Marshall, Stuart, 'Strategies of Dissemblance', Colin Campbell, Media Works Catalogue,, Winnipeg Art Gallery, 1991 Mays, John Bentley, "Retrospective Dallies with Sexuality' The Globe and Mail, (Toronto), Dec. 26, 1991 Mays, John Bentley, 'Feminist foibles target of Campbell's satiric video'. The Globe and Mail, (Toronto) March 3, 1989 Mays, John Bentley, "Tapes by a tribe of story tellers', The Globe and mail, (Toronto), Sept. 1, 1984 McFadden, Sarah, Bosporus Dialogues, Art in America No. 6, June 1993, pg. 57 Pointbriand, Chantel, ed. Performance Text(s) & Documents. Parachute, Montreal. 1981 Schneider, Ira and Korot, eds, Video Art, New York, Harcourt Brace Jovanovitch. 1976 Steenburgh, Phil van, Habits' Toronto, 'Invitation to a screening' Cinema Canada, pp 132, July-August. 1986 Video By Artists 2, ed. Town, Elke. Art Metropole. Toronto 1986 Sorfleet, Andrew, 'The Colonization of Canada,' C Magazine, Spring, 1992 Szyllinger, Irene. Reviews; Vanguard (Vancouver) Vol ii No 7 pp36-37 Tuer, Dot, 'Perspectives of The Body In Canadian Video Art,' C Magazine, pp 29-36, Winter 1993 Tuer, Dot, 'Transsexing The Feminine, Video In Drag,' Paralellogram, February, 1989 Tuer, Dot, 'Requiem for a Modern Love' Colin Campbell Media Works, Catalogue, Winnipeg Art Gallery, 1991 Van der Spek, Corneil, 'Mapping '70's Angst,' Xtra, Jun 20, 1996 Interview 'Metro Morning' CBC Radio, "Picturing Toronto-The Queen Street Years" Interview 'The Arts Today' CBC Radio (with Joyce Mason, Editor of C Magazine,) 'Picturing Toronto-The Queen Street Years' MIX Magazine, Winter 1998/1999, VOL.24.3, pg. 20 PERFORMANCE WORK 1987 "Late Night Fuse," Lee's Palace, Toronto 1986 "You Taste American," John Greyson. "Six of 1001 Nights," A Space Gallery. catalogue. "Pure Sin," Tanya Mars. A Space Gallery, Toronto. 1984 X's and O's In The Dead Of Winter'. Margaret Dragu. Off Centre Centre, Calgary. "My Wireless Is Running," (with Dragu, Margaret), Danceworks, Harbourfront, Toronto "A Piece Of The Action," John Greyson, A Space Gallery. Toronto. "You Taste American," Greyson, John. Moment 'homme' Tangent, Montreal. Catalogue. 1982 "I AM Already Changing My Mind" (with Dragu, Margaret), Danceworks, Harbourfront, Toronto. 1980 "Peripheral Blur," Adelaide Court Theatre, Toronto. prod. Factory Theatre Lab. "Peripheral Blur," Parachute Performance Festival, Montreal. prod. Parachute. Catalogue. 1972 Hickey, b&w, 15 minutes, silent Self-Hickey, b&w, 15 minutes, silent True-False, b&w, 9 minutes I've Got Rhythm, I've Got Natural Rhythm, b&w, 10 minutes, silent. Sackville, I'm Yours, b&w, 14:40 minutes Smile, b&w, 10 minutes Real Split, b&w, 12 minutes, silent. 1973 This Is The Way I Really Am, b&w, 20 minutes Jaanus, b&w, 18 minutes, silent Shoot, b&w, 20 minutes Relationships, b&w, 2 channels, 20 minutes 1974 Correspondence 1, b&w, 20 minutes Correspondence 2, b&w, 20 minutes This Is An Edit-This Is Real, b&w, 16:50 minutes Love-Life, b&w, 25 minutes California, Myth and Reality, b&w, 25 minutes I'm a Voyeur, b&w, 15:20 minutes Secrets, b&w, 30 minutes 1975 Hindsight, b&w, 22:20 minutes Insight, b&w, 25 minutes Foresight, b&w, 25 minutes 1976 Passage, b&w, 17 minutes Flight, b&w, 10 minutes The Woman From Malibu, b&w, 12 minutes The Temperature In Lima, b&w, 10 minutes 1977 Culver City Limits, b&w, 12 minutes Shango Botanica, b&w, 41:30 minutes Last Seen Wearing, b&w, 24 minutes Hollywood and Vine, b&w, 20 minutes 1978 Rat's Country, b&w, 12 minutes Modern Love, b&w, 90 minutes 1980 Bad Girls, color, 65 minutes L.A. Flex, color, 20 minutes Peachland, color, 40 minutes He's A Growing Boy-She's Turning Forty, color, 40 minutes 1981 Dangling By Their Mouths, color, 60 minutes Conundrum Clinique, color, 2 channels, 36 minutes Conundrum Clinique, color, 14 minutes Snip Snip, color, 30 minutes (with Rodney Werden) 1983 White Money, color, 8 minutes Un Mois en Languedoc, color 21 minutes 1984 The Woman Who Went Too Far, color, 10 minutes 1986 No Voice Over, color, 27 minutes Bennies From Heaven, color, 25 minutes 1987 Black and Light, color, 53 minutes 1988 Fiddle Faddle, color, 26 minutes 1990 Skin, color, 16 mm, 18 minutes 1997 Rendez-vous, color, 13 minutes 1998 Deja Vu, color, 18 minutes 2000 Dishevelled Destiny, color, 29 minutes Mischief, color 15 minutes ARTIST'S PUBLICATIONS File (Toronto) Vol 4 No 2 pp 53-54, Fall 1979 The Woman from Malibu, Toronto, Art Metropole, 1986 Modern Love, Toronto-Los Angeles, Foundation For Art Resources- Art Metropole 'Noise', Colin Campbell Media Works 1972-90, Winnipeg Art Gallery/National Gallery of Canada, 1990 'Marden and Alty,' Homogenous 2, Toronto. Artist publication. Limited edition. 1992 NOVELS The Lizard's Bite (1995) Disappearance (1998) CURATORIAL WORK Danceworks, Toronto, Performance Curator, 1983-84 Moment'homme, Tangent, Performance Curator, Montreal. 1984 A Space Performance Series, Toronto, 1985-87 Rio de Janeiro Biennale, 'Toronto Video', 1985 'Video Tapes From Chile and Brazil', The National Gallery of Canada, Ottawa. Catalogue. 1985 FILM AND VIDEO FESTIVAL JURIES AND PROGRAMMING Images 1990 Film and Video Festival (Toronto). Programmer, Open Call. Festival du Nouveau Cinema, 1990. Montreal, Quebec. Video jury. COLLECTIONS Museum of Modern Art, New York. The National Gallery of Canada, Ottawa. Art Gallery of Ontario, Toronto. Art Bank, Ottawa. Vancouver Art Gallery, Vancouver, B.C. Canadian Cultural Centre, Paris. Donnelly Library, New York. Museum of Modern Art, Ghent. Kunsthalle, Berlin. Concordia University, Montreal. Oakville Galleries, Oakville. Art Metropole, Toronto. Centre Georges Pompidou, Paris, France. Pyramid Arts Centre, Rochester, New York. Brock University, St. Catherine, Ont. Owens Art Gallery, Sackville, N. B. AWARDS Canada Council 'A' Grants: 1988, 1984, 1982, 1980 Canada Council 'B' Grants: 1979, 1977, 1975, 1967, 1966 Ontario Arts Council Grants: 1987, 1986, 1984, 1982, 1981 Bell Canada Lifetime Achievement Award in Video, 1996 JURIES AND ADVISORY PANELS Canada Council (Video, Performance, Media Arts) 1993 Ontario Arts Council Jury, 1994 Metro Cultural Affairs Media Arts Panel, 1991-93 (Toronto) Manitoba Arts Council Video and Film Jury 1993 Selection of curators for 1987 Sao Paolo Biennale, and 1988 Venice Biennale. Ontario Arts Council Advisory Panel in Film Video Photography, 1990, 1994 Canada Council Video Jury, 1995 Manitoba Arts Council Jury Arts Organisations, 1995 RELATED ACTIVITIES President of Board, V Tape Distribution 1986 to present. Making Scenes Council Member, Ottawa Ontario 1997 to present. Advisory Board, Northern Visions Images Festival of Independent Film and Video, Toronto. 1994-to present Advisory Board, Inside Out Collective, Gay and Lesbian Film and Video Festival, Toronto 1994-to present Interview Committee for Video Audio Integrated Media Arts Officer, The Canada Council, Ottawa 1992 Board Member, Trinity Square Video, 1990-92 20th Anniversary Catalogue Trinity Square Video Committee, 1992 Founding Member Beaver Hall Artist Housing Co-op Toronto 1986-91 Performance Committee, A Space Gallery, 1986-87
Applying Alternating Structure Optimization to Word Sense Disambiguation Rie Kubota Ando IBM T.J. Watson Research Center Yorktown Heights, NY 10598, U.S.A. firstname.lastname@example.org Abstract This paper presents a new application of the recently proposed machine learning method Alternating Structure Optimization (ASO), to word sense disambiguation (WSD). Given a set of WSD problems and their respective labeled examples, we seek to improve overall performance on that set by using all the labeled examples (irrespective of target words) for the entire set in learning a disambiguator for each individual problem. Thus, in effect, on each individual problem (e.g., disambiguation of “art”) we benefit from training examples for other problems (e.g., disambiguation of “bar”, “canal”, and so forth). We empirically study the effective use of ASO for this purpose in the multi-task and semi-supervised learning configurations. Our performance results rival or exceed those of the previous best systems on several Senseval lexical sample task data sets. 1 Introduction Word sense disambiguation (WSD) is the task of assigning pre-defined senses to words occurring in some context. An example is to disambiguate an occurrence of “bank” between the “money bank” sense and the “river bank” sense. Previous studies e.g., (Lee and Ng, 2002; Florian and Yarowsky, 2002), have applied supervised learning techniques to WSD with success. A practical issue that arises in supervised WSD is the paucity of labeled examples (sense-annotated data) available for training. For example, the training set of the Senseval-2$^1$ English lexical sample task has only 10 labeled training examples per sense on average, which is in contrast to nearly 6K training examples per name class (on average) used for the CoNLL-2003 named entity chunking shared task$^2$. One problem is that there are so many words and so many senses that it is hard to make available a sufficient number of labeled training examples for each of a large number of target words. On the other hand, this indicates that the total number of available labeled examples (irrespective of target words) can be relatively large. A natural question to ask is whether we can effectively use all the labeled examples (irrespective of target words) for learning on each individual WSD problem. Based on these observations, we study a new application of Alternating Structure Optimization (ASO) (Ando and Zhang, 2005a; Ando and Zhang, 2005b) to WSD. ASO is a recently proposed machine learning method for learning predictive structure (i.e., information useful for predictions) shared by multiple prediction problems via joint empirical risk minimization. It has been shown that on several tasks, performance can be significantly improved by a semi-supervised application of ASO, which obtains useful information from unlabeled data by learning automatically created prediction problems. In addition to such semi-supervised learning, this paper explores ASO multi-task learning, which learns a number of WSD problems simultaneously to exploit the inherent predictive structure shared by these WSD problems. Thus, in effect, each individual problem (e.g., disambiguation of “art”) benefits from labeled training examples for other problems (e.g., disambiguation of “bar”, disambiguation of “canal”, and so forth). The notion of benefiting from training data for other word senses is not new by itself. For instance, $^1$http://www.cs.unt.edu/~rada/senseval/. WSD systems have been evaluated in the series of Senseval workshops. $^2$http://www.cnts.ua.ac.be/conll2003/ner/ on the WSD task with respect to WordNet synsets, Kohomban and Lee (2005) trained classifiers for the top-level synsets of the WordNet semantic hierarchy, consolidating labeled examples associated with the WordNet sub-trees. To disambiguate test instances, these coarse-grained classifiers are first applied, and then fine-grained senses are determined using a heuristic mapping. By contrast, our approach does not require pre-defined relations among senses such as the WordNet hierarchy. Rather, we let the machine learning algorithm ASO automatically and implicitly find relations with respect to the disambiguation problems (i.e., finding shared predictive structure). Interestingly, in our experiments, seemingly unrelated or only loosely related word-sense pairs help to improve performance. This paper makes two contributions. First, we present a new application of ASO to WSD. We empirically study the effective use of ASO and show that labeled examples of all the words can be effectively exploited in learning each individual disambiguator. Second, we report performance results that rival or exceed the state-of-the-art systems on Senseval lexical sample tasks. 2 Alternating structure optimization This section gives a brief summary of ASO. We first introduce a standard linear prediction model for a single task and then extend it to a joint linear model used by ASO. 2.1 Standard linear prediction models In the standard formulation of supervised learning, we seek a predictor that maps an input vector (or feature vector) $ x \in X $ to the corresponding output $ y \in Y $. For NLP tasks, binary features are often used – for example, if the word to the left is “money”, set the corresponding entry of $ x $ to 1; otherwise, set it to 0. A $ k $-way classification problem can be cast as $ k $ binary classification problems, regarding output $ y = +1 $ and $ y = -1 $ as “in-class” and “out-of-class”, respectively. Predictors based on linear prediction models take the form: $ f(x) = w^T x $, where $ w $ is called a weight vector. A common method to obtain a predictor $ \hat{f} $ is regularized empirical risk minimization, which minimizes an empirical loss of the predictor (with regularization) on the $ n $ labeled training examples $ \{(X_i, Y_i)\} $: \[ \hat{f} = \arg \min_f \left( \sum_{i=1}^{n} L(f(X_i), Y_i) + r(f) \right). \] A loss function $ L(\cdot) $ quantifies the difference between the prediction $ f(X_i) $ and the true output $ Y_i $, and $ r(\cdot) $ is a regularization term to control the model complexity. 2.2 Joint linear models for ASO Consider $ m $ prediction problems indexed by $ \ell \in \{1, \ldots, m\} $, each with $ n_\ell $ samples $ (X^\ell_i, Y^\ell_i) $ for $ i \in \{1, \ldots, n_\ell\} $, and assume that there exists a low-dimensional predictive structure shared by these $ m $ problems. Ando and Zhang (2005a) extend the above traditional linear model to a joint linear model so that a predictor for problem $ \ell $ is in the form: \[ f_\ell(\Theta, x) = w^\ell_T x + v^\ell_T \Theta x, \quad \Theta \Theta^T = I, \] where $ I $ is the identity matrix. $ w_\ell $ and $ v_\ell $ are weight vectors specific to each problem $ \ell $. Predictive structure is parameterized by the structure matrix $ \Theta $ shared by all the $ m $ predictors. The goal of this model can also be regarded as learning a common good feature map $ \Theta x $ used for all the $ m $ problems. 2.3 ASO algorithm Analogous to (1), we compute $ \Theta $ and predictors so that they minimize the empirical risk summed over all the problems: \[ [\Theta, \{\hat{f}_\ell\}] = \arg \min_{\Theta, \{\hat{f}_\ell\}} \sum_{\ell=1}^{m} \left( \sum_{i=1}^{n_\ell} \frac{L(f_\ell(\Theta, X^\ell_i), Y^\ell_i)}{n_\ell} + r(f_\ell) \right). \] It has been shown in (Ando and Zhang, 2005a) that the optimization problem (3) has a simple solution using singular value decomposition (SVD) when we choose square regularization: $ r(f_\ell) = \lambda \\|w_\ell\\|_2^2 $ where $ \lambda $ is a regularization parameter. Let $ u_\ell = w_\ell + \Theta^T v_\ell $. Then (3) becomes the minimization of the joint empirical risk written as: \[ \sum_{\ell=1}^{m} \left( \sum_{i=1}^{n_\ell} \frac{L(u^\ell_T X^\ell_i, Y^\ell_i)}{n_\ell} + \lambda \\|u_\ell - \Theta^T v_\ell\\|_2^2 \right). \] This minimization can be approximately solved by repeating the following alternating optimization procedure until a convergence criterion is met: 1. Fix $(\Theta, \{v_\ell\})$, and find $m$ predictors $\{u_\ell\}$ that minimizes the joint empirical risk (4). 2. Fix $m$ predictors $\{u_\ell\}$, and find $(\Theta, \{v_\ell\})$ that minimizes the joint empirical risk (4). The first step is equivalent to training $m$ predictors independently. The second step, which couples all the predictors, can be done by setting the rows of $\Theta$ to the most significant left singular vectors of the predictor (weight) matrix $U = [u_1, \ldots, u_m]$, and setting $v_\ell = \Theta u_\ell$. That is, the structure matrix $\Theta$ is computed so that the projection of the predictor matrix $U$ onto the subspace spanned by $\Theta$'s rows gives the best approximation (in the least squares sense) of $U$ for the given row-dimension of $\Theta$. Thus, intuitively, $\Theta$ captures the commonality of the $m$ predictors. ASO has been shown to be useful in its semi-supervised learning configuration, where the above algorithm is applied to a number of auxiliary problems that are automatically created from the unlabeled data. By contrast, the focus of this paper is the multi-task learning configuration, where the ASO algorithm is applied to a number of real problems with the goal of improving overall performance on these problems. 3 Effective use of ASO on word sense disambiguation The essence of ASO is to learn information useful for prediction (predictive structure) shared by multiple tasks, assuming the existence of such shared structure. From this viewpoint, consider the target words of the Senseval-2 lexical sample task, shown in Figure 1. Here we have multiple disambiguation tasks; however, at a first glance, it is not entirely clear whether these tasks share predictive structure (or are related to each other). There is no direct semantic relationship (such as synonym or hyponym relations) among these words. \| Nouns \| art, authority, bar, bum, chair, channel, child, church, circuit, day, detention, dyke, facility, fatigue, feeling, grip, hearth, holiday, lady, material, month, nation, nature, post, restraint, sense, spade, stress, yew \| \|-------\|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\| \| Verbs \| begin, call, carry, collaborate, develop, draw, dress, drift, drive, face, ferret, find, keep, leave, live, match, play, pull, replace, see, serve, strike, train, treat, turn, use, wander, wash, work \| \| Adjectives \| blind, colourless, cool, faithful, fine, fit, free, graceful, green, local, natural, oblique, simple, solemn, vital \| Figure 1: Words to be disambiguated: Senseval-2 English lexical sample task. \| Local context \| word uni-grams in 5-word window, word bi- and tri-grams of $(w_{-2}, w_{-1}), (w_{+1}, w_{+2}), (w_{-1}, w_{+1}), (w_{-3}, w_{-2}, w_{-1}), (w_{+1}, w_{+2}, w_{+3}), (w_{-1}, w_{+1}, w_{+2}, w_{+3})$. \| \|---------------\|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\| \| Syntactic \| full parser output; see Section 3 for detail. \| \| Global \| all the words excluding stopwords. \| \| POS \| uni-, bi-, and tri-grams in 5-word window. \| Figure 2: Features. $w_i$ stands for the word at position $i$ relative to the word to be disambiguated. The 5-word window is $[-2, +2]$. Local context and POS features are position-sensitive. Global context features are position insensitive (a bag of words). The goal of this section is to empirically study the effective use of ASO for improving overall performance on these seemingly unrelated disambiguation problems. Below we first describe the task setting, features, and algorithms used in our implementation, and then experiment with the Senseval-2 English lexical sample data set (with the official training / test split) for the development of our methods. We will then evaluate the methods developed on the Senseval-2 data set by carrying out the Senseval-3 tasks, i.e., training on the Senseval-3 training data and then evaluating the results on the (unseen) Senseval-3 test sets in Section 4. Task setting In this work, we focus on the Senseval lexical sample task. We are given a set of target words, each of which is associated with several possible senses, and their labeled instances for training. Each instance contains an occurrence of one of the target words and its surrounding words, typically a few sentences. The task is to assign a sense to each test instance. Features We adopt the feature design used by Lee and Ng (2002), which consists of the following four types: (1) Local context: $n$-grams of nearby words (position sensitive); (2) Global context: all the words (excluding stopwords) in the given context (position-insensitive; a bag of words); (3) POS: parts-of-speech $n$-grams of nearby words; (4) Syntactic relations: syntactic information obtained from parser output. To generate syntactic relation features, we use the Slot Grammar-based full parser ESG (McCord, 1990). We use as features syntactic relation types (e.g., subject-of, object-of, and noun modifier), participants of syntactic relations, and bigrams of syntactic relations / participants. Details of the other three types are shown in Figure 2. Implementation Our implementation follows Ando and Zhang (2005a). We use a modification of the Huber’s robust loss for regression: \[ L(p, y) = \begin{cases} (\max(0, 1 - py))^2 & \text{if } py \geq -1; \\ -4py & \text{otherwise;} \end{cases} \] with square regularization $ \lambda = 10^{-4} $, and perform empirical risk minimization by stochastic gradient descent (SGD) (see e.g., Zhang (2004)). We perform one ASO iteration. 3.1 Exploring the multi-task learning configuration The goal is to effectively apply ASO to the set of word disambiguation problems so that overall performance is improved. We consider two factors: feature split and partitioning of prediction problems. 3.1.1 Feature split and problem partitioning Our features described above inherently consist of four feature groups: local context (LC), global context (GC), syntactic relation (SR), and POS features. To exploit such a natural feature split, we explore the following extension of the joint linear model: \[ f_\ell(\{\Theta_j\}, x) = w_\ell^T x + \sum_{j \in F'} v_\ell^{(j)^T} \Theta_j x^{(j)}, \] where $ \Theta_j \Theta_j^T = I $ for $ j \in F $, $ F $ is a set of disjoint feature groups, and $ x^{(j)} $ (or $ v_\ell^{(j)} $) is a portion of the feature vector $ x $ (or the weight vector $ v_\ell $) corresponding to the feature group $ j $, respectively. This is a slight modification of the extension presented in (Ando and Zhang, 2005a). Using this model, ASO computes the structure matrix $ \Theta_j $ for each feature group separately. That is, SVD is applied to the sub-matrix of the predictor (weight) matrix corresponding to each feature group $ j $, which results in more focused dimension reduction of the predictor matrix. For example, suppose that $ F = \{SR\} $. Then, we compute the structure matrix $ \Theta_{SR} $ from the corresponding sub-matrix of the predictor matrix $ U $, which is the gray region of Figure 3 (a). The structure matrices $ \Theta_j $ for $ j \notin F $ (associated with the white regions in the figure) should be regarded as being fixed to the zero matrices. Similarly, it is possible to compute a structure matrix from a subset of the predictors (such as noun disambiguators only), as in Figure 3 (b). In this example, we apply the extension of ASO with $ F = \{SR\} $ to three sets of problems (disambiguation of nouns, verbs, and adjectives, respectively) separately. ![Figure 3: Examples of feature split and problem partitioning.](image) To see why such partitioning may be useful for our WSD problems, consider the disambiguation of “bank” and the disambiguation of “save”. Since a “bank” as in “money bank” and a “save” as in “saving money” may occur in similar global contexts, certain global context features effective for recognizing the “money bank” sense may be also effective for disambiguating “save”, and vice versa. However, with respect to the position-sensitive local context features, these two disambiguation problems may not have much in common since, for instance, we sometimes say “the bank announced”, but we rarely say “the save announced”. That is, whether problems share predictive structure may depend on feature types, and in that case, seeking predictive structure for each feature group separately may be more effective. Hence, we experiment with the configurations with and without various feature splits using the extension of ASO. Our target words are nouns, verbs, and adjectives. As in the above example of “bank” (noun) and “save” (verb), the predictive structure of global context features may be shared by the problems irrespective of the parts of speech of the target words. However, the other types of features may be more dependent on the target word part of speech. ThereTherefore, we explore two types of configuration. One applies ASO to all the disambiguation problems at once. The other applies ASO separately to each of the three sets of disambiguation problems (noun disambiguation problems, verb disambiguation problems, and adjective disambiguation problems) and uses the structure matrix $\Theta_j$ obtained from the noun disambiguation problems only for disambiguating nouns, and so forth. Thus, we explore combinations of two parameters. One is the set of feature groups $F$ in the model (5). The other is the partitioning of disambiguation problems. ### 3.1.2 Empirical results ![Figure 4: F-measure on Senseval-2 English test set. Multi-task configurations varying feature group set $F$ and problem partitioning. Performance at the best dimensionality of $\Theta_j$ (in $\{10, 25, 50, 100, \cdots\}$) is shown.](image) In Figure 4, we compare performance on the Senseval-2 test set produced by training on the Senseval-2 training set using the various configurations discussed above. As the evaluation metric, we use the F-measure (micro-averaged)$^3$ returned by the official Senseval scorer. Our baseline is the standard single-task configuration using the same loss function (modified Huber) and the same training algorithm (SGD). The results are in line with our expectation. To learn the shared predictive structure of local context (LC) and syntactic relations (SR), it is more advantageous to apply ASO to each of the three sets of problems (disambiguation of nouns, verbs, and adjectives, respectively), separately. By contrast, global context features (GC) can be more effectively exploited when ASO is applied to all the disambiguation problems at once. It turned out that the configuration $F = \{\text{POS}\}$ does not improve the performance over the baseline. Therefore, we exclude POS from the feature group set $F$ in the rest of our experiments. Comparison of $F = \{\text{LC} + \text{SR} + \text{GC}\}$ (treating the features of these three types as one group) and $F = \{\text{LC}, \text{SR}, \text{GC}\}$ indicates that use of this feature split indeed improves performance. Among the configurations shown in Figure 4, the best performance (67.8%) is obtained by applying ASO to the three sets of problems (corresponding to nouns, verbs, and adjectives) separately, with the feature split $F = \{\text{LC}, \text{SR}, \text{GC}\}$. ASO has one parameter, the dimensionality of the structure matrix $\Theta_j$ (i.e., the number of left singular vectors to compute). The performance shown in Figure 4 is the ceiling performance obtained at the best dimensionality (in $\{10, 25, 50, 100, 150, \cdots\}$). In Figure 5, we show the performance dependency on $\Theta_j$’s dimensionality when ASO is applied to all the problems at once (Figure 5 left), and when ASO is applied to the set of the noun disambiguation problems (Figure 5 right). In the left figure, the configuration $F = \{\text{GC}\}$ (global context) produces better performance at a relatively low dimensionality. In the other configurations shown in these two figures, performance is relatively stable as long as the dimensionality is not too low. ![Figure 5: Left: Applying ASO to all the WSD problems at once. Right: Applying ASO to noun disambiguation problems only and testing on the noun disambiguation problems only. x-axis: dimensionality of $\Theta_j$.](image) ### 3.2 Multi-task learning procedure for WSD Based on the above results on the Senseval-2 test set, we develop the following procedure using the feature split and problem partitioning shown in Figure 6. Let $\mathcal{N}$, $\mathcal{V}$, and $\mathcal{A}$ be sets of disambiguation problems whose target words are nouns, verbs, and adjectives, respectively. We write $\Theta_{(j,s)}$ for the struc- --- $^3$Our precision and recall are always the same since our systems assign exactly one sense to each instance. That is, our F-measure is the same as ‘micro-averaged recall’ or ‘accuracy’ used in some of previous studies we will compare with. Figure 6: Effective feature split and problem partitioning. ture matrix associated with the feature group $j$ and computed from a problem set $s$. That is, we replace $\Theta_j$ in (5) with $\Theta_{(j,s)}$. - Apply ASO to the three sets of disambiguation problems (corresponding to nouns, verbs, and adjectives), separately, using the extended model (5) with $F = \{LC, SR\}$. As a result, we obtain $\Theta_{(j,s)}$ for every $(j,s) \in \{LC, SR\} \times \{\mathcal{N}, \mathcal{V}, \mathcal{A}\}$. - Apply ASO to all the disambiguation problems at once using the extended model (5) with $F = \{GC\}$ to obtain $\Theta_{(GC, \mathcal{N} \cup \mathcal{V} \cup \mathcal{A})}$. - For a problem $\ell \in P \in \{\mathcal{N}, \mathcal{V}, \mathcal{A}\}$, our final predictor is based on the model: $$f_\ell(x) = w^T_\ell x + \sum_{(j,s) \in T} v^{(j,s)^T}_\ell \Theta_{(j,s)} x^{(j)},$$ where $T = \{(LC, P), (SR, P), (GC, \mathcal{N} \cup \mathcal{V} \cup \mathcal{A})\}$. We obtain predictor $\hat{f}_\ell$ by minimizing the regularized empirical risk with respect to $w_\ell$ and $v_\ell$. We fix the dimension of the structure matrix corresponding to global context features to 50. The dimensions of the other structure matrices are set to 0.9 times the maximum possible rank to ensure relatively high dimensionality. This procedure produces 68.1% on the Senseval-2 English lexical sample test set. 3.3 Previous systems on Senseval-2 data set Figure 7 compares our performance with those of previous best systems on the Senseval-2 English lexical sample test set. Since we used this test set for the development of our method above, our performance should be understood as the potential performance. (In Section 4, we will present evaluation results on the unseen Senseval-3 test sets.) Nevertheless, it is worth noting that our potential performance (68.1%) exceeds those of the previous best systems. Our single-task baseline performance is almost the same as LN02 (Lee and Ng, 2002), which uses SVM. This is consistent with the fact that we adopted LN02’s feature design. FY02 (Florian and Yarowsky, 2002) combines classifiers by linear average stacking. The best system of the Senseval-2 competition was an early version of FY02. WSC04 used a polynomial kernel via the kernel Principal Component Analysis (KPCA) method (Schölkopf et al., 1998) with nearest neighbor classifiers. 4 Evaluation on Senseval-3 tasks In this section, we evaluate the methods developed on the Senseval-2 data set above on the standard Senseval-3 lexical sample tasks. 4.1 Our methods in multi-task and semi-supervised configurations In addition to the multi-task configuration described in Section 3.2, we test the following semi-supervised application of ASO. We first create auxiliary problems following Ando and Zhang (2005a)’s partially-supervised strategy (Figure 8) with distinct feature maps $\Psi_1$ and $\Psi_2$ each of which uses one of $\{LC, GC, SR\}$. Then, we apply ASO to these auxiliary problems using the feature split and the problem partitioning described in Section 3.2. Note that the difference between the multi-task and semi-supervised configurations is the source of information. The multi-task configuration utilizes the label information of the training examples that are labeled for the rest of the multiple tasks, and the semi-supervised learning configuration exploits a large amount of unlabeled data. 1. Train a classifier $C_1$ only using feature map $\Psi_1$ on the labeled data for the target task. 2. Auxiliary problems are to predict the labels assigned by $C_1$ to the unlabeled data, using the other feature map $\Psi_2$. 3. Apply ASO to the auxiliary problems to obtain $\Theta$. 4. Using the joint linear model (2), train the final predictor by minimizing the empirical risk for fixed $\Theta$ on the labeled data for the target task. Figure 8: Ando and Zhang (2005a)'s ASO semi-supervised learning method using partially-supervised procedure for creating relevant auxiliary problems. ### 4.2 Data and evaluation metric We conduct evaluations on four Senseval-3 lexical sample tasks (English, Catalan, Italian, and Spanish) using the official training / test splits. Data statistics are shown in Figure 9. On the Spanish, Catalan, and Italian data sets, we use part-of-speech information (as features) and unlabeled examples (for semi-supervised learning) provided by the organizer. Since the English data set was not provided with these additional resources, we use an in-house POS tagger trained with the PennTree Bank corpus, and extract 100K unlabeled examples from the Reuters-RCV1 corpus. On each language, the number of unlabeled examples is 5–15 times larger than that of the labeled training examples. We use syntactic relation features only for English data set. As in Section 3, we report micro-averaged F measure. ### 4.3 Baseline methods In addition to the standard single-task supervised configuration as in Section 3, we test the following method as an additional baseline. Output-based method The goal of our multi-task learning configuration is to benefit from having the labeled training examples of a number of words. An alternative to ASO for this purpose is to use directly as features the output values of classifiers trained for disambiguating the other words, which we call ‘output-based method’ (cf. Florian et al. (2003)). We explore several variations similarly to Section 3.1 and report the ceiling performance. ### 4.4 Evaluation results Figure 10 shows F-measure results on the four Senseval-3 data sets using the official training / test splits. Both ASO multi-task learning and semi-supervised learning improve performance over the single-task baseline on all the data sets. The best performance is achieved when we combine multi-task learning and semi-supervised learning by using all the corresponding structure matrices $\Theta_{(j,s)}$ produced by both multi-task and semi-supervised learning, in the final predictors. This combined configuration outperforms the single-task supervised baseline by up to 5.7%. Performance improvements over the supervised baseline are relatively small on English and Spanish. We conjecture that this is because the supervised performance is already close to the highest performance that automatic methods could achieve. On these two languages, our (and previous) systems outperform inter-human agreement, which is unusual but can be regarded as an indication that these tasks are difficult. The performance of the output-based method (baseline) is relatively low. This indicates that output values or proposed labels are not expressive enough to integrate information from other predictors effectively on this task. We conjecture that for this method to be effective, the problems are required to be more closely related to each other as in Florian et al. (2003)’s named entity experiments. A practical advantage of ASO multi-task learning over ASO semi-supervised learning is that shorter computation time is required to produce similar performance. On this English data set, training for multi-task learning and semi-supervised learning takes 15 minutes and 92 minutes, respectively, using a Pentium-4 3.20GHz computer. The computation time mostly depends on the amount of the data on which auxiliary predictors are learned. Since our experiments use unlabeled data 5–15 times larger than labeled training data, semi-supervised learning takes longer, accordingly. \| \| #words \| #train \| avg #sense per word \| avg #train per sense \| \|-------\|--------\|--------\|---------------------\|----------------------\| \| English \| 73 \| 8611 \| 10.7 \| 10.0 \| Senseval-3 data sets \| Language \| #words \| #train \| avg #sense per word \| avg #train per sense \| \|----------\|--------\|--------\|---------------------\|----------------------\| \| English \| 57 \| 7860 \| 6.5 \| 21.3 \| \| Catalan \| 27 \| 4469 \| 3.1 \| 53.2 \| \| Italian \| 45 \| 5145 \| 6.2 \| 18.4 \| \| Spanish \| 46 \| 8430 \| 3.3 \| 55.5 \| \| methods \| English \| Catalan \| Italian \| Spanish \| \|----------------------------------------------\|----------\|---------\|---------\|---------\| \| ASO \| \| \| \| \| \| multi-task learning \| 73.8 (+0.8) \| 89.5 (+1.5) \| 63.2 (+4.9) \| 89.0 (+1.0) \| \| semi-supervised learning \| 73.5 (+0.5) \| 88.6 (+0.6) \| 62.4 (+4.1) \| 88.9 (+0.9) \| \| multi-task-semi-supervised \| 74.1 (+1.1) \| 89.9 (+1.9) \| 64.0 (+5.7) \| 89.5 (+1.5) \| \| baselines \| \| \| \| \| \| output-based \| 73.0 (0.0) \| 88.3 (+0.3) \| 58.0 (-0.3) \| 88.2 (+0.2) \| \| single-task supervised learning \| 73.0 \| 88.0 \| 58.3 \| 88.0 \| \| previous systems \| \| \| \| \| \| SVM with LSA kernel [GG05] \| 73.3 \| 89.0 \| 61.3 \| 88.2 \| \| Senseval-3 (2004) best systems \| 72.9 (goo) \| 85.2 (sgg04) \| 53.1 (sgg04) \| 84.2 (sgg04) \| \| inter-annotator agreement \| 67.3 \| 93.1 \| 89.0 \| 85.3 \| Figure 10: Performance results on the Senseval-3 lexical sample test sets. Numbers in the parentheses are performance gains compared with the single-task supervised baseline (italicized). [G04] Grozea (2004); [SGG04] Strapparava et al. (2004). GGS05 combined various kernels, which includes the LSA kernel that exploits unlabeled data with global context features. Our implementation of the LSA kernel with our classifier (and our other features) also produced performance similar to that of GGS05. While the LSA kernel is closely related to a special case of the semi-supervised application of ASO (see the discussion of PCA in Ando and Zhang (2005a)), our approach here is more general in that we exploit not only unlabeled data and global context features but also the labeled examples of other target words and other types of features. G04 achieved high performance on English using regularized least squares with compensation for skewed class distributions. SGG04 is an early version of GGS05. Our methods rival or exceed these state-of-the-art systems on all the data sets. 5 Conclusion With the goal of achieving higher WSD performance by exploiting all the currently available resources, our focus was the new application of the ASO algorithm in the multi-task learning configuration, which improves performance by learning a number of WSD problems simultaneously instead of training for each individual problem independently. A key finding is that using ASO with appropriate feature / problem partitioning, labeled examples of seemingly unrelated words can be effectively exploited. Combining ASO multi-task learning with ASO semi-supervised learning results in further improvements. The fact that performance improvements were obtained consistently across several languages / sense inventories demonstrates that our approach has broad applicability and hence practical significance. References Rie Kubota Ando and Tong Zhang. 2005a. A framework for learning predictive structures from multiple tasks and unlabeled data. Journal of Machine Learning Research, 6(Nov):1817–1853. An early version was published as IBM Research Report (2004). Rie Kubota Ando and Tong Zhang. 2005b. High performance semi-supervised learning for text chunking. In Proceedings of ACL-2005. Radu Florian and David Yarowsky. 2002. Modeling consensus: Classifier combination for word sense disambiguation. In Proceedings of EMNLP-2002. Radu Florian, Abe Ittycheriah, Hongyan Jing, and Tong Zhang. 2003. Named entity recognition through classifier combination. In Proceedings of CoNLL-2003. Cristian Grozea. 2004. Finding optimal parameter settings for high performance word sense diambiguation. In Proceedings of Senseval-3 Workshop. Upali S. Kohomban and Wee Sun Lee. 2005. Learning semantic classes for word sense disambiguation. In Proceedings of ACL-2005. Yoong Keok Lee and Hwee Tou Ng. 2002. An empirical evaluation of knowledge sources and learning algorithms for word sense disambiguation. In Proceedings of EMNLP-2002. Michael C. McCord. 1990. Slot Grammar: A system for simpler construction of practical natural language grammars. Natural Language and Logic: International Scientific Symposium, Lecture Notes in Computer Science, pages 118–145. Bernhard Schölkopf, Alexander Smola, and Klaus-Rober Müller. 1998. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, 10(5). Carlo Strapparava, Alfio Gliozzo, and Claudio Giuliano. 2004. Pattern abstraction and term similarity for word sense disambiguation at IRST at Senseval-3. In Proceedings of Senseval-3 Workshop. Dekai Wu, Weifeng Su, and Marine Carpuat. 2004. A kernel PCA method for superior word sense disambiguation. In Proceedings of ACL-2004. Tong Zhang. 2004. Solving large scale linear prediction problems using stochastic gradient descent algorithms. In ICML ’04, pages 919–926.
This article was originally published in a journal published by Elsevier, and the attached copy is provided by Elsevier for the author's benefit and for the benefit of the author's institution, for non-commercial research and educational use including without limitation use in instruction at your institution, sending it to specific colleagues that you know, and providing a copy to your institution's administrator. All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution's website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier's permissions site at: http://www.elsevier.com/locate/permissionusematerial Do expert surveys produce consistent estimates of party stances on European integration? Comparing expert surveys in the difficult case of Central and Eastern Europe Stephen Whitefield\textsuperscript{a}, Milada Anna Vachudova\textsuperscript{b}, Marco R. Steenbergen\textsuperscript{b}, Robert Rohrschneider\textsuperscript{c}, Gary Marks\textsuperscript{b,d,}, Matthew P. Loveless\textsuperscript{e}, Liesbet Hooghe\textsuperscript{b,d} \textsuperscript{a} Oxford University, Oxford, UK \textsuperscript{b} University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3449, USA \textsuperscript{c} Indiana University, Bloomington, IN, USA \textsuperscript{d} Vrije Universiteit Amsterdam, Amsterdam, The Netherlands Abstract Expert surveys have been subject to a number of criticisms concerning their ability to produce accurate estimates of party positions. Such criticisms have particular \textit{prima facie} credibility in new post-Communist democracies of Central and Eastern Europe, where party development is regarded as weak. This paper compares data from two expert surveys independently conducted between 2002 and 2004. We find, contrary to expectations, that there is a remarkable overlap in positions assigned to parties. This suggests the usefulness of expert surveys even in the ‘most difficult’ case of post-Communist party systems. It also suggests that parties in these countries have developed effective means of communicating their positions on major issues. © 2006 Elsevier Ltd. All rights reserved. Keywords: Survey; Expert data; Cross-validation; Central and Eastern Europe; European integration 1. Introduction Expert surveys are frequently used by political scientists interested in measuring the positions of political parties on various issues (Castles and Mair, 1984; Huber and Inglehart, 1995; Ray, 1999; Marks and Steenbergen, 2004). Such surveys offer a number of advantages, but they have also been subject to a range of criticisms regarding the validity and reliability of the measures produced (Budge, 2001). To date, expert surveys have been conducted in the relatively stable, historically established and information-rich environments of Western party systems. Concerns about the validity of expert surveys in the arguably unstable and information-poor environments of new post-Communist democracies, therefore, are likely to be even sharper. This article considers expert surveys in a ‘most difficult’ context by cross-validating measures obtained from two surveys of party stances towards European integration. One was undertaken by a team at the University of North Carolina, Chapel Hill (henceforth Chapel Hill);\textsuperscript{1} the other was conducted by Robert \textsuperscript{} Corresponding author. University of North Carolina at Chapel Hill, Center for European Studies, 223 E. Franklin Street, Chapel Hill 27599-3449, USA. Fax: +1 919 962 5375. E-mail address: firstname.lastname@example.org (G. Marks). \textsuperscript{1} Liesbet Hooghe, Gary Marks, Marco Steenbergen, Milada Vachudova, Erica Edwards, Moira Nelson, and David Scott. Rohrschneider and Stephen Whitefield at Indiana and Oxford Universities (henceforth RW). A comparison of these measures reveals the capacity of expert surveys to obtain useful estimates of party positioning even in the difficult circumstances of the fluid party landscape of Central and Eastern European states. 1.1. The two expert surveys The Chapel Hill survey commenced in September 2002 and was completed in May 2003; the RW survey took place between November 2003 and March 2004. Of the 87 parties in the RW survey and 73 in the Chapel Hill study, 57 parties from nine East European countries are common to both studies. For the purpose of this article, we therefore merged the data on these parties from the two data sets. The distribution of parties across countries is as follows: Bulgaria (4), Czech Republic (5), Hungary (5), Latvia (7), Lithuania (6), Poland (8), Romania (5), Slovakia (9), and Slovenia (8). 1.2. Sampling Respondents were expected to know about the full range of electorally significant parties in a given country and to be familiar with the broad thrust of literature on party systems, though not of course with the particular hypotheses of the researchers. Master lists of respondents were compiled independently by each survey team. In both cases, this list included persons who fulfilled one or more of the following criteria: (1) they were employed at an academic institution and were known experts on the party system of the country in question; (2) they worked in a non-partisan think-tank and analyzed European integration in their country of expertise; (3) they had published two or more articles considering the domestic politics of the country’s road to EU membership in English or in the local language; (4) they were recommended by known experts in the field. RW identified 264 experts of whom 111 (or 42 percent) completed the survey questionnaire for the 13 countries surveyed. Chapel Hill identified 291 experts for ten countries of whom 98 (or 34 percent) provided valid responses. The average number of experts for the overlapping countries included in both surveys is 8.7 (RW) and 9.9 (Chapel Hill). Both teams were somewhat surprised and gratified—given the results presented below—to find that only a total of five experts responded to both surveys. Any overlap in the positions ascribed to parties is thus not an effect of overlapping samples of experts. 2. Sources of error Expert surveys are relatively flexible and inexpensive—virtues that help explain their popularity among researchers. But experts, like the rest of us, make mistakes. Depending on the phenomena the experts are asked to evaluate, their knowledge of the subject in question, and the way in which their expertise is tapped, their evaluations may not be accurate. Some degree of inaccuracy can therefore be taken for granted, and this is our point of departure. In evaluating the relative validity of expert data it is useful to make a distinction between random error and systematic error (Marks, 2007). Random error arises when the sources of the error do not replicate across measurement instruments, i.e. experts. One expert may place a party to the left of where it is truly located, while another expert may locate it to the right of the actual location. If one aggregates these errors, no systematic pattern can be found; on the average, the errors cancel each other out. Random error appears to be at the heart of Budge’s (2001) concerns about expert surveys. Budge fears that different experts may in fact be making different judgments. They may be evaluating different segments of the party (activists, leaders, or voters), different facets of the party (e.g. economic versus social ideology), different behaviors (rhetoric versus actual voting behavior in legislatures), or different time points. To the extent that experts differ in the judgments they render, random error in the placement of the parties would be likely. In statistical terms, a measure of this error would be given by the variance across experts. Systematic error arises when errors are replicated across experts. For example, there may be a systematic tendency to place parties to the left (or to the right) of their actual positions. This kind of error can be described statistically as ‘bias’, which measures the discrepancy between the average expert judgment and the true score. The problem of bias is connected to the concerns raised by McDonald and his collaborators in this issue. They worry that experts tend to rely on party family as a shortcut that allows them to place parties. To the extent that many, if not most, experts do this, one would expect to see a replication of errors that, in turn, would produce bias. For example, if experts were to think of party X as a member of the socialist family, they might attribute leftist orientations to that party on all sorts of issues, even if in reality the party may be moderate or to the right on some issues. We can combine both error conceptions in the mean squared error (MSE—compare Groves, 1989). Let $\theta$ denote the placement of a party by experts, then MSE = $E[(\hat{\theta} - \theta)^2]$, where $\theta$ is the true position of the party. As any statistical textbook will show, the mean squared error is equal to the sum of the unsystematic error—i.e. the variance—and systematic error—i.e. the squared bias. As a general rule, the variance will decline as the number of experts increases. (Or, in the terms of classical test theory, the reliability of a composite of expert judgments increases with the number of experts.) The more experts there are, the greater the chance that random errors will cancel each other out. The squared bias, by contrast, is immune to the number of experts. If experts are making the same systematic errors, then adding to their number will not reduce the error of their estimates. These logical properties are illustrated in Fig. 1. Whereas the mean squared error recedes to zero where there is no bias, it will always remain positive where there is bias (Marks, 2007). There is good reason to believe that McDonald et al.’s concern about biased expert evaluations is less relevant for Central and Eastern Europe. As we noted earlier, party families are less cohesive and less useful indicators in Central and Eastern Europe than in Western Europe. Since the party family heuristic seems to be the most plausible source of bias, we have little reason to believe that expert respondents will misjudge the same political parties in the same direction. However, random measurement error is as likely to be present in Eastern as in Western Europe. Indeed, there is good reason to believe that expert judgment is rather more difficult in Central and Eastern Europe than in Western Europe. Thus, Budge’s concerns about expert surveys remain valid (Budge, 2001). We restate these concerns under three rubrics; (1) ambiguity in the assessment criteria, (2) too many small or new parties, and (3) a lack of policy positions in general. 2.1. Ambiguous assessment criteria Error will creep into expert evaluations to the extent that the survey instrument is ambiguous (Steenbergen and Marks, in press). With this in mind, both the Chapel Hill and RW questionnaires cue respondents to evaluate the formal position of the party (RW) or the positioning of the party leadership (Chapel Hill), and both questionnaires specify the year of evaluation. Neither survey explicitly asks experts to disregard the actual behavior of parties, but they do state that they are interested in the policy positions taken by the party in question. Given the rapidity of political change in Central and East European countries, the nine-month lag between the surveys could produce divergence in estimates, even if each survey were accurate. Central and East European parties are less institutionalized and have tended to shift positions with greater rapidity than their western counterparts. Presumably, these characteristics would amplify ambiguities in the surveys. One might also argue that since party elites in Central and Eastern Europe may be less effective in communicating the stances of their parties, the best way to measure these would be to analyze their formal party programs. 2.2. Too many small or new parties Even if the questions posed to experts were unambiguous, it is still possible that experts would not have sufficient knowledge to provide accurate answers. Political parties in the post-Communist party systems of Central and Eastern Europe tend to be numerous, small, and new. While some level of continuity exists among most Communist-successor parties and, in some states, pre-war ‘revival’ parties (some of which nominally operated during the Communist period itself), the majority of parties are post-Communist creations (Lewis, 2001). Moreover, many post-Communist party systems are characterized by comparatively high levels of organizational instability, party failure, and new party formation, even more than a decade after democratization (Kreuzer and Pettai, 2003). The number of effective parties is also comparatively high, with many small parties that, additionally, compete as part of broader electoral blocs. The median vote of the 57 parties in the combined data set is 9.6 percent, while the average vote is 13.9 percent. So our data set contains mostly small parties with a tail of larger parties that dominate their respective party systems. Hence, the informational challenges facing experts are severe even when the range of parties is restricted—as in the two surveys analyzed here—to those that are represented in national parliaments. 2.3. Lack of policy positions in general A number of scholars have noted a systematic de-emphasis of clear issue positions among political parties in the post-Communist context. Political parties are said to have loose ideological moorings on account of the absence or weakness of political cleavages (Lawson et al., 1999). Unlike parties in Western Europe, those in the East are not easily placed in party families characterized by durable constituencies and programmatic commitments. Rather, many Eastern European parties are believed to appeal to voters on non-programmatic grounds, by virtue of the charisma of their leaders or by stressing their ability to deliver effective governance, economic growth, or patronage (Evans and Whitefield, 1993; Kitschelt et al., 1999). As a consequence, respondents may find it difficult to identify how parties position themselves on EU policy. On the other hand, there are also scholars who argue that political cleavages in post-Communist societies are actually well established, and that party competition in the East has a surprisingly high degree of structure (Whitefield, 2002; Jasiewicz, 2003). If this is the case, parties have strong incentives to adopt unambiguous policy positions on major issues and to convey these positions to voters—and to experts. 2.4. Lack of policy positions on European integration A distinct and important question for the value of the RW and Chapel Hill surveys of the positions of Central and East European political parties is whether these parties have developed identifiable stances on different aspects of European integration (Taggart and Szcerbiak, 2004). Concern about this might have been valid for the early and mid-1990s. Until joining the EU was fully on the table, many political parties had only superficial positions on European integration. By 1997, all countries in our dataset had applied for EU membership, and by 2002, the process of joining the EU had come so far along that virtually all political parties were induced to spell out positions on the topic (Vachudova, 2005). Moreover, recent work shows a strong connection between European integration and the main issue dimensions in domestic politics (Rohrschneider and Whitefield, 2004; Marks et al., 2005; Vachudova, 2005), thus reducing the informational challenge for experts. Future surveys will undoubtedly show a further ‘filling in’ of party positions as EU membership forces political parties to take refined stands across a range of EU policies. 2.5. Empirical consequences If the expectation is that placing political parties in Central and Eastern Europe is a difficult task, then random error in expert judgments should run rampant. If so, we would expect the correlations between measures of the same phenomena across the instruments to be attenuated. We now evaluate the extent to which the two expert surveys do, or do not, correspond. We begin with the stances of political parties on European integration and turn next to their stances on the left–right ideological dimension. 3. Party positioning on European integration In this section we consider questions designed to capture variation among political parties on various aspects of European integration. While not identical, the survey questions appear to measure the same, or a similar, thing. 3.1. Political integration We begin the discussion by comparing measures of party positioning on integration as a whole. The broadest measure of party positioning on European integration in the Chapel Hill survey is the following question: ‘How would you describe the general position to European integration that the party’s leadership has taken over the course of 2002?’ Experts placed the position of parties on a scale from 1 (strongly opposed to European integration) to 7 (strongly in favor of European integration).’ The RW survey asked experts a similar question: ‘How about the EU? Regardless of the specific form that integration may take, where do parties stand on creating a politically unified Europe? (1–7 Strongly Oppose/Strongly Support)?’ The Chapel Hill question is a general one that encompasses political and market integration; the RW survey focuses directly on political integration. Despite this difference in conceptual range, both questions conceive European integration in broad terms. To what extent do experts place parties at similar locations? Fig. 2 reveals an extraordinary degree of convergence given that we have similar, but not identical, indicators and that there is a nine-month time gap between the two surveys. The correlation coefficient is 0.96—i.e. over 91 percent of the variance of party positions is shared (see Appendix A). This is as strong a relationship as one is likely to find in the empirical world. At the country level, the two surveys produce nearly identical rankings of parties. Note also that the surveys converge in their estimates of the absolute positioning of parties. The regression line runs almost perfectly through the origin of the two-dimensional space in Fig. 2.\footnote{We did not force the regression line through the origin.} The country-specific patterns in Fig. 2 neatly duplicate the overall relationship. Despite the fact that the number of cases in each party system is small for the purposes of statistical analyses, the two surveys produce consistently similar estimates for parties’ policy stances on Europe. 3.2. Market integration Both surveys measure party positions on market integration. The Chapel Hill survey asks: ‘Some parties want to strengthen EU powers to eliminate market barriers (i.e. free movement of goods, services, capital, and labor). Other parties oppose strengthening EU powers in this area. Where does the leadership of the following parties stand?’ Experts placed the position of parties on a scale from 1 (strongly opposes expanding EU powers on the internal market) to 7 (strongly favors expanding EU powers on the internal market).’ The RW survey asks experts to consider party positions on market integration in the following way: ‘Where do the parties in [country x] stand on creating a Europe-wide, integrated market for the European Union? (1–7 Strongly Oppose/Strongly Support)?’ Again, the conceptual reach of both questions differs slightly. The Chapel Hill survey focuses on the authority of the European Union with respect to market integration; the RW survey focuses on economic integration per se. Still, both surveys overlap to a substantial degree (Fig. 3). The correlation coefficient is a strong 0.87, which again indicates that a substantial amount of variance in party stances is shared between the estimates (75.3 percent). It is also worth noting that the country patterns are consistent across the surveys. 4. Party positioning on left/right ideological placement The surveys of party positioning also provide information about how political parties place themselves on the basic ideological dimensions that structure political competition in Central and Eastern European countries. To what extent do the two surveys reveal similar pictures? Both surveys include a question asking experts to place parties on a general left/right spectrum. The RW survey asks: ‘In [country x], parties may be located to the left or the right of the political spectrum. In general terms, please locate each party on the ideological spectrum in [country x], with 1 standing for left wing, and 7 standing for right wing.’ The Chapel Hill question is worded: ‘First, we would like you to classify the parties in terms of their broad ideology. On the scale below, 0 indicates that a party is at the extreme left of the ideological spectrum, 10 indicates that it is at the extreme right, and 5 means that it is at the center. For each party, please circle the ideological position that best describes a party’s overall ideology.’ In the post-Communist context, the terms ‘left’ and ‘right’ may have nationally distinctive, unstable or unclear meanings (Whitefield and Evans, 1998), given the complexity of the political cleavages with which the terms might be associated (Whitefield, 2002). The general left/right dimension might also be considered to comprise two distinct components: an economic left/right dimension oriented to state intervention versus market outcomes, and a more diffuse, non-economic dimension summarizing conflict on communal, religious, and life-style issues (Kitschelt et al., 2002; Hooghe et al., 2002). Given these potential ambiguities how much overlap, then, do we find in expert placement of parties on the left/right dimension? As Fig. 4 indicates, the two surveys produce similar information both for all political parties and for parties within individual countries. The overall association for the 57 common parties is 0.87, which indicates that around three-quarters of the variance is shared. We find a similarly strong association when we compare where parties stand regarding their more specific policy position on the role of government and on distributional issues. This contradicts the view that many post-Communist parties lack clear stances on economic policy issues (Grzymala-Busse and Innes, 2003) or operate in countries with weak economic cleavages (Lawson et al., 1999). Here, the RW survey asks: Beginning with [issue 1], could you now situate parties in [country x]? Please use a seven-point scale to score the position of a party. A score of 7 indicates the most strongly liberal position and a score of 1 indicates the least liberal position on any particular issue. If a party has no stance on a given issue, please give it a score of 99. (1) ‘A. Economy: redistributional issues (for example, tax levels, welfare state spending)’, with 7 indicating an ‘anti-tax, anti-welfare’ policy, and 1 indicating ‘pro-welfare distribution’. (2) ‘B. Economy: State-run versus market economy’, with 7 indicating ‘pro-market’ and 1 ‘anti-market’. We averaged the two indicators and rescaled it to the original metric (1–7). The Chapel Hill survey asks: ‘Political scientists often classify parties in terms of their ideological stance on economic issues. Parties to the right emphasize a reduced economic role for government. They want privatization, lower taxes, less regulation, reduced government spending, and a leaner welfare state. Parties to the left want government to play an active role in the economy. Using these criteria, indicate where parties are located in terms of their economic ideology.’ Experts were asked to place the parties on a scale ranging from 0 (extreme left) through 5 (center) to 10 (extreme right). Again, the results in Fig. 5 demonstrate a strong association between the two surveys’ economic position measures (0.86). As before, around three-quarters of the variance in the data is shared. Country patterns are convergent. 5. Modeling differences While one might be content to note the high degree of overlap in party stances, we would like to take our investigation one step further in order to explore the structure of error in the two surveys. To do this, we calculate the absolute difference between Chapel Hill and RW scores for individual political parties for each of the three questions discussed above. The greater the random error in measuring the stance of a party, the greater, on average, should be the difference between the two measures. To what extent, if at all, are these differences patterned by type of party, by issue, or by country? There are several grounds for suspecting that experts’ ability to pinpoint the stances of political parties will vary predictably, and we conduct an OLS analysis to see if they are associated with differences across the surveys. We suspect, for example, that experts can more easily evaluate parties that are large (i.e. win a greater share of the vote), that participate in government, that are centrist rather than extremist in left/right terms, that are united on the issue of concern, and that regard the issue as salient. There are also grounds for expecting that parties in smaller countries are more difficult to track. Perhaps, in addition, the sheer number of political parties in a country makes their positioning less easy to pinpoint. To evaluate these priors, our model includes the following variables at the party level: the standard deviation of expert scores in each survey; party vote share; incumbency; left/right placement of parties; and left/right extremism. We include two variables that tap the character of the issue: issue salience for the party; and intra-party dissent. At the country level we include population and the number of parties evaluated in a country. Appendix C provides information about these variables. Table 1 presents the results of this analysis for political integration, internal market integration, and left–right placement of parties. We use the original 7-point metric for political and economic integration indicators, and we standardize scores for the ideological placement indicator because the metric of the ideological placement indicators varies. The standard deviations of expert judgments are robustly associated with the gap between survey measures. However, the estimate for the effect of disagreement among Chapel Hill experts for political integration is negative, not positive as we expect.$^3$ With respect to other possible sources of error, our findings are spotty. We find that differences among our measures of political integration are significantly greater for smaller parties, and for parties that do not regard European integration as a salient issue. Smaller countries generate greater differences in expert evaluations of the left/right positioning of parties. But these associations are not robust across the three issues dealt with here, and we find also that they are not particularly robust in alternative model specifications. Our attempt to pin down the sources of difference between the surveys leaves us with the impression that we are dealing mainly with white noise. This is consistent with the high correlations that we have found across the measures. Differences across the measures are relatively small, and as Appendix B reveals, in two of three --- $^3$ These estimates for expert disagreement on left/right are highly correlated across the surveys ($r = 0.73$). We therefore enter only that for the RW survey in our equation. Table 1 Explaining absolute differences between the Chapel Hill and the RW expert data \| \| Political integration \| Internal market \| Left/right ideology \| \|--------------------------------\|-----------------------\|-----------------\|---------------------\| \| Constant \| 0.713 (0.589) \| −0.399 (0.740) \| 0.167 (0.209) \| \| Party characteristics \| \| \| \| \| CH expert disagreement (SD) \| −0.186* (0.106) \| 0.431** (0.168) \| – \| \| RW expert disagreement (SD) \| 0.335* (0.168) \| 0.052 (0.203) \| 0.306 (0.076) \| \| Vote share \| 0.012 (0.005) \| −0.012 (0.007) \| −0.001 (0.004) \| \| Government party \| −0.113 (0.113) \| 0.116 (0.168) \| −0.068 (0.085) \| \| Left/right position \| 0.134 (0.056) \| 0.053 (0.083) \| 0.060 (0.041) \| \| Left/right extremism \| 0.080 (0.106) \| 0.146 (0.171) \| −0.043 (0.090) \| \| Issue characteristics** \| \| \| \| \| Salience of issue \| −0.129* (0.064) \| −0.041 (0.095) \| – \| \| Dissent on issue \| −0.070 (0.044) \| 0.088 (0.066) \| – \| \| Country characteristics \| \| \| \| \| Population size \| −0.060 (0.056) \| 0.072 (0.088) \| −0.088 (0.039) \| \| Number of parties \| 0.019 (0.020) \| 0.051 (0.034) \| 0.023 (0.016) \| \| $R^2$ \| 0.374 \| 0.246 \| 0.447 \| \| Adjusted $R^2$ \| 0.238 \| 0.082 \| 0.368 \| B-coefficients with standard errors in parentheses. $N = 57$. $p < 0.05$, $p < 0.10$. cases, these differences bear little association with each other. 6. Conclusions Expert surveys have been subject to a number of criticisms about their capacity to produce accurate estimates of party positions. Such criticisms have even greater prima facie credibility in the new democracies of Central and Eastern Europe, where party development has been regarded as weak. The evidence presented in this article, however, demonstrates that two distinct groups of experts, at somewhat different points in time, converge to a remarkable degree in their assessments of party positions. This suggests that expert surveys are useful even in the challenging case of post-Communist party systems. At the same time, our analysis sheds light on some substantive properties of post-Communist politics. Obtaining valid and reliable measures speaks to our capacity to know the world, and therefore to the object of our knowledge. Against those who argue that post-Communist party systems are too fluid and ill-defined to be knowable, the fact that we obtain valid and reliable measures of post-Communist party positioning says something important about the capacity of parties to make themselves understood. CEE parties appear to compete in an intelligible competitive space, which facilitates information signaling and communication between parties and experts (and perhaps voters too). Acknowledgements L.H., G.M., M.S., and M.V. would like to thank Erica Edwards and Moira Nelson for compiling the Chapel Hill 2002 data set and Hans-Dieter Klingemann for supplying additional data. The project was funded by the UNC Center for European Studies and the North Carolina European Union Center. Data for the Rohrschneider/Whitefield expert survey were obtained with a grant from the Nuffield Foundation project, ‘Political Mobilization and Elite Framing in Generating Support for Supra-National Institutions and European Union Enlargement in Post-Communist Eastern Europe’. Grant Number: SGS/00827/G. S.W. and R.R. would like to thank the Foundation for its generous support and Matt Loveless for his research assistance, and R.R. would like to acknowledge the warm hospitality of the Transatlantic Center of the German Marshall Fund of the United States in Brussels. S.W. would like to acknowledge the Leverhulme Trust for its great generosity in supporting research leave to work on this project. Appendix A. Correlation matrix of six measures \| \| Political integration CH \| Internal market CH \| Left/right CH \| Political integration RW \| Internal market RW \| \|----------------------\|--------------------------\|-------------------\|--------------\|--------------------------\|-------------------\| \| Internal market CH \| 1.000 \| \| \| \| \| \| Left/right CH \| –0.047 \| 0.030 \| \| \| \| \| Political integration RW \| 0.956 \| 0.889 \| –0.112 \| \| \| \| Internal market RW \| 0.901 \| 0.868 \| 0.067 \| 0.907 \| \| \| Left/right RW \| 0.007 \| 0.097 \| 0.865 \| –0.040 \| 0.135 \| Appendix B. Correlation matrix of absolute differences (dependent variables in OLS regression) \| Difference on political integration \| Difference on internal market \| Difference on ideology \| \|-------------------------------------\|-------------------------------\|-------------------------\| \| Difference on political integration \| 1.000 \| \| \| Difference on internal market \| –0.019 \| 1.000 \| \| Difference on ideology \| 0.297* \| 0.099 \| 1.000 \| Note: Differences are absolute differences between CH and RW scores. N = 57. P < 0.05. Appendix C. Description of independent variables - CH expert disagreement (SD)** - Standard deviation among experts on party’s position on (a) political integration, (b) internal market, and (c) left/right positioning across the 57 parties. Mean = 0.85 (political integration); 1.01 (internal market); 1.20 (left/right). Source: Chapel Hill 2002 expert data - RW expert disagreement (SD) - Standard deviation among experts on party’s position on (a) political integration, (b) internal market, and (c) left/right positioning across the 57 parties. Mean = 0.96 (political integration); 1.07 (internal market); 0.88 (left/right) Source: Rohrschneider/Whitefield data - Vote share - Percentage of vote in 2003 or the most recent national election prior to 2003. Range is from 2.9 to 43.1 percent. Mean = 13.9. Source: http://www.electionworld.org - Government party - A three-category variable that takes on the value of 0 if the party was not in government during 2002, 0.5 if it was in government for part of the time, and 1 if it was in government for the duration of 2002. Mean = 0.40. Source: http://www.electionworld.org - Left/right position - Mean of left/right scores of the Chapel Hill and Rohrschneider/Whitefield dataset for each party. Standardized values - Left/right extremism - Absolute distance of a political party from the median left/right position for all parties in the expert dataset (calculated from left/right position above) - Salience - Salience for a particular party on (a) political integration and (b) internal market on a seven-point scale, whereby 1 stands for ‘not important at all’ and 7 for ‘very important’ in ‘how a party appeals to the public.’ Source: Rohrschneider/Whitefield data - Dissent - Degree of internal dissent in a party on European integration in 2002, ranging from 1 (party is completely united) to 10 (party is extremely divided). Source: Chapel Hill 2002 expert data - Size of country - Population in millions (natural log function). Source: CIA world facts - Number of parties - The number of parties in a country that obtained 3 percent or more of the vote or at least one seat in the national parliament during the latest parliamentary elections. Values for countries range from 5 in the Czech republic to 14 in Lithuania. Source: http://www.electionworld.org References Budge, I., 2001. Validating party policy placements. British Journal of Political Science 31, 210–223. Castle, E.G., Mair, P., 1984. Left-right political scales: some expert judgments. European Journal of Political Research 12, 73–88. Evans, G., Whitefield, S., 1993. Identifying the bases of party competition in Eastern Europe. British Journal of Political Science 23, 521–548. Groves, R.M., 1989. Survey Errors and Survey Costs. Wiley, New York. Grzymala-Busse, A., Innes, A., 2003. Great Expectations: the EU and domestic political competition in East Central Europe. East European Politics and Societies 17, 64–73. Hooghe, L., Marks, G., Wilson, C., 2002. Does left/right structure party positions on European integration? Comparative Political Studies 35, 965–989. Jasiewicz, K., 2003. Pocketbook or Rosary? Economic and Identity Voting in 2000–2001 Elections in Poland. Studies in Public Policy Number 379. Centre for the Study of Public Policy, University of Strathclyde, Glasgow. Kitschelt, H., et al., 1999. Post-Communist Party Systems: Competition, Representation, and Inter-Party Cooperation. Cambridge University Press, Cambridge. Kreuzer, M., Petrai, V., 2003. Patterns of political instability: affiliation patterns of politicians and voters in post-communist Estonia, Latvia and Lithuania. Studies in Comparative International Development 38, 76–98. Lawson, K., Rommele, A., Karasimeonov, G. (Eds.), 1999. Cleavages, Parties and Voters: Studies from Bulgaria, the Czech Republic, Poland and Romania. Praeger, London. Lewis, P. (Ed.), 2001. Political Parties in Post-Communist Eastern Europe. Routledge, New York. Marks, G., Steenbergen, M. (Eds.), 2004. European Integration and Political Conflict. Cambridge University Press, Cambridge. Marks, G., Liesbet, H., Nelson, M., Edwards, E., 2005. Party ideology and European integration: an east-west comparison. In: Rohrschneider, R., Whitefield, S. (Eds.), Public Opinion, Party Competition and European Union Integration in Post-Communist Eastern Europe. Palgrave, New York. Marks, G., Hooghe, L., Steenbergen, M.R., Bakker, R., 2007. Cross-validation data on party positioning on European integration. Electoral Studies 26, 1–14. Ray, L., 1999. Measuring party orientation towards European integration: results from an expert survey. European Journal of Political Research 36, 283–306. Rohrschneider, R., Whitefield, S., 2004. Responsible party government? Explaining party stances on European integration in post-communist Eastern Europe. Under review. Steenbergen, M.R., Marks, G. Evaluating expert judgments. European Journal of Political Research, in press. Taggart, P., Szczerskiak, A., 2004. Contemporary euroscepticism in the systems of the European Union candidate states of Central and Eastern Europe. European Journal of Political Research 43, 1–27. Vachudova, M.A., 2005. Europe Undivided: Democracy, Leverage, and Integration After Communism. Oxford University Press, Oxford. Whitefield, S., 2002. Political cleavages and post-communist politics. Annual Review of Political Science 5, 181–200. Whitefield, S., Evans, G., 1998. The evolution of left and right in post-Soviet Russia. Europe-Asia Studies 45, 1023–1042.
Adams County Public Hospital District No. 3 Regular Meeting of the Board of Commissioners Thursday May 26, 2016 CPR Training: 4:00 p.m. Light Supper: 5:30 p.m. Board Meeting: 6:00 p.m. CONFERENCE ROOM Regular Board Meeting 1. Call to Order 2. Comments from Audience 3. Auxiliary Report 4. Medical Staff Report 5. Department Report/Guest Presentation: Othello School District TISSAM Students 6. Conflict of Interest Declarations 7. Consent Agenda A. April 28, 2016 Regular Board; May 6, 2016 Special Board; May 19 Board Policy; May 24, 2016 Finance; May 25, 2016 Personnel Committee Meetings B. Claims and payroll C. Bad debts and charity care. 8. Patient Satisfaction Survey 9. Financial and Statistical Report 10. Old Business: A. Ambulance Committee 11. New Business: A. Malpractice Insurance 12. Administrative Report Nursing Report Patient Story CEO Report 13. Articles 14. Process Feedback 15. Executive Session: The Board will meet in closed Executive Session per: 1) RCW 70.44.062(2) whereby all meetings, proceedings, and deliberations of a quality improvement committee established under RCW 4.24.250, 43.70.510, or 70.41.200 and all meetings, proceedings, and deliberations of the board of commissioners, its staff or agents, to review the report or the activities of a quality improvement committee established under RCW 4.24.250, 43.70.510, or 70.41.200 may, at the discretion of the quality improvement committee or the board of commissioners, be confidential and may be conducted in executive session. Any review conducted by the board of commissioners or quality improvement committee, or their staffs or agents, shall be subject to the same protections, limitations, and exemptions that apply to quality improvement activities under RCW 4.24.240, 4.24.250, 43.70.510, and 70.41.200. However, any final action of the board of commissioners on the report of the quality improvement committee shall be done in public session; and 2) per RCW 42.30.110(g) to evaluate the qualifications of an applicant for public employment or to review the performance of a public employee. However, subject to RCW 42.30.140(4), discussion by a governing body of salaries, wages, and other conditions of employment to be generally applied within the agency shall occur in a meeting open to the public, and when a governing body elects to take final action hiring, setting the salary of an individual employee or class of employees, or discharging an employee, that action shall be taken in a meeting open to the public. The Executive Session is expected to last 20 minutes. 16. Adjournment Adams County Public Hospital District No. 3 Board of Commissioners Meeting Minutes May 26, 2016 Present: Shirley McCullough, Lynda Bowers, Bob Carlson, Larry McCourtie, and Terry Thompson Others Present: Connie Agenbroad, Mark Bunch, Tina Bernsen, Rhonda Golladay, Mindy Gonzalez, and Luann Morgan (Othello Outlook) Prior to the meeting being called to order all Commissioners attended a 1 ½ hr CPR class and passed. Way to go! Call to Order: President McCullough called the meeting to order at 6:00 p.m. President McCullough announced that the meeting is being recorded. President McCullough asked if everyone could hear her. All present acknowledged they could hear her. Comments from Audience: None Auxiliary Report: The Auxiliary’s Golf Outing is Saturday June 25, 2016. Total donations so far are $6,950.00. Two teams have signed up to golf. It appears that we are on target compared to last year. Medical Staff Report: None Department Report: The “Take it Seriously: Sex, Abstinence, and Media (TISSAM)” representatives gave an excellent presentation on teen pregnancy prevention. April Williamson, OCH RN, also helps with the group. Conflict of Interest Declarations: None Approval Consent Agenda: Bob Carlson made a motion to approve the Consent Agenda comprised of the; 1) April 28, 2016 Regular Board; May 6, 2016 Special Board; May 19, 2016 Board Policy; May 24, 2016 Finance; and May 25, 2016 Personnel Committee minutes. 2) Claims and payroll in the amounts of $1,299,260.65; 3) Bad Debts and Charity care in the amounts of $131,511.61 and $73,785.93 respectively. Larry McCourtie seconded the motion. The motion passed unanimously. Patient Satisfaction: Mindy Gonzalez presented patient satisfaction surveys for ER services and Inpatient services. Financial Report: For the month ending, April, 2016, OCH incurred an operating loss (including approximately $50,573 in hospital levy; $39,765 in EMS levy) in the amount of $62,987 with investment gain of $33,137 for a total net loss of $29,850; year-to-date operating loss (including approximately $202,292 in hospital levy; $159,060 in EMS levy;) was $446,571 with investment gain of $123,350 for a year-to-date loss of $323,221. Old Business: Ambulance Selection Committee: A Special Board will be held Wednesday June 8, 2016 at 5:30 p.m. Committee members will consist of all Commissioners, EMT’s Corey Shearer, Jim Lomax, Tiffany Cutforth, Shannon Cecil, and Hallie Miller. Rhonda Golladay and Connie Agenbroad will also be on the committee. New Business: Approval of Malpractice Renewal Resolution 16-01: Larry McCourtie made a motion to approve Resolution 16-01, expending $134,322.00 as follows: renewing $44,515.00 for Professional/General Liability and EBL coverage; $4,482.00 for Directors and Officers coverage; $11,704.00 Employment Practices Liability; $30,971.00 Excess Liability Coverage; $27,144.00 for broker services through Parker Smith & Feek; and $15,506.00 for Cyber and Regulatory Liability. Bob Carlson seconded the motion. The motion passed unanimously. Replacement of Dietary Oven: Lynda Bowers made a motion to approve replacing the dietary oven not to exceed $8,500.00. Terry Thompson seconded the motion. The motion passed unanimously. Nursing Report: Rhonda reported that in April there were two (2) reported restrictions with no diversions. Patient Story: No particular patient story to reports. The hospital has been busy delivering babies. Lynda Bowers commented on her personnel experience with the ambulance and hospital staff. She commented on what excellent care she received and how much she appreciated getting a phone the following day for follow up on how she was feeling. CEO Report: On May 3, 2016, State Representative Joe Schmick met with Ermila Saucedo, Tina Bernsen, Rhonda Golladay, Mindy Gonzalez, and Connie Agenbroad to discuss telemedicine. Representative Schmick has been visiting area hospitals in his district to hear about ideas using telemedicine in rural areas. Connie reported that there might be a need for per diem CRNA coverage during the summer as one CRNA will be out for medical leave for about four to six weeks. WSHA CEO rounding video was viewed. Each Commissioner received a copy of the 2014-2015 Strategic Vision plan. Connie asked that they take some time to review this plan, as she would like to discuss at the June meeting. Save the date: WSHA annual Meeting October 12th, 13th and 14th, 2016. A thank you card was received from CBHA to the Radiology department. Thanking the department for outstanding service. Articles: The Board reviewed articles regarding, “Yikes! Live white python falls from ceiling in Tacoma General Hospital” from q13fox.com, May 9, 2016; “Paramedic program keeping hospital visits at minimum” from dailysunnews.com, May 12, 2016; “Legislation offers more support to sexual assault victims” from nwcn.com, May 18, 2016; “Feds: Insurers can’t discriminate on LGBT Obamacare coverage” from kitsapsum.com, May 13, 2016; “One step closer: Samaritan approves structure of potential Confluence affiliation” from columbiabasinherald.com, May 12, 2016; “Lawmakers to address medical bill sticker shock” from krem.com, May 11, 2016; “More affordable health insurance option urged for Washington”, from king5.com, May 2, 2016 Process Feedback: None Executive Session: President McCullough stated that the Board would meet in closed Executive Session for approximately 15 minutes per: 1) RCW 70.44.062(2) to review and discuss quality improvement activities and 2) RCW 42.30.110(g) to review and discuss the performance of a public employee. Any final action of the board of commissioners would be taken in open session. A complete legal citation of the above was available in the agenda to this meeting, upon request, and as cited below: 1) RCW 70.44.062(2) whereby all meetings, proceedings, and deliberations of a quality improvement committee established under RCW 4.24.250, 43.70.510, or 70.41.200 and all meetings, proceedings, and deliberations of the board of commissioners, its staff or agents, to review the report or the activities of a quality improvement committee established under RCW 4.24.250, 43.70.510, or 70.41.200 may, at the discretion of the quality improvement committee or the board of commissioners, be confidential and may be conducted in executive session. Any review conducted by the board of commissioners or quality improvement committee, or their staffs or agents, shall be subject to the same protections, limitations, and exemptions that apply to quality improvement activities under RCW 4.24.240, 4.24.250, 43.70.510, and 70.41.200. However, any final action of the board of commissioners on the report of the quality improvement committee shall be done in public session, and 2) RCW 42.30.110(g) to evaluate the qualifications of an applicant for public employment or to review the performance of a public employee. However, subject to RCW 42.30.140(4), discussion by a governing body of salaries, wages, and other conditions of employment to be generally applied within the agency shall occur in a meeting open to the public, and when a governing body elects to take final action hiring, setting the salary of an individual employee or class of employees, or discharging an employee, that action shall be taken in a meeting open to the public. The Executive Session is expected to last 20 minutes. The Executive Session convened at 7:15 p.m. Resumption of Open Session: The open session resumed at 7:35 p.m. Adjournment: The meeting adjourned at 7:35 p.m. Bob Carlson, Secretary
Flow-induced vibrations of high mass ratio flexible filaments freely hanging in a flow Lionel Schouveiller, Christophe Eloy, and Patrice Le Gal Institut de Recherche sur les Phénomènes Hors Equilibre, UMR 6594 CNRS—Universités d’Aix-Marseille, Technopôle de Château-Gombert 49, rue Frédéric Joliot Curie, Boite Postale 146, F-13384 Marseille Cedex 13, France (Received 24 June 2004; accepted 2 February 2005; published online 31 March 2005) The behavior of high mass ratio flexible filaments freely hanging in steady horizontal uniform flows is experimentally and theoretically investigated. When the flow velocity is small, static equilibrium states, where the filaments are inclined to the flow, are observed. Then, above a critical value of the wind velocity, the filaments exhibit periodic oscillations in the vertical plane. The problem is theoretically addressed considering the beam theory equations for the filament dynamics where the action of the flowing fluid is modeled using semiempirical expressions. These equations are first solved for the stationary equilibrium states. Then, the stability of these steady solutions relative to small perturbations is analyzed. A good agreement between experimental and theoretical results is found. © 2005 American Institute of Physics. [DOI: 10.1063/1.1878292] I. INTRODUCTION Interactions of flows with flexible elongated structures are encountered in many industrial fields such as ocean engineering (towing cables, mooring lines, risers) or civil engineering (cables of suspension bridges or of hanging roofs). Moreover, because they can induce structure vibrations that are a potential cause of damage, they have received considerable attention. Flow-induced vibrations can arise from different mechanisms which were classified by Blevins. For steady flows, efforts have mainly concerned the vortex-induced vibrations (referred to, hereafter, as VIV) and comprehensive reviews were published by Bearman and more recently by Williamson and Govardhan. Most of the studies devoted to flow-induced vibrations of elongated bodies are motivated by ocean engineering applications for which the mass ratio $\mu$ is usually less than 15. This parameter $\mu$ is defined as the ratio of the “oscillating mass,” including the mass of the structure and the added fluid mass, divided by the displaced fluid mass. The mass ratio $\mu$ is the primary parameter of VIV (e.g., Ref. 2): the lower $\mu$, the higher the susceptibility of structures to VIV. On the contrary, in the present study, we consider flexible filaments freely hanging in wind. This configuration leads to a high value of $\mu$ of the order of 1000. Note that similar investigations have recently been reported (Facchinetti et al.) for a mass ratio $\mu$ slightly above 2. Comparison with our results will emphasize the strong influence of $\mu$ on the filament dynamics. Finally, we should also mention the extensive work of Paidoussis and co-workers, whose references can be found in the recent paper of Paidoussis et al., devoted to the stability of a flexible cylinder in an axial flow, that is, with an initial incidence angle equal to zero. The theoretical analysis reported in the present paper generalizes these studies to the case of a flexible cylinder inclined to the flow, i.e., for arbitrary angles of incidence. In the following section, details on the experimental setup and on our experimental observations are presented. In Sec. III, we derive the governing equations describing the filament equilibrium and dynamics. A theoretical linear stability analysis is also performed whose results are compared with experiments. Finally, brief complementary discussions are given in Sec. IV together with the conclusion. II. EXPERIMENTS A. Experimental setup Our experiments are performed in a low turbulence wind tunnel. Its horizontal working section is 1 m long and has a square cross section of $0.5 \times 0.5$ m$^2$. The free stream velocity $U$ is measured with a conventional Pitot tube and can be continuously varied up to 25 m s$^{-1}$. The quality of the tunnel has been quantified, prior to this study, with a constant temperature hot wire anemometer. At 25 m s$^{-1}$ the free stream uniformity was better than 0.5% over 80% of the tunnel width and the turbulence level measured on the tunnel axis was close to 0.1%. The filaments are made of silicone and have a circular and constant cross section $A = \pi d^2/4$ and a variable length $l$. To be maintained in the wind, each filament is inserted in a steel tube having the same inner diameter $d$. This support is fixed outside of the test section and crosses perpendicularly the wind tunnel top wall. It is bent at right angle in such a way that its final part is parallel to the free stream, this part is 10 cm long. The filament freely hangs at this tube end. This setup assures a clamped boundary condition at the upstream end with the filament parallel to the free stream, the other filament end being free. For the present study, visualizations of the filament are performed through the transparent side walls of the wind... tunnel test section. Images are captured with a video camera operating at 60 Hz and then processed to get quantitative results. For instance, the flapping frequencies of the filament are obtained by a fast Fourier transform (FFT) analysis of space-time images. All the filaments are made of silicone with a Young’s modulus $E=2.5$ MPa. Their flexural rigidity is $EI$ where $I=\pi d^4/64$. Extensive experiments are carried out for filaments of diameters $d=1.2$ and $1.8$ mm, but some measurements have been validated with filaments of diameter $d=1.6$ mm. The length $l$ of these filaments is varied up to about 30 cm. The density $\rho$ of silicone is about 950 kg m$^{-3}$ and experiments are performed at a temperature close to 20 °C in such a way that the values $\nu_g=15\times 10^{-6}$ m$^2$ s$^{-1}$ and $\rho_f=1.2$ kg m$^{-3}$ are considered for the kinematical viscosity and density of the flowing air, respectively. This results in a mass ratio $\mu$ of about 800. Note that for a nonconfined cylinder the added fluid mass is equal to the displaced fluid mass (see the detailed analysis of Lopes \textit{et al.}\cite{17}) in such a way that the mass ratio simply writes $\mu=(\rho+\rho_f)/\rho_f$. B. Experimental observations The first experiment consists in gradually increasing the free stream velocity $U$ for a given filament (i.e., for given $d$ and $l$). For low wind velocities, the filament appears to be stationary. Then, when $U$ exceeds a well-defined critical value $U_c$, it exhibits a flapping motion. A visual inspection reveals that this motion takes place in a vertical plane, at least for the wind velocities considered here. The static equilibrium positions mainly correspond to the balance between the gravity force and the fluid load. Therefore they depend on the flow velocity $U$. Figure 1 presents superimposed views of steady positions of a filament of diameter $d=1.2$ mm, length $l=12$ cm and for ten values of $U$ ranging from 0 to 13.6 m s$^{-1}$. For this particular filament, we find $U_c=13.7$ m s$^{-1}$. For all these static equilibria, the filament axis is totally contained in a vertical plane (parallel to gravity and free stream). As $U$ is increased, the filament tends to align with the free stream, so its incidence angle decreases. Moreover, images of Fig. 1 show that except close to the clamped end, the filament is nearly rectilinear. Unsteady behavior of the filament spontaneously takes place when the free stream velocity $U$ exceeds a threshold $U_c$. Critical values $U_c$ are determined for numerous combinations of $d$ and $l$ and are reported in Fig. 2 for $d=1.2$ mm and 1.8 mm. It appears that $U_c$ depends on the length $l$ and diameter $d$ of the considered filament. We first note the increase of $U_c$ with $d$. Besides, for both diameters, we can observe that the threshold $U_c$ strongly decreases with $l$ for short filaments, whereas the dependence of $U_c$ on $l$ is weaker for $l$ larger than 15 cm. The unsteady behavior consists in a regular periodic flapping of the filament in a vertical plane, i.e., a plane parallel to free stream and gravity. These vertical oscillations are illustrated by Fig. 3 which shows pictures of ten superimposed successive views captured at 60 Hz during flapping motion. In contrast to short filaments [Fig. 3(a)], one node can be distinguished in the envelope of the flapping motion of long filaments. One of these nodes has been marked by an arrow in Fig. 3(b). But the motion amplitude being quite ![FIG. 1. Superimposed views of steady states of the filament $l=12$ cm, $d=1.2$ mm for $U$ from 0 to 13.6 m s$^{-1}$ ($U=0.0$, 1.1, 2.2, 3.4, 4.5, 6.0, 7.5, 9.5, 11.5, and 13.6 m s$^{-1}$, flow from right to left).](image1) ![FIG. 2. Experimental thresholds $U_c$ for the appearance of flapping as function of the filament length $l$ for $d=1.2$ mm (+) and 1.8 mm (○).](image2) ![FIG. 3. Superimposed successive views of the flapping filament for $l=12$ cm, $d=1.2$ mm, $U=16.9$ m s$^{-1}$ (a) and $l=21$ cm, $d=1.8$ mm, $U=14.7$ m s$^{-1}$ (b). Flow from right to left.](image3) small, except at the free end, it was not always possible to determine precisely if one or more nodes are present along the filaments and consequently to measure accurately the boundary between zero- and one-node flapping. The flapping of the filament is periodic. In Fig. 4, its frequency $f$ is plotted as a function of $U$ for various filaments. Frequency $f$ is calculated by the spectral analysis of sequences of 360 images captured at 60 Hz. This gives a resolution of about 0.17 s$^{-1}$. The frequency is found to be constant along the filament. In Fig. 4, it appears that $f$ increases with $U$ and decreases for increasing $d$ or $l$. Note also that plateaus are sometimes observed at threshold. As already mentioned and in contrast to the VIV where vibrations are mainly transverse to the free stream, and perpendicular to the gravity, in the present study strictly vertical, i.e., parallel to the gravity, oscillations are always observed at threshold. III. THEORETICAL ANALYSIS A. Static equilibrium states We introduce the curvilinear coordinate $s$ along the filament axis, with $s=0$ at the free end and $s=l$ at the upstream extremity. The incidence angle $\theta(s)$ is the local angle between the free stream and the filament axis. We consider a small filament element $ds$; forces and moments acting on this small element are displayed in Fig. 5. They consist in a tension $T$ and a shear $Q$ in a cross section, a bending moment $M$, the gravity force, and the fluid dynamics load. We consider separately the inviscid force $F$ and the viscous force of components $L$ and $N$. $F$, $L$, and $N$ are forces per unit length. We assume that a cross section of the straight filament remains plane during all deformations in such a way that the Bernoulli–Euler beam theory can be applied. Moreover, because no notable lengthening is detected during experiments, the center line of the filament is considered as inextensible. The force balance for static equilibrium writes in the $s$-wise and transverse direction, respectively, $$\frac{dT}{ds} = L - (\rho - \rho_f)gA \sin \theta,$$ $$\frac{dQ}{ds} = F + N - (\rho - \rho_f)gA \cos \theta.$$ The moment balance gives $$Q = \frac{dM}{ds} = EI \frac{d^2\theta}{ds^2},$$ where the second equality results from the expression of $M$ given by the Bernoulli–Euler beam theory. Because the curvature radius of the filament is very large relative to its diameter $d$, the flow is considered as locally two dimensional. Therefore, for both components $N$ and $L$ of the viscous force, we use the semiempirical expressions deduced by Taylor$^8$ from force measurements on a smooth circular cylinder set at various incidence angles in a wind: $$N = \frac{1}{2}\rho_f dU^2(C_D \sin^2 \theta + 4 \text{Re}^{-1/2} \sin^{3/2} \theta),$$ $$L = \frac{1}{2}\rho_f dU^2(5.4 \text{Re}^{-1/2} \cos \theta \sin^{1/2} \theta),$$ where $\text{Re} = Ud/\nu_{f,l}$ is the Reynolds number and the pressure drag coefficient $C_D$ is treated as an adjustable parameter of the model. The inviscid fluid dynamic force $F$ is equal in magnitude but opposite to the rate of change of the fluid momentum. Using the expression deduced from an elongated-body potential flow theory by Lighthill,$^8$ we find that $$F = -\rho_f A U^2 \frac{d\theta}{ds} \cos^2 \theta,$$ where $\rho_f A$ is the added mass of fluid per unit length of filament (e.g., Lopes et al.$^9$). Thus $F$ is proportional to the filament curvature $d\theta/ds$. It results that the drag of a straight cylinder in a potential flow is zero, this is known as the d’Alembert’s paradox. At the clamped end the filament is maintained parallel to the wind so that $$\theta(s=l) = 0.$$ Forces and moment acting on the free end cross section are supposed to be negligible giving $$T(s=0) = 0,$$ \[ Q(s = 0) = 0, \] (9) \[ M(s = 0) = 0, \quad \text{i.e.,} \quad \frac{d\theta}{ds}(s = 0) = 0. \] (10) The above equation system (1)–(3) together with the force expressions (4)–(6) and the boundary conditions (7)–(10) are solved numerically using a shooting method with a standard Runge–Kutta integration scheme. The shape $ \theta(s) $ of the filament axis is thus deduced as a function of the control parameter $ U $. Calculations are repeated for various values of the only free parameter $ C_D $. Results obtained for the same conditions as that of the experiments of Fig. 1 ($ l = 12 $ cm and $ U $ from 0 to 13.6 m s$^{-1}$) and calculated with $ C_D = 0.8 $ and 0.9 are plotted in Fig. 6. Beyond the qualitative agreement revealed by comparing Figs. 1 and 6, a systematic quantitative comparison between the experiments and the theory is performed to determine the suitable value of the drag coefficient $ C_D $. For this purpose, the incidence angle at the free end $ \theta(s = 0) $ is used and experimental data are compared to the theoretical results obtained for various values of $ C_D $. The good agreement between theory and experiments is illustrated by the example in Fig. 7 where $ l = 12 $ cm, $ d = 1.2 $ mm, and $ C_D = 0.8 $ and 0.9 for the calculations. It allows to validate Taylor’s semiempirical expressions (4) and (5) used for the viscous fluid load. Comparisons are repeated for several combinations of $ d, l, $ and $ U $ and the best agreement is found for $ C_D $ ranging between 0.8 and 0.9. In the following these two values of the drag coefficient will be considered. B. Linear stability analysis In contrast to VIV, the observed flapping motion results directly from an instability and not from a structural response to hydrodynamical forcing. Linear stability analysis of the static equilibrium positions is then performed to determine the theoretical critical values of the free stream velocity and flapping characteristics at, or close to, the threshold. 1. Governing equation for small displacements We first derive the governing equation for the filament element $ ds $ undergoing a small displacement $ y(s, t) $ as seen in Fig. 8. $ y(s, t) $ is measured from the stationary position defined by $ \theta(s) $ the solution of Eqs. (1)–(3) at a given $ U $. According to the experimental observations, only vertical motions are considered. In this case, the normal and tangential components of the fluid velocity relative to the filament are, respectively, \[ v_N(s, t) = U \sin \theta + U \cos \theta \frac{\partial y}{\partial s} - \frac{\partial y}{\partial t}, \] (11) \[ v_T(s, t) = -U \cos \theta + \left( U \sin \theta - \frac{\partial y}{\partial t} \right) \frac{\partial y}{\partial s}. \] (12) The incidence angle $ \theta_i $ between the free stream and the filament axis is then \[ \theta_i = \theta + \frac{\partial y}{\partial s} - \frac{\cos \theta}{U} \frac{\partial y}{\partial t}. \] (13) The structural damping of the filament material is neglected here in such a way that the forces and moment acting on the filament element are the same as listed in Sec. III A (see also Fig. 5). Therefore, to first order in $ y $, Newton’s second law projected onto the $ y $ direction writes \[ \rho A \frac{\partial^2 y}{\partial t^2} = \frac{\partial}{\partial s} \left( T \frac{\partial y}{\partial s} \right) + F + f + N + n - L \frac{\partial y}{\partial s} - \frac{\partial (Q + q)}{\partial s} \] \[ - (\rho - \rho_f) g A \cos \theta, \tag{14} \] and moment balance is \[ Q + q = \frac{\partial (M + m)}{\partial s} = EI \left( \frac{d^2 \theta}{ds^2} + \frac{\partial^2 y}{\partial s^2} \right), \tag{15} \] where upper case letters $ T, Q, F, N, L, $ and $ M $ denote the quantities (functions of $ s $ only), to order zero in $ y $, given by the static equilibrium equations (1)–(6). Lower case letters $ q, f, n, $ and $ m $ correspond to the supplementary terms appearing at first order in $ y $, and are functions of $ s $ and $ t $. Combining the derivative of Eq. (15) with (2), relation (14) becomes \[ \rho A \frac{\partial^2 y}{\partial t^2} = \frac{\partial}{\partial s} \left( T \frac{\partial y}{\partial s} \right) + f + n - L \frac{\partial y}{\partial s} - EI \frac{\partial^4 y}{\partial s^4}. \tag{16} \] The inviscid fluid force is (Lighthill$^b$) \[ F_A = \rho_f A \left( \frac{\partial}{\partial t} - U \cos \theta \frac{\partial}{\partial s} \right) v_N, \tag{17} \] where the normal velocity $ v_N(s,t) $ is given by (11); hence \[ f = \rho_f A \left( - \frac{\partial^2 y}{\partial t^2} + 2U \cos \theta \frac{\partial^2 y}{\partial t \cdot \partial s} + U^2 \cos \theta \sin \theta \frac{d\theta}{ds} \frac{\partial y}{\partial s} \right. \] \[ \left. - U^2 \cos^2 \theta \frac{\partial^3 y}{\partial s^3} \right). \tag{18} \] In Eq. (16) $ n $ is the first order term of expression (4) in which $ \theta $ has been replaced by the incidence angle $ \theta $ given by (13) and $ U $ by $ (v_N^2 + v_T^2)^{1/2} $ where the two velocity components are given by (11) and (12). It results that \[ n = \frac{1}{2} \rho_f dU^2 \left[ (C_D \sin 2\theta + 6 \text{Re}^{-1/2} \cos \theta \sin^{1/2} \theta) \Psi \right. \] \[ \left. + (2C_D \sin^2 \theta + 6 \text{Re}^{-1/2} \sin^{1/2} \theta) \frac{u}{U} \right], \tag{19} \] where \[ \Psi = (\partial y / \partial s) - (\cos \theta / U)(\partial y / \partial t) \quad \text{and} \quad u/U = -(\sin \theta / U)(\partial y / \partial t). \] An equation similar to (16) has been derived by Páidoussis$^{9,10} $ to describe the small lateral motions of a filament in an axial flow, i.e., small displacements around the static equilibrium state $ \theta(s)=0 $ for all $ s $ between 0 and $ l $. In this case, and contrary to the present configuration, the governing equation (16) can be linearized with respect to the (small) incidence angle $ \theta $. This situation has been extensively studied and references can be found in the recent series of papers by Páidoussis and co-workers (Páidoussis et al.$^5$ Lopes et al.$^6$ Semler et al.$^13$). Moreover, it should be pointed out that, for the lengths considered here, the filaments in static equilibrium are rectilinear on most of their length, specially close to the flapping threshold (Figs. 1 and 6). Hence, in the following we restrict our stability analysis to a filament having a constant incidence angle equal to the value at free end $ \theta(s=0) $. From Eqs. (1) and (5), a linear variation of the tension $ T $ along the filament results. ### 2. Results and comparisons For the analysis we consider periodic perturbations of the form \[ y(s,t) = \text{Re}[Y(s)e^{i\omega t}], \tag{20} \] where $ \omega $ is a complex frequency. Assuming the filament is initially rectilinear (i.e., $ \theta $ is constant therefore $ T $ is linear in $ s $) and using the filament length $ l $ to make the spatial variables dimensionless (let $ \xi = s/l $ and $ \eta = Y/l $) the governing equation for small displacements (16) becomes \[ \frac{d^4 \eta}{d\xi^4} + a \frac{d^2 \eta}{d\xi^2} + b \frac{d^2 \eta}{d\xi^2} + (c_1 + i\omega c_2) \frac{d \eta}{d\xi} + (\omega^2 e_1 + i\omega e_2) \eta = 0, \tag{21} \] where the coefficients $ a, b, c_1, c_2, e_1, $ and $ e_2 $ depend nonlinearly on $ U $ and $ \theta $, their expressions are given in the Appendix. A dramatic simplification occurs when $ \theta=0 $, which corresponds to the classical case of a filament in an axial flow (e.g., see the work of Páidoussis$^{9,10} $). For the situation considered here, the coefficient $ c_1 $ is nonzero; this induces a new destabilizing term whose effect has never been into account before. The boundary conditions at the clamped end, $ \xi=1 $, are \[ \eta = \frac{d \eta}{d \xi} = 0, \tag{22} \] and at the free end, $ \xi=0 $, are \[ \frac{d^2 \eta}{d\xi^2} = \frac{d^3 \eta}{d\xi^3} = 0. \tag{23} \] To solve the whole governing equation (21) we use a Galerkin method involving the eigenfunctions $ \phi_i(\xi) $ of the filament without flow. This is applicable because we assume that the fluid load is negligible at the free end in such a way that the boundary conditions (22) and (23) are identical with or without flow. Then, the method consist to approximate the filament displacement $ \eta(\xi) $ by the truncated series \[ \eta(\xi) = \sum_{i=1}^{n} a_i \phi_i(\xi), \tag{24} \] where \[ \phi_i(\xi) = \frac{\text{ch } k_i(1-\xi) - \cos k_i(1-\xi)}{\text{ch } k_i + \cos k_i} \] \[ - \frac{\text{sh } k_i(1-\xi) - \sin k_i(1-\xi)}{\text{sh } k_i + \sin k_i}. \tag{25} \] From the boundary conditions (22) and (23) we deduce that to obtain nontrivial solutions, the wavenumbers $ k_i $ must verify $ \cos k_i \text{ch } k_i = -1 $. Relations (24) and (25) are substituted into the governing equation (21), and then the resulting equation is projected onto the free modes $ \phi_i $. The complex eigenfrequencies $\omega$ are then determined from the solvability condition of the linear system that is obtained. Each eigenfrequency $\omega$ corresponds to a specific eigenmode of the instability, made of a combination of the free modes $\phi_i$. The real part $\text{Re}[\omega]$ is the angular frequency $2\pi f$ and its temporal growth rate is given by $-\text{Im}[\omega]$. It follows that a negative value of $\text{Im}[\omega]$ for at least one mode is indicative of a flapping instability of the filament. The computations are repeated with an increasing number $n$ of modes in (24) in order to ensure the convergence of the solution with a good enough accuracy. The theoretical results in Figs. 9 and 10 are obtained with $n=23$ resulting in angular frequencies $\text{Re}[\omega]$ and growth rates $-\text{Im}[\omega]$ evaluated with an accuracy better than 0.2 s$^{-1}$ for $l=35$ cm, and better than 0.1 s$^{-1}$ for $l<20$ cm. For shorter filaments the convergence could be achieved with a decreasing number of modes (typically $n<10$ for $l<10$ cm). The theoretical values for the critical velocity $U_c$ are plotted in Fig. 9 as a function of the filament length $l$ for the diameters $d=1.2$ mm [Fig. 9(a)] and 1.8 mm [Fig. 9(b)]. A comparison with experimental results is also shown and reveals in both cases a good qualitative and quantitative agreement. It is noteworthy that this agreement is obtained without adjustable parameter. However, a slight shift of the calculated curves towards the lower length values can be noted. This may be due to the fact that the filament is considered as rectilinear on its whole length $l$ whereas the actual rectilinear part is shorter. The instability mode destabilized at the threshold $U_c$ depends on the filament length $l$. Thus, the characteristic lobes of the theoretical curves $U_c(l)$ in Fig. 9 correspond to various instability modes that differ in their shapes and frequencies. Note that these instability modes are combinations of the Galerkin modes $\phi_i$ (and should not be confused with them). Theoretical and experimental measurements of flapping frequencies are also compared in Fig. 9 for three filaments ($l=12$ and 20 cm for $d=1.2$ mm, $l=12$ cm for $d=1.8$ mm). In all cases, and although they are of the same order of magnitude, we note that the frequencies predicted by the model are systematically overestimated when compared to the experimental results. In the same manner, their dependence with the filament length, as it is observed during the experiments, is not fully recovered by the model. Nevertheless, qualitative agreements can be found such as the dependence of the frequency $\omega$ on the free stream velocity $U$ which is predicted with a very good approximation. Also, when considering results for filaments of a given length ($l=12$ cm) but of different diameters ($d=1.2$ and 1.8 mm) we can see that, in agreement with the experimental observations, the theory gives a flapping frequency $\omega$ that decreases as $d$ is increased. IV. DISCUSSIONS AND CONCLUSION A. Effect of mass ratio Although the mass ratio $\mu$ has not been varied in the present experiments, a comparison with the recent study of Facchinetti et al.\textsuperscript{4} invalidates a VIV scenario to interpret our observations. These authors present indeed results on a freely hanging flexible cable of aspect ratio $l/d=250$, towed in a water tank with a mass ratio $\mu$ of about 2.07. They report the observations of waves propagating along the cable and consisting in traveling transverse (i.e., perpendicular to the free stream) displacements of the cable. These waves appear as soon as the Reynolds number $\text{Re}=Ud/\nu$ exceeds the critical value of the appearance of the vortex shedding (of about 50) and consequently are locked on these vortex shedding. These are evidences that these waves are excited by the periodic vortex shedding and a model of these VIV emphasizes the resonance of the structure with the periodic fluid loading. In contrast, for the same aspect ratio (a filament with $l=30$ cm and $d=1.2$ mm, for instance, resulting in an aspect ratio $l/d$ of 250) we observe a critical value $\text{Re}_c \approx 610$ (see Fig. 2) with vertical displacements of the filament. This flapping frequency appears clearly not to be locked on the vortex shedding frequency that would correspond to a Strouhal number $\text{St}=fd/U$ of 0.21 (see, e.g., the recent review by Norberg\textsuperscript{12}), while St is only of about $4 \times 10^{-3}$ for the flapping oscillations (considering the total excursion of the free end of the filament instead of the filament diameter as length scale, we obtain a Strouhal number of order $10^{-1}$); in this estimation, the incidence of the filament versus the flow has been taken into account with the cosine correction proposed by Williamson.\textsuperscript{13} Therefore the flow-induced vibrations presented here appear to be clearly of a different nature that of the classical VIV. Our theoretical study shows that the flapping of the filaments results directly from a loss of stability of the filament stationary states under the action of external steady forces. Besides, this is also consistent with the VIV models (e.g., see the work by Facchinetti et al.\textsuperscript{4}) that predict the narrowing of the synchronization domain when the mass ratio is increased. In our case, the VIV resonance domain is negligibly small and cannot be invoked to explain our observations. B. Beyond the linear regime Finally we should also mention that our experiments show that the transition from a static state to flapping exhibits hysteresis and bistability. Actually, when experiments, as described in Sec. II B, are repeated with a decreasing velocity $U$, the flapping motion persists up to a value $U_{\text{--}}$ lower than $U_c$. This is illustrated in Fig. 11 that shows the angular frequency of the flapping for increasing and decreasing $U$, for the filament $d=1.2$ mm and $l=20$ cm. While flapping appears at a critical value $U_c$ of the wind velocity between 9.4 and 9.6 m s$^{-1}$, it is maintained to $U_{\text{--}}$ between 8.7 and 8.9 m s$^{-1}$ when $U$ is decreased. In the bistability domain [$U_{\text{--}}, U_c$], the filament can easily be forced to either a static state or flapping by external perturbations. For $U < U_{\text{--}}$ and $U > U_c$ all attempts to change the spontaneous state of the filament are fruitless. Such an hysteretical cycle has also been observed in numerous flow-induced vibration problems as for in experiments on a filament in an axial flow (Paidousis et al.\textsuperscript{7}) or in a flowing soap film (Zhang et al.\textsuperscript{14}). This bistability is supposed to be responsible for the dispersion of the experimental critical velocities observed in Fig. 2. C. Conclusion Observations of flow-induced vibrations for a filament freely hanging in a vertical plane are reported. In contrast to low mass ratio situation, no VIV have been observed. The persistence of steady states up to Reynolds number of several hundreds, results in the appearance of another kind of instability where the filaments oscillate in the vertical plane. We have shown that this flapping results from a loss of stability of the filament static equilibrium states under the action of the steady fluid flow loads. Our theoretical results confirm the appearance of this instability and, in particular, the threshold values for its observation. The different characteristics of the flapping are also correctly predicted by the model and, in particular, the exchanges between different flapping modes when the length of the filaments is increased. Finally, a strong nonlinear behavior with the existence of a bistable regime has been detected experimentally. This would demand a nonlinear theoretical analysis in order to describe the saturation of flapping amplitude as $U$ is increased and the observed hysteretical behavior at threshold. APPENDIX Expressions of the coefficients of Eq. (21): $$a = \frac{l^2}{EI}\rho_fAU^2\cos^2\theta,$$ $$b = -\frac{l^3}{EI}[L+(\rho-\rho_f)gA\sin\theta],$$ $$c_1 = -\frac{l^3}{EI}\left[(\rho-\rho_f)gA\sin\theta + \frac{1}{2}\rho_fdu^2(C_D\sin 2\theta\right.$$ $$\left.+ 6\text{ Re}^{-1/2}\cos\theta\sin^{1/2}\theta)\right],$$ \[ c_2 = -\frac{l^3}{EI} 2 \rho_f A U \cos \theta, \] \[ e_1 = -\frac{l^4}{EI} (\rho + \rho_f) A, \] \[ e_2 = \frac{l^4}{EI} \rho_f dU [C_D \sin^2 \theta + 3 \text{Re}^{-1/2} \sin^{1/2} \theta]. \] --- 1. R. D. Blevins, Flow-Induced Vibration (R. E. Krieger, Malabar, FL, 1986). 2. P. W. Bearman, “Vortex shedding from oscillating bluff bodies,” Annu. Rev. Fluid Mech. 16, 195 (1984). 3. C. H. K. Williamson and R. Govardhan, “Vortex-induced vibrations,” Annu. Rev. Fluid Mech. 36, 413 (2004). 4. M. L. Faccinetti, E. de Langre, and F. Biotley, “Vortex-induced travelling waves along a cable,” Eur. J. Mech. B/Fluids 23, 199 (2004). 5. M. P. Paidoussis, E. Grinevich, D. Adamovic, and C. Semler, “Linear and non-linear dynamics of cantilevered cylinders in axial flow. Part 1: Physical dynamics,” J. Fluids Struct. 16, 691 (2002). 6. J. L. Lopes, M. P. Paidoussis, and C. Semler, “Linear and non-linear dynamics of cantilevered cylinders in axial flow. Part 2: The equation of motion,” J. Fluids Struct. 16, 715 (2002). 7. G. I. Taylor, “Analysis of the swimming of long and narrow animals,” Proc. R. Soc. London, Ser. A 214, 158 (1952). 8. M. J. Lighthill, “Note on the swimming of slender fish,” J. Fluid Mech. 9, 307 (1960). 9. M. P. Paidoussis, “Dynamics of flexible slender cylinder in axial flow. Part 1. Theory,” J. Fluid Mech. 26, 717 (1966). 10. M. P. Paidoussis, “Dynamics of cylindrical structures subjected to axial flow,” J. Sound Vib. 29, 365 (1973). 11. C. Semler, J. L. Lopes, N. Augu, and M. P. Paidoussis, “Linear and non-linear dynamics of cantilevered cylinders in axial flow. Part 3: Nonlinear dynamics,” J. Fluids Struct. 16, 739 (2002). 12. C. Norberg, “Fluctuating lift on a circular cylinder: Review and new measurements,” J. Fluids Struct. 17, 57 (2003). 13. C. H. K. Williamson, “Vortex dynamics in the cylinder wake,” Annu. Rev. Fluid Mech. 28, 477 (1996). 14. J. Zhang, S. Childress, A. Libchaber, and M. Shelley, “Flexible filaments in a flowing soap film as a model for one-dimensional flags in a two-dimensional wind,” Nature (London) 408, 835 (2000).
On the Computational Complexity of the Manufacturing Job Shop and Renetrant Flow Line F.L. Lewis Follow this and additional works at: https://digitalrepository.unm.edu/ece_rpts Recommended Citation Lewis, F.L., "On the Computational Complexity of the Manufacturing Job Shop and Renetrant Flow Line." (1996). https://digitalrepository.unm.edu/ece_rpts/48 This Technical Report is brought to you for free and open access by the Engineering Publications at UNM Digital Repository. It has been accepted for inclusion in Electrical & Computer Engineering Technical Reports by an authorized administrator of UNM Digital Repository. For more information, please contact email@example.com. ON THE COMPUTATIONAL COMPLEXITY OF THE MANUFACTURING JOB SHOP AND REENTRANT FLOW LINE F.L. Lewis Automation and Robotics Research Institute The University of Texas at Arlington 7300 Jack Newell Blvd. S, Ft. Worth, Texas 76118-7115 and B.G. Horne NEC Research Institute 4 Independence Way Princeton, NJ 08540 and C.T. Abdallah EECE Department University of New Mexico Albuquerque, NM 87131, USA UNM Technical Report: EECE96-002 Report Date: April 6, 1996 # Contents 1 Introduction 1 2 Complexity Theory Overview 3 3 Structure and Modeling of Flexible Manufacturing Systems 5 3.1 Flexible Manufacturing Systems (FMS) 5 3.2 Manufacturing System Structures 5 3.3 Petri Net Representation of FMS 7 3.3.1 Incidence Matrix and Marking Transition Equation 7 3.3.2 Resource Loops and $p$-Invariants 8 4 Computational Complexity of Finding the $p$-Invariants in the Job Shop 9 5 Computational Complexity of Finding the $p$-Invariants in the Flow Line 11 5.1 Structure of the Reentrant Flow Line 11 5.1.1 Definition of a General Class of Reentrant Flow Lines 12 5.1.2 Special Form of the Incidence Matrices 13 5.2 Algorithm for Computation of the $p$-Invariants 13 6 Conclusion 15 List of Tables List of Figures 3.1 Reentrant flow line with 4 machines and 2 parts. ........................................... 6 3.2 PN representation of the reentrant flow line. .................................................. 7 Abstract This paper presents a comparison study of the computational complexity of the general job shop protocol and the flow line protocol in a flexible manufacturing system. It is shown that a certain representative problem of finding resource invariants is $NP$-complete in the case of the job shop, while in the flow line case it admits a closed-form solution. The importance of correctly selecting part flow and job routing protocols in flexible manufacturing systems to reduce complexity is thereby conclusively demonstrated. Key Words Job Shop problem, Flow Line, Complexity Theory, Petri Net Chapter 1 Introduction In a general flexible manufacturing system (FMS) where resources are shared, a key role in part routing, job selection, and resource assignment is played by the FMS controller. Given the same resources of machines, robots, fixtures, tooling, and so on, different structures result under different routing/assignment strategies by the controller. Unstructured strategies are generally classified as the so-called job shop organization, while structured protocols result in various sorts of flow lines, with or without assembly. The importance of structure in determining complexity has not been rigorously addressed in FMS. The theory of NP-completeness [7] potentially provides a comprehensive approach for analysis of computational complexity in FMS. This possibility has not been rigorously explored. There are many distinct analysis and design problems to be solved in FMS, including scheduling with optimality, computation of the Petri net (PN) p-invariants to determine resource loops, analysis of deadlocks and circular waits, design and implementation of deadlock avoidance strategies, and design/selection of dispatching and routing algorithms. These problems have varying degrees of complexity, and complexity varies as well depending on whether one has a flow line, assembly line, or job shop protocol structure. Many traditional scheduling and sequencing problems have been found to be in NP, thus it has been necessary to develop heuristics or approximate methods for analysis and solution. It has been shown, for instance that, even for the flow line with 2 processors, scheduling while minimizing the maximum flow time is NP-complete for both nonpreemptive and preemptive schedules [8]. For the general job shop protocol the situation is even worse (see for example page 242 of [7]). Branch and bound algorithms are generally used in this case. For the flow line, the lot-sizing problem is polynomial, while for the flow line with assembly it is exponential. On the other hand, determining circuits in a graph, as required, e.g., to find the wait relations in an FMS, is polynomial. The complexity of many problems, including the determination of the PN p-invariants, has not yet been determined. There is currently no comprehensive theory that provides a categorization of the complexity of analysis problems for the flow line, assembly line, and job shop. There is no formal theory describing how to impose structured flow and command protocols on an FMS to simplify its complexity. Petri nets (PN) [16, 18] have been extensively used in the analysis of manufacturing systems, with quite variable results. Though, ad hoc applications abound, PN have a body of theoretical results on liveness, boundedness, reachability, and so on that make them very useful in studies of FMS when seriously applied. Applications of PN are found in [3, 6, 9, 23]. PN approaches to the design of FMS sequencing/dispatching controllers are found in [10, 11, 17]. The PN incidence matrix has been used for analysis applications in FMS. It has been shown that it can be used to study structural properties of FMS, including determination of the siphons [1] and deadlock avoidance [14]. In these papers, the problem of finding a binary basis for the nullspace of W is important, for such a basis defines a special class of siphons known as the p-invariants or resource loops, which must be known for effective deadlock avoidance. However, matrix applications in PN have not been fully exploited. In this paper we show that it is possible by judicious means to reveal a special structure of the PN incidence matrix. in a very general class of reentrant flow lines that can include assembly operations. To reveal the importance of structure in the study of complexity for FMS, we select the representative problem of determining the $p$-invariants. It is shown that for unstructured job shop protocols this problem is $\mathcal{NP}$-complete, while for reentrant flow line protocols it is polynomial. The importance of selecting suitable controller sequencing protocols to reduce complexity in FMS is thereby shown. This paper is organized as follows. Chapter 2 presents an overview of computational complexity theory and of the reduction approach. Chapter 3 presents various manufacturing structures including the job shop, assembly line, and flow line, and introduces some PN notions. The computational complexity of finding the $p$-invariants is shown to be $\mathcal{NP}$-hard for the job shop in Chapter 4 and polynomial for a general class of reentrant flow lines in Chapter 5. Until recently, it was felt that decidable problems are practically solved and thus not very interesting. The introduction of computational complexity theory has since changed this misconception. Computational complexity theory is often used to establish the tractability or intractability of computational problems, and is concerned with the determination of the intrinsic computational difficulty of these problems [7]. In order to discuss the complexity of an algorithm, one must begin with a model of computation, for which the Turing Machine is the most commonly used. The simplicity of the Turing machine model appears to make it of little practical value; however, the Church-Turing Thesis holds that the class of problems that are tractable on a Turing machine are also tractable on any other reasonable model of computation (including the computers we use). One important concept in this theory is that of a polynomial-time algorithm, i.e. an algorithm whose running time can be bounded by a polynomial in the size of the description of the problem. In practice, such an algorithm can be feasibly implemented on a real computer. This is in contrast to an exponential-time algorithm, which is only feasible if the problem being solved is extremely small. The complexity class $P$ consists of all decision problems that can be decided in polynomial-time, while the class $EXP$ consists of those that can be decided in exponential-time. The complexity class $NP$ lies inbetween consisting of all decision problems that can be decided algorithmically in nondeterministic polynomial-time. An algorithm is nondeterministic if it is able to choose or guess a sequence of choices that will lead to a solution, without having to systematically explore all possibilities. This model of computation is not realizable, but it is of theoretical importance. In practice, problems in $NP$ are those for which a candidate solution can be verified to be a valid solution in polynomial-time, but the best known algorithms to find such a solution run in exponential time. Many practical problems belong to $NP$ and it is as of yet unknown whether $P = NP$. In other words, these two complexity classes form an important boundary between the tractable and intractable problems. A problem is said to be $NP$-hard if it is as hard as any problem in $NP$. Thus, if $P \neq NP$, the $NP$-hard problems can only admit deterministic solutions that take an unreasonable (i.e., exponential) amount of time, and they require (unattainable) nondeterminism in order to achieve reasonable (i.e., polynomial) running times. The central idea used to demonstrate $NP$-hardness evolves around the $NP$-complete problems. A problem is said to be $NP$-complete if every decision problem in $NP$ is polynomial-time reducible to it. This means that the $NP$-complete problems are as hard as any decision problem in $NP$. Given two decision problems $\Pi_1$ and $\Pi_2$, $\Pi_1$ is said to be polynomial-time reducible to $\Pi_2$ (written as $\Pi_1 \leq_p \Pi_2$), if there exists a polynomial time algorithm $R$ which transforms every input $x$ for $\Pi_1$ into an equivalent input $R(x)$ for $\Pi_2$. By equivalent we mean that the answer produced by $\Pi_2$ on input $R(x)$ is always the same as the answer $\Pi_1$ produces on input $x$. Thus, any algorithm which solves $\Pi_2$ in polynomial time can be used to solve $\Pi_1$ on input $x$ in polynomial time by simply computing $R(x)$, and then running $\Pi_2$. In order to show that a particular decision problem $\Pi_2$ is $\mathcal{NP}$-complete, one starts with a problem $\Pi_1$ which is known to be $\mathcal{NP}$-complete, and shows that $\Pi_1 \leq_p \Pi_2$. This proves that $\Pi_2$ is $\mathcal{NP}$-hard. To complete the proof that $\Pi_2$ is $\mathcal{NP}$-complete, it must be demonstrated that a candidate solution can be verified in polynomial time. In this paper, we use the One-In-3SAT problem which is known to be $\mathcal{NP}$-complete [7] in order to show that solving a certain problem for the general job shop is $\mathcal{NP}$-complete. We then use the special structure of the reentrant flow line problem to show that the same problem can be efficiently obtained for the flow line. This highlights the importance of structure in flexible manufacturing systems. The One-In-3SAT problem is as follows: One-In-3SAT: Instance: Given a set $U$ of variables, a collection $C$ of clauses over $U$ such that each $c \in C$ has $\|c\| = 3$. Question: Is there a truth assignment for $U$ such that each clause in $C$ has exactly one true literal? Example 1 Let $U = \{a, b, c, d\}$ and $C = \{\bar{a}\bar{b}c, \bar{a}bd, \bar{b}cd\}$. Then a solution is $a = b = \text{true}$ and $c = d = \text{false}$. □ Chapter 3 Structure and Modeling of Flexible Manufacturing Systems In this chapter we discuss flexible manufacturing systems with several sorts of structures, including the reentrant flow line, the assembly line, and the job shop. The importance of structure and protocol in flexible manufacturing systems is highlighted. Some Petri Net modeling techniques are introduced. 3.1 Flexible Manufacturing Systems (FMS) To meet competition in a global marketplace and provide flexible manufacturing in today’s high-mix low-volume manufacturing environment, manufacturing systems have gone away from old-style fixed-hardware sequential assembly lines with dedicated workstations. The trend for several years has been towards flexible manufacturing systems (FMS), which have four major components [2]: a set of machines or work stations, an automated material handling system that allows flexible job routing, distributed buffer storage sites, and a computer-based supervisory controller for monitoring the status of jobs and directing part routing and machine job selections. With this change in style, the emphasis has shifted towards the design of sophisticated decision-making controllers that include functions of job sequencing and dispatching, parts routing, job release, deadlock avoidance, etc. Unfortunately, rigorous approaches to FMS in problems such as dispatching and routing, steady-state analysis, queueing stability, bottleneck studies, and so on have focused on simple types of systems including single-server, flow line without assembly, serial forms, etc. Systems with finite buffers and nonserial systems (e.g. systems with assembly, etc.) have few results, with fewer still for general job shop structures and large-scale interconnected systems. It is by now known that many manufacturing problems are in NP so that significant increases in computing power do not significantly improve computational capabilities. There is no general approach for taking advantage of the FMS structure to reduce computational complexity. 3.2 Manufacturing System Structures The physical portion of an FMS is comprised of its resources: the set of machines or work stations, the automated material handling system, and the distributed buffers. We call these the manufacturing facility. Given the same resource facilities in the FMS, different sequencing algorithms by the controller produce different flow/protocol structures, including the reentrant flow line, the assembly line, and the general job shop protocol. Not only should the controller provide guaranteed performance, but it should impose a suitable structural protocol to achieve prescribed performance specifications, and it should be easily reconfigurable to change the FMS structure, dispatching rules, routing algorithms, etc. as products or performance requirements change. Disciplines such as discrete event (DE) systems are emerging to confront such problems [19]. A major issue is that the structure imposed by the controller should avoid or reduce NP-complexity problems. Formally, a manufacturing facility is a set $ \mathcal{R} = \{r_i\} $ of resources (e.g. machines, tools, fixtures, robots, transport devices, etc.), each of which has a distinct function. Each $ r_i $ can denote a pool of more than one machine that performs the same function. The resources operate on parts; parts of the $ j $-th type are denoted $ p_j $. A job sequence for part type $ p_j $ is a sequence of $ P_j $ jobs $ J_j = \{J_{1j}, J_{2j}, \ldots J_{P_j j}\} $ required to produce a finished product. We distinguish between jobs in the part sequence even if, for instance $ J_{2j} $ and $ J_{5j} $ are both drilling operations. The sequence of jobs may be determined from a task decomposition, bill of materials, assembly tree, or precedence matrix (c.f. Steward’s sequencing matrix [20]). If each job is performed on a single part and delivers a single part there is said to be no assembly. Once the sequence of jobs for a part type has been assigned, resources must be assigned to perform the jobs. This is performed by a manufacturing engineer based on the facilities available. If a single resource is needed for each job, for instance, this corresponds to a pairing $(J_{k j}, r_i)$ of the $ k $-th job for part $ p_j $ with a resource $ r_i $. The ordering of the jobs for a given part type can be either fixed or variable. For instance, in an application it may be allowable to either drill then machine a part, or to machine and then drill the part. Likewise, the resources assigned to each job can be either fixed or variable. For instance, either of two machines of different types (e.g. from different resource pools) might be capable of performing a given drilling job. In the general job shop the sequence of jobs is not fixed, or the assignment of resources to the jobs is not fixed. The effect is that part routing decisions must be made during processing. In the flow line the sequence of jobs for each part type is fixed and the assignment of resources to the jobs is fixed. The result is that each part type visits the resources in the same sequence, though different part types may have different sequences. The flow line is also known as the “job shop with fixed part routing”. The sequence in which part type $ p_j $ visits the resources in a flow line will be called the $ j $-th part path. A flow line is said to reentrant if any part type revisits the same resource more than once in its job sequence [12, 15]. This occurs if the same resource is assigned to different jobs in the part’s sequence. A sample reentrant flow line is given in Fig. 3.1. In this figure, $ R1 $ and $ R2 $ could be transport robots, for instance, that move the parts between certain jobs; $ B1, B2 $ could be buffers; and $ M1, M2, M3, M4 $ could be machines. Thus, the resources may include machines, robots, buffers, transport devices, fixtures, tools, and so on. 3.3 Petri Net Representation of FMS Some knowledge of Petri nets is assumed. A Petri net (PN) is a bipartite (e.g., having two sorts of nodes) digraph described by $(P, T, I, O)$, where $P$ is a set of places, $T$ is a set of transitions, $I$ is a set of (input) arcs from places to transitions, and $O$ is a set of (output) arcs from transitions to places. In our application, the PN places represent manufacturing resources and jobs, and the transitions represent decisions or rules for resource assignment/release and starting jobs. Operation duration times and resource setup times are captured in timed places, as opposed to the timed transition approach. For instance, a standard representation for a reentrant flow line is given in Fig. 3.1. The PN representation for the same system is shown in Fig. 3.2, where the places are drawn as circles and the transitions as bars. The flow line structure is evident in the parallel part type paths, interconnected by shared resource places (e.g., B1, M2) that service jobs for several part types. Note that along one part path, some resources (e.g., R1, R2) are used more than once, so that this flow line is reentrant. Each part path in the figure has a set of pallets denoted by PA1, PA2; one pallet is needed to hold each part entering the cell. Places ending in $P$, all on the job paths, correspond to jobs in progress. Places ending in $A$ correspond to the availability of resources. 3.3.1 Incidence Matrix and Marking Transition Equation It is common in PN theory [18] to represent the sets of arcs $I$ and $O$ in the PN description $(P, T, I, O)$ as matrices. Thus, element $I_{ij}$ of matrix $I$ is equal to 1 if place $j$ is an input to transition $i$. Element $O_{ij}$ of matrix $O$ is equal to 1 if place $j$ is an output of transition $i$. Otherwise the elements of $I, O$ are set to 0. Matrix $I$ is called the input incidence matrix, and $O$ the output incidence matrix. Both matrices are considered as maps from $P$ to $T$. Then, the PN incidence matrix is defined as \[ W = O - I. \] (3.1) A column vector $p$ indexed by the set of places $P$ is called the PN $p$-vector (place vector). The PN marking vector is the marking vector $m(p)$ defined as follows. Definition 1 (Marking and Support) Given a PN, the PN marking is the number of tokens in each place in the net. Given a place $p \in P$, the marking of $p$, $m(p)$, is the number of tokens in $p$. Given a vector of places $p = [p_1 \ p_2 \ldots p_q]^T$, the marking $m(p)$ is the vector $m(p) = [m(p_1) \ m(p_2) \ldots m(p_q)]^T$ of markings of the individual places. The support of a vector is the set of its elements having nonzero values. It is common to simplify the notation so that $m(t)$ denotes the marking vector $m(p)$ at time $t$. Then, in terms of the PN incidence matrix, one can write the PN marking transition equation $$m(t_2) = m(t_1) + W^T \tau = m(t_1) + (O - I)^T \tau,$$ (3.2) where $m(t)$ is the PN marking vector at time $t$, $t_1 < t_2$, and $\tau$ is a vector denoting which transitions have fired between times $t_1$ and $t_2$; element $\tau_i = n_i$ if the $i$-th transition has fired $n_i$ times in the interval. ### 3.3.2 Resource Loops and $p$-Invariants Central to the study of resource allocation in FMS are the following notions. Definition 2 ($p$-Invariant and Resource Loop) A $p$-invariant is a place vector $p$ having elements of zeros and ones that is in the nullspace of $W$, that is $$Wp = 0.$$ The set of places corresponding to the support of $p$ is known as a resource loop, also loosely called a $p$-invariant. The complete set of $p$-invariants of a PN gives a great deal of information. In [14] it is shown that they provide the basis for deadlock avoidance algorithms. The importance of $p$-invariants may be understood by noting that, beginning with (3.2), for any $p$-invariant $p$ one has $$p^T m(t_2) = p^T m(t_1) + p^T W^T \tau = p^T m(t_1).$$ (3.4) Noting that premultiplication by $p^T$ simply sums up the tokens in the positions of $m(\cdot)$ corresponding to the support of $p$, this is seen to be a statement that the total number of tokens in positions of $m(\cdot)$ corresponding to the support of $p$ is conserved. That is the $p$-invariants define those loops in the PN within which the numbers of tokens are conserved. These conservative loops defined by the $p$-invariants are the resource loops. Chapter 4 Computational Complexity of Finding the $p$-Invariants in the Job Shop The resource loops of an FMS contain information of great value in deadlock avoidance, shared resource conflict resolution using dispatching techniques, and so on. Unfortunately, to find the $p$-invariants it is necessary solve (3.3), determining a basis for the nullspace of $W$ that has only ones and zeros. In this chapter, we show that finding such a binary basis is an $\mathcal{NP}$-complete problem for the general job shop structure. Then, in Section 5 it is shown that for the reentrant flow line, with or without assembly, an analytic solution can be given for the problem. Theorem 1 The problem of finding a binary basis for $W$ in the general job shop is $\mathcal{NP}$-Complete. Proof: In order to solve the general job shop problem, we need to find a basis of the nullspace of the incidence matrix $W$. Since $W$ contains coefficients $w_{ij} \in \{-1, +1, 0\}$ and since a meaningful basis of its nullspace will have vectors $p$ whose entries $p_i$ also belong to \{0, +1\}, the problem is equivalent to finding $p_i$ such that $\sum_{i=1}^{n} w_{ij} p_i = 0; \quad \forall j = 1, \cdots n$. Note however, that the zero vector $p_i = 0, \forall i$ should be excluded. We shall then define the following problem Matrix Basis Instance: An $n \times 2n$ matrix $A \neq 0$ with entries in \{-1, 0, 1\}. Question: Does there exist a vector $x \neq 0$ with entries in \{0, 1\} such that $Ax = 0$? and prove that Matrix Basis is $\mathcal{NP}$-complete by transformation from One-In-3Sat. We begin with a proof for $A$ of size $n \times m$ and then later show how to augment the matrix to make it of size $n \times 2n$. Let $n = \|U\| + \|C\|$ and $m = 2\|U\| + 1$, where $U$ and $C$ are the sets of variables and clauses in the instance of One-In-3Sat. The columns of $A$ (and thus the components of the vector $x$) will correspond to complemented and uncomplemented assignments of the $\|U\|$ literals and an auxiliary variable $z$, i.e. $$x = \left[ \begin{array}{cccccc} x_1 & \bar{x}_1 & x_2 & \bar{x}_2 & \ldots & x_n & \bar{x}_n & z \\ \end{array} \right]' .$$ A valid solution vector will correspond to each component of $x$ being equal to 0 or 1 depending on whether the corresponding literal is true or false. All nontrivial solutions will have $z = 1$. The first $\|U\|$ rows of $A$ are used to insure that the solution vector is a valid truth assignment to the literals, i.e. so that value assigned to $x_i$ will be the logical complement of the value assigned to $\bar{x}_i$. Specifically, the first $\|U\|$ rows are configured as, $$a_{i,j} = \begin{cases} 1 & j \in \{2i-1, 2i\} \\ -1 & j = 2\|U\| + 1 \\ 0 & \text{otherwise} \end{cases}$$ The remaining $\|C\|$ rows are used to satisfy the requirement that exactly one literal in each clause is true. Specifically, denote a literal by $\bar{x}_i$ (i.e. $\bar{x}_i \in \{x_i, \bar{x}_i\}$), and denote the $i$th clause by $c_i = \bar{x}_{p_i} \bar{x}_{q_i} \bar{x}_{r_i}$. Then set $$a_{\|U\|+i,j} = \begin{cases} 1 & j = 2s - 1, \quad \bar{x}_s = x_s, \quad s \in \{p_i, q_i, r_i\} \\ 1 & j = 2s, \quad \bar{x}_s = \bar{x}_s, \quad s \in \{p_i, q_i, r_i\} \\ -1 & j = 2\|U\| + 1 \\ 0 & \text{otherwise} \end{cases}$$ Example 2 Let $U = \{x_1, x_2, x_3, x_4\}$ and let $C = \{x_1 \bar{x}_2 x_3, x_2 x_3 \bar{x}_4, \bar{x}_1 x_2 x_4\}$. Then the matrix $A$ is given by $$A = \begin{bmatrix} x_1 & \bar{x}_1 & x_2 & \bar{x}_2 & x_3 & \bar{x}_3 & x_4 & \bar{x}_4 & z \\ 1 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & -1 \\ 0 & 0 & 1 & 1 & 0 & 0 & 0 & 0 & -1 \\ 0 & 0 & 0 & 0 & 1 & 1 & 0 & 0 & -1 \\ 0 & 0 & 0 & 0 & 0 & 0 & 1 & 1 & -1 \\ 1 & 0 & 0 & 1 & 1 & 0 & 0 & 0 & -1 \\ 0 & 0 & 1 & 0 & 1 & 0 & 0 & 1 & -1 \\ 0 & 1 & 1 & 0 & 0 & 0 & 1 & 0 & -1 \end{bmatrix}$$ Every solution besides the trivial solution must have $z = 1$ since if $z = 0$ then the first $\|U\|$ rows of $A$ will guarantee that every other entry will also be equal to zero. The same rows will guarantee that for nontrivial solutions exactly one of $x_i$ and $\bar{x}_i$ will be equal to one. The last $\|C\|$ rows of $A$ will only be satisfied by nontrivial solutions such that exactly one literal of each clause is true. We can easily make $A$ of size $n \times 2n$ by adding one additional row and $2\|C\| + 1$ additional columns, i.e. construct the augmented matrix $$A' = \begin{bmatrix} A & B \\ C & D \end{bmatrix}$$ where $B$ and $C$ are matrices of zeros of sizes $(\|U\| + \|C\|) \times (2\|C\| + 1)$ and $1 \times (2\|U\| + 1)$ respectively, and $D$ is a matrix of ones of size $1 \times (2\|C\| + 1)$. The last row insures that the last $2\|C\| + 1$ components of the solution vector must be equal to zero, but these variables in no way interfere with the construction above. The augmented matrix is of size $n \times 2n$ where $n = \|U\| + \|C\| + 1$. The transformation is easily done in time linear in the size of the matrix, which is quadratic in $\|U\|$ and $\|C\|$. Therefore, we have shown that Matrix Basis is NP-Hard. On the other hand, one can easily verify the existence of $p_i$ as a member of the nullspace of $W$ which then proves that the problem is NP-Complete. □ Chapter 5 Computational Complexity of Finding the $p$-Invariants in the Flow Line Like many other problems, finding the $p$-invariants in a general job shop protocol is $\mathcal{NP}$-complete, as seen in the previous chapter. In this chapter, a special job flow protocol is imposed that allows one to give an analytical solution to this problem, so that the complexity is polynomial. This protocol corresponds to a large class of reentrant flow lines with or without assembly. The importance of structure in an FMS is thereby shown in regards to computational complexity, so that care should be taken in selecting job sequencing and routing strategies in FMS. The flow line structure allows one to model and analyze large-scale interconnected FMS in a polynomial number of operations using block matrices. 5.1 Structure of the Reentrant Flow Line In the reentrant flow line with or without assembly, e.g. Fig. 3.2, denote the set of jobs for part type $j$ as $J_j$ and the set of all the jobs as $J = \bigcup_j J_j$. It is noted that the part input places $PI$ and part output places $PO$ are not included as jobs. Places that occur off the part paths represent the availability of resources; denote by $R$ the set of all such places. The set of resources may be partitioned as $R = R_{ns} \cup R_s$, with $R_{ns}$ the nonshared resources and $R_s$ the shared resources. The set of PN places is given by $P = J \cup R$, the set of resources plus the set of jobs. Note that all transitions occur along the part paths. Partition the PN marking vector $p$ as $$p = \begin{bmatrix} v \\ r \end{bmatrix},$$ where $v$ is the vector of places corresponding to the jobs $J$ and $r$ is the vector of places corresponding to the resources $R$. Then, the PN incidence matrix has the structure $$W = [W_v \ W_r] \equiv S^T - F = [S_v^T - F_v \ S_r^T - F_r]$$ where $S_v^T, S_r^T$ are the output incidence matrices of the jobs and resources respectively, and $F_v^T, F_r^T$ are the input incidence matrices of the jobs and resources respectively. This formalizes some discussions in [3] concerning places of type $A, B, C$. Matrix $F_v$ is called the Steward sequencing matrix [20] or the Bill of Materials (BOM) [4] in manufacturing; it has element $(i, j) = 1$ if job $j$ is an immediate prerequisite for job $i$. Matrix $F_r$ is the resource requirements matrix used in [13]; it has element $(i, j) = 1$ if resource $j$ is required for job $i$. It is important to order the job places correctly to obtain a lower triangular matrix $F_v$ [5, 21], for then the sequencing of the jobs is causal. A causal ordering is also important in taking advantage of structure to reduce complexity. The special structure of matrices $F_v, F_r, S_v, S_r$ for a general class of reentrant flow lines is revealed in terms of the following constructions. ### 5.1.1 Definition of a General Class of Reentrant Flow Lines Definition 3 (Complete and Partial Part Paths) Given a reentrant flow line with assembly, define a complete part path as one that terminates in an output product (e.g. a PO place in the PN), and a partial part path as one that merges with another part path in an assembly operation. Note that each complete part path terminates in an extra transition that is required to produce the product output equations and to release the pallets, if any are used in that corresponding part path. To obtain a causal ordering of the jobs, number the job places sequentially from left to right along each single part path. Suppose part path $j_1$ is a complete path, with a partial part path $j_2$ merging into path $j_1$ at the assembly point, represented by a transition on that path. In this situation, one may number the jobs of partial path $j_2$ from left to right, stopping at the last job prior to the assembly transition. Then, return to the beginning of path $j_1$, picking up the place ordering by numbering the the job places of path $j_1$ from left to right. The transitions should be numbered corresponding to the job places they feed into. This procedure corresponds to numbering the jobs from bottom to top as is standard in a manufacturing assembly tree [22]. The subsequent analysis deals with the class of reentrant flow lines now defined. This class is more general than the one in [6] as it allows assembly operations as well as the use of more than one resource per job (e.g. tool, fixture, and machine) as in [10]. Definition 4 (Dot Notation for Input and Output Sets of a Node) Given a transition $t \in T$, define by $t\bullet$ the set of places that are inputs to $t$, and by $t\bullet$ the set of places that are outputs of $t$. Given a place $p \in P$, define by $p\bullet$ the set of transitions that are inputs to $p$, and by $p\bullet$ the set of transitions that are outputs of $p$. Given a set of nodes $\mathcal{S} = \{v_i\}$ (either places or transitions), define $\bullet\mathcal{S} = \{\bullet v_i\}$ and $\mathcal{S}\bullet = \{v_i\bullet\}$. Definition 5 (Pallet Places) Let the set of transitions along the $j$-th part path be $x_{j1}, x_{j2}, \ldots, x_{jL_j}$. Then, if part path $j$ is complete, it may have a pallet place $p_{j0}$. If so, it should be selected such that $p_{j0} \in \bullet x_{j1}, p_{j0} \notin \bullet x_{j\ell}, \ell \neq 1$, and $p_{j0} \in x_{jL_j}\bullet, p_{j0} \notin x_{j\ell}\bullet, \ell \neq L_j$. That is, if present, pallets are used for all jobs on a complete part path. Definition 6 (Set of Jobs of a Given Resource) Given a reentrant flow line with jobs $\mathcal{J}$ and resources $\mathcal{R}$, define the jobs associated with resource $r \in \mathcal{R}$ as $$J(r) = r\bullet\bullet \cap \mathcal{J}. \quad (5.3)$$ In terms of these constructions, the class of FMS studied here is given as follows. Denote the set of resources minus the pallets as $\mathcal{R}_{-0} = \mathcal{R} - \{p_{j0}\}$. Definition 7 (Definition of a Class of Reentrant Flow Lines) Define the class of reentrant flow lines with or without assembly as those satisfying the following properties. 1. For all places $p \in P$, one has $p\bullet \cap p\bullet = \phi$ the empty set. (No self-loops.) 2. For each part path $j$, the first transition satisfies $x_{j1}\bullet \cap \mathcal{R} = \phi$ and, if the path is complete the last transition satisfies $\bullet x_{jL_j} \cap \mathcal{R} = \phi$. (Each part path has a well-defined beginning and end.) 3. For each resource $r \in \mathcal{R}_{-0}$, one has $r \in p\bullet\bullet \cap \mathcal{R}$ for all $p \in J(r) = r\bullet\bullet \cap \mathcal{J}$. (Unity job duration—each job is described by only one job place along the part path.) This definition results in the following facts, easily derivable using definition. Lemma 2 (Properties of the Class of Reentrant Flow Lines) The class of reentrant flow lines considered satisfies the following properties: 1. The job set of $r$ is given by $J(r) = r \bullet \bullet \cap J = \bullet \bullet r \cap J$ for all resources $r \in R_{-0}$. 2. $p \bullet \bullet \cap R = \bullet \bullet p \cap R$ for all jobs $p \in J$. 3. If there are pallets for part path $j$, then $p_{j0} \bullet \bullet \cap R = \phi$, $\bullet \bullet p_{j0} \cap R = \phi$. (This follows directly from Definition 5 and Definition 7 item 2). 4. Let $p \in P = R \cup J$. Then $p \in p_\ell$ for some $p$-invariant $p_\ell$. That is, the flowline is covered by $p$-invariants. ### 5.1.2 Special Form of the Incidence Matrices. The reentrant flow line Definition and Lemma mean that the PN matrices in (5.2) have a particular form. Matrices $F_v, S^T_v$ consist of diagonal blocks, one per part path, which in $S^T_v$ are identity matrices, and in $F_v$ have a subdiagonal of 1’s. If there is assembly there will be some 1’s in $F_v$ below the diagonal blocks, where a 1 in element $(i,j)$ means that place $j$ is the last place in a partial part path and joins transition $i$ in another part path. Matrices $F_r, S^T_r$ are related as follows. If the $i$-th transition is not the last transition in a partial part path, and there is an entry of 1 in position $(i,j)$ of $F_r$, meaning resource $j$ is committed at transition $i$, then there is an entry of 1 in position $(i+1,j)$ of $S^T_r$, meaning that the resource is released at the next transition. If the $i$-th transition is the last transition in a partial part path, and there is an entry of 1 in position $(i,j)$ of $F_r$, then there is an entry of 1 in position $(k,j)$ of $S^T_r$, meaning that the resource is released at the assembly transition $k$. This structure results in a particularly convenient form of the PN incidence matrix $W = [S^T_v - F_v \ S^T_r - F_r] \equiv [W_v \ W_r]$. Block $W_v$ has diagonal blocks having 1’s on the diagonal and -1’s on the subdiagonal, with some -1’s below these blocks in the case of assembly operations. In each column, matrix $W_r$ has a -1 immediately followed by a 1, except in the case of assembly where the occurrence of the following 1 is shifted down to the assembly transition. In the case of shared resources, there is more than one -1,1 pair in the column. In columns corresponding to pallets, the 1 occurs at the beginning of the associated diagonal block of $W_v$ and the -1 at its end. ### 5.2 Algorithm for Computation of the $p$-Invariants For the reentrant flow line, an algorithm for determining all the $p$-invariants in a polynomial number of operations is given by the following theorem. Theorem 2 (Computation of a Set of Independent $p$-Invariants) Let there be given the PN matrices (5.2) for a flow line satisfying Definition 7, with places in the job vector $v$ ordered in the causal ordering specified in Section 5.1. Form matrices $\hat{F}_v, \hat{F}_r$ by deleting the rows of $F_v, F_r$ corresponding to the extra terminating transitions in each complete part path. Form matrices $\hat{S}_v, \hat{S}_r$ by deleting the columns of $S_v, S_r$ corresponding to the extra terminating transitions in each complete part path. Then, the complete set of $p$-invariants (resource loops) is given by the columns of the matrix $$P = \left[ \begin{array}{c} -(\hat{S}^T_v - \hat{F}_v)^{-1} (\hat{S}^T_r - \hat{F}_r) \\ I \end{array} \right]$$ \hspace{1cm} (5.4) where $I$ is the identity matrix. proof: The $p$-invariants are defined using (3.3) where $W$ is given by (5.2) and, for the reentrant flow line, $W_v, W_r$ have the special form noted in Section 5.1.2. This shows that the $p$-invariants are defined by $$[W_v \quad W_r] \begin{bmatrix} v \\ r \end{bmatrix} = 0,$$ with $v$ a vector of job places and $r$ a vector of resource places, or $$W_v v = -W_r r.$$ To construct a special left inverse of $W_v$ to solve this equation for $v$, delete the extra last transitions in the complete part paths to define $$\hat{W} = \hat{S}^T - \hat{F} = [\hat{S}_v^T - \hat{F}_v \quad \hat{S}_r^T - \hat{F}_r] \equiv [\hat{W}_v \quad \hat{W}_r].$$ This makes matrix $\hat{W}_v$ square. This is allowed as the deleted rows of $W_v$ are in the row space of the remaining rows. Then, the $p$-invariants are defined by $$\hat{W}_v v = -\hat{W}_r r,$$ so that $v = -\hat{W}_v^{-1} \hat{W}_r r$ for any $r$. To obtain a basis for nullspace $W$, set $r = I$, the identity, resulting in (5.4). It is required now to show that the resulting $v$ is binary. According to the discussion in Section 5.1.2 on the special structure of the DE matrices, $\hat{W}_v$ is lower block triangular with blocks on the diagonal corresponding to each part path and having the form $$\begin{bmatrix} 1 & 0 & 0 & 0 \\ -1 & 1 & 0 & 0 \\ 0 & -1 & 1 & 0 \\ 0 & 0 & -1 & 1 \end{bmatrix}.$$ The inverse of such a block is $$\begin{bmatrix} 1 & 0 & 0 & 0 \\ 1 & 1 & 0 & 0 \\ 1 & 1 & 1 & 0 \\ 1 & 1 & 1 & 1 \end{bmatrix}, \tag{a}$$ which appears as the corresponding diagonal block of $\hat{W}_v^{-1}$. In the case of assembly, there are some entries in $\hat{W}_v^{-1}$ below these diagonal blocks. Specifically, if there is a subdiagonal entry of -1 in position $(i, j)$ of $W_v$ the meaning is that there is a partial part path $j_1$ whose last place $j$ feeds into an assembly transition $i$ in a part path $j_2$. In this event, the lower off-diagonal block corresponding to the diagonal blocks $j_1$ and $j_2$ (e.g. block $(j_2, j_1)$) is zero, but filled with 1's on rows $i$ and below. Now one must turn to the structure of $-\hat{W}_r$. Since resources are always committed prior to their release, and all jobs have unity duration, the entries in any column of $-\hat{W}_r$ consist in the case of no assembly of 1's followed immediately by -1's. It is easy to see that such entries multiplied by blocks such as (a) always result in elements of 0 or 1 in $v$. In the case of an assembly with partial part path $j_1$ feeding into part path $j_2$, an entry of 1 on the row corresponding to the last transition of partial path $j_1$ is followed in any column $j$ by a -1 in row $i$, where transition $i$ is the assembly transition in path $j_2$. However, this corresponds to the beginning of the fill of 1's in block $(j_2, j_1)$ of $\hat{W}_v^{-1}$, and hence $\hat{W}_v^{-1} \hat{W}_r$ can be seen to yield only entries of 0 or 1 in $v$. Chapter 6 Conclusion We have shown by reduction from the One-In-3Sat problem that finding a binary basis for the nullspace of the p-invariant matrix is NP-complete in the general job shop problem. This implies that the job shop deadlock analysis problem will be at least as hard as the subproblem of finding the p-invariants. In the case of the reentrant flow line with assembly, however, we exhibited a closed-form solution for a binary basis. The importance of correctly selecting part flow and job routing protocols in flexible manufacturing systems is thereby conclusively demonstrated. The job routings and resource allocations should follow the structural protocols developed Section 5.1 to simplify the complexity of shared resource dispatching analysis of the FMS. [1] E.R. Boer and T. Murata, “Generating basis siphons and traps of Petri nets using the sign incidence matrix,” IEEE Trans. Circuits and Systems, vol. 41, no. 4, pp. 266-271, April 1994. [2] J.A. Buzacott and D.D. Yao, “Flexible manufacturing systems: a review of analytical models,” Management Sci., vol. 32, no. 7, pp. 890-905, July 1986. [3] A.A. Desrochers, Modeling and Control of Automated Manufacturing Systems, IEEE Computer Society Press, 1990. [4] E.A. Elsayed and T.O. Boucher, Analysis and Control of Production Systems, 2nd ed., Prentice-Hall, Englewood Cliffs, NJ, 1994. [5] S.D. Eppinger, D.E. Whitney, and R.P. Smith, “Organizing the tasks in complex design projects,” Proc. ASME Int. Conf. Design Theory and Methodology, pp. 39-46, Sep. 1990. [6] J. Ezpeleta, J.M. Colom, and J. Martinez, “A Petri net based deadlock prevention policy for flexible manufacturing systems,” IEEE Trans. Robotics and Automation, vol. 11, no. 2, pp. 173-184, Apr. 1995. [7] M.R. Garey, and D.S. Johnson, Computers and Intractability: a Guide to the Theory of NP-completeness. Freeman, San Francisco, CA, 1979. [8] S.C. Graves, “A review of production scheduling,” Operations Research, vol. 29, pp. 646-675, Aug. 1981. [9] M.D. Jeng and F. DiCesare, “A synthesis method for Petri net modeling of automated manufacturing systems with shared resources,” Proc. IEEE Conf. Decision and Control, pp. 1184-1189, Dec. 1992. [10] M.D. Jeng and F. DiCesare, “Synthesis using resource control nets for modeling shared-resource systems,” IEEE Trans. Robotics and Automation, vol. 11, no 3, pp. 317-327, June 1995. [11] B.H. Krogh and L.E. Holloway, “Synthesis of feedback control logic for discrete manufacturing systems,” Automatica, vol. 27, no. 4, pp. 641-651, July 1991. [12] P.R. Kumar and S.P. Meyn, “Stability of queueing networks and scheduling policies,” Proc. IEEE Conf. Decision and Control, pp. 2730-2735, Dec. 1993. [13] A. Kusiak, “Intelligent scheduling of automated machining systems,” in Intelligent Design and Manufacturing, ed. A. Kusiak, New York: Wiley, 1992. [14] F. L. Lewis, H.-H. Huang, A. Gürel, O.C. Pastravanu, and D. Tacconi, “Analysis of deadlocks and circular waits using a matrix-based approach,” submitted, 1995. [15] S.H. Lu and P.R. Kumar, “Distributed scheduling based on due dates and buffer priorities,” IEEE Trans. Automat. Control, vol. 36, no. 12, pp. 1406-1416, Dec. 1991. [16] T. Murata, “Petri nets: properties, analysis and applications,” Proc. IEEE, vol. 77, no. 4, pp. 541-580, Apr. 1989. [17] T. Murata, N. Komoda, K. Matsumoto, and K. Haruna, “A Petri net-based controller for flexible and maintainable sequence control and its applications in factory automation,” IEEE Trans. Ind. Electronics, vol. IE-33, no. 1, pp. 1-8, Feb. 1986. [18] Peterson, J.L. (1981), Petri Net Theory and the Modeling of Systems, Prentice-Hall, Englewood Cliffs, NJ. [19] Ramadge, R.J.G. and W.M. Wonham, “The control of discrete event systems,” Proceedings of the IEEE, vol. 77, no. 1, pp. 81–98, 1989. [20] D.V. Steward, “On an approach to techniques for the analysis of the structure of large systems of equations,” SIAM Review, vol. 4, no. 4, pp. 321-342, Oct. 1962. [21] J.N. Warfield, “Binary matrices in system modeling,” IEEE Trans. Systems, Man, Cybern., vol. SMC-3, no. 5, pp. 441-449, Sept. 1973. [22] J. Wolter, S. Chakrabarty, and J. Tsao, “Methods of knowledge representation for assembly planning,” Proc. NSF Design and Manuf. Sys. Conf., pp. 463-468, Jan. 1992. [23] M.-C. Zhou, F. DiCesare, A.D. Desrochers, “A hybrid methodology for synthesis of Petri net models for manufacturing systems,” IEEE Trans. Robotics and Automation, vol. 8, no. 3, pp. 350-361, Jun. 1992.
Strong Controllability of Temporal Networks with Decisions Matteo Zavatteri\textsuperscript{1}, Romeo Rizzi\textsuperscript{1}, and Tiziano Villa\textsuperscript{1} \textsuperscript{1}University of Verona, Italy \textsuperscript{1}\{matteo.zavatteri,romeo.rizzi,tiziano.villa\}@univr.it Abstract A number of extensions of simple temporal networks have been proposed over the last years to face several sources of uncertainty, either in isolation or simultaneously. This paper focuses on a hierarchy of simple temporal networks where the top-level formalism is that of conditional simple temporal networks with uncertainty and decisions (CSTNUDs), a formalism dealing with controllable and uncontrollable durations and controllable and uncontrollable conditional constraints simultaneously. We propose an algorithm to check strong controllability of CSTNUDs. We prove that strong controllability of temporal networks in this hierarchy is NP-complete if controllable conditional constraints are considered. 1 Introduction: A Hierarchy of Simple Temporal Networks Temporal networks are a framework to model temporal plans and check the coherence of their temporal constraints imposing minimal and maximal temporal distances between the occurrences of the events in the plan. Temporal plans mainly divide in plans having everything under control and plans having something out of control. The main components of a temporal network are time points and constraints. Time points are real variables modeling events. Events occur when their corresponding time points are executed (i.e., assigned values). Constraints are linear inequalities, and more precisely difference inequalities over pairs of events, bounding the minimal and maximal temporal distance between the respective scheduling times. When events and constraints are fixed in advance, we deal with a standard consistency problem (STN). Definition 1 (STN). A simple temporal network (STN) is a pair $(\mathcal{T}, \mathcal{C})$, where $\mathcal{T} = \{X, \ldots, Z\}$ is a finite set of time points and $\mathcal{C}$ is a finite set of temporal constraints $Y - X \leq k$, where $Y, X \in \mathcal{T}$ and $k \in \mathbb{R}$. STNs model fully controllable temporal plans. An STN is consistent if there exists a schedule $t : \mathcal{T} \to \mathbb{R}$ satisfying all constraints. We write $t(X)$ for the time at which $X$ was executed. Consistency of STNs is in P [11]. However, STNs cannot deal with controllable and uncontrollable conditional constraints nor with uncontrollable durations. To bridge such gaps simple temporal networks with uncertainty (STNUs, [18]) and conditional simple temporal networks (CSTNs, [14]) were proposed. Definition 2 (STNU). A simple temporal network with uncertainty (STNU) is a triple $(\mathcal{T}, \mathcal{L}, \mathcal{C})$ extending an STN with a finite set of contingent links $\mathcal{L}$ each one having the form $(A, x, y, C)$ where $A, C \in \mathcal{T}$ and $x, y \in \mathbb{R}$ with $0 < x < y < \infty$. $A$ is the activation time point and it is under control, $C$ is the contingent one and it is not. Once $A$ is executed, the execution time of $C$ is revealed by Nature at at $t(C)$ guaranteeing that $t(C) \in [t(A) + x, t(A) + y]$. Contingent links do not share contingent time points. STNUs model temporal plans with uncontrollable (but bounded) durations. STNUs (and all other formalisms specifying uncontrollable parts) bring with them three main notions of controllability: weak, strong and dynamic. Weak controllability is when, for each combination of uncontrollable parts known in advance, there exists a way to operate on the controllable part satisfying all constraints. Strong controllability is the opposite case and says that there exists a way to operate on the controllable part satisfying all constraints no matter what will happen. Dynamic controllability requires the existence of a strategy refining how we operate on the controllable part in real time depending on what is going on. STNUs do not specify controllable nor uncontrollable conditional constraints. To model uncontrollable conditional constraints in isolation, the formalism of CSTNs [13] (formerly conditional temporal problem (CTP, [17])) was proposed and subsequently extended to the formalism of CSTNUs [12, 13] in order to augment CSTNs with uncontrollable durations. Conditionals are expressed as labels, conjunctions of literals over a finite set of Boolean variables saying when the components labeled by them are relevant. Initially, labels were on both time points and constraints but later it was proved that having labels on constraints only does not limit the expressiveness of the network [3]. Temporal networks with labels on constraints only are called streamlined. In what follows, we will only consider streamlined networks. Let $ \mathcal{B} = \{a, b, \ldots, z\} $ be a finite set of Boolean variables, a label $ \ell = \lambda_1 \ldots \lambda_m $ is any finite conjunction of literals $ \lambda_i $ over the variables in $ \mathcal{B} $ (we omit the $ \land $ connective to ease reading). The empty label is denoted by $ \Box $. The label universe of $ \mathcal{B} $, denoted by $ \mathcal{B}^* $, is the set of all possible (consistent) labels drawn from $ \mathcal{B} $. For instance, if $ \mathcal{B} = \{a, b\} $, then $ \mathcal{B}^* = \{\Box, a, b, \neg a, \neg b, ab, a \neg b, \neg a b, \neg a \neg b\} $. Two labels $ \ell_1, \ell_2 \in \mathcal{B}^* $ are consistent if and only if their conjunction $ \ell_1 \ell_2 $ is satisfiable. Definition 3 (CSTN/CSTNU). A conditional simple temporal network (CSTN) is a tuple $ \langle T, O, B, O, C \rangle $ extending an STN with a finite set of observation time points $ O \subseteq T = \{A?, \ldots, Z?\} $, a finite set of Boolean variables $ \mathcal{B} $, a bijection $ O : \mathcal{B} \to O $ assigning a unique Boolean variable to each observation point (we write $ O^{-1} : O \to \mathcal{B} $ for the inverse), and turning $ C $ into a finite set of conditional constraints $ \ell \to Y - X \leq k $ each meaning that if $ \ell $ is true, then $ Y - X \leq k $ must hold; with $ \ell \in \mathcal{B}^* $, $ Y, X \in T $ and $ k \in \mathbb{R} $. Once we execute an observation time point $ A? $, Nature sets instantaneously the uncontrollable truth value of the associated Boolean variable $ O^{-1}(A?) = a $. A conditional simple temporal network with uncertainty (CSTNU) is a tuple $ \langle T, O, B, O, L, C \rangle $ extending a CSTN with a finite set of contingent links. CSTNUs deal with controllable and uncontrollable durations and uncontrollable conditionals simultaneously. However, they fail to model controllable conditionals. To bridge this gap conditional simple temporal network with uncertainty and decisions (CSTNUDs, [19, 27]) were recently proposed. Definition 4 (CSTNUD). A conditional simple temporal network with uncertainty and decisions (CSTNUD) is a tuple $ \langle T, O, D, B, O, L, C \rangle $, extending a CSTNU with a finite set of decision time points $ D = \{D!, \ldots, Z!\} $ and in which the set of Boolean variables $ \mathcal{B} $ is partitioned in two disjoint subsets $ B_D \cup B_O $, representing the set of controllable and uncontrollable Boolean variables, respectively. The mapping $ O $ is turned into a bijection $ O : \mathcal{B} \to O \cup D $. Once we execute a decision time point $ D! $, we also set instantaneously the controllable truth value of the associated Boolean variable $ O^{-1}(D!) = d $. A CSTNUD where $ L = \emptyset $ is a conditional simple temporal network with decisions (CSTND, [2, 19, 27]). A CSTNUD where $ O = \emptyset $ is a simple temporal network with uncertainty and decisions (STNUD, [19, 27]). A CSTNUD where $ L = O = \emptyset $ is a simple temporal network with decisions (STND, [2, 19, 27]). Fig. 1a shows such a hierarchy. We graphically represent temporal networks as directed graphs whose sets of nodes coincide with the set of time points (decision and observation time points are suffixed with ! and ?, respectively). A double red edge $ A \Rightarrow C $ labeled by $ [x, y] $ models a contingent links $ \langle A, x, y, C \rangle $. An edge $ X \rightarrow Y $ labeled by $ \langle k, \ell \rangle $ models a constraint $ \ell \to Y - X \leq k $. If $ \ell = \Box $ we just use $ k $ as a label. ## 2 Strong Controllability of CSTNUDs: Super Projections Strong controllability of STNUs, CSTNs and CSTNUs is in P [4, 17, 18] and boils down to check the consistency of a reduced STN. For STNUs the algorithm in [18] computes an STN by “rewriting” all constraints involving contingent time points as constraints knowing the corresponding activation only. For CSTNs the algorithm in [17] checks consistency of the underlying STN ignoring all labels. For CSTNUs Figure 1: A hierarchy of temporal networks (a), an STNU (b), its STN super-projection (c), a CSTNUD (d) its super-projections (e,f) and STND projections (g,h). We highlight uncontrollable parts in red. both methods are applied one after the other: first conditionals, then contingent links [1]. In this section we propose an alternative (simpler reduction) to check strong controllability of STNUs and then we extend it to CSTNUDs as an initial approach. Consider the STNU in Fig. 1b. That network is strongly controllable. A strong schedule is $t(B) = 0$, $t(D) = 1$, $t(A) = 4$ because if the contingent link $(A, 2, 4, C)$ takes its minimal duration, then $C$ will occur at $t(C) = t(A) + 2 = 6$ and the constraint requiring that $C$ happens at least 5 since $D$ will be satisfied. We claim that an STNU is strongly controllable iff its super-projection is consistent. The super-projection is an STN obtained by applying the simple local substitution reduction illustrated in Fig. 1c. The super-projection is constructed as follows. We generate an STN whose set of time points consists of all non-contingent time points of the original STNU plus as many pairs of time points $C_m$ and $C_M$ as the number of contingent time points $C$ in the original STNU. The set of constraints is generated as follows. All original constraints not involving contingent time points belong to this STN. For each contingent link $(A, x, y, C)$ in the original STNU, we add four constraints $C_m - A \leq x$, $A - C_m \leq -x$, $C_M - A \leq y$, $A - C_M \leq -y$ in order to enforce that the corresponding $C_m$ and $C_M$ are executed exactly $x$ and $y$ after $A$, respectively. Each original constraint involving a contingent time point $C$ is duplicated in a pair of constraints: one involving $C_m$ and the other involving $C_M$. To give an example, consider Fig. 1b and note that the original $(A, 2, 4, C)$ is modeled in Fig. 1c by a triple of time points $A, C_m$ and $C_M$ plus $C_m - A \leq 2$, $A - C_m \leq -2$, $C_M - A \leq 4$, $A - C_M \leq -4$ and that the original constraint $D - C \leq -5$ is modeled by $D - C_m \leq -5$ and $D - C_M \leq -5$. Algorithm 3 extends this reduction in order to deal with STNUDs. To patch it for getting $\text{StnuSP}(N)$ we make these modifications. The input and output become “An STNU $N = (\mathcal{T}, \mathcal{L}, \mathcal{C})$” and “The STN super-projection $N^* = (\mathcal{T}^, \mathcal{C}^)$”. Lines 8-11 remain the same without any “$\ell \rightarrow$” (note that line 7 omits “$\Box \rightarrow$” on constraints rewriting contingent links). The return statement becomes “$(\mathcal{T}^, \mathcal{C}^)$”. Theorem 1. Any STNU is strongly controllable if and only if its super-projection is consistent. Proof. Let $N = (\mathcal{T}, \mathcal{L}, \mathcal{C})$ be any STNU and $N^* = (\mathcal{T}^, \mathcal{C}^)$ its super-projection. Assume that $N$ is strongly controllable. Let $\mathcal{T}_X = \mathcal{T} \cap \mathcal{T}^$ be the set of non-contingent time points and let $\mathcal{T}_C = \mathcal{T} \setminus \mathcal{T}_X$ be the set of contingent time points $N$. Let $t : \mathcal{T}_X \rightarrow \mathbb{R}$ be a feasible scheduling of $N$ (i.e., a scheduling satisfying all constraints no matter which durations Nature chooses for contingent links). Let $t^$ be the extension of $t$ on the new domain $\mathcal{T}^$ (i.e., set of time points of $N^$) defined by the following three rules: Rule 1: $t^$ is an extension of $t$, thus, for each non-contingent time point $X \in \mathcal{T}_X$, $t^(X) = t(X)$. Rule 2: if $X = C_m$ for some contingent node $C \in \mathcal{T}_C$, then $t^(X) = t(\text{act}(C)) + L(C)$. Rule 3:* If $X = C_M$ for some contingent node $C \in T_C$, then $t^(X) = t(act(C)) + U(C)$. where $act(C)$ is the activation time point of $C$ (in the original STNU) and $L()$ and $U()$ are the lower and upper bound vectors on the delays of contingent time points (i.e., all minimal and all maximal durations). Fact 1* For each constraint $Y - X \leq k \in C$ where $Y$ and $X$ are not contingent time points, then $Y - X \leq k \in C^$, $t^(Y) - t^(X) \leq k$ holds by assumption and since by Rule 1 $t(Y) = t^(Y)$ and $t(X) = t^(X)$, then $t^(Y) - t^(X) \leq k$ holds as well. Fact 2* For each contingent link $(A, x, y, C) \in L$, $t^(C_m) \leq t(C) \leq t^(C_M)$ because of Rules 2 and 3. Fact 3 For each constraint $X - C \leq -k \in C$, with $C$ a contingent time point, $t(X) - t(C) \leq -k$ holds by assumption and $t^(X) - t^(C_m) \leq -k$ and $t^(X) - t^(C_M) \leq -k$ hold because of Fact 2. Fact 4 For each constraint $C - X \leq k \in C$, with $C$ a contingent time point, $t(C) - t(X) \leq k$ holds by assumption and $t^(C_m) - t^(X) \leq k$ and $t^(C_M) - t^(X) \leq k$ hold because of Fact 2. Therefore, if $t$ is a feasible schedule for $\mathcal{N}$, then $t^$ is a feasible schedule for $\mathcal{N}^$ (Facts 1, 2, 3 and 4). Likewise, if $t^$ is a feasible schedule for $\mathcal{N}^$, then $t^\|_X$ is a feasible schedule for $\mathcal{N}$, where $t^\|_X$ is the projection of $t^$ over the non-contingent time points of $\mathcal{N}$. Indeed, let $d : T_C \rightarrow \mathbb{R}$ be a delay vector assigning a duration to each contingent time point $C \in T_C$. For every possible delay vector $d$ and for every possible contingent time point $C$ of $\mathcal{N}$, it holds that $t(C_m) = t(act(C)) + L(C) \leq t(act(C)) + d(C) \leq t(act(C)) + L(C) = t(C_M)$ since $L \leq d \leq U$. That is, we only need to cope with delay vectors which are sandwiched in-between the lower bound vector and the upper bound vector. Without loss of generality consider a binary constraint $Y - C \leq k$ involving $C$ in the original STNU $\mathcal{N}$. Since $t^(Y) - t^(C_m) \leq k$ and $t^(Y) - t^(C_M) \leq k$ hold, then the original $t(Y) - t(C) \leq k$ holds as well. □ Likewise, a CSTNUD is strongly controllable if and only if its STND super projection is consistent (the proof extends that of Theorem 1 to accommodate uncontrollable conditionals as well). Algorithm 1 provides an approach to solve strong controllability of CSTNUDs. Algorithm 1 first applies Algorithm 2 to get rid of uncontrollable conditional constraints, then applies Algorithm 3 to rewrite contingent links maintaining controllable conditional constraints and finally calls a StndSolver (see [23] for some possible algorithms) to check consistency of the STND super-projection (i.e., to find a truth value assignment to the controllable Boolean variables such that the STN-projection according to that assignment is consistent). Due to lack of space we only show a graphical execution of Algorithm 1 that starting on the input Fig. 1d, first removes uncontrollable conditionals getting Fig. 1e, then rewrites contingent links getting Fig. 1f and finally returns a solution for Fig. 1h which is the only consistent STND-projection (note the negative loop between $B$ and $D$ in Fig. 1g, the STND-projection according to $d$). Thus, we fix $\neg d$ as a decision and get the strong schedule $t(B) = 0$, $t(D) = 2$, $t(A) = 5$ for Fig. 1d. Algorithm 1: CstnudStrongSchedule($\mathcal{N}$)* - Input: A CSTNUD $\mathcal{N} = (\mathcal{T}, \mathcal{O}, \mathcal{D}, \mathcal{B}, O, \mathcal{L}, C)$ - Output: A strong temporal plan - 1 return StndSolver(StnudSP(CstnudSP($\mathcal{N}$))) Algorithm 2: CstnudSP($\mathcal{N}$) - Input: A CSTNUD $\mathcal{N} = (\mathcal{T}, \mathcal{O}, \mathcal{D}, \mathcal{B}, O, \mathcal{L}, C)$ - Output: The STNUD super-projection $\mathcal{N}^$ - 1 for $\ell \in \mathcal{B}_O$ do - 2 $O^+(\ell) = O(\ell)$ - 3 $C^ \leftarrow \emptyset$ - 4 for $\ell \rightarrow Y - X \leq k \in C$ do - 5 Let $\ell_D$ be $\ell$ without literals over vars in $\mathcal{B}_O$ - 6 $C^* \leftarrow C^* \cup \{\ell_D \rightarrow Y - X_i \leq k\}$ - 7 return $(\mathcal{T}, \mathcal{D}, \mathcal{B}_D, O^-, O^+, \mathcal{L}, C^)$ Algorithm 3: StnudSP($\mathcal{N}$)* - Input: An STNUD $\mathcal{N} = (\mathcal{T}, \mathcal{D}, \mathcal{B}, O, \mathcal{L}, C)$ - Output: The STND super-projection $\mathcal{N}^$ - 1 $T^ \leftarrow \emptyset$ - 2 for $X \in \mathcal{T}$ do - 3 if $X$ is non-contingent then $\Delta(X) = \{X\}$; - 4 else $\Delta(X) = \{X_m, X_M\}$; - 5 $T^* \leftarrow T^* \cup \Delta(X)$ - 6 for $(A, x, y, C) \in \mathcal{L}$ do - 7 $C^* \leftarrow C^* \cup \{C_m - A_i \leq x, A_i - C_m \leq -x,$ - 8 $C_M - A_i \leq y, A_i - C_M \leq -y\}$ - 9 for $\ell \rightarrow Y - X \leq k \in C$ do - 10 for $Y_i \in \Delta(Y)$ do - 11 $C^* \leftarrow C^* \cup \{\ell \rightarrow Y_i - X_i \leq k\}$ - 12 return $(T^, \mathcal{D}, \mathcal{B}, O, C^)$ Theorem 2. Strong controllability of STNUDs, CSTNDs and CSTNUDs is NP-complete. Proof. Hardness: Consequence of the fact that deciding consistency of STNDs is NP-hard and it is a special case of deciding strong controllability of STNUds, CSTNDs and CSTNUds (when $ \mathcal{L} = \emptyset $, $ \mathcal{O} = \emptyset $ and $ \mathcal{L} = \mathcal{O} = \emptyset $, respectively). Membership: Regardless of the formalism, a certificate of YES consists of a truth value assignment $ s $ for $ \mathcal{B}_D $ and a schedule $ t $ for $ T_X $ (set of non-contingent time points). To verify it we use the reductions in Algorithm 2 to get rid of $ \mathcal{B}_C $ (if $ \neq \emptyset $) and Algorithm 3 to rewrite $ \mathcal{L} $ (if $ \neq \emptyset $). If $ \mathcal{L} \neq \emptyset $, then for each $ (A, x, y, C) \in \mathcal{L} $, $ t(C_m) = t(A) + x $ and $ t(C_M) = t(A) + y $. Finally, we verify that $ (s, t) $ is a YES certificate for the resulting STNDs. We know that this check is polynomial [2]. ### 3 Conclusions and Future Work We provided a simple approach to check strong controllability of CSTNUds based on an elementary local reduction. The resulting algorithm’s running time is upper bounded by (i) the product of a low-degree polynomial where the magnitude of the numbers does not occur and (ii) a singly exponential term of the form $ 2^{\|D\|} $, where $ D $ is the set of binary decisions. With small modifications the algorithm we gave can be adapted to all formalisms in Fig. 1a. The idea is to reduce a strong controllability problem to a consistency problem by computing super-projections. All formalisms in Fig. 1a are reducible to their super-projections in polynomial time. Strong controllability of STNUds, CSTNDs, CSTNUds is NP-complete. As future work, we plan to deepen research on complexity of strong controllability for these and for other classes of (temporal)-constraint networks such as those discussed (or employed) in [8, 9, 10, 20, 21, 22, 24, 25, 26, 28, 29] other than comparing with strong controllability of disjunctive formalisms [4, 5, 6, 7, 15]. Also, a line of research started in [16] provided an approach for the optimal design of consistent STNs. Considering the new way to compute a super-projection of an STNU, this line of research can be extended to provide an approach for the optimal design of strongly controllable STNUs. ### References [1] N. Bhargava and B. C. Williams. Complexity bounds for the controllability of temporal networks with conditions, disjunctions, and uncertainty. Artificial Intelligence, 271:1–17, 2019. [2] M. Cairo, C. Combi, C. Comin, L. Hunsberger, R. Posenato, R. Rizzi, and M. Zavatteri. Incorporating decision nodes into conditional simple temporal networks. In TIME 2017, volume 90 of LIPIcs, pages 9:1–9:17. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2017. [3] M. Cairo, L. Hunsberger, R. Posenato, and R. Rizzi. A streamlined model of conditional simple temporal networks - semantics and equivalence results. In TIME 2017, volume 90 of LIPIcs, pages 10:1–10:19. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2017. [4] A. Cimatti, M. Do, A. Micheli, M. Roveri, and D. E. Smith. Strong temporal planning with uncontrollable durations. Artificial Intelligence, 256:1–34, 2018. [5] A. Cimatti, L. Hunsberger, A. Micheli, R. Posenato, and M. Roveri. Sound and complete algorithms for checking the dynamic controllability of temporal networks with uncertainty, disjunction and observation. In TIME 2014, pages 27–36. IEEE CPS, 2014. [6] A. Cimatti, L. Hunsberger, A. Micheli, R. Posenato, and M. Roveri. Dynamic controllability via timed game automata. Acta Informatica, 53(6-8):681–722, 2016. [7] A. Cimatti, A. Micheli, and M. Roveri. Dynamic controllability of disjunctive temporal networks: Validation and synthesis of executable strategies. In AAAI 2013, pages 3116–3122. AAAI Press, 2016. [8] C. Combi, R. Posenato, L. Viganò, and M. Zavatteri. Access controlled temporal networks. In ICAART 2017, pages 118–131. INSTICC, ScitePress, 2017. [9] C. Combi, R. Posenato, L. Viganò, and M. Zavatteri. Conditional simple temporal networks with uncertainty and resources. Journal of Artificial Intelligence Research, 64:931–985, 2019. [10] C. Combi, L. Viganò, and M. Zavatteri. Security constraints in temporal role-based access-controlled workflows. In CODASPY 2016, pages 207–218. ACM, 2016. [11] R. Dechter, I. Meiri, and J. Pearl. Temporal constraint networks. Artif. Intell., 49(1-3):61–95, 1991. [12] L. Hunsberger and R. Posenato. Sound-and-complete algorithms for checking the dynamic controllability of conditional simple temporal networks with uncertainty. In TIME 2018, volume 120 of LIPIcs, pages 14:1–14:17. Schloss Dagstuhl - Leibniz-Zentrum fuer Informatik, 2018. [13] L. Hunsberger, R. Posenato, and C. Combi. The Dynamic Controllability of Conditional STNs with Uncertainty. In PlanEx at ICAPS-2012, pages 1–8, 2012. [14] L. Hunsberger, R. Posenato, and C. Combi. A sound-and-complete propagation-based algorithm for checking the dynamic consistency of conditional simple temporal networks. In TIME 2015, pages 4–18. IEEE CPS, 2015. [15] A. Micheli. Disjunctive temporal networks with uncertainty via SMT: recent results and directions. Intelligenza Artificiale, 11(2):155–178, 2017. [16] R. Rizzi and R. Posenato. Optimal design of consistent simple temporal networks. In TIME 2013, pages 19–25. IEEE CPS, 2013. [17] I. Tsamardinos, T. Vidal, and M. E. Pollack. CTP: A new constraint-based formalism for conditional, temporal planning. Constraints, 8(4):365–388, 2003. [18] T. Vidal and H. Fargier. Handling contingency in temporal constraint networks: from consistency to controllabilities. Journal of Experimental & Theoretical Artificial Intelligence, 11(1):23–45, 1999. [19] M. Zavatteri. Conditional simple temporal networks with uncertainty and decisions. In TIME 2017, volume 90 of LIPIcs, pages 23:1–23:17. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2017. [20] M. Zavatteri. Temporal and Resource Controllability of Workflows Under Uncertainty. PhD thesis, University of Verona, Italy, 2018. [21] M. Zavatteri. Temporal and resource controllability of workflows under uncertainty. In Proceedings of the Dissertation Award, Doctoral Consortium, and Demonstration Track at BPM 2019, volume 2420 of CEUR Workshop Proceedings, pages 9–14. CEUR-WS.org, 2019. [22] M. Zavatteri, C. Combi, R. Posenato, and L. Viganò. Weak, strong and dynamic controllability of access-controlled workflows under conditional uncertainty. In BPM 2017, pages 235–251. Springer, 2017. [23] M. Zavatteri, C. Combi, R. Rizzi, and L. Viganò. Hybrid sat-based consistency checking algorithms for simple temporal networks with decisions. In TIME 2019, volume 147, page 2:1–2:17. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2019. [24] M. Zavatteri, C. Combi, and L. Viganò. Resource controllability of workflows under conditional uncertainty. In Business Process Management Workshops, pages 68–80. Springer, 2019. [25] M. Zavatteri, R. Rizzi, and T. Villa. Complexity of weak, strong and dynamic controllability of encus. In OVERLAY 2019 (to appear). CEUR-WS.org, 2019. [26] M. Zavatteri and L. Viganò. Constraint networks under conditional uncertainty. In ICAART 2018, pages 41–52. SciTePress, 2018. [27] M. Zavatteri and L. Viganò. Conditional simple temporal networks with uncertainty and decisions. Theoretical Computer Science, 797:77–101, 2019. [28] M. Zavatteri and L. Viganò. Conditional uncertainty in constraint networks. In Agents and Artificial Intelligence, pages 130–160. Springer, 2019. [29] M. Zavatteri and L. Viganò. Last man standing: Static, decremental and dynamic resiliency via controller synthesis. Journal of Computer Security, 27(3):343–373, 2019.
Land Reform, Collectivisation and the Peasants in North Korea Author(s): Chong-Sik Lee Source: The China Quarterly, No. 14 (Apr. - Jun., 1963), pp. 65-81 Published by: Cambridge University Press on behalf of the School of Oriental and African Studies Stable URL: http://www.jstor.org/stable/651343 Accessed: 27/04/2011 15:00 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=cup. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact firstname.lastname@example.org. Land Reform, Collectivisation and the Peasants in North Korea By CHONG-SIK LEE In a speech delivered at the rally commemorating the fifteenth anniversary of the liberation of Korea, the North Korean Premier announced that the peasants in North Korea were now the owners of large-scale collectivised farms and that they had the firm technical foundation for bumper crops every year without strenuous efforts. He declared: "This is the beginning of a world for our farm villages." Another spokesman of the North Korean régime has stated: "It is easy (or good) to work and enjoyable to live in the co-operativised North Korean farms. There is a bumper crop every year in the constantly changing collectivised fields and the peasants' work and living are literally song and dances." Careful reading of even the propaganda materials reveals, however, that the millennium is yet to come in North Korea and that the régime has been undergoing a considerable degree of strain in the agricultural sphere. Peasants have been under the severest regimentation and the life of "song and dances" is a remote possibility, if that. What, then, is the actual condition of North Korean peasants? What transformation have North Korean farms undergone, and how have the changes been accomplished? What is the lot of the peasants on the "co-operativised" farms in North Korea? When the Russian army marched into North Korea in August 1945 it faced a number of major problems in connection with the task of occupation. One of these was establishing a régime of native personnel sympathetic to Communism, and another was clothing this régime with a semblance of legitimacy and public support. The first problem was solved with relative ease. The Russian command initially endorsed a popular local leader, Cho Man-sik, as the head of the native governing body over its entire zone of occupation. When the institution of the Five Provinces Administration Bureau --- 1 Kim Il-Song, Choson inmin ui minjokjok myongiol 8.15 haebang 15 chunyon kyongch'uk taehoe eso han pogo. (A Report Delivered at the Rally Commemorating the Fifteenth Anniversary of the Liberation of August 15 which is the national Holiday of the Korean People) (Pyongyang: Korean Workers' Party Press, 1960), p. 7. Kim Kwang-hyon, in his Ch'ollima Choson (Korea of the Flying Horse) (Pyongyang: National Press, 1961), p. 61. gained a degree of legitimacy in North Korea, Cho was replaced by Kim Il-song, the Russian choice for North Korean leadership. For the second and the more difficult task of winning popular support for the Russian-endorsed régime, such progressive measures as land reform, promulgation of the labour ordinance, which stipulated an eight-hour work day for workers, and the implementation of free and compulsory education were used. The land reform programme served a particularly important political function for the Russian authorities. In spite of the heavier concentration of industry in North Korea, the Russian zone of occupation was still predominantly agricultural. According to a recent North Korean source, 74-1 per cent. of the North Korean population in 1946 was engaged in agriculture.$^2$ But the majority of the peasants had little or no land of their own. According to Kim Il-song’s report of 1947, 68 per cent. of the farmers held 54 per cent. of the tilled land in North Korea, while 56-7 per cent. were classified as “poor farmers” and held only 5-4 per cent. of the land.$^3$ Except for approximately twenty per cent., who were self-employed, the North Korean farmers belonged either to the “pure” tenant-farmer category and had no land of their own or to the “poor farmer” category and worked their own small plots along with the land held by wealthier landlords. This lopsided distribution of land, coupled with an iniquitous system of farm rent (farm rent averaged from 50 per cent. to 60 per cent. of the crops, often mounting to 70 per cent.) had caused much unrest among the peasants under Japanese rule. Many peasants were forced to abandon their farms and tenancies to migrate to Manchuria. Strikes of tenant farmers had been frequent throughout Korea, even though the Japanese government discouraged them. For these reasons, Communist agitation and propaganda among the peasants during the 1930s, particularly in the more mountainous Hamgyong Pukto, was very effective.$^4$ Although the situation in the other provinces in North Korea was considerably better than that in Hamgyong Pukto, grievances did exist among the poorer elements of the agricultural population, and the Communist-led “peasants union” movement had a fairly strong following. It is not surprising, therefore, to find that the Russian $^2$ Facts about Korea (Pyongyang: Foreign Languages Publishing House, 1961), p. 9. $^3$ Kim Il-song, “What Should the Parties and Social Groups Demand on the Occasion of the Establishment of the Democratic People’s Republic of Korea,” dated June 14, 1947, in Kin Nichisei Senshu (Selected Works of Kim Il-song), (Kyoto: Sanichi Shobo, 1952), I, p. 217. In his speech of April 20, 1948, Kim revised the statistics to 68 per cent. of the landlords holding 58-8 per cent. of land: ibid. Supp. Vol., p. 114. $^4$ See Thought Section, Prosecutor’s Bureau, High Court, Shiso iho (Ideological Report Series), No. 11, (June 1937), pp. 146-170. authorities in North Korea assigned priority to winning the support of the peasants and that the land reform programme was employed for this task. On February 8, 1946, the North Korean Interim People's Committee was established in Pyongyang after a conference of the "political parties, social organisations and local people's committees." This conference also decided upon the "impending duties" of the new Interim People's Committee, which, of course, included the land reform programme. The second of the eleven articles stated: "Preparations [will] be made to dispose of, within a short period, the land and forests confiscated from the Japanese aggressors and pro-Japanese reactionary elements. Land and forests subject to confiscation from Korean landlords will be nationalised, the land-tenant system will be abolished, and the land will be distributed to the peasants without compensation." On March 5, the committee, ostensibly adopting the proposals of the Peasants Federation which had demanded the agricultural reform, proclaimed the ordinance on land reform. The ordinance stipulated that all land formerly possessed by Japanese imperialists (including the government, civilians and organisations), national traitors, Korean landlords with more than five chongbo (a chongbo is 2.45 acres), absentee landlords, and churches, monasteries and other religious organisations with more than five chongbo, and all land continuously held in tenancy, be confiscated without compensation and be distributed without charge to landless peasants or peasants with little land. The ordinance further stipulated that all debts owed by peasants to landlords would be voided and that all farm animals, farm machinery, and houses formerly owned by landlords should be confiscated and then distributed among the peasants. The government, however, retained the option of transferring all the buildings to schools, hospitals, and other social groups. Further, the peasants were prohibited from selling, renting, or mortgaging their newly acquired land. Irrigation facilities formerly owned by expropriated landlords and all forests, except small forests owned by peasants, were confiscated and transferred to state ownership. 5 "Political and Organisational Policies of the Workers' Party," Dec. 25, 1947, Kin Nichiei Senshu, I, pp. 264-265. 6 Kwahakwon Yoksa Yongsuo (Academy of Science, Centre for Historical Studies), Choson tongsa (Outline History of Korea) (Pyongyang: Academy of Science Press, 1958) [Hak-u Sobang reprint edition, Tokyo, 1959], III, p. 31. 7 Ibid. p. 32. For an English text of the land reform ordinance see Report of the United Nations Commission for the Unification and Rehabilitation of Korea (New York: United Nations, General Assembly, Official Records: Sixth Session, Supplement No. 12 (A/1881), 1951), pp. 59-60. Table I. Origins of the Confiscated Land, 1946 \| Owners \| Area (chongbo) \| \|----------------------------------------------------------------------\|----------------\| \| Japanese Government, nationals and organisations \| 100,797 \| \| National traitors \| 21,718 \| \| Landlords with more than five chongbo \| 285,692 \| \| Landlords who rent all of their land \| 338,067 \| \| Landlords who continuously rent their land \| 239,650 \| \| Churches, monasteries and other religious groups \| 14,401 \| \| Total \| 1,000,325 \| Source: Kin Nichisei Senshu, Supp. Vol., p. 115; also in Choson Tongsa, III, p. 34, quoted from Choson Chungang Nyongam (Korean Central Almanac), 1949, pp. 71–72. Table II. Disposition of the Confiscated Land, 1946 \| Recipients \| Number of families \| Area (chongbo) \| \|---------------------------------------------------------------------------\|--------------------\|----------------\| \| Farm labourers \| 17,137 \| 22,387 \| \| Tenants without land \| 442,973 \| 603,407 \| \| Farmers with little land \| 260,501 \| 345,974 \| \| Landlords wishing to farm in new localities \| 3,911 \| 9,622 \| \| Total \| 724,522 \| 981,390 \| Source: Kin Nichisei Senshu, Supp. Vol., p. 115; also in Choson Tongsa, III, pp. 34–35. How was it that the peasants with admittedly retarded class consciousness were “aroused” in such a short period to make the land reform successful? What were the actual functions of the cadres dispatched from Pyongyang and the provincial headquarters of the Party? How was the resistance of the landlords overcome? The tactics employed in executing the land reform deserve to be elaborated beyond the bare outlines and the statistics provided by Communist sources. The North Korean experience deserves scrutiny particularly because, unlike that of neighbouring China, it was accompanied by little bloodshed. A prize-winning novelist in South Korea, Hwang Sun-won, depicts the entire process of land reform masterfully in fictionalised form in his novel K’ain Ui Huye (The Heirs of Cain). Hwang, a North Korean, escaped to the south in May 1946, and evidently had first-hand knowledge. The cadres dispatched from Pyongyang, according to Hwang, were mindful to execute the land reform with the least disturbance and resistance. The programme was to be carried out in the name of the peasants and, therefore, it required their direct participation. The cadres began by nurturing an atmosphere conducive to smooth functioning. Their first course of action was effectively to ostracise the landlords and to isolate them from the other peasants. Mere intimation that it would not be to their own advantage to associate closely with their former landlords was sufficient to make most of the tenant-farmers shun the landlords, who in many cases had developed friendly relationships with their tenants. The landlords were then constantly harassed by frequent questioning by the cadres and the police and were placed under close observation. Local tenant-farmers, particularly those with a record of close collaboration with the landlords, such as former tenant-foremen, were chosen as heads of the peasants' committees to aid the cadres. This system provided those with dubious records an opportunity to rectify their past mistakes; the chosen ones, who had the best knowledge of the community, showed enthusiasm and ruthlessness in their new duties, not realising that they would also be purged at a later date. The peasants were at first sceptical of the land reform ordinance. They would not believe that the land could be distributed without compensation to the owner. They were excited by the possibility that the land could be theirs, but they were embarrassed by their own excitement: How could one take the property of someone else without compensation? Would it not amount to robbery? The peasants were not easily convinced by the cadres' argument that the land originally belonged to them. Many of the peasant families had never owned a plot of land for generations. The peasants' curiosity, however, was aroused. Landlords seemed to be powerless before the new government. It was obvious to all that the land would be taken away from the former owners, regardless of the local peasants' attitudes. Should one resist the temptation of free land, resist the government in power and, indeed, resist what appeared to be the current of the time? But, if so, why? Was it not clear that the landlords were doomed, regardless of one's efforts? Under these circumstances, would it not be prudent to follow the cadres' instructions more willingly, thereby perhaps obtaining a better piece of land? The peasants' consciences were clear; they did not initiate the land reform programme, and, in fact, they loathed it. They were as helpless as the landlords. When the first part of the actual land reform was begun—that is, the confiscation of landlords' properties—the cadres were careful not to involve any large number of landlords on any one occasion. Undue agitation of landlords as a class should be avoided, lest they should organise themselves for revolt. Confiscations of land would take place sporadically; confiscations would not even take place simultaneously within a prefecture. Only the properties of the absentee landlords and the large landowners would be confiscated at first. Another round would reach the lower level. This scheduling left landlords with smaller properties a faint hope that they might be spared. It assured the régime of minimal resistance at each stage and even some co-operation from those spared at a particular stage. Human greed and instinct for survival seem to know no bound; and when the price for survival was mere acquiescence in others' destruction, the price was willingly paid. There was no recourse for the purged landlords. They could gain freedom only by fleeing from Communist jurisdiction and escaping south of the 38th Parallel. In North Korea they would either be arrested and sentenced to an indefinite period of hard labour, transferred to another locality with a small plot of land, or maltreated by the new local authorities and simply chased away. Crossing the "border" at the 38th Parallel became more and more difficult as time went by. There was no time to be wasted. Many purged and yet-to-be purged landlords chose South Korea as their future home.* Being assured of the success of the land reform programme, the authorities in Pyongyang immediately moved towards further consolidation of its bases. This phase of the programme entailed two areas of activity: organisation and discipline of the peasants, and the increase of food production. Organisation of the peasants had, in fact, been begun even before land reform was put into effect. Domestic-faction Communists had appealed to those of the "poor peasants" class immediately after the Japanese surrender on August 15, 1945, and carried on a broad campaign to enlist them in the Communist Party. We do not know how many peasants were inducted at various stages of the campaign, but, according to Kim Il-song, more than 105,000 members of the North Korean Workers' (Communist) Party in July 1946, were of "poor peasant" origin.* Efforts to enlarge the membership of the Party continued, and by March 1948, Kim reported the total to be more than 700,000 members, of which 374,000 were of "poor peasant" origin. It is significant that while the total rose by 90-3 per cent, during the --- * The estimated number of refugees in South Korea of North Korean origin on Dec. 31, 1947, was 1,116,600. First Part of the Report of the United Nations Temporary Commission on Korea (New York: General Assembly, Official Records: Third Session, Supplement No. 9 (A/375), 1948), I, p. 23. * Kim Nichizel Senshu, III, p. 48. The party members of worker (proletariat) origin at this time were 73,000 and the total membership was 366,000. Ibid. p. 47. LAND REFORM, COLLECTIVISATION AND THE PEASANTS twenty months, the membership of "poor peasant" origin showed more than a threefold increase. The Party was not the only organ used to organise the peasants. Peasants' federations, the Democratic Youth League and the Women's League carried out equally vigorous campaigns for membership. Membership of an individual in the Party or any one of the organisations did not make him ineligible for membership in another group. The intensity of the membership campaign is indicated by the following two sets of figures presented by Kim Il-song: \| \| April 1946 \| July 1946 \| \|--------------------------\|------------\|-----------\| \| Peasants' federations \| ... \| 800,000 \| 1,800,000 \| \| Workers' Leagues \| ... \| 350,000 \| 350,000 \| \| Democratic Youth League \| ... \| 500,000 \| 1,000,000 \| \| Women's League \| ... \| 350,000 \| 600,000 \| \| Total \| \| 2,000,000 \| 3,750,000 \| It is important to note that the membership of the workers' leagues did not increase at all during the three-month period, while that of the peasants' federations more than doubled. The rise in the membership of the other two organisations also doubled or nearly doubled. It is reasonable to assume that most of the new members of the young persons' and women's organisations were from the rural areas. The Pyongyang government immediately began to utilise the peasants' organisations for the purpose of solving economic problems. Kim Il-song declared on February 19, 1947, that "the immediate task of the North Korean agricultural economy is the solution of the food problem. North Korea must be transformed from a region of food shortage into a region of abundance. This can be done by enlarging the area of cultivation and increasing the rate of production." Kim specified further that in 1947 the area of cultivation would be increased by 300,172 chongbo (735,421 acres), an increase of 15.5 per cent. as compared to 1946, and that 155,000,000 Won would be invested by the People's Committee in the expansion of irrigation facilities. The use of chemical fertilisers would also be accelerated, from 109,000 tons in 1946 to 187,000 tons in 1947. Implicitly, the peasants in North Korea were warned to brace themselves for a period of hard labour. Kim Il-song's speech of April 13, --- 10 The figures for April are from his speech on land reform, Apr. 13, 1946, at the first enlarged committee meeting of the North Korean Provisional People's Committee. Kin Nichisei Senshu, I, p. 18. The figures for July are from his report on the establishment of Democratic People's United Front, July 22, 1946. Ibid., p. 64. 11 A report on the People's Economic Development Plan for 1947, at the conference of North Korean provincial, municipal and prefectural People's Committees. Kin Nichisei Senshu, I, p. 180. 1946, summing up the results of the land reform programme, included the following paragraph: The peasants must help each other and mobilise all labour power for the sake of the [successful completion of the] first movement to increase agricultural production since the liberation. The economic and financial foundations of Korea are still weak. Therefore, it is important for us to mobilise labour power fully. We must organise "Build the Nation" labour-service corps and systematically organise and guide the construction and repair of the irrigation facilities... ¹² Later Kim reported that the peasants had shown creativity in developing irrigation works on a broad scale, completing 58 irrigation projects in 1947 alone, for irrigating 19,753 chongbo of ricefields. ¹³ In January 1950, Kim made it known that the irrigation works at Anju must be completed. "For this purpose," he declared, "the entire people and all civil engineers must be mobilised." ¹⁴ In spite of these duties required of the peasants, they were promised a prosperous life. The ordinance on tax in kind for farm products, proclaimed June 27, 1946, specified that the peasants would be "exempt from all taxes on land and rent," and would submit instead only 25 per cent. of their total harvests (rice, other grains, beans and potatoes), being permitted to dispose of the surplus at free markets. On May 1, 1947, the ordinance was amended so that the tax in kind would be 27 per cent. on paddy land, 23 per cent. on dry land and only 10 per cent. on "fire field." ¹⁵ Compared to the 50 to 60 per cent. of the crops that tenants paid to their landlords before the land reform, this was indeed a significant improvement. The peasants, however, were not permitted to enjoy their improved economic status for long. Ambitious economic development plans instituted by the Pyongyang government demanded more than token co-operation from the peasants. The economic plan for 1947, the first of the development plans to be instituted in North Korea, called for a rapid reconstruction of industries to double the production of 1946. The workers were called to increase their productivity by 48 per cent. An increase in coal production and an improvement in transportation facilities were also demanded. ¹⁶ Should not the peasants be permitted to participate in the movement for progress? ¹² Ibid. p. 16. ¹³ Ibid. Supp. Vol., p. 116, speech of Apr. 20, 1948. ¹⁴ Ibid. p. 198, speech of Jan. 25, 1950. ¹⁵ For the text of the ordinance see Report of the United Nations Commission for the Unification and Rehabilitation of Korea, 1951, p. 60. See also U.S. Department of State, North Korea: A Case Study in the Techniques of Takeover, p. 57. "Fire field" indicates the areas cleared for cultivation by fire but abandoned in a few years when fertility of soil was exhausted. ¹⁶ For Kim Il-song's report on the economic development plan of 1947 see Kin Nichisei Senshu, I, pp. 170–183. Kim noted in this connection that the number of Already, in 1947, Kim Il-song had announced publicly that an extensive movement for the donation of "patriotic rice" had been started among the farmers by Kim Che-won and was making a great contribution towards national development. This "manifestation of passionate patriotism by the North Korean peasants," Kim said, had brought in such large donations of rice that there was "more than enough for the construction of the fatherland." He declared that the government intended to use the 1,500 ton surplus of rice in 1947 towards the building of a large-scale modern university in Pyongyang.$^{17}$ We can assume that the amount of "patriotic rice" donations would be expected to increase as the "peasants' patriotism" was intensified during the successive years. Official sources revealed that the peasants began to donate "large sums in foodstuffs and money" to enable the government to purchase military aircraft.$^{18}$ The peasants, along with the rest of the society, were required to purchase assessed amounts of the public bonds which the government began to issue in large quantities in late 1949. As the Korean War drew near and the Pyongyang régime mobilised all available resources in preparation for war, the material benefits that had accrued to the peasants by the land reform programme were reduced to virtually nil. The opening of the war in 1950 brought about a major disruption in North Korean agriculture, as in every other sector of the economy. Many villages that lay in the path of war were destroyed. Many peasants, young and old, were conscripted for army service before and during the war, and the casualty rate during the war was very high. Furthermore, a great number of North Korean peasants abandoned their farms during the short period of the United Nations occupation of North Korea and moved to South Korea. In short, the socio-political and economic fabric of North Korean agriculture was broken down almost completely. If the war meant economic disaster for North Korea, however, it at least provided the leaders with an opportunity to review their past performance and plan for future development. We have no means to determine the exact thoughts of the North Korean leadership, but Kim Il-song's speech of December 21, 1950, at the third congress of the central committee of the Communist Party rings with bitterness against $^{17}$ Ibid. p. 246. $^{18}$ Cho Chung ch'inson nongop hyoptong chohap nongmin dul ui munhwa wa p'ungsup (Culture and Customs of the Peasants in the Korea-China Friendship Agricultural Co-operative) (Pyongyang: Academy of Science Press, 1960), Folklore Research Series, No. 4, p. 211. For a discussion of the inflated assessment of the crop for tax purposes and the quality of crop required for tax, etc., see U.S. Department of State, op. cit., pp. 57–59. poor performance and weak discipline among the Party cadres. Nor was the Premier satisfied with the cadres' achievements among the masses. Despite the rapid increase of membership in the Party and the auxiliary organisations, the Party had not been able to muster the support it needed from the people during the most trying period of its existence. Indeed, the weakness of the Party's educational activities was manifested everywhere. By the Premier's own admission, some North Koreans co-operated with the enemy. "Some joined the local security units [formed against the Communists by the Republic of Korea authorities], anti-Communist groups, and other reactionary groups to murder, insult, and suppress the members of the Workers' Party and its supporters." 19 There are reasons to believe that the Party decided to rectify these weaknesses, particularly the paucity of propaganda and "political education" directed towards the masses, by changing the economic structure in North Korea. The land reform may have eliminated the "feudalistic components" of society, but this was far from creating necessary conditions for the transition to the Socialist stage of economy. It was only natural that the peasants would retain and even develop "capitalistic thought" under the system of private ownership of land. When the "poor peasants" were allotted pieces of land, their appetite for land and material goods was heightened to a new degree and their pecuniary interest was sharpened more than ever. Instead of arousing the revolutionary consciousness which the Communist leaders seem to have hoped for, the peasants under the new system tended to develop political complacency. The Communist theoreticians should have realised that the land reform merely eliminated the contradiction between the means of production and the productive forces that had existed before and did not create the conditions necessary for the growth of "workers' consciousness" or "revolutionary consciousness." Thus the land reform in fact dichotomised the society into two distinct sectors: the socialised urban sector of the proletariat and the more capitalistic rural sector. Only by depriving the peasants of their private ownership of the means of production could the Party be assured that the gap would be filled and workers' consciousness developed among the peasants. 20 Stalin's "law of development" also required the subordination of agriculture to the direction of Socialist industry. 19 Kin Nichisei Senshu, II, p. 141. 20 The North Korean Communists were hard put to justify the collectivisation programme. Thus a North Korean theoretician, Cho Chae-son, was forced to admit that "The Socialistic transformation of the agriculture is a special (or unusual) application of the law of (interrelationship between) the characteristics of productive LAND REFORM, COLLECTIVISATION AND THE PEASANTS While these ideological and political considerations were compelling North Korean leaders to re-examine the régime's agrarian policies, economic conditions in North Korea after the war seem to have compelled them to take immediate action. Grain output decreased during the war$^{21}$ and the food shortage was severe. North Korea found it necessary to import at least 5 to 10 per cent. of its consumption needs not only during the war, but for several years afterward. Even in 1961, when the North Korean radio was jubilantly announcing bumper crops in the collectivised fields, the régime found it necessary to import 45,000 metric tons of grain from Australia.$^{22}$ In addition to these problems there was an acute labour shortage. The North Korean loss of population during the war, including migration to the south and war casualties, was estimated to be about 2-1 million persons.$^{23}$ Industrial development required the mobilisation of labourers from rural areas. According to official statistics, the proportion of peasants in the population declined from 74-1 per cent. in 1946 to 66-4 per cent. in 1953 and to 44-4 per cent. in 1960. In other words there was a shift of approximately 20 per cent., or 200,000 persons from rural areas to the urban centres between 1953 and 1960.$^{24}$ Serious forces and the productive relations during the transitional period toward Socialism." (Italics added.) In his explanation of this "special application" of the Marxist law, Cho cited the "positive reaction of the new superstructure," i.e., the dictatorship of the proletariat, and the harmful effect of the imbalance between the Socialist industrial sector and the privately managed agriculture. He added further that the imbalance has been noticeable at the latter part of the two-year People's Economic Plan (1949–50). Cho Chae-son, Choson Minjujuui Inmin Konghwaguk Sahoe Kyongje Chedo (The Socio-Economic System in the Democratic People's Republic of Korea) (Pyongyang: Korean Workers' Party Press, 1958), pp. 38–39. $^{21}$ According to a North Korean source, the sown area and grain output decreased as follows: \| \| 1948 \| 1951 \| 1952 \| 1953 \| \|----------------\|------\|------\|------\|------\| \| Sown area \| 100 \| 89.2 \| 95.6 \| 97.4 \| \| Grain output \| 100 \| 84.9 \| 91.9 \| 87.2 \| Data from Agricultural Co-operativisation in D.P.R.K. (Pyongyang: Foreign Languages Publishing House, 1958), p 8. $^{22}$ U.S. Department of State, Bureau of Intelligence and Research, "Research Memorandum," June 21, 1962, RSB-105, p. 13. (Processed.) $^{23}$ Ibid. p. 14. $^{24}$ The percentages by occupation divisions of and total population of North Korea are reported to be as follows: \| \| 1946 \| 1949 \| 1953 \| 1956 \| 1960 \| \|----------------------\|------\|------\|------\|------\|------\| \| Workers and office employees \| 18.7% \| 26.0% \| 29.7% \| 40.9% \| 52.0% \| \| Farmers \| 74.1 \| 69.3 \| 66.4 \| 56.6 \| 44.0 \| \| Others \| 5.0 \| 2.9 \| 2.4 \| 2.0 \| 3.3 \| \| Total \| 100.0% \| 100.0% \| 100.0% \| 100.0% \| 100.0% \| \| Population (thousands) \| 9,257 \| 9,022 \| 8,491 \| 9,359 \| 10,789 \| Facts About Korea (Pyongyang: Foreign Languages Publishing House, 1961), p. 9. The population figures were taken from U.S. State Department, "Research Memorandum," p. 14. as the agrarian problems were, industrial development took priority over agriculture. It is against this background that the "co-operativisation" movement was initiated. After a brief "experimental stage" between August 1953 and October 1954, the Pyongyang régime decided to launch a full-scale collectivisation movement at the Communist Party's central committee's plenum in early November 1954. Within the first year—by December 1955—49 per cent. of the peasants were collectivised. Another 30 per cent. were collectivised during the following year. By December 1957, three years after the co-operative movement had been initiated, 95-6 per cent. of the peasants were organised into 16,032 co-operatives with an average of 64 households each. By August 1958, the entire farm population in North Korea had been inducted into co-operatives and minor adjustments of the size of co-operatives had been made; there were then 13,309 co-operatives with an average of 80 households each. In October, only two months after the announcement of the "victorious completion of the agricultural co-operativisation movement," the régime proclaimed that the co-operatives would be reorganised into larger units. Henceforth, the ri—originally the lowest administrative unit, roughly equivalent to a precinct, but enlarged by the North Korean régime to encompass several villages—would serve as the unit for co-operatives. Instead of 13,309 co-operatives there would be only 3,843 and each would consist of an average of 300 households rather than 80. The average acreage of a co-operative would be 500 chongbo (1,225 acres) rather than 130 chongbo. There is little room for doubt that the sudden reversal of the previous policy was brought about by the North Korean leaders' fascination with and idealisation of the commune movement in China. It is to be noted that the Chinese Communists began the merger of co-operatives in April 1958, and in August the Politburo of the Chinese Communist Party ordered the adoption of the commune system throughout China. The enthusiasm of the Chinese Communists was promptly echoed by North Koreans. --- 25 Pak Mun-gyu, the Minister of Agriculture from September 1948 to March 1954, was replaced by Kim Il-song's comrade of the partisan days, Kim Il, in March 1954. Kim Il served as Vice-Premier and Minister of Agriculture until September 1957. Kim Il became First Vice-Premier in 1959. Pak Mun-gyu, an agrarian economist of long standing, was relegated to a less important cabinet post. 26 For a detailed treatment of the process of collectivisation see my article, "The 'Socialist Revolution' in the North Korean countryside," Asian Survey, II, No. 8 (Oct. 1962), pp. 9-22. 27 Kim Il-song Sonjip (Selected Works of Kim Il-song) (Pyongyang: Korean Workers' Party Press, 1960), VI, p. 185 28 See "Chungguk eso ui sahoejuui konsol ui taeyakchin" ("The Great Leap Forward in the Socialist Construction in China"), Kulloja (The Worker), No. 155, Oct. 15, 1958, p. 77. LAND REFORM, COLLECTIVISATION AND THE PEASANTS Just as in the Chinese communes, the organ of the government, ri, was merged with the organisation in charge of production, the co-operative. The co-operative became at once a basic social unit and a basic organisation of state power in that it is designed to integrate rural industry, agriculture, trade, culture and education, and military affairs into a single whole. Even official North Korean sources admitted that the bulldozing of the collectivisation programme had been accompanied by outright coercion and intimidation along with propaganda. "In some places," according to a North Korean booklet, organisational work was undertaken in a bureaucratic manner or as a campaign, on the pretext of overtaking and surpassing those areas where the co-operative movement had advanced faster. In so doing, the degree of preparedness of the peasants was not taken into consideration and various unfair methods were used in drawing some irresolute peasants into co-operatives... Such deviations were also to be seen in the practice of showing contempt for individual farmers and encroaching upon their interests under the pretext of consolidating and developing co-operatives.$^{29}$ Resistance of the farmers to the radical collectivisation programme was by no means negligible. Although no violent uprisings were reported by North Korean sources, cases of evasion and sabotage were not infrequent. Premier Kim summed up the situation in a speech delivered before a conference of the leaders of the co-operatives in 1959 in pointed language: Class enemies slandered our Party's policies and spread reactionary rumours opposing the agricultural co-operativisation movement. They also fought to destroy our agricultural co-operatives by damaging common properties or obstructing production. Especially when agricultural co-operativisation was near completion and Socialist victory was becoming more consolidated in cities and farms, the anti-revolutionary elements' infiltration, destruction, and obstructive behaviour became more vicious.$^{30}$ North Korean authorities also encountered considerable difficulty in executing the collectivisation programme because of the lack of well-trained cadres. Violation of democratic management principles, inadequate planning, poor utilisation of co-operatives' properties and financial mistakes were cited as some of the mistakes committed by the "executives." An official source also reported that among the managerial personnel of some co-operatives were found subversive elements who had crept into the co-operatives and managed to occupy the leading positions. The defects of officials not only led to $^{29}$ Agricultural Cooperativization in D.P.R.K., pp. 37–38. $^{30}$ Kim Il-song Sonjip, VI, p. 175. waste of labour but also a drop in the rate of co-op members' attendance at work, causing a hindrance to important farm work.\textsuperscript{31} Collectivised farms in North Korea are still called co-operatives, but in fact they differ only very slightly from the artel type \textit{kolkhozy} (collective farms) in the Soviet Union and the communes in China. All land formerly held by individual farmers is now controlled by co-operatives which theoretically own the land. Individual farmers, who work as members of "work teams" or "brigades," under the co-operatives, are paid in kind and cash at the end of the year according to the work days contributed. Farmers are allowed to own a private garden plot, fruit trees, cattle and bees, although data on the extent of these possessions are not available. In theory, each co-operative is an independent entity and the farmer has full authority to control the affairs of his co-operative through either the co-operative general assembly or the representative assembly which elect the chairman and the members of the co-operative committees. In practice, however, the co-operatives form an integral part of the national economy, and each co-operative is rigidly controlled by the "management committees" at the \textit{kun} (prefectural) level which were created by the Party's central committee plenum in November-December, 1961. It should be noted that the central committee of the Workers' Party admitted, at the time of the establishment of the management committees, the inadequacy of the co-operative officials in meeting the demand for increased production and the deficiency of the "administrative (or bureaucratic) method" used by prefectural people's committees in guiding the co-operatives. The prefectural people's committees still exist as administrative agencies of government but they are deprived of power over agricultural production and management. The new management committees are staffed by agro-technicians from central government organs.\textsuperscript{32} Under the "co-operative system," the state exercises rigid control over the farmers' income and consumption. Since grain dealers were eliminated by decree in October, 1954, and the state-operated stores constitute the only channel for farmers' disposal of their surplus products, state control and supervision is easily facilitated. The régime has instituted a progressive scale of tax in kind, compulsory purchase of farm products with price differentials, and compulsory accumulation of funds and grains at the co-operative level. Although the official figure for tax in kind was reduced from an average of 20-1 per cent. between 1956 to 1958 to 8-4 per cent. after 1959, actual rates of tax collection \textsuperscript{31} \textit{Agricultural Cooperativization in D.P.R.K.}, p. 43. \textsuperscript{32} \textit{Cf. Nodong Shinmun}, Dec. 25, 1961, editorial. and compulsory savings at the co-operative level cannot be accurately determined. The life of co-operative peasants is regimented, disciplined and organised to the minute. The demands heaped upon peasants seem to have no limit. A Japanese reporter sympathetic to the North Korean régime describes the daily schedule of the North Korean farmers as it was related to him by the head of a co-operative: [Members of the co-operative] gather on the farm (or in the workshops) around 8.30 in the morning at the signal of a bell from the co-operative. Each member receives instructions from his group leader regarding the day's work. At noon, the bell will ring again and members can go either to the public dining hall or their own homes for lunch. There are circle meetings for drama, music and dancing during the lunch hour. Peasants rest ten to twenty minutes after fifty minutes of work. During the breaks, there are "news reading meetings" when newspapers or magazines will be read aloud. After a day's work, members have their work evaluated and they receive certain points which are recorded in their Labour Notebook. Distribution of crops or cash at the end of the year is done according to the total points accumulated. . . . After work, members with their Labour Notebooks either return home or go to a common bathhouse. They see movies, listen to the radio, or join circle activities.$^{33}$ Even this idealised version does reveal the extent to which the peasants are regimented. In terms of daily work, the above story omits a vital part, i.e., the "voluntary" work undertaken by the peasants. In December 1959, for instance, the Democratic Youth League in the co-operative visited by the Japanese reporters is supposed to have decided to collect three tons of "mud coal" a month per person. The "mud coal" was to be used either as fuel or fertiliser.$^{34}$ Earlier in 1958, when the régime had decided to execute a "people's movement for light industry," a variety of "factories" had been established at co-operatives to produce consumers' products.$^{35}$ The labour required for irrigation works, construction of schools, roads and houses had to be furnished by the peasants on a "voluntary," hence unpaid, basis. The régime has evidently decided that overwork and the lack of material incentives can be offset by further indoctrination and intimidation of the peasants. For instance, in the "Korea-China Friendship Agricultural Co-operative," which consisted of 750 families with 1,227 co-operative members as of December 1959, there were fifteen party-policy study groups with some 600 members "guided" by fifty-nine $^{33}$ Ho-Cho Kishadan (Reporters' Group Visiting Korea), Kita-Chosen no Kiroku (Record of North Korea) (Tokyo: Shin Dokusho-sha, 1960), pp. 200-201. $^{34}$ Ibid. p. 206. $^{35}$ Tera Goro, 38 Dosen no Kita (North of the 38th Parallel) (Tokyo: Shin Nippon Shuppansha, 1959), p. 106. agitators and seven "conversation leaders." The co-operative maintained six "Korean Workers' Party History Study Rooms" along with sundry kinds of circles and "mass political edification networks." The function of the "agitators" and "conversation leaders" is to visit the working teams during their rest periods and "explain" the Party's policy and the domestic and international developments. The co-operative, which is one of the "model co-operatives" in North Korea, has also been favoured by the frequent visits of "concentrated guidance groups" dispatched by the central committee of the Party. In order to guarantee the effect of the Party's activities in the co-operatives, the Party interspersed indoctrination activities with occasional "struggles against the counter-revolutionary elements" in the co-operatives, which exposed "liquidated landlords, collaborators with the enemy during the temporary retreat period, and Christian ministers and elders." Some of the purged are alleged to have attempted to collaborate with espionage agents dispatched by the enemy. These purges (the report on the "Korea-China" co-operative mentions purges in 1957 and 1959) were followed by intensive "edification activities" in order to instil "revolutionary awareness" among those influenced by the purged elements. North Korean publications of the past few years have been exuberant in asserting the success of the collectivisation programme and the continuous increase (except in 1959 when the total output declined below the 1958 level) in agricultural production. The Seven-Year Development Plan (1961-67) calls for an increase of total agricultural production by 2-4 times, although the grain output is aimed at 1-7 times the 1960 norm. The plan calls for 5,000,000 metric tons of grain in 1963. Obviously some progress has been made in agricultural production during the last several years, although the food problem in North Korea is far from being solved. As of the spring of 1963, there is no sign of retreat in the North Korean collectivisation programme in spite of the virtual abandonment of the commune system in China. According to official sources, 1962 was another "bumper crop year" despite drought and floods. Is there --- 36 Cho Chung Ch'inson Nongop Hyoptong Chohap . . ., p. 241. 37 Ibid. p. 237. 38 Ibid. p. 240. 39 Ibid. p. 241. 40 See Control Figures for the Seven-Year Plan (1961-67) for the Development of the National Economy of the Democratic People's Republic of Korea (Pyongyang: Korean Central News Agency, 1961), p. 18. The officially reported grain outputs in 1960 and 1961 were 3,803,000 tons and 4,830,000 tons, respectively. Approximately 40 per cent. of the total was rice. The adjusted estimates of the U.S. Department of State for these years were 2,781,000 tons and 3,378,000 tons, respectively. See Department of State, Bureau of Intelligence and Research, "Research Memorandum," RSB-105, June 21, 1962, p. 11. anything inherently superior in the North Korean system vis-à-vis the Chinese? Space does not permit detailed analysis of this question but a few factors are obviously in favour of North Korea: smaller territory (46,539 sq. mi. vs. 3,691,502 sq. mi. in China), smaller population (10,700,000 vs. 700,000,000 in China), relatively abundant supply of chemical fertilisers, and extensive irrigation facilities. The fact that the North Korean collectivisation programme was instituted immediately after a devastating war during which a great proportion of farmers had been uprooted from their farms and many others had abandoned their farms to flee to South Korea was also advantageous to the initial stage of collectivisation. The success of agricultural policies in North Korea is not to be measured by the increase in production alone, however. As has been repeatedly emphasised by North Korean authorities, the major purpose of the socialisation movement has been to bring about the equilibrium between the industrial and agricultural sectors of the economy and to prepare the peasants for the transition to the Communist stage. Judged by this criterion, the land reform programme before the war was an obvious failure. But will collectivisation and intensive indoctrination successfully bring about a change in the minds of the peasants? Will the change be significant enough to avoid the kind of catastrophic defeat that the Communist régime suffered in 1950 with the mass exodus of the peasants? No one, including the Communist leaders themselves, can offer a definite answer to this question. One thing, however, is certain. The Communist régime will never be able to win over the minds of the peasants as long as it regards the rural population as a means to an end rather than an end in itself. As the Communist leaders should know, the Korean peasants have been exploited far too long not to recognise exploitation for what it is, no matter what it is called.
Cross Sections and Transverse Single-Spin Asymmetries in Forward Neutral-Pion Production from Proton Collisions at $\sqrt{s} = 200$ GeV J. Adams, C. Adler, M. M. Aggarwal, Z. Ahammed, J. Amonett, B. D. Anderson, M. Anderson, D. Arkhipkin, G. S. Averichev, S. K. Badyal, J. Balewski, O. Barannikova, L. B. Barnby, J. Baudot, S. Bekele, V. V. Belaga, R. Bellwied, J. Berger, B. I. Bezyerkhny, S. Bhardwaj, P. Bhaskar, A. K. Bhati, H. Bichsel, A. Billmeier, L. C. Bland, C. O. Blyth, B. E. Bonner, M. Botje, A. Boucham, A. Brandin, A. Bravar, R. V. Cadman, X. Z. Cai, H. Caines, M. Calderón de la Barca Sánchez, J. Carroll, J. Castillo, M. Castro, D. Cebrá, P. Chaloupka, S. Chattopadhyay, H. F. Chen, Y. Chen, S. P. Cherlenko, M. Cherney, A. Chikianian, B. Choi, W. Christie, J. P. Coffin, T. M. Cormier, J. G. Kramer, H. J. Crawford, D. Das, S. Das, A. A. Derevchikov, L. Didenko, T. Dietel, W. J. Dong, X. Dong, J. E. Draper, F. Du, K. D. Dubey, V. B. Dunin, J. C. Dunlop, M. R. Dutta Majumdar, V. Eckardt, L. G. Efimov, V. Emelianov, J. Engelage, G. Eppley, B. Erasmus, M. Estienne, P. Fachini, V. Faine, J. Faivre, R. Fatemi, K. Filimonov, V. Filip, E. Finch, Y. Fisyak, D. Fliert, K. J. Foley, J. Fu, C. A. Gagliardi, N. Gagunashvili, J. Gans, M. S. Ganti, L. Gaudichet, M. Germain, F. Geurts, V. Gakhzianikian, P. Ghosh, J. E. Gonzalez, O. Grachov, V. Grigoriev, S. Gronstad, D. Grosnick, M. Guedon, S. M. Guertin, E. Gushin, T. D. Gutierrez, T. J. Hallman, D. Hardtke, W. Harris, M. Heinz, T. W. Henry, S. Heppelmann, T. Herston, B. Hippolyte, A. Hirsch, E. Hjort, G. W. Hoffmann, M. Horsley, H. Z. Huang, S. L. Huang, T. J. Humenick, G. Igo, A. Ishihara, P. Jacobs, W. W. Jacobs, M. Janik, H. Jiang, I. Johnson, P. G. Jones, E. G. Judd, S. Kabana, M. Kaneta, M. Kaplan, D. Keane, V. Yu. Khodyrev, J. Kiryluk, A. Kisiel, J. Klay, S. R. Klein, A. Klyachko, D. D. Koetke, T. Kollegger, M. Kopytine, L. Kotchenda, A. D. Kovalenko, M. Kramer, P. Kravtsov, V. I. Kravtsov, K. Krueger, C. Kuhn, A. I. Kulikov, A. Kumar, G. J. Kunde, C. L. Kunz, R. Kh. Kutuev, A. A. Kuznetsov, M. A. C. Lamont, M. J. Landgraf, S. Lange, C. P. Lansdell, B. Laslik, F. Laue, J. Laurel, A. Lebedev, R. Lednicky, M. J. LeVine, C. Li, Q. Li, S. J. Lindenbaum, M. A. Lisa, F. Liu, L. Liu, Z. Liu, Q. J. Liu, T. Ljubicic, W. H. Llope, H. Long, R. S. Longacre, M. Lopez-Noriega, W. A. Love, T. Ludlam, D. Lynn, J. Ma, Y. G. Ma, D. Magestro, S. Mahajan, L. K. Mangotra, D. P. Mahapatra, R. Majka, R. Manweiler, S. Margitis, C. Markert, L. Martin, J. Marx, H. S. Matis, Yu. A. Matulevich, T. S. McShane, F. Meissner, Yu. Melnick, A. Meschanin, M. Messer, M. L. Miller, Z. Milosevich, N. G. Minave, C. Mironov, D. Mishra, J. Mitchell, B. Mohanty, L. Molnar, C. F. Moore, M. J. Mora-Corral, D. A. Morozov, V. Morozov, M. M. de Moura, M. G. Munhoz, B. K. Nandi, S. K. Nayak, T. K. Nayak, J. M. Nelson, P. Nevski, V. A. Nikitin, L. V. Nogach, B. Norman, S. B. Nurushev, G. Odyniec, A. Ogawa, V. Okorokov, M. Oldenburg, D. Olson, G. Paic, S. U. Pandey, S. K. Pal, Y. Panebratsev, S. Y. Panitkin, A. I. Pavlinov, T. Pawlak, V. Petrovichovizh, C. Perkins, W. Perry, V. A. Petrov, S. C. Phatak, R. Picha, M. Planinic, J. Pluta, N. Porile, J. Porter, A. M. Poskanzer, M. Potekhin, E. Potrebenikova, B. V. K. S. Potukuchi, D. Prindle, C. Pruneau, J. Putschke, G. Rai, G. Rakness, R. Raniwala, S. Raniwala, Q. Ravel, R. L. Ray, S. V. Razin, D. Reichhold, J. G. Reid, G. Renault, F. Retiere, A. Ridiger, H. G. Ritter, J. B. Roberts, O. V. Rogachevski, J. L. Romero, A. Rose, C. Roy, L. J. Ruan, R. Sahoo, I. Sakrejda, S. Salur, J. Sandweiss, I. Savin, J. Schambach, R. P. Scharenberg, S. Schmitz, L. S. Schroeder, K. Schweda, S. Seliverstov, P. Seyboth, E. Shahaliiev, M. Shao, M. Sharma, K. E. Shestermanov, S. S. Shimanskiii, R. N. Singaraju, F. Simon, G. Skoro, N. Smirnov, R. Snellings, G. Sood, P. Sorensen, J. Sowinski, H. M. Spinka, B. Srivastava, S. Stanislaus, R. Stock, A. Stolpovsky, M. Strikhanov, B. Stringfellow, C. Struck, A. A. P. Suade, E. Sugarbaker, C. Suire, M. Sumbera, B. Surrow, T. J. M. Symons, A. Szanto de Toledo, P. Szarwas, A. Tai, J. Takahashi, A. H. Tang, D. Thein, J. H. Thomas, V. Tikhomirov, M. Tokarev, M. B. Tonjes, T. A. Trainor, R. Trentalange, R. E. Tribble, M. D. Trivedi, V. Trofimov, O. Tsai, Ullrich, D. G. Underwood, G. Van Buren, A. M. VanderMolen, A. N. Vasiliev, M. Vasiliev, E. V. Vigdor, Y. P. Viyogi, S. A. Voloshin, W. Waggoner, F. Wang, G. Wang, X. L. Wang, Z. M. Wang, H. Ward, J. W. Watson, R. Wells, G. D. Westfall, C. Whitten, H. Wiemann, R. Willson, S. W. Wissink, R. Witt, J. Wood, J. Wu, N. Xu, Z. Xu, Z. Yamamoto, P. Yepes, V. I. Yurevich, Y. V. Zanevsky, I. Zborovský, H. Zhang, W. M. Zhang, Z. P. Zhang, A. P. Zolnierczuk, Z. Zoukarneeva, J. Zoukarneeva, and A. N. Zubarev. Measurements of the production of forward high-energy $\pi^0$ mesons from transversely polarized proton collisions at $\sqrt{s} = 200$ GeV are reported. The cross section is generally consistent with next-to-leading order perturbative QCD calculations. The analyzing power is small at $x_F$ below about 0.3, and becomes positive and large at higher $x_F$, similar to the trend in data at $\sqrt{s} \leq 20$ GeV. The analyzing power is in qualitative agreement with perturbative QCD model expectations. This is the first significant spin result seen for particles produced with $p_T > 1$ GeV/c at a polarized proton collider. DOI: 10.1103/PhysRevLett.92.171801 PACS numbers: 13.85.Ni, 12.38.Qk, 13.88.+e An early qualitative expectation from perturbative quantum chromodynamics (pQCD) was that the chiral properties of the theory would make the analyzing power for inclusive particle production be very small [1]. The analyzing power ($A_N$) is the azimuthal asymmetry in particle yields from a transversely polarized beam... incident on an unpolarized target. Earlier experiments studied polarized proton collisions ($p_1 + p$) at center-of-mass energies $\sqrt{s} \leq 20$ GeV and measured $A_N$ for pion production at moderate transverse momentum (0.5 $\leq p_T \leq 2.0$ GeV/$c$). Contrary to the naive expectation, $A_N$ was found to be 20%–40% for pions produced at large values of Feynman $x$ ($x_F = 2 \cdot p_L / \sqrt{s}$, where $p_L$ is the longitudinal momentum of the pion) [2–5]. Similarly, elastic proton [6] and recent semi-inclusive deep-inelastic lepton scattering experiments [7,8] have reported transverse single-spin asymmetries which differ significantly from zero. These results have sparked substantial theoretical activity to understand transverse spin effects within the framework of pQCD [9]. Perturbative QCD calculations of pion production involve the convolution of parton distribution and fragmentation functions with a hard partonic interaction. The reliability of calculations in the pQCD framework is expected to increase with $p_T$. In this framework, forward $\pi$ production in $p + p$ collisions is dominated by scattering of a valence quark in one proton from a soft gluon in the other. At large pseudorapidities ($\eta$) and $\sqrt{s} \leq 20$ GeV, there may be significant contributions to particle production from soft hadronic processes collectively known as beam fragmentation. At a collider, $\sqrt{s}$ is significantly larger, leading to the expectation that the origin of forward pions will shift towards collisions of the partonic constituents of the proton, consistent with the PYTHIA Monte Carlo generator [10]. Measurements of the cross section for forward pion production are important to establish that pQCD is a suitable framework for treating polarization observables in these kinematics. Different mechanisms have been identified in the pQCD framework by which one might expect transverse spin effects [11–17], all of which may contribute to some degree. With only data at $\sqrt{s} \leq 20$ GeV for comparison, these models are not well constrained. Despite this, the models have been extrapolated by an order of magnitude in $\sqrt{s}$ and approximately a factor of 2 in $p_T$, and all predict that sizable transverse spin effects will persist at $\sqrt{s} = 200$ GeV. This Letter addresses the question if $A_N$ is sizable at $\sqrt{s} = 200$ GeV, as predicted by these models. We present measurements of the cross section and $A_N$ for the production of forward $\pi^0$ mesons having $p_T > 1$ GeV/$c$ from $p_1 + p$ collisions at $\sqrt{s} = 200$ GeV. Data were collected by the STAR experiment (Solenoid Tracker at RHIC) at the Brookhaven National Laboratory Relativistic Heavy Ion Collider (RHIC) in January 2002. RHIC is the first polarized proton collider. Polarization is produced by optical pumping of an atomic-beam source [18] and is partially preserved through an accelerator complex to reach RHIC [19]. In RHIC, a pair of helical dipole magnets in each ring serves as the first use of full “Siberian snakes” [20] in a high-energy accelerator to preserve polarization during beam acceleration. The stable spin axis of the RHIC rings is vertical. Beam bunches crossed the STAR interaction region (IR) every 213 ns. The polarization direction alternated between up and down for successive bunches of one beam and after every two bunches of the other beam. Data were sorted according to the spin direction of the beam corresponding to positive $x_F$ pion production. Summing all bunches in the other beam resulted in negligible remnant polarization. Typical luminosities were $10^{30}$ cm$^{-2}$ s$^{-1}$, and the integrated luminosity was 150 nb$^{-1}$ for these data. The average beam polarization for each fill, $P_{\text{beam}}$, was determined using a Coulomb-Nuclear Interference (CNI) polarimeter located in RHIC [21,22]. At 24.3 GeV, the RHIC injection energy, the analyzing power of the CNI reaction is $A_N^{\text{CNI}} = 0.0133 \pm 0.0041$ [23,24], and can be used to deduce the absolute polarization of the proton beam. However, at 100 GeV, the beam energy used for RHIC collisions, $A_N^{\text{CNI}}$ has not yet been measured. The CNI asymmetries measured at injection and collision energies were nearly equal for many fills. Since the beam acceleration process is unlikely to increase $P_{\text{beam}}$, this suggests that $A_N^{\text{CNI}}$ at 100 GeV is no smaller than at 24.3 GeV. For the present analysis, we assume there is no change in $A_N^{\text{CNI}}$ between these two energies, giving an average value of $\langle P_{\text{beam}} \rangle = 0.16$. A prototype forward $\pi^0$ detector (PFPD) was installed at STAR 750 cm from the IR to identify $\pi^0$ mesons. At this time, STAR does not have the capability to identify large rapidity charged pions. The PFPPD consisted of a Pb-scintillator sampling calorimeter [25], placed with its edge $\approx 30$ cm left of the oncoming polarized proton beam (beam left). The PFPPD was 21 radiation lengths deep and subdivided into $4 \times 3$ towers. To measure the transverse profiles of photon showers, the PFPPD had a shower-maximum detector (SMD) approximately 5 radiation lengths deep, comprising two orthogonal layers of $100 \times 60$ scintillator strips spaced at 0.37 cm. To address systematic errors associated with measuring left-right asymmetries with a single arm detector, an array of Pb-glass detectors with no SMD was placed to the right of the oncoming polarized proton beam (beam right). Similar arrays were placed above and below the vertically polarized beam, where no asymmetries are expected. The luminosity was measured at STAR using beam-beam counters (BBC) [26] composed of segmented scintillator annuli mounted around the beam at longitudinal positions $z = \pm 370$ cm, spanning $3.3 < \|\eta\| < 5.0$. Proton collision events were identified by requiring the coincidence of at least one BBC segment fore and aft of the IR. Absolute luminosity was determined by measuring the transverse size of the colliding beams and the number of protons colliding at STAR. The cross section measured for the BBC coincidence condition is $26.1 \pm 0.2(\text{stat}) \pm 1.8(\text{syst})$ mb [27], consistent with simulation [10,28]. The BBC observes 87% $\pm$ 8% of the inelastic, nonsingly diffractive cross section. All forward calorimeters were read out when the energy deposited in any one calorimeter was greater than that from an electron of $\sim 15$ GeV. The BBC coincidence requirement was imposed to select $p + p$ collisions. The asymmetry measured at beam left is $$P_{\text{beam}} A_N = \frac{N_+ - RN_-}{N_+ + RN_-}. \quad (1)$$ The number of $\pi^0$ mesons detected when the beam spin vector was oriented up (down) is $N_{+(-)}$. The spin-dependent relative luminosity ($R = L_+/L_- = 1.15$) was measured with the BBC. Background contributions to $R$ were reduced by increasing the coincidence requirements to at least two BBC segments on each side of STAR. The systematic errors on $R$, primarily coming from the change in $R$ when the background is corrected, are of the order of $10^{-3}$ [26] and are a factor of 10 to 20 smaller than $P_{\text{beam}} A_N$ measured with the PFPD. Neutral pions are reconstructed utilizing the formula $$M_{\gamma\gamma} = E_\pi \sqrt{1 - z_\gamma^2} \sin(\phi_{\gamma\gamma}/2) = E_{\text{tot}} \sqrt{1 - z_\gamma^2} d_{\gamma\gamma}/2z_{\text{vtx}},$$ using events with at least two clusters in the SMD. The energy of the leading $\pi^0$, $E_\pi$, is taken to be the total energy deposited in all of the towers, $E_{\text{tot}}$. The opening angle between the two photons, $\phi_{\gamma\gamma}$, is determined by $z_{\text{vtx}}$, the distance between the collision vertex and the PFPD, and the separation of the photons at the detector, $d_{\gamma\gamma}$. Both $d_{\gamma\gamma}$ and $z_\gamma = \|E_{y1} - E_{y2}\|/(E_{x1} - E_{x2})$ are measured by an analysis of the energy deposited in the strips of the SMD. The value of $d_{\gamma\gamma}$ is determined from the fitted centroids of the peaks, while $z_\gamma$ is derived from the ratio of the fitted areas under the peaks. A fiducial volume is defined by requiring the SMD peaks to be more than 12 strips from the detector edge. Figure 1 shows the $M_{\gamma\gamma}$ spectra for two energy bins. The mass resolution is 20 MeV/$c^2$ (rms) for $15 < E_{\text{tot}} < 80$ GeV, limited by the measurement of $\phi_{\gamma\gamma}$. The centroid of the $\pi^0$ peak is used to determine the calibration for each tower for each fill to an accuracy of the order of 1%. The calibration is found to have negligible dependence on energy or spin state. The $\pi^0$ detection efficiency is determined in a matrix of $E_\pi$ and $\eta$ from a Monte Carlo (MC) simulation of $p + p$ collisions [10] and the detector response [28]. The open histograms in Fig. 1 are MC events which undergo the same reconstruction and selection as the data. The MC matches the data well for several variables, including $p_T$, $E_{\text{tot}}$, and $\eta$. The $\pi^0$ detection efficiency is dominated by the geometrical acceptance of the calorimeter. The $\pi^0$ sample is distorted by coincident particles from the jets containing them. The PFPD is about one hadronic interaction length deep. When two photons from $\pi^0$ decay overlap with other particles, the PFPD response to the other particles tends to increase $E_{\text{tot}}$ relative to $E_\pi$ and broaden the $\phi_{\gamma\gamma}$ resolution. This results in a broad $M_{\gamma\gamma}$ distribution peaked at a value larger than $M_\pi$. The average value of $E_{\text{tot}}$ is about 3 GeV larger than $E_\pi$, independent of $E_\pi$. MC events with $\|E_{\text{tot}} - E_\pi\| > 2$ GeV are shown as the hatched histograms in Fig. 1. Events with only one photon from $\pi^0$ decay plus other particles exist predominantly at small $M_{\gamma\gamma}$, and are suppressed by requiring $z_\gamma < 0.3$. The $E_\pi$-dependent systematic error in the cross section is about 20%, dominated by the jet correction. The MC simulation includes $\pi^0$ events from forward jets. The uncertainty includes the difference when these effects are explicitly corrected in both the data and the simulation, and in neither. Noncollision background is suppressed to the level of 1% by requiring the coincidence from the BBC in the offline analysis. Following our simulations, the cross section is corrected by 10% to account for the bias introduced by the BBC coincidence condition. Hadronic background comprising events with no leading $\pi^0$ in the acceptance of the calorimeter is predominantly at small $M_{\gamma\gamma}$, and is reduced by constraining $z_\gamma$. The hadronic background amounts to about 2% of the yield underneath the $\pi^0$ peak at $0.09 < M_{\gamma\gamma} < 0.22$ GeV/$c^2$. The inclusive $\pi^0$ production cross section for $30 < E_\pi < 55$ GeV in 5 GeV bins is presented in Fig. 2. Data with $3.4 < \eta < 4.0$ were selected, giving $\langle \eta \rangle = 3.8$ independent of $E_\pi$; in this range the detector efficiency is well understood. The dominant contributions to the normalization error come from knowledge of the absolute transverse position of the detector (10%), the absolute luminosity determination (8%), and the model dependence of the BBC efficiency (8%). The data are plotted at the average $E_\pi$ of the bin. The curves on the plot are next-to-leading order (NLO) pQCD calculations [29] evaluated at $\eta = 3.8$, using the CTEQ6M [32] parton distribution functions and equal FIG. 2. Inclusive $\pi^0$ production cross section versus leading $\pi^0$ energy ($E_\pi$). The average transverse momentum $\langle p_T \rangle$ is correlated with $E_\pi$, as the PFPD was at a fixed pseudorapidity ($\eta$). The inner error bars are statistical, and are smaller than the symbols for most points. The outer error bars combine these with the $E_\pi$-dependent systematic errors. The curves are NLO pQCD calculations evaluated at $\eta = 3.8$ [29–31]. The curves on the plot are predictions from the pQCD models, fitted to data at $\sqrt{s} = 20$ GeV, extrapolated to $\sqrt{s} = 200$ GeV and evaluated at $p_T = 1.5$ GeV/c [14–17]. One model attributes single-spin effects to the convolution of the transversity distribution function with a spin-dependent Collins fragmentation function [14]. The Sivers model adds explicit spin-dependent $k_T$ dependence to the parton distribution functions [15]. Other models ascribe the effects to twist-3 parton correlations in the initial or final state [16,17]. The data are qualitatively consistent with all of these predictions. The trend of $A_N$ at lower $\sqrt{s}$ is to increase from zero beginning at a value of $x_F$ which depends on $\sqrt{s}$ [5]. The significance of the increase for these data is 4.7 $\sigma$ (including statistical and point-to-point systematic errors) from a linear fit to the open circles in Fig. 3 for $x_F > 0.27$, with $\chi^2 = 0.9$ for 3 degrees of freedom. This is the first significant spin result seen for particles with $p_T > 1$ GeV/c at a polarized proton collider. In summary, high-energy $\pi^0$ mesons have been observed at forward angles from $p_t + p$ collisions at $\sqrt{s} = 200$ GeV. The differential cross section is, in general, consistent with NLO pQCD calculations. The analyzing power is small at $x_F$ below about 0.3, and becomes positive and large at higher $x_F$, similar to the trend observed in fixed-target data at $\sqrt{s} \leq 20$ GeV. The analyzing power at $\sqrt{s} = 200$ GeV is in qualitative agreement with pQCD model predictions. Higher precision measurements of $A_N$ as a function of both $x_F$ and $p_T$ may help to... differentiate among the models. Future measurements may attempt to determine the Collins fragmentation function in $p_t + p$ collisions, as well as to look at jet production and Drell-Yan scattering to isolate potential contributions to transverse spin effects. We thank the RHIC Operations Group and RCF at BNL, and the NERSC Center at LBNL for their support. This work was supported in part by the HENP Divisions of the Office of Science of the U.S. DOE; the U.S. NSF; the BMBF of Germany; IN2P3, RA, RPL, and EMN of France; EPSRC of the United Kingdom; FAPESP of Brazil; the Russian Ministry of Science and Technology; the Ministry of Education and the NNSFC of China; Grant Agency of the Czech Republic, DAE, DST, and CSIR of the Government of India; the Swiss NSF. *URL: www.star.bnl.gov [1] G. L. Kane, J. Pumplin, and W. Repko, Phys. Rev. Lett. 41, 1689 (1978). [2] R. D. Klem et al., Phys. Rev. Lett. 36, 929 (1976); W. H. Dragset et al., Phys. Rev. D 18, 3939 (1978). [3] S. Saroff et al., Phys. Rev. Lett. 64, 995 (1990); B. E. Bonner et al., Phys. Rev. D 41, 13 (1990). [4] B. E. Bonner et al., Phys. Rev. Lett. 61, 1918 (1988); A. Bravar et al., ibid. 77, 2626 (1996); D. L. Adams et al., Phys. Lett. B 261, 201 (1991); 264, 462 (1991); Z. Phys. C 56, 181 (1992). [5] K. Krueger et al., Phys. Lett. B 459, 412 (1999); C. E. Allgower et al., Phys. Rev. D 65, 092008 (2002). [6] P. R. Cameron et al., Phys. Rev. D 32, 3070 (1985); D. G. Crabb et al., Phys. Rev. Lett. 65, 3241 (1990). [7] A. Airapetian et al., Phys. Rev. Lett. 84, 4047 (2000); Phys. Lett. B 535, 85 (2002); 562, 182 (2003). [8] A. Bravar et al., Nucl. Phys. Proc. Suppl. 79, 520 (1999). [9] For a review, see V. Barone, A. Drago, and P. G. Ratcliffe, Phys. Rep. 359, 1 (2002). [10] T. Sjöstrand, Comput. Phys. Commun. 82, 74 (1994). [11] J. Collins, Nucl. Phys. B396, 161 (1993). [12] D. Sivers, Phys. Rev. D 41, 83 (1990); 43, 261 (1991). [13] A. Efremov and O. Teryaev, Phys. Lett. 150B, 383 (1985). [14] M. Anselmino, M. Boglione, and F. Murgia, Phys. Rev. D 60, 054027 (1999); M. Boglione and E. Leader, Phys. Rev. D 61, 114001 (2000). [15] M. Anselmino, M. Boglione, and F. Murgia, Phys. Lett. B 362, 164 (1995); M. Anselmino and F. Murgia, ibid. 442, 470 (1998); U. D’Alesio and F. Murgia, AIP Conf. Proc. 675, 469 (2003). [16] J. Qiu and G. Sterman, Phys. Rev. D 59, 014004 (1998). [17] Y. Koike, AIP Conf. Proc. 675, 449 (2003). [18] A. Zeleznik et al., in Proceedings of the Particle Acceleration Conference (IEEE, New York, 1999), p. 106. [19] G. Bunce et al., Annu. Rev. Nucl. Part. Sci. 50, 525 (2000); H. Huang et al., Phys. Rev. Lett. 73, 2982 (1994); M. Bai et al., ibid. 80, 4673 (1998). [20] Ya. S. Derbenev et al., Part. Accel. 8, 115 (1978). [21] I.G. Alekseev et al., AIP Conf. Proc. 675, 812 (2003). [22] H. Spinka, AIP Conf. Proc. 675, 807 (2003). [23] O. Jimnouchi et al., AIP Conf. Proc. 675, 817 (2003). [24] J. Tojo et al., Phys. Rev. Lett. 89, 052302 (2002). [25] C. Allgower et al., Nucl. Instrum. Methods Phys. Res., Sect. A 499, 740 (2003). [26] J. Kiryluk, AIP Conf. Proc. 675, 424 (2003). [27] J. Adams et al., Phys. Rev. Lett. 91, 172302 (2003); A. Drees and Z. Xu, in Proceedings of the Particle Acceleration Conference (IEEE, Chicago, IL, 2001), p. 3120. [28] GEANT 3.21, CERN program library. [29] F. Aversa et al., Nucl. Phys. B327, 105 (1989); B. Jager et al., Phys. Rev. D 67, 054005 (2003); D. de Florian, ibid. 67, 054004 (2003). [30] B. A. Kniehl et al., Nucl. Phys. B597, 337 (2001). [31] S. Kretzer, Phys. Rev. D 62, 054001 (2000). [32] J. Pumplin et al., J. High Energy Phys. 07 (2002) 012. [33] P. Aureneche et al., Eur. Phys. J. C 13, 347 (2000). [34] S. S. Adler et al., Phys. Rev. Lett. 91, 241803 (2003).
1.0 VISION and MISSION 1.1 The Foundation Vision is inspired philanthropy and enhanced student success. 1.2 The Foundation Mission is dedicated to advancing the Nova Scotia Community College’s Vision, Mission and Promise through philanthropy and prudent investment management. 2.0 GOALS OF THE POLICY 2.1 The goals of this policy are: (i) To maximize investment returns of the endowment funds, consistent with accepted risk tolerance levels supporting consistent annual cash flow requirements. (ii) To preserve in real dollar terms the capital of endowment assets. (iii) To at least earn the expected required rate of return as per the Endowment and Management Spending policy (3.5% net of inflation and fees) over a four year rolling period for the portfolio to accommodate the Foundation’s annual income disbursement. 3.0 RESPONSIBILITIES/ACCOUNTABILITIES: 3.1 The Board has the responsibility to oversee the management and financial reporting processes of the Foundation while satisfying itself that the resulting information reasonably reflects the organization’s financial position. 3.2 The AF&I Committee acts in a financial advisory capacity to propose recommendations to the Board regarding the: - Annual development and revisions to this policy including formulation of investment philosophy and asset mix strategy - Monitoring of investment performance quarterly to ensure expectations are met - Monitoring of the investment manager at least annually and recommending the appointment of a new investment manager as required 3.3 The Foundation’s Management Team is responsible for preparing financial reports, ensuring the effectiveness and efficiency of financial processes and developing effective internal controls. 3.4 More detailed roles and responsibilities are outlined in the Endowment Management and Spending Policy for reference. 4.0 PERMITTED INVESTMENTS 4.1 In investing Foundation capital, an investment manager must exercise the care, skill, diligence and judgment that a prudent investor would exercise in making investments. Permitted investment and prudent investor is referenced from the Trustee Act (Nova Scotia). 4.2 The investment manager may hold any part of the portfolio in one or more pooled funds managed by the Manager. It is expected that the pooled funds be operated within constraints reasonably similar to those contained in this Policy. The Manager shall notify the Committee in the event that a pooled fund exhibits, or may exhibit, any significant departure from this Policy. 4.3 Fund investments include: - Cash investments (T-bills, commercial paper and bankers acceptances) will be rated no lower than R-1 low or its equivalent, or will be in no bond rated below BBB or its equivalent by the DBRS. No more than 10% of the cash component may be invested in paper of a single issuer. - The credit quality of the fixed income portion of the portfolio must have an average credit quality of BBB - or better. No more than 25% of the overall fixed income allocation may be invested in instruments with a credit quality less than BBB - and no more than 20% of the fixed income portion of the portfolio may be invested outside of Canada. - Investment in Canadian conventional mortgages is permitted. No more than 50% of the overall fixed income allocation may be invested in domestic conventional (non-guaranteed) mortgages on income producing commercial, residential and industrial properties. - All equity investments (Canadian, US and foreign) will be in companies which trade on a recognized major exchange. At no time can shares in a single company represent more than 5% of the overall portfolio based on market value. - Alternative investments such as real estate, infrastructure and hedge strategies are permitted investments as part of the Foundation’s overall investment strategy. 4.4 Exception: Gift of Shares - Normal business practice is to redeem gifts of shares to cash on the day received. As per the approval of the Board and NSCC Executive, 38,000 of TD Bank common shares were gifted by the Donald R. Sobey Foundation and held in a non-fee account with the investment manager. - Investment income generated by these shares will fund the Donald R. Sobey Bursary and will be automatically received to the NSCC Foundation bank account quarterly. - The market value of the shares will increase or decrease in the applicable quarter and be reflected in unrealized gains or losses on the financial statements accordingly. - These gift of shares will not be redeemed to cash in whole or in part unless the Donald R. Sobey Foundation is in agreement (i.e. no encroachment of capital) 4.5 Derivatives - The Manager may use derivatives, such as swaps, options, futures and forward contracts, for hedging purposes, to protect against losses from changes in interest rates and market indices, for non-hedging purposes, and as a substitute for direct investment. - With prior written approval of the Committee, the Manager may use derivatives for speculative trading or to create a portfolio with leverage. Any use of derivatives for leverage or speculative purposes is limited to alternative strategies (such as Absolute Return) and limited to no more than 15% of the Foundation’s portfolio. - The Committee will review the investment manager’s hedging policy on an ongoing bases and approve any changes to their hedging policy. 5.0 RETURN AND RISK OBJECTIVES 5.1 The pooled endowment fund must be constructed in such a manner as to target the long term desired outcome of the Foundation as indicated in section 2.0. 5.2 The Foundation’s pooled endowment fund return objective is to generate a total investment return based on the following; 15% FTSE Canada Short Term Overall Bond Index 15% FTSE Canada Universe Bond Index 20% S&P/TSX Capped Composite Index 40% MSCI World Net Index ($C) 5% MSCI Emerging Markets Net Index ($C) 5% FTSE Canada 30-day T-bill Index 5.3 Diversification amongst asset classes is provided through the asset allocation guidelines in section 6.0. Diversification within each asset class is provided by limiting the percentage of the market value of Foundation assets invested in a single issuer noted in section 4.3. 5.4 In maintaining acceptable risk levels, the pooled endowment fund can assume a medium to medium low investment risk in a balanced portfolio. 6.0 ASSET ALLOCATION 6.1 The investment manager must consider the following criteria in planning the investment of Foundation assets, in addition to any others that are relevant to the circumstances: (a) General economic conditions (b) Possible effects of inflation or deflation (c) Impact of each investment or course of action to the overall Foundation portfolio (d) Expected total return from income and the appreciation of capital (e) Investment fee structure (f) Liquidity needs, regularity of income and preservation or appreciation of capital 6.2 Target asset mix and ranges to meet investment goals are as follows: \| Asset Class \| Target Asset Mix \| Minimum \| Maximum \| \|------------------------------\|------------------\|---------\|---------\| \| Cash/Money Market \| 0% \| 0% \| 10% \| \| Fixed Income \| \| \| \| \| Canadian Universe Bond \| 7.5% \| 0% \| 20% \| \| High Yield Bond \| 0% \| 0% \| 10% \| \| Core Plus Bond \| 7.5% \| 0% \| 20% \| \| Conventional Mortgages \| 15% \| 0% \| 20% \| \| Total Fixed Income \| 30% \| 25% \| 45% \| \| Diversified Equities \| \| \| \| \| Canadian \| 20% \| 15% \| 35% \| \| Foreign \| 45% \| 30% \| 50% \| \| Low Volatility Global \| 20% \| 15% \| 35% \| \| Global Equity \| 20% \| 15% \| 35% \| \| Emerging Markets \| 5% \| 0% \| 10% \| \| Total Diversified Equities\| 65% \| 50% \| 70% \| \| Alternative Investments \| \| \| \| \| Multi-Strategy Alpha \| 5% \| 0% \| 15% \| \| Total Alternative Investments \| 5% \| 0% \| 15% \| 7.0 MONITORING AND REPORTING 7.1 On a quarterly basis, the investment manager will provide the following written reports: 7.2 On a quarterly basis, the investment manager will provide the following written reports: - Portfolio Summary - Performance Review 7.3 The investment manager will meet with the AF&I Committee in-person on a regular basis (at least bi-annually) to provide commentary and presentation of: - an economic outlook and summary of portfolio strategy including successes and failures of each asset class - forward-looking investment views and plans considering the Foundation’s position - any necessary changes in the fund’s investment strategy, including changes in the status of the firm and its personnel - an explanation of investment performance in comparison to the Foundation’s mission and return objectives 7.4 Each quarter, the AF&I Committee will evaluate the pooled endowment fund performance based on the benchmark portfolio outlined in section 5.2, fund objectives, strategy, long-term expected risk, returns and other qualitative factors that may impact the ability to achieve desired investment goals. 7.5 Annually, the AF&I Committee will evaluate the Investment Manager’s performance based on the four year rolling average returns as compared to the funds objectives, investment performance, and meeting reporting requirements and risk taken to achieve such performance. 7.6 The Investment Manager will meet with the Board during the Foundation’s AGM. 8.0 CONFLICT OF INTEREST 8.1 A conflict of interest, whether actual or perceived, is defined for the purpose of this policy as any event in which the Board, an employee of the Foundation, any Manager or delegate, the custodian, or any person directly related to any of the foregoing, knowingly permits his/her interest to conflict with his or her duties or powers relating to the investment of the endowment assets or to any other matter relating to the Endowment Fund which may benefit materially from the knowledge of and participation in an investment decision. 8.2 Any actual or perceived conflict of interest shall immediately be disclosed in writing to the AF&I Committee. The AF&I Committee will then determine whether the conflict of interest exists and, if determined that it does exist, will take all necessary and appropriate measure to remedy the situation. All disclosures pertaining to a conflict of interest will be recorded in the AF&I Committee minutes. 9.0 VOTING RIGHTS 9.1 The responsibility of exercising and directing voting rights acquired through Foundation investments are delegated to the investment manager who will at all times act in the best interest of the Foundation. 10.0 POLICY REVIEW 10.1 Policy review will occur every year in order to assess the expectations and outcomes of this policy and ensure investments are maximizing returns to achieve the best benefit to donors and to NSCC’s strategic priorities. 11.0 DEFINITIONS AF&I Committee: Refers to the Foundation’s Audit, Finance, and Investment Committee; an advisory sub-committee of the Foundation’s Board. Board: Refers to the Foundation’s Board of Directors. Capital: The original donation(s) to capital of an endowment and any additional appreciation from inflation with the intent of being held in perpetuity. DBRS: Dominion Bond Rating Service. A globally recognized provider of timely credit rating opinions that offer insight and transparency across a broad range of financial institutions, corporate entities and government bodies and various structured finance product groups. Endowment: A restricted donation made to the Foundation by a donor in which the principal value is invested. A portion of the earned income from the investment supports a specified purpose as defined by the donor through a formal gift agreement. Foundation: Refers to the NSCC Foundation. Foundation Management Team: Refers to the Executive Director, Senior Development Officer, and Manager, Donor Care. FTSE Canada Universe Bond index: provides benchmark performance standards for Canadian fixed income investments. Investment earnings: The realized gains, unrealized gains and interest income earned or received from the pooled endowment fund in a specified period of time. Investment risk: The probability of occurrence of losses relative to the expected return on any particular investment due to the unpredictability of market direction. Investment manager: Refers to the appointed investment manager who manages the endowment on behalf of the Foundation. Investment management fees: All direct expenses incurred to properly manage, control, and report on the investment activities of the endowment, including investor relations expenses and administration costs. Market value: The current value of the pooled endowment fund, including capital and market stabilization (net investment earnings) and cash available for disbursement. MSCI Emerging Markets Net Index ($C): Morgan Stanley Capital International stock market index, captures large and mid-cap representation across 24 Emerging Markets (EM) countries. MSCI World Net Index ($C): Morgan Stanley Capital International stock market index, captures large and mid-cap representation across 23 Developed Markets countries. NSCC: Refers to the Nova Scotia Community College. Pooled endowment fund: Allows multiple donor endowments to invest in the same pool of assets. Funds are invested as a single pool in order to optimize net return and diversification of risk. The pooled endowment fund includes all individual endowment funds reported on a pro-rata basis. S&P/TSX Capped Composite Index: is the benchmark Canadian index, representing approximately 70% of the total market capitalization on the Toronto Stock Exchange (TSX) with about 250 companies included in it. Dave Saxton Chair, NSCC Foundation Board of Directors Dale Noseworthy Treasurer, NSCC Foundation Board of Directors Next Review Date: March 2020
The Use of LANDSAT Data in a Large Area Crop Inventory Experiment (LACIE) R. B. MacDonald F. G. Hall R. B. Erb Follow this and additional works at: http://docs.lib.purdue.edu/lars_symp MacDonald, R. B.; Hall, F. G.; and Erb, R. B., "The Use of LANDSAT Data in a Large Area Crop Inventory Experiment (LACIE)" (1975). LARS Symposia. Paper 46. http://docs.lib.purdue.edu/lars_symp/46 This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact firstname.lastname@example.org for additional information. Reprinted from Symposium on Machine Processing of Remotely Sensed Data June 3 - 5, 1975 The Laboratory for Applications of Remote Sensing Purdue University West Lafayette Indiana IEEE Catalog No. 75CH1009-0-C Copyright © 1975 IEEE The Institute of Electrical and Electronics Engineers, Inc. Copyright © 2004 IEEE. This material is provided with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the products or services of the Purdue Research Foundation/University. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to email@example.com. By choosing to view this document, you agree to all provisions of the copyright laws protecting it. THE USE OF LANDSAT DATA IN A LARGE AREA CROP INVENTORY EXPERIMENT (LACIE) R.B. MacDonald, F.G. Hall, and R.B. Erb Lyndon B. Johnson Space Center, National Aeronautics and Space Administration, Houston, Texas I. ABSTRACT A Large Area Crop Inventory Experiment (LACIE) has been undertaken jointly by the U.S. Department of Agriculture, the National Oceanic and Atmospheric Administration (NOAA) of the Department of Commerce and the National Aeronautics and Space Administration (NASA) to prove out an economically important applica- tion of remote sensing from space. At the outset LACIE will concentrate on wheat grown in the North American area. The experiment will combine crop area measurements obtained from LANDSAT data and meteoro- logical information from NOAA satellites and from ground stations designed to relate weather conditions to yield assessment and ultimately to production fore- casts. The Department of Agriculture will study the utilization of the experimentally derived production estimates in its crop reports. These reports are made public as a routine service to the domestic and international agriculture community. If this activity is successful and the results prove useful the applica- tion will be extended to other regions and ultimately to other crops. II. INTRODUCTION A series of experimental investigations, utilizing multispectral and meteorological data to identify and measure the areal extent of major crop types and to estimate their yields, has established a base of technology, which if properly expanded, can satisfy the requirements of a major agricultural application objective, i.e., crop production inventories over large areas. The Large Area Crop Inventory Experiment (LACIE) will expand on this available tech- nology base and assemble an experimental system for demonstrating a crop production inventory in a meaningful "quasi-operational" environment. The experiment is supported by a parallel research, test and evaluation (RT&E) effort designed to develop solutions in areas where additional technology may be required. The operation and evaluation of the experimental system will be iterated with the RT&E effort to develop a technology which either satisfies the applications objectives or demonstrates that the applications objectives can be satisfied, given that solutions can be developed for specific key problems. Thus, the expected accomplishments of LACIE will be the development and testing of the technology to produce agricultural crop production inventories on a global scale or, alternatively, the definition of any key problems to be solved prior to the implementation of an operational system. This paper describes the background of events which shaped the LACIE design, the technical approach being pursued, the details of the implementation of this approach and initial results of the experiment. The authors wish to acknowledge the many contributions to this experiment, its planning, implementation and execution, by their colleagues not only in NASA but more particularly in the U.S. Department of Agriculture (USDA) and in the National Oceanic and Atmospheric Administration (NOAA) of the U.S. Department of Commerce. This is truly a joint venture among the three agencies and many have played important roles in bringing the experiment to its present state. III. BACKGROUND To appreciate the particular approach chosen, of the many possible approaches, the general context surrounding the design of LACIE must be understood. The major factors which influenced the design decisions were: (1) the applications objectives and requirements to be satisfied by the technology; (2) the status of the existing remote sensing technology; (3) the time frame imposed to accomplish the stated objectives; (4) the estimated available resources for the experiment and (5) the constraints, both self-imposed and indigenous to the agencies participating in LACIE. Each of these factors will be treated in the following paragraphs. APPLICATIONS OBJECTIVES AND REQUIREMENTS A general application objective shaping the overall LACIE design was to develop, test and prove an economically important application of remote sensing from space. The crop inventory application was chosen since it represented an economically important application which could feasibly be accomplished near term with a system built from existing technology. Wheat was chosen as the crop for the experiment because of its importance in human nutrition and international trade. From this general definition of the objective followed more specific requirements. To prove out this technology, wheat production, area and yield estimates would be necessary on a regional and national level. The system would have to be capable of producing periodic reports from plots taken through post harvest containing wheat area, yield and production estimates with a quantitative assessment of the confidence and accuracy of these estimates. In addition the reports should identify the wheat growth stage at which the estimate was made and list all source data used to derive the estimate. The accuracy and/or timeliness of this information must improve upon the accuracy and/or timeliness already obtainable by the USDA from areas outside the U.S. and Canada. In addition all information would need to be relatable to geographic coordinates. The experiment would have to demonstrate and produce experience and information applicable to the design of an eventual operational system. The system design itself was required to conform to existing USDA information security requirements and where feasible to existing standard specifications to facilitate technology transfer. Finally, the system would have to be designed to provide a basis for a potentially cost effective operational system. STATUS OF THE AVAILABLE TECHNOLOGY At the outset of LACIE, a careful review and analysis of the status of the remote sensing technology indicated that wheat production inventories over large areas were feasible. The major task facing LACIE was and is to expand the technology from the relatively local areas for which it had been developed and tested to the large areas over which it would be applied. For area estimation, the major efforts would have to focus on the development of methodology to obtain training statistics for inaccessible locations and to extend these statistics over large regions; of equal importance, operations procedures would have to be developed to (1) minimize the impact of cloud cover interference on LANDSAT data acquisition and (2) to manage the analysis of the large volumes of data required for the large area application. Historically, a number of key developments contributed to the base technology. These developments are described in the remainder of this section. In 1966 scientists coupled computerized pattern recognition analysis techniques with remotely-sensed multispectral electromagnetic energy measurements to classify major agricultural crops automatically. (Reference 1). In particular it was learned that mature wheat could be quite accurately recognized in this way. The first major application of this technology occurred during the Corn Blight Watch Experiment (Reference 2) in 1971. This experiment provided the first demonstration that a single agricultural crop could be identified over large areas using computer-aided analysis of multispectral information. This feasibility demonstration was designed around a remote sensing technology base developed during more than six years of research into the computer-aided analysis of multispectral data acquired from aircraft. In addition during this period experiments were conducted from space with remote sensors carried on the Mercury, Gemini and Apollo satellite systems. A most significant investigation conducted on Apollo 9 was the SO-65 experiment. This was a simulation of the future LANDSAT multispectral scanner (MSS). Photographs were acquired by an array of four cameras having film/filter combinations chosen to simulate the then-planned spectral bands of LANDSAT. Scientists digitized these photographs, conducted computer processing and demonstrated that agricultural crops could in fact be automatically recognized with LANDSAT-like measurements acquired from space (Reference 3). In July 1972 the first Earth Resources Technology Satellite, initially known as ERTS-1 and since renamed LANDSAT 1, was launched and computer-aided analysis of multispectral data from a space platform commenced. The characteristics of the spacecraft acquired MSS data as compared to aircraft-acquired data (narrower scan angles, near simultaneous coverage over larger areas and repetitive coverage) led to the simplification of many of the problems encountered with aircraft platforms and opened the way toward the development of a greatly expanded remote sensing technology base. A series of critical agricultural feasibility investigations using LANDSAT 1 MSS data acquired for a diverse set of test sites and environmental circumstances (References 4 through 8) generally established that major crops could be identified and measured with reasonable accuracy. The degree to which the feasibility had been established was somewhat limited however, in that these investigations were of limited scope, i.e., results were obtained at a single time over a limited area and were conducted against somewhat simple experimental designs which did not establish the performance indices needed to predict the performance associated with an areal inventory. These early results were promising enough to convince some within the remote sensing community that a technology base existed which could be expanded to conduct gross inventories of the areal extent of agricultural crops over larger areas. A follow on effort, the Large Area Project, was proposed by NASA/JSC in mid-1973, which would intensify the LANDSAT 1 feasibility effort and establish classification accuracy and mensuration accuracy to identify major types of wheat grown in selected U.S. test sites. In parallel, the repeatability of the prior LANDSAT 1 investigations for selected row and small grain crops would be verified. At the same time the USDA and Canada entered upon a joint study for spring wheat identification. In addition, a carefully controlled experiment was initiated over test sites in the U.S. to assess the capabilities of the most promising ADP techniques to identify several major food crops. This latter effort, known as the Crop Identification Technology Assessment for Remote Sensing (CITARS) is being concluded and will be documented in the future. Concomitant to these events, interest was developing within USDA in having a more effective and timely method to inventory the available world food supply. Consequently, in the final quarter of 1973, discussions were begun among personnel of NASA and the USDA centering around the most reasonable approach to the development and demonstration of a major application of remote sensing technology. It was agreed first that, for several reasons, the gross inventory of a single crop, wheat, would represent an application of considerable value. Wheat was chosen primarily because of the considerable experience which had been acquired in the remote sensing of wheat (identification and area mensuration) and because of its importance as a crop. Wheat exceeds in production and in areal extent any other grain crop in the world. From a technical standpoint, the simplification offered by focusing on the identification and mensuration of a single crop was attractive. In this way, the major problem to be faced, i.e., expanding the technology to a large area application, could be achieved with a minimum of distractions. Subsequently, based on the successful demonstration of this expanded technology, the focus could be shifted to solving problems related to multiple-crop applications. The status of the technology in relation to the identification and mensuration of wheat was still somewhat uncertain so preliminary feasibility investigations for wheat were conducted at the JSC and as part of CITARS at the Environmental Research Institute of Michigan (ERIM). These investigations generally indicated that in major wheat-producing regions the identification performance characteristics for wheat was similar to other crop identification performances reported. However, there were indications of difficulties in inventorying marginal wheat-producing areas such as southern Indiana and Illinois. The JSC investigation utilized LANDSAT 1 data acquired over Hill County, Montana, for three biological phases of wheat (greening, heading and mature) and for two biophases over Burke County, North Dakota (emergence and jointing). These data were processed using only some of the algorithms to be used in LACIE and with much more analyst intervention than planned for LACIE. In addition, multiple LANDSAT passes were spatially registered and processed to evaluate the effect of multitemporal data on crop identification performance. The results of this investigation are shown in Table I. The single pass results were similar to those reported in a variety of other investigations; as can be seen, the use of multipass data provided a considerable improvement in performance. However, the relationship that these performance numbers would bear to the expected performance for an area inventory were not obvious. The obscure relationship between the reported performance indices and the area estimation accuracy, which could be expected to result from an application of these techniques, was due primarily to three factors: (1) the relationship between the performance indices of the per pixel classifier and the area estimation accuracy was not completely established and the majority of performance results reported were not in terms of areal estimation accuracy. (2) the unknown degree to which the classifier performance would be degraded over larger areas by factors known to affect signatures such as differences in atmospheric haze and sun angle, soil color, growing seasons, and agricultural practice and (3) the error interactions between the areal estimation model to be used for LACIE and the "per pixel," maximum likelihood classifier to be used were unknown. The latter two of these three factors would have to be answered by an experiment over larger areas than had previously been examined and thus would have to be addressed by the experimental system in LACIE. The first of these, was partially addressed, both empirically and theoretically in early follow-on feasibility investigations for LACIE. For the per pixel classifier to be used in LACIE the estimate, $P_{ew}$, of the areal proportion of wheat in a sample would be the ratio of the number of pixels $P_w$ classified as wheat to the total number of pixels in the sample. Thus in terms of the $P(w/w)$ and $P(w/o)$ in Table I, $P_{ew}$ for a sample would be $$P_{ew} = P(w/w)P_w + P(w/o)(1-P_w) \quad (1)$$ where $P(w/w)$, $P(w/o)$ are as defined by Table I and $P_w$, is the actual proportion by area of wheat in the sample. Thus the fractional difference $D$ between the estimate $P_{ew}$ and $P_w$ would be related to $P(w/w)$, $P(w/o)$ and $P_w$ by $$D = \frac{P_{ew} - P_w}{P_w} = \frac{P(w/w)P_w - P(w/o)(1-P_w)}{P_w} - P_w \quad (2)$$ This expression indicates the sources of the difficulties in relating the per pixel probabilities for error to the overall area estimation accuracy. The magnitudes of the terms $P(w/w)$ and $P(w/o)$ depend on the confusion crops present which can vary considerably over regions planted to wheat. In turn, the fractional difference between the estimated and actual proportion as can be seen from equation (2) depends on the relative amount of wheat present in the scene. The values of $P(w/w)$ and $P(w/o)$ reported in the literature however, were for specific confusion crops with specific relative abundancies often unknown or unreported by the investigator. For these reasons, a study was initiated within JSC to obtain preliminary estimates of $D$ resulting from the per segment areal estimation scheme to be utilized for LACIE. For this study three classification runs were made using LANDSAT 1 data obtained over the Hill County, Montana (for a 2 x 6 mile) test site. Two of the classifications were made using single LANDSAT passes acquired on April 16, 1973, and May 23, 1973, during the green and headed biophases respectively. The third classification utilized twelve channels, a spatially registered combination of these two single passes, plus a pass acquired on June 27, 1973. For this segment within Hill County the areal proportion of wheat $P_w$ is 0.302 as determined from insitu observations. Table II gives the results of this study. Notice that the values for $P(w/w)$ and $P(w/o)$ are somewhat "worse" than the values shown (Table I) for the earlier JSC studies conducted over the same site. These latter values were obtained using processing procedures more representative of the LACIE procedures in that analyst intervention and iterations were greatly reduced. Even with these reported per pixel classification accuracies of from 50.8 percent to 85.1 percent, the proportion estimates for the segment were accurate to within ten percent of the observed values, ranging from a maximum difference of 8.9 percent to as small as a one percent difference. These relatively small values of D result from the fact that for these cases the errors of omission, P(w/w) - P_w, which lead to underestimates of the amount of wheat, are largely balanced by the errors of commission, P(w/o) - (1-P_w), which lead to overestimates of the amount of wheat. If this error-cancelling tendency could be maintained operationally for all LACIE segments processed and D tended to fluctuate randomly about zero for these segments, then the classifier tested above would provide large area estimates with accuracies exceeding 30 percent. For large area inventories, the rms error of the estimate given by equation (1) would also be important since for a given sample unit the total area estimation error would be given by \[ E = \left( b^2 + v^2 \right)^{1/2} \] where $ b $ is the average bias associated with the estimator and $ v $ is the variance component associated with the estimator. Most investigations prior to LACIE had not been designed to quantify $ V $ and it would be important to know the relative magnitudes of $ b $ and $ V $ in judging the ability of a particular classification scheme to satisfy a particular criterion for area estimation accuracy. A desirable classifier would be one leading to a small or known value for the bias. With such a classifier the area estimation error $ E $, obtainable could then be decreased to an acceptable level by increasing the number of samples examined; or if the bias were known the estimate could be corrected. Without this property, the area estimate over a region would have a residual unknown error, whose magnitude could not be reduced by increasing the amount of data processed. The CITARS effort initiated prior to LACIE was designed to provide estimates of the performance parameters discussed above. The results from this investigation indicate that certain of the classification procedures tested do not have a negligible bias and that there may be some difficulty in correcting this bias. However, certain other classification procedures, evaluated for performance on corn and soybeans at optimum times for discrimination of these crops, have associated values of bias and variance which would render them acceptable as area estimators of these crops. A test and evaluation effort similar to the CITARS investigations, has been designed as an integral part of LACIE and will provide estimates of bias and variance specifically for wheat. Preliminary results of this effort will be presented subsequently in this paper. Since total production estimates are the desired end product of a crop inventory, investigations into available data and literature were begun to determine: (1) to what extent wheat yield (production per unit area) would need to be monitored and (2) the status of the available technology for wheat yield estimation. An examination of the agricultural statistics reported by various countries indicated that yearly variations in reported yield accounted for as much of the yearly variation in reported production as did yearly variations in reported area. Thus it would be necessary to monitor both area and yield if remote sensing was to provide production estimates which improved upon those estimates obtainable by existing methods. An examination of the status of the available technology for wheat yield estimation indicated that the key technical issue to be faced in yield estimation was the development and testing of available yield models over regions important to world wheat production. This in turn required a considerable effort devoted to the development of the historic yield and meteorological data base required for yield model development, and a careful evaluation of the effect on the accuracy of the yield predictions of factors not directly accounted for in the available models, such as fertilizer practice and catastrophic events such as insects, disease, etc. To establish the status of yield estimation technology, personnel from NASA/JSC conferred with investigators in the field and with personnel from the National Oceanic and Atmospheric Administration (NOAA) who were already planning to monitor meteorological conditions for crop condition reports. Of the methods investigated for wheat yield determination, agro-meteorological (agromet) models appeared to be the most promising approach for wheat yield estimation in a large scale application. The application of multispectral sensing to determine crop condition was (and is still in the early stages of development). The relationship between spectral data and yield had not been quantified, although multispectral imagery is considered to be one of several sources of information regarding crop condition. Agromet models existed which permitted yield estimates based upon measured values of key meteorological parameters. The feasibility of these models for yield estimation was indicated by physiological studies which demonstrated that factors such as temperature and moisture played key roles in influencing wheat yield. In addition to these studies, statistical analyses in operational settings (such as described in References 9 through 12) verified the strength of these relationships. Values for the required meteorological variables could also be obtained on a near real time basis from ground stations and the potential existed for supplementing these data with environmental satellite data. From the investigations described above it appeared that an experimental crop inventory system utilizing remote sensing technology, could be developed over a two to three year period. This system would demonstrate that remote sensing could upgrade existing information-gathering capabilities and would demonstrate that the crop inventory application requirements could be satisfied given the solutions to certain key problems. TIME CONSTRAINING FACTORS The time frame and schedules imposed on LACIE resulted from several considerations. Of these the major ones were: (1) the remote sensing community had a technical requirement to maintain continuity in the development and use of satellite MSS data for practical applications which in view of the anticipated lifetime of the LANDSAT-1 sensors dictated a LANDSAT-2 launch in early 1975 with the possible LANDSAT-3 launch in 1977; (2) the time anticipated to extend the technology and develop an inventory system and (3) the particular timing for the wheat crop cycles over the LACIE regions of interest. Considering all this, a phased approach was chosen which consisted of 3 phases spanning approximately three and one half years. Phase I concentrated on (1) a test of the capability of an area estimation system built from existing technology to determine wheat area within regions of the U.S. and to classify wheat in other areas and (2) development and testing of yield and production estimation models over regions in the U.S. Phases II and III would test LACIE capabilities to determine wheat area, yield and production in the U.S. and other wheat-producing regions. RESOURCE AVAILABILITY The three major determinants of resources required to execute LACIE were: (1) the design and construction of the experimental inventory system from available technology; (2) the provision of people and computers to analyze the large volumes of data required to support the application objectives and (3) the funding for the research, test and evaluation required to evaluate and fill gaps in the available technology. The agencies involved placed a high priority on accomplishing the stated objectives of LACIE as soon as feasible, and resources were made available to pursue the LACIE objectives vigorously with a goal of delivering, at the culmination of LACIE, proven technology plus a definition of the key problems to be solved prior to the implementation of an operational inventory system. OTHER CONSTRAINING FACTORS Last but not least in the consideration of the LACIE design were self-imposed constraints and certain organizational and institutional factors peculiar to the agencies participating in LACIE. Certain constraints were self-imposed by the project designers to simulate the structure of the anticipated operational situation. The two most important constraints arising from this philosophy were: (1) the use of insitu acquired ground observations for the current year would be restricted, (2) the experiment would be conducted in a quasi-operational manner with analysis done in real-time (14 days from data acquisition to completion of processing) with reporting geared to current crop reporting schedules. Constraint (1) above restricted the acquisition of insitu crop type identifications to limited areas in the U.S. and further limited the use of this data to the evaluation of LACIE and to the development of classifier training statistics for regions outside the U.S. This restriction required that two new elements of technology be developed. (1) analyst procedures for manual analysis of LANDSAT imagery for crop type identification over a small fraction of the data (2 percent of the LANDSAT data processed) and (2) signature extension procedures for extension of the spectral signatures to the remaining 98 percent of the LANDSAT data. Since the LACIE effort required a joint application of NASA, USDA and NOAA resources, constraints were imposed upon the LACIE design which reflected the existing programs, capabilities, facilities, policies and goals of each agency. The structure within NASA which shaped the design of LACIE in a major way was the nature of the existing implementation for ADP analysis of multispectral data, and the extent to which this implementation could be modified or augmented given the projected resources and schedule. The Earth Resources Interactive Processing System (ERIPS) contained a majority of the implemented data processing technology readily available to LACIE. ERIPS was resident on IBM 360/75 machines in the Houston Real Time Computer Complex (Mission Control Center). To meet the schedules imposed upon LACIE, a decision was made to develop the LACIE experimental system around ERIPS, changing or augmenting it only where necessary. The influence of this decision is strongly reflected in the current LACIE subsystem for processing LANDSAT data to provide area estimations. Another set of factors which strongly influenced the LACIE design was the decision to utilize multitemporal data, the associated state of the art in registration technology and the projected available resources for registering the LACIE MSS data. A reasonable upper limit on registration accuracy required for multitemporal recognition processing was estimated at one pixel rms. The registration technology which could readily be implemented for use by LACIE would permit a "one-job" registration (to the required accuracy) of roughly one-one hundredth of a LANDSAT frame, or less than about a 10 x 10 mile square on the ground. The requirement for one registration job per frame, with the projected available resources, precluded an optimally efficient sampling strategy with units of smaller size scattered throughout each LANDSAT frame; that is, the quantity of 10 x 10 mile portions required to cover the geographic area under consideration could not be registered within the projected available resources. Within these constraints a sample unit of 5 x 6 nautical miles (nm) was chosen as adequate. The major factors within the USDA which influenced the design of LACIE were: (1) the level of resources available to acquire the ground observations and measurements required to support critical development and evaluation tasks within LACIE and (2) the USDA desire to facilitate the transferability from NASA to USDA, of technology validated by LACIE. USDA projected sufficient resources to obtain observations and measurements within about 28 intensive test sites totaling about 285,000 acres. In addition, data published in USDA crop reports would be available. Although the resources are considerable to obtain even this amount of ground data, considerable ingenuity was required to develop an experimental design which could use this data set to evaluate properly new developments and to quantify the performance of the experimental system. In early 1974 project planning and certain key developments were begun to initiate the Large Area Crop Inventory Experiment. The year was spent developing the management approach, designing the experiment, designing and building the system to support the experiment and working out the necessary interagency agreements for joint USDA, NOAA and NASA participation. This work culminated in late 1974 in a Joint Project Plan, which was reviewed by persons representing the leading technical expertise within the remote sensing community. In November 1974, initial operations began with a preliminary data system for classification and mensuration of wheat using LANDSAT 1 data acquired over selected segments in Kansas. IV. DESCRIPTION OF THE LACIE The description of the experiment will be treated in two parts. The first section will describe the technical approach, i.e., the basic design of the experiment including the approach to area, yield and production estimation, the experimental data system design, the flow of the data through the system, the evaluation of the system output produced, and details of the needed test and evaluation facilities supporting LACIE. The second section will describe the functional approach, i.e., the functions required to support the development, test, operation and evaluation of the LACIE data system, the organization developed to execute these functions and the phases and schedules for implementing the basic approach. TECHNICAL APPROACH The crop identification and mensuration is carried out with LANDSAT multispectral scanner (MSS) data and the inventory is being performed on a sampling rather than an exhaustive coverage basis. Data is being acquired through the normal Goddard Space Flight Center (GSFC) processes over the geographic areas under study. A stratified random sampling strategy has been developed employing 5 x 6 nautical mile segments randomly allocated to strata according to the 1969 agricultural census data specifying areas planted to wheat in the U.S. A total of 637 sample segments were allocated to the U.S. to obtain a sample error of approximately 2 percent. This criteria is to some degree arbitrary since the precision and accuracy of the production estimates are the final concern in accuracy. However, at this stage of development not enough is known regarding the overall error interactions between the LACIE system components to design a sample strategy against a specific set of performance criteria. Preliminary analysis of the sample strategy indicates that the associated sample error (precision) is less than 2 percent at a 2σ confidence level. As will be discussed in section V a preliminary performance assessment of LACIE area estimates in Kansas provides no reason to alter the sample strategy design. In the RT&E effort alternate sample strategies are being investigated to minimize the cloud cover interference problems. The LACIE sample strategy for the U.S. allocates the total of 637 segments to counties such that each county receives sample units in proportion to the product of the total area of the county times the estimate of the standard deviation of wheat areal density in the county. The estimate of the standard deviation is derived from the 1969 agricultural census estimate of the wheat areal density for the county by assuming the wheat density to be binomially distributed within the county. This strategy usually gives from 0 to 5 segments to a county and for each stratum (crop reporting district (CRD) for the U.S.) a maximum of about 15. Figure 1 is a facsimile of segment locations within a CRD. Area estimates for counties receiving no segments will be ratio estimates to counties which do contain segments. Data for these 5 x 6 nm segments (containing 112 lines of 196 pixels each) will be extracted from the full scenes by GSFC and transmitted to JSC. Roughly 20 percent of the segments will be "training" segments. In the LANDSAT imagery acquired from these segments, training "fields" will be manually located to train the classification algorithm to identify wheat in the "training" and "ordinary" segments. An ordinary segment is one which does not contain training fields are selected. Training segment data is acquired at every opportunity to maximize the probability of acquiring acceptable data. Acceptable is defined to GSFC in terms of minimally tolerable interference from cloud cover (< 30 percent of segment area obscured). GSFC will similarly preprocess ordinary segments which will be transmitted only 4 times during the growing season, once for each of the biological phases of crop establishment, growth, heading, and maturity. The first data taken in each phase which meets quality criteria will be used. For each segment, the first take of the season will become a reference set and subsequent data takes will be registered to the reference set to form a multitemporal set of up to 16 channels. The analysis of the MSS data is being intentionally carried out without the use of current ground data. Ground observations will, of course, be used for evaluation purposes but they only are used operationally when they are typically available in real time over large areas from existing ground stations. This self-imposed constraint means it is necessary to train the classifier using LANDSAT data itself together with crop calendar information. To enhance the chances for success, "seasonally adjusted" crop calendars, developed from "normal year" calendars modulated by current year weather data are being used. The initial crop calendar update model implemented for LACIE (operated at Washington by NOAA) is based on the Robertson biopace model (Reference 13). This model, based on real time measurements of maximum and minimum daily temperature and USDA estimates of planting start dates, will be used to provide bi-weekly updated estimates of the actual times for occurrence of the LACIE biophases for each of 60 crop reporting districts in the U.S. In addition to use in training field identifications the outputs of this model will be used to specify LANDSAT data acquisition windows to GSFC for each LACIE biophase. Following receipt in Houston, LANDSAT digital data is converted to film image form and analyst interpreters select from training segments 40 to 50 training fields for wheat and other agricultural categories and provide a definition of the boundary of such fields to the ADP analyst for the classification. This represents manual analysis of about 2 percent of all LANDSAT data acquired. In these segments the analyst interpreter (AI) will rely mainly on interpretative keys which distinguish wheat from nonwheat based on tonal appearance and change over the growing season and spatial information such as texture and shape. In addition the AI will be provided with historical cropping calendar data for each segment. In regions where wheat has a crop calendar distinctly different from other crops, the AI should be able to accurately distinguish wheat from other, provided he has an accurate knowledge of the current year's cropping calendars for the various crops. The AI will be provided with weather summary data (from NOAA) each week summarizing meteorological events known to affect crop appearance. Snowfall, heavy precipitation or drought and temperature extremes will be the key variables of interest. While the AI procedures described above will be the backbone of the initial LACIE effort, the concept of extending signatures between regions known to be "analogous" to each other is being investigated in the RT&E activity. Image analysis will also be used for another purpose. Previously, difficulty has been experienced in classifying accurately such non-cropland areas as forests and towns. Frequently, even in areas with highly stereotyped data at the LANDSAT resolution, poor classification has resulted. Therefore, major non-cropland areas will be identified by image analysis and will be manually excluded. Similarly, areas in which data quality is poor due to such factors as clouds and noise can be identified and manually excluded. Major wheat-growing regions will be partitioned into smaller areas over which signature extension is expected to be successful. This partitioning will be accomplished based upon such ancillary data as crop calendars, meteorology, and soil color, as well as on the basis of trial classifications. Each signature extension region will contain one or more training segments. If cloud cover or other operational problems prevent acquisition of the training segment for a given signature extension region, signature extension will be attempted from a neighboring region. This "neighbor region" approach is expected to provide reduced classification accuracies in the region without a training segment. It is anticipated that signature extension regions will be typically about the size of a LANDSAT scene. The classification subsystem design is based upon the judgment that wheat can be separated adequately from other crops by analysis of up to four acquisitions of LANDSAT data during the biological development of wheat. The biophases chosen are: a. Crop establishment: planting to booting (with a gap during dormancy for winter wheat) b. Green: booting to heading c. Heading: heading to soft dough d. Mature: soft dough to harvest Signatures obtained on one calendar date within a biophase are not necessarily expected to be valid for other acquisition dates within the same phase; however, training field boundaries generated for one date will usually be valid on other acquisition dates within the same biophase. Therefore, by acquiring the training segments on each LANDSAT pass and using training field boundaries located on one acquisition date for that biophase statistics appropriate to a different acquisition date can be computed without expending additional image analysis effort. It is planned to conduct major image analyses only upon training segments and only once per biophase. Procedures for the classification were based in part on the CITARS procedures described in another Section (Background). Basically, the training data are clustered to aid in selecting suitable training classes, sun angle and mean level adjustments are made if required, a feature selection process can be employed to reduce the number of channels using the Mahalanobis distance as a separability measure. The segment is classified with a maximum likelihood classifier into wheat and nonwheat classes. The details of this procedure are covered in another paper (Trichel, et al.). A high degree of analyst interaction will be possible but the intent of the system is eventually to automate the classification as completely as possible. The fraction of each segment's area classified as wheat will be determined by ratioing wheat pixel count to total pixel count. The key RT&E issues being addressed in the classification of LANDSAT data are methods for developing training signatures, methods for extending the developed signatures over large areas and improved methods for estimating wheat proportions within the sample segments. The technology required for LACIE represents to some extent a departure from the existing technology base in that insitu ground observations have been traditionally used to "train" the classifier. The LACIE RT&E effort in this regard is focused on improving analyst interpreter techniques and the development of the "analog" area concept discussed earlier. Signature extension is key to LACIE in that the manpower expended per segment decreases drastically with increasing ability to apply training statistics over larger areas. Signature extension RT&E is focused on: (1) definition of methods to determine signature strata, i.e., geographic regions for which multispectral signatures are sufficiently homogeneous so as not to significantly degrade classification performance. Such strata will most probably be uniform in soil spectral characteristics, crop biophase and in the agricultural practices employed. (2) development of algorithms which permit signature extension between areas with different environmental conditions i.e., atmospheric conditions or sun angles. Finally, improved proportion estimation schemes are key to developing unbiased estimations of wheat proportions within segments. The importance of this is pointed out in the background section of this paper. Yield projections will be made from models which involve weather data, typically, precipitation and temperature. Such data will be obtained from current (ground) networks. The development of these yield models is being carried out by NOAA at their Center of Climatic and Environmental Assessment at Columbia, Missouri. The initial models which will be statistical in nature, i.e., expressions for yield as a function of key meteorological parameters will be derived from regression analyses using historical yield and weather data over each of a number of yield strata. For the U.S. in Phase I, a stratum will be a Crop Reporting District. Within any one stratum the same set of coefficients in the model will apply, however, varying weather conditions at various locations within the stratum will result in different projected yields. The operation of this model will take place at NOAA facilities in Washington D.C. and the results will be supplied to JSC. Basic meteorological parameters currently available on the World Meteorological Organization (WMO) network will provide the input data to the models. The later phases of LACIE may employ yield models of a more sophisticated type, such as the Baier model, in which plant growth phenomena are taken into account explicitly. Also, observations from environmental satellites may be utilized to extend and interpolate the meteorological data from the WMO network. The area and yield determinations per stratum provide the basic inputs to a production estimate. The LACIE system will produce monthly yield estimates, area estimates and production estimates for each major wheat-producing Crop Reporting District, state and region in the U.S. For each crop reporting district the area estimate will be computed by the relationship \[ A_j = A_{1j} + A_{2j} + A_{3j} \] (4) where $ A_{1j} $ is the estimate of the area in the counties within the $ j $th CRD which had no segments allocated, $ A_{2j} $ is the estimate for those counties which were allocated fractions of segments PPS (probability proportional to size without replacement) and $ A_{3j} $ is the estimate for counties allocated one or more segments. For counties falling into the first class the area estimate is \[ A_{1j} = (A_{2j} + A_{3j}) \frac{x_j}{w_j} \] (5) where $x_j$ is the agricultural census wheat area for counties in group 1 and $w_j$ is the agricultural census area for the CRD. For counties with total area $A_2$ falling into class 2 containing $m_j$ sample segments $$A_{2j} = A_2 \frac{p_j}{m_j} \sum_{k=1}^{m_j} \frac{\hat{p}_{jk}}{p_{jk}}$$ (6) where $\hat{p}_{jk}$ is the LACIE wheat proportion estimate within the $k$th county, $p_{jk}$ is the agricultural census wheat proportion estimate in that county and $p_j$ is the census estimate for the $j$th CRD. For the $m_j$ counties falling into class 3, $A_{3j}$ is simply the product of the average areal proportion of wheat in each county as estimated from the sample segments multiplied by the area of the counties containing the segments i.e., $$A_{3j} = \frac{A_3}{m_j} \sum_{k=1}^{m_j} p_k$$ (7) where $\hat{p}_k$ is the wheat areal proportion in the $k$th county. Methods for wheat yield aggregation are still being investigated as a part of the LACIE research test and evaluation effort as is production aggregation. The final step in each phase of LACIE is an evaluation of the output products of the LACIE system. This evaluation begins with an assessment of the "accuracy" of the output estimates of area, yield and production. These "accuracy" figures, are included with the area, yield and production estimates and transmitted in a report to an "Information Evaluation (IE) Group" within the USDA in Washington. The IE group will compare the LACIE estimates to conventional estimates and note differences in these. In addition the LACIE reports will be assessed for their value to the normal operations of USDA. The IE group will provide feedback to LACIE personnel in Houston. Any discrepancies between the LACIE estimates and conventional estimates will be investigated by USDA analysts at Houston to determine if differences are due to faults in the LACIE approach (data, techniques, etc.) For area "accuracy" three main quantifiers are proposed: (1) the percent difference between conventional estimates and the LACIE estimates i.e., percent bias, (2) the precision of the area estimate $P = [\text{Var}(\hat{p})]^{1/2}/\hat{p}$ where $\hat{p}$ is the LACIE wheat areal density estimate for a given strata, region or country, (3) the confidence level $\alpha$ that the LACIE estimate $A$ is within ten percent of the conventional estimate $A$. Over a large area, such quantifiers are more easily defined than obtained. In the U.S. the only information available for comparison to the LACIE estimates are: (1) crop identification data obtained by the USDA from the 28 Intensive Test Sites (ITS), ranging from 5 x 6 n. miles to 3 x 3 miles in extent, and located in major wheat growing regions, (2) SRS estimates of area at the county, crop reporting district, state, regional and national level, (3) historic area data compiled by the SRS at these levels, (4) insitu information from the SRS sample segments (approximately 1 mile square). If, as was the case in LACIE development testing, retrospective data for prior crop years is being analyzed, the accuracy of the LACIE estimates may be assessed by comparing them to the SRS data. It is generally believed that while the SRS data is not one hundred percent accurate, its accuracy is sufficient to use as a standard against which to evaluate how well the LACIE accuracy is meeting its goals. Real time however, estimates of the percent bias of the LACIE estimates will need to be inferentially determined without reference to the SRS data since it will not be available to the LACIE system. Such methods are being developed within LACIE which will utilize intensive test site and other data. The precision of the LACIE estimates can be determined from an examination of real time LACIE area estimates and historic SRS area data. The variance of $p$, $\text{Var}(p)$ is given for an area composed of $K$ strata by: \[ \text{Var}(p) = \sum_{k=1}^{K} c_k^2 \text{Var}(p_k) \] (8) where $ \text{Var}(p_k) $ is the variance of the estimate of the wheat areal proportion in the $ k $th stratum and $ c_k^2 $ are areal density weighting functions for the strata within the area. Within a given stratum (assuming all counties are class 3 counties in the particular example) $ \text{Var}(p_k) $ is given by: \[ \text{Var}(p_k) = \sum_{j=1}^{n} D_{kj}^2 \text{Var}(\hat{p}_{kj}) \] (9) where $ \text{Var}(\hat{p}_{kj}) $ is the variance of the estimate of the wheat areal proportion contained in the $ j $th county within the $ k $th crop reporting district and $ D_{kj}^2 $ are areal density weighting functions for the counties. Estimating the variance $ \text{Var}(p_k) $ is not straightforward in LACIE since many counties contain only one sample segment. A good estimate of this quantity is critical since it will determine to what degree the "accuracy" of the LACIE estimates can be determined. The figures quoted previously for the sample error associated with the LACIE sample strategy were based on the assumption that the wheat proportion distribution relative to the 5 x 6 nm sample segment would be binomial so that $ \text{Var}(p_{kj}) = p_{kj}(1-p_{kj}) $. This estimate will be somewhat conservative (i.e., will create overestimates of the precision) since LACIE sample segments will be considerably larger than the average wheat field and the wheat fields tend to be distributed somewhat uniformly throughout a county as opposed to being conglomerated in just one portion of the county. Latest estimates of the variance of the LACIE area estimates (see Section V) are based on the use of historic area data and real time LACIE proportion estimates to determine the within-county variance. For a stratum, historic SRS area figures are used to determine the between-county variance of $ p $ within the stratum. This figure is subtracted from the estimate of the variance of the LACIE proportion estimates over the stratum to obtain an estimate of the LACIE within-county variance. Other methods to estimate this variance are being investigated in the RTAE effort. Given an estimate of the bias and precision, an estimate of the confidence level can be obtained. For LACIE $ \alpha $ will be defined for a stratum, state, etc. as \[ \alpha = P(\|\hat{A} - A\| < 0.1A) \] (10) where $ \hat{A} $ is the LACIE estimate of the area and $ A $ is the conventional SRS estimate. As a performance objective of the LACIE experimental system, an $ \alpha $ of 0.9 at a national level has been chosen. Such a choice requires that the LACIE area estimate be, with a ninety percent confidence, within ten percent of the conventional estimate. This criteria has been referred to as the "90/90" criteria. By assuming the LACIE estimate to be normally distributed about its mean value, the estimate of the bias and the $ \text{Var}(p) $ can be used to determine $ \alpha $. Methods for similarly quantifying the performance of the yield and production estimates are being developed in the RTAE effort. The focus in this area is to develop an error simulation/propagation model which describes the contribution of yield and area errors to the production estimation error. IMPLEMENTATION APPROACH To understand the implementation approach taken for LACIE it must be realized that the experiment calls for simultaneous execution of activities which, given a more leisurely schedule, would normally be undertaken sequentially. Specifically, the application of the new technology had to proceed in full recognition of the fact that many components of this Application Evaluation System (AES) had not been fully developed and thoroughly tested. At the same time, research test and evaluation had to proceed to strengthen the suspect areas and to conduct thorough tests of new system components prior to their introduction into the mainstream AES. The way these project elements relate is depicted graphically in figure 2. As this figure makes clear the role of LACIE is, for one important agricultural application, to bridge the gap between the multitude of feasibility tests and exploratory studies and the eventual operational systems. The functional organization for LACIE involves numerous elements of the three participating agencies. Each agency maintains, of course, its own administrative control over those resources allocated to LACIE. However, at the project level the technical staff is fully integrated (figure 3). A number of implementing organizations may be in collaboration for a particular subsystem. For example in the case of the DAPTS (Data Acquisition, Preprocessing and Transmission Subsystem) NASA Goddard, USDA field staff (Agricultural Stabilization and Conservation Service) and NOAA (Environmental Data Service, National Weather Service and National Environmental Satellite Service) all acquire and preprocess portions of the data for LACIE in response to requirements specified by the subsystem manager. In other cases a single implementing organization such as the NASA Data Systems and Analysis Directorate will implement hardware and software that supports the functional requirements of several subsystems. For example, the portion of the LACIE data system which was derived from the ERIPS referred to earlier, supports the functional requirements of the electronic part of the ISRRS (Information Storage, Retrieval and Reformating Subsystem) and the CAMS (Classification and Mensuration Subsystem). To as great an extent as possible the personnel assigned to LACIE have no other duties. In each case the responsibility for a particular area is made in consonance with the respective agency roles. For example, the CAS manager is from USDA, the YES manager from NOAA. To the extent possible some personnel from USDA are assigned to each functional area since USDA will eventually wish to operate an operational follow-on system. NOAA participation is limited (by resource availability) to YES, DAPTS, and project management activity. The great majority of LACIE effort and resources is allocated to the AES (Application Evaluation System). This system provides the actual mechanisms, including ADP equipment, software, personnel, procedures and facilities with which LACIE data is processed to produce wheat inventory reports and evaluations by the user of the utility of those reports. A simplified schematic of the AES is given in figure 4. A full treatment of the functioning of this system is beyond the scope of this paper. It should be noted; however, that as complex a system as the AES requires the exercise of careful control over requirements, configuration and operation. The implementation of the system is, accordingly, carried out in response to formally documented detailed requirements. Interfaces between collaborating organizations are rigorously defined by interface control documentation and formal change control procedures are in force to eliminate uncoordinated change. The research test and evaluation effort, while modest in comparison with the AES effort, is fully as important in reaching the overall goal of a successful application of remote sensing technology. Research of a highly focused and applied nature is initiated pursuant to requirements developed by the AES personnel who perceive the technological gaps. In conjunction with research personnel who are in touch with the most current developments in the remote sensing community. This research is carried on under contract by academic and research institutions long experienced in the field. A tabulation of the major task areas and the institutions working on each is given in Table 3. In each case a task team monitors the contract and assures its congruity with LACIE needs. Both research and AES personnel participate. In each research effort specific goals are spelled out with deliverable products scheduled at appropriate times. Generally such a product would be an alternate algorithm or procedure for some part of the LACIE system. During the development of such a product some testing will, of course, have been done. If the results appear promising then further more rigorous testing will be done either by the developing institution according to an approved test plan or by the test and evaluation personnel in the project. Implementation of the new procedure in the AES will take place in parallel with the testing or following it depending upon the urgency of the requirement and new procedure and implementation resources available. It should also be noted that RT&E effort can contribute in an important way to eventual follow-on systems by providing solutions for key problems that may be identified but not resolved during the lifetime of LACIE. The development of the LACIE system and its operation is being conducted in three phases, each tied to the wheat-growing cycle and expanding in scope as capability increases. The first phase covers the 1974-75 crop year (in the U.S.) and addresses area estimates for nine wheat-growing states in the Great Plains. In a smaller area or areas, likely one or two states, yield models will be tested and production estimates made. Classification tests will be conducted on representative segments in other wheat-growing regions and the 28 intensive test sites in North America will be analyzed to provide one basis for performance assessment. The second phase from the fall of 1975 to the spring of 1977 will include area, yield and production estimates for a large area, likely the entire U.S., a continuation of tests on representative sites elsewhere and intensive test sites and will cover a longer crop cycle to include both northern and southern hemisphere wheat crops. The third phase from the fall of 1976 to the spring of 1978, will provide area, yield and production estimates over one or more large area regions and will incorporate those refinements to the technology developed during the RT&E efforts of the earlier phases. The schedule for LACIE is clearly very success-oriented. This was considered necessary, however, to exploit fully the LANDSAT 2 capability and to demonstrate a large-scale application of space remote sensing at the earliest possible time. V. SUMMARY OF INITIAL PERFORMANCE ASSESSMENT At the writing of this paper an initial assessment of the performance of the initial LACIE/CAMS system had been performed by conducting analyses of the results of processing LANDSAT 1 data acquired during 1973 and 1974. The segments in Kansas used for this study, the LACIE sample segments, were moved within each county to encompass either an intensive test site or an SRS sample unit. Of these 33 segments, 5 were LACIE intensive test site segments for which "wall to wall" insitu observations of crop type were available and the remaining 28 each contained a 1 x 1 mile section of crops observed for crop types by the Statistical Reporting Service of USDA. For three of the five ITS segments, LANDSAT data was available for all biological phases. For the other two, cloud cover resulted in the data for two biological phases not being acquired. For this 1973-74 data set, biophase 4 is postharvest. For the 28 remaining segments, 22 were acquired during biophase 1, 4 acquired during biophase 2, and 2 acquired during biophase 3. The analyst interpreter (AI) performance was evaluated on the intensive test sites by comparing their identifications of wheat and nonwheat to ground observations. The AIs picked approximately 15 wheat and 15 nonwheat fields for classifier training. The figures in Table 4 are percent correct identifications of these fields when compared to the ground observed identifications. The classification performance for wheat obtained by using the analyst-interpreter provided fields as training data was determined. The classification performance varied considerably from biophase to biophase and from segment to segment (see Table 5). These performances are for single pass data only. Multidate analyses in the intensive test sites resulted in a considerable performance improvement. Estimates for the three parameters used in LACIE to quantify the area estimation accuracy were computed for the 28 segments analyzed. Two estimates of the bias were obtained for 12* of the 28 segments. The wheat areal proportion estimates obtained by LACIE were compared to the proportions as determined from the SRS sample units (1 x 1 mile) within the 5 x 6 nm segments. The results shown in Table 6, indicate the difference of -0.0035 to result in a negative "bias" (underestimate) in the LACIE estimate of -1.02%, of the SRS mean value. In Table 7, an aggregation of the LACIE results over the 6 crop reporting districts containing the LACIE segments agree to -3% with the 1973-74 SRS estimate over the same CRD's. The accuracy of the LACIE area estimate was also estimated in two ways. Returning to Table 6 the standard error associated with the difference between the LACIE and SRS estimates in the 12* sample segments is about 12% of the SRS mean value over these segments. This is an arbitrary subset of the 28. Projected to the area consisting of 6 CRD the associated standard error would be roughly 8%. In Table 8, the precision of the LACIE estimate is computed based on all 28 segments processed for each of the 6 CRD's examined and is computed for the aggregate of these. These are the precisions associated with the LACIE area estimation discussed in Section IV. Notice that the precisions fluctuate with the CRD. This is a result of the fact that each CRD had a different number of acceptable sample segments and that the error of the ratio estimators discussed earlier enters into these calculations. The precision of the LACIE estimate for the 6 CRD's is 8.3%. Based on these preliminary results, an estimate can be made of the performance to be expected at a state level in Kansas. Assuming the bias of -3% to remain about the same at the state level and (based on sample theory) assuming the precision will decrease in proportion to the square root of the ratio of the number of samples to be acquired cloud free at the state level in Kansas (50) to the number examined here (28), a rough estimate of the precision at the state level would be \[ \sqrt{\frac{\text{Var}(\hat{p})}{\hat{p}}} = \sqrt{\frac{28}{50}} \times 8.3 = 6.2\% \] (11) For the bias figure of 3%, the confidence that the LACIE Kansas estimate will be to within 10% of the SRS state estimate would be approximately 85%. Projecting this estimate beyond Kansas to the national level is probably not too meaningful at this time since: (1) the Kansas results were obtained using only local signatures (i.e., signatures developed for each segment without including signature extension), (2) conditions in other states will vary considerably, (3) of the 28 segments examined, data for 22 was acquired in biphase I (bare soil). However, the very preliminary evidence examined indicates that the confidence at a national level could increase considerably as a result of increased sampling. It is concluded that this preliminary assessment shows every indication that the "90/90" criteria nationally is a reasonable goal to be achieved by the LACIE system and one which will productively stimulate the development of the LACIE technology. VI. REFERENCES 1. Purdue University Bulletin #844, Laboratory for Application of Remote Sensing (LARS) Volume 3 Annual Report, "Remote Multispectral Sensing in Agriculture," dated September 1968. 2. "Corn Blight Watch Experiment Final Report, Experimental Results," NASA/JSC, Vol. III, May 1974. 3. Anuta, Paul E., and MacDonald, Robert B., "Crop Surveys from Multiband Satellite Photography Using Digital Techniques," LARS Information Note 032371, Purdue University, Lafayette, Indiana. 4. Williams, D. L., Morain, S. A., Barker, B. and Corner, S. C., "Identification of Winter Wheat from ERTS-1 Imagery," Symposium on Significant Results Obtained from the ERTS-1, Vol. I, p. 11, 1973. 5. Thomson, Frederick J., "Crop Species Recognition and Mensuration in the Sacramento Valley," ERTS Symposium on Significant Results Obtained from the ERTS-1, Vol. I, p. 181, 1973. 6. Bizzell, R. M., Wade, L. C., Prior, H. L., and Spiers, B., "The Results of an Agricultural Analysis of the ERTS-1 MSS Data at the Johnson Space Center," ERTS Symposium on Significant Results Obtained from the ERTS-1, Vol. I, p. 189, 1973. 7. Bauer, M. E. and Cipra, J. E., "Identification of Agricultural Crops by Computer processing of ERTS MSS Data," ERTS Symposium on Significant Results Obtained from the ERTS-1, Vol. I, p. 206, 1973. 8. Erb, R. Bryan, "The Utility of ERTS-1 Data for Applications in Agriculture and Forestry," Third ERTS-1 Symposium, Vol. I, p. 75, 1973. 9. Thompson, L. M., "Weather and Technology in the Production of Wheat in the United States," J. of Soil and Water Conservation 24:219-24, 1969. 10. Haun, T. R., "Prediction of Spring Wheat Yields from Temperature Precipitation Data," South Carolina Agricultural Experimentation Report, 1973. 11. Williams, G. D. V., "Estimate of Prairie Provincial Wheat Yields Based on Precipitation and Potential Evapotranspiration," Can. J. Plant Science 53:17-30, 1973. 12. Williams, G. D. V., "Weather and Prairie Wheat Production," Can. J. of Agricultural Economics 17:1:99-109, 1973. 13. Robertson, George W., "A Biometeorological Time Scale for a Cereal Crop Involving Day and Night Temperatures and Photoperiod," International Journal of Biometeorology 12:191-223, 1968. VII. TABLES AND ILLUSTRATIONS Table 1. Probabilities for Correct Classification and Probabilities for Commission Error Hill County, Montana \| \| Single Pass \| Multidate \| \|----------------\|-------------\|-----------\| \| \| $t_1$ \| $t_2$ \| $t_3$ \| $(t_1, t_2)$ \| $(t_1, t_3)$ \| $(t_1, t_2, t_3)$ \| \| P(w/w) \| 0.70 \| 0.90 \| 0.80 \| 0.90 \| 0.90 \| 0.95 \| \| P(w/o) \| 0.20 \| 0.15 \| 0.05 \| 0.05 \| 0.05 \| 0.00 \| $t_1 = \text{greening}$ $t_2 = \text{heading}$ $t_3 = \text{mature}$ Burke County North Dakota \| \| $t_1$ \| $t_2$ \| $(t_1, t_2)$ \| \|----------------\|-------------\|-----------\|---------------\| \| P(w/w) \| 0.75 \| 0.85 \| 0.90 \| \| P(w/o) \| 0.10 \| 0.10 \| 0.05 \| $t_1 = \text{emergence}$ $t_2 = \text{jointing}$ P(w/w) = Probability with which wheat test pixels are classified as wheat P(w/o) = Probability for classifying non-wheat test pixel as wheat (commission error) Table 2. Area Estimation Accuracy Results for Hill County Study \| \| Single Pass \| Multidate \| \|----------------\|-------------\|-----------\| \| \| Greening \| Heading \| Three Passes \| \| P(w/w) \| 0.586 \| 0.698 \| 0.851 \| \| P(w/o) \| 0.141 \| 0.155 \| 0.060 \| \| Bias Error D \| -0.089 \| 0.056 \| -0.01 \| Table 3. Major Task Areas in LACIE Research, Test and Evaluation and Performing Organization \| Task Area \| Performer \| \|----------------------------\|---------------------------------------------------------------------------\| \| o Area \| \| \| o Definition of Training Statistics \| UCB (U. of Calif. Berkeley), LARS (Laboratory for Application of Remote Sensing), ERIM (Environmental Research Institute of Michigan) \| \| o Registration \| LARS \| \| o Classification \| \| \| o Temporal Sampling Strategy \| TAMU (Texas A&M Univ.) \| \| o Multitemporal Classification \| UTD (U. of Texas, Dallas) \| \| o Proportion Estimation \| ERIM, UTD, TAMU, Rice Univ. \| \| o Feature Selection \| UH (U. of Houston), TAMU \| \| o Signature Extension \| UCB, LARS, ERIM, Colo. (Colorado State) \| \| o Estimation of Unharvested Wheat \| TAMU \| \| o Sampling and Aggregation \| UCB, TAMU \| \| o Field Measurements \| LARS, TAMU, ERL (Earth Resources Laboratory, FOD (Flight Operations Directorate) \| \| o Yield \| \| \| o Yield Models \| KSU (Kansas State Univ.), TAMU, ERIM, Clemson \| \| o Production \| \| \| o Error Model \| TRW \| * Supports Classification, Estimation Unharvested Wheat, Signature Extension Table 4. Intensive Study Site Summary of AI Performance Percent of Fields Correctly Identified \| Biological Stage \| Morton \| Finney \| Ellis \| Saline \| Rice \| \|------------------\|--------\|--------\|-------\|--------\|------\| \| \| W \| NW \| W \| NW \| W \| NW \| W \| NW \| W \| NW \| \| IA \| 27% \| 100% \| \| \| \| \| \| \| \| \| \| IB \| 100% \| 100% \| 100% \| 80% \| 100% \| 66% \| 100% \| 100% \| 100% \| 100% \| \| II \| 89% \| 100% \| 100% \| 100% \| 100% \| 80% \| 100% \| 100% \| 100% \| 100% \| \| III \| \| 100% \| 100% \| 100% \| 100% \| 80% \| \| \| \| \| \| IV \| 100% \| 100% \| 100% \| 100% \| 100% \| 100% \| 100% \| 100% \| 100% \| 100% \| Table 5. Classification Performance on 5 Kansas Intensive Test Sites \| Segment \| 1034 \| 1042 \| 1106 \| 1111 \| 1114 \| \|---------\|------\|------\|------\|------\|------\| \| \| W \| NW \| W \| NW \| W \| NW \| W \| NW \| W \| NW \| \| 1A \| 75.72\| 89.64\| \| \| \| \| \| \| \| \| \| 1B \| 54.09\| 76.35\| 72.46\| 94.61\| 21.60\| 85.00\| \| \| \| \| \| 2 \| 86.80\| 61.20\| 75.35\| 78.88\| 66.05\| 31.30\| \| \| \| \| \| 3 \| 68.80\| 88.90\| \| \| 80.27\| 18.54\| \| \| \| \| \| 4 \| 38.30\| 84.70\| 37.50\| 54.80\| \| \| 85.13\| 48.00\| 78.10\| 67.50\| \| \| \| \| \| \| \| \| 78.00\| 65.00\| \| \| Number Biophase 1A Fall Seedbed Preparation 1B Spring Reemergence 2 Booting through Heading 3 Soft Dough to Harvest (Mature) 4 Post Harvest Segment Number County 1034 Finney 1042 Morton 1106 Ellis 1111 Rice 1114 Saline Table 6. Comparisons of SRS and LACIE Estimates \| Segment \| County \| Phase \| Wheat Proportion \| \|---------\|--------\|-------\|------------------\| \| \| \| \| ERTS (Y) \| SRS (X) \| \| 1040 \| Kearny \| 2 \| .322 \| .337 \| \| 1109 \| Marion \| 1 \| .254 \| .222 \| \| 1036 \| Grant \| 1 \| .520 \| .321 \| \| 1118 \| Reno \| 2 \| .180 \| .434 \| \| 1018 \| Graham \| 2 \| .127 \| .182 \| \| 1029 \| Scott \| 1 \| .400 \| .411 \| \| 1037 \| Gray \| 1 \| .290 \| .266 \| \| 1045 \| Stevens\| 2 \| .249 \| .209 \| \| 1065 \| Haskell\| 1 \| .300 \| .321 \| \| 1104 \| Barton \| 3 \| .380 \| .605 \| \| 1106 \| Ellis \| 2 \| .607 \| .404 \| \| 1110 \| McPherson \| 2 \| .526 \| .401 \| Mean: .34625 .34275 Difference in means = .0035 Associated Standard Error = .0407 Table 7. A Map Comparison of USDA and LACIE Wheat Proportion Estimates for Six Kansas Crop Reporting Districts \| Kansas CRD Number \| Wheat Proportion \| Relative Difference Percent \| \|-------------------\|------------------\|----------------------------\| \| \| USDA (1973-74) \| LACIE \| \| 1 \| .244 \| .189 \| -2 \| \| 2 \| .228 \| .230 \| 5 \| \| 4 \| .255 \| .214 \| 16 \| \| 5 \| .307 \| .390 \| 27 \| \| 7 \| .267 \| .321 \| 20 \| \| 8 \| .385 \| .282 \| -27 \| \| All 6 CRD's \| .286 \| .278 \| -3 \| Table 8. Precision of LACIE Area Estimations \| Kansas CRD Number \| No. of Segments \| Precision \| \|-------------------\|-----------------\|-----------\| \| \| Allocated \| Classified\| \| \| 1 \| 8 \| 1 \| 60% \| \| 2 \| 11 \| 3 \| 21.2% \| \| 4 \| 9 \| 2 \| 26% \| \| 5 \| 11 \| 8 \| 8.7% \| \| 7 \| 14 \| 10 \| 10% \| \| 8 \| 14 \| 4 \| 16% \| \| Aggregated \| 67 \| 28 \| 8.3% \| Figure 1. Segment Locations Within a Crop Reporting District - GENERAL RESEARCH AND TECHNOLOGY DEVELOPMENT - LANDSAT-I INVESTIGATION - CITARS SPECIFIC PROCEDURES APPLICATION SYSTEM PHASE I PHASE II TEST AND EVALUATION LACIE R&D USER EVALUATION OF APPLICATION SYSTEM SPECIFICATIONS FOR USER ORIENTED SYSTEMS PROTOTYPE FOR EVENTUAL OPERATIONAL SYSTEMS GENERAL TECHNIQUES LACIE Figure 2. LACIE Project Elements Figure 3. LACIE Project Organization LACIE MANAGER - LACIE PROJECT OFFICER - DEPUTY PROJECT MANAGER - ASSISTANT DEPUTY PROJECT MANAGER - USDA - ASSISTANT DEPUTY PROJECT MANAGER - NOAA TECHNICAL STAFF - INTEGRATION MANAGER - PROJECT SCIENTIST - SYSTEMS ENGINEER - PROGRAM OFFICE MANAGER - DATA SYSTEMS & ANALYSIS DIRECTOR - JURISDICTIONAL DIRECTOR - JURIS SYSTEMS DESIGNER RESEARCH AND DEVELOPMENT COORDINATION WORKING GROUP RESEARCH TEST & EVALUATION - CHIEF - A3ST. CHIEF RESEARCH TEST & EVALUATION APPLICATIONS SYSTEMS EVALUATION - CHIEF - DEPUTY FOR DESIGN - A3ST. DEPUTY FOR DESIGN (USDA) - DEPUTY FOR OPERATIONS - A3ST. DEPUTY FOR OPERATIONS (USDA) OPERATIONS WORKING GROUP INFORMATION STORAGE, RETRIEVAL & REFORMATING DATA ACQUISITION, PREPROCESSING & TRANSMISSION CLASSIFICATION AND MENSURATION YIELD ESTIMATION - YIELD TASK FORCE CROP ASSESSMENT - CLOUD COVER TASK FORCE Figure 4. LACIE INTEGRATED APPLICATIONS EVALUATION SYSTEM LEGEND: - STAFF INTERFACES - DATA FLOW & INTERFACES - SYSTEM ACRONYMS: - DAPTS: DATA ACQUISITION PREPROCESSING AND TRANSMISSION SUBSYSTEM - YES: YIELD ESTIMATE SUBSYSTEM - CAMS: CROP ASSESSMENT MANAGEMENT SUBSYSTEM - CAS: CROP ASSESSMENT SUBSYSTEM - ERTS: EARTH RESOURCES TECHNOLOGY SATELLITE - IRS: INFORMATION RECOVERY SUBSYSTEM - ICS: INTERFACE CONTROL DOCUMENT - IRIMS: INFORMATION STORAGE, RETRIEVAL, AND REFORMATING SUBSYSTEM DATA BASES: - CAMS - FIELDS - METEOROLOGICAL & INDICES - RESULTS - MET INTEGRATED NON-ADP DATA SYSTEM INTEGRATED ADP DATA SYSTEM
A dark matter solution for the XENON1T electron excess and the galactic center 511 keV line Yasaman Farzan School of physics, Institute for Research in Fundamental Sciences (IPM) P.O.Box 19395-5531, Tehran, Iran E-mail: firstname.lastname@example.org http://physics.ipm.ac.ir/ The excess of the 511 keV line from the Milky Way galactic bulge, confirmed by the INTEGRAL detector, is a longstanding mystery. The morphology of the line appears to be proportional to the square of the dark matter density, hinting towards a dark matter origin. On the other hand, in 2020, XENON1T has reported an excess of electrons with a recoil energy of 2 – 4 keV. We present a model based on a dark matter of a few MeV mass that decays into a pair of pico-charged particles with a lifetime much larger than the age of the universe. The magnetic field of the galaxy suppresses these relativistic pico-charged particles whose scattering on the electrons can explain the signal reported by XENON1T. The annihilation of the pico-charged particles in the galactic bulge leads to $e^- e^+$ production and therefore to an excess of the 511 keV line. We review the present observational bounds and the strategies to test the model. Keywords: Dark matter, pico-charged particles, XENON1T electron excess, 511 keV line, galactic bulge 1. Introduction Various observations such as galactic rotation curves, structure formation, dynamics of the galaxy clusters, collision of bullet clusters and anisotropies of Cosmic Microwave Background (CMB) point towards existence of a new form of matter all over the universe known as Dark Matter (DM). The constituent particles of DM should be either electrically neutral or their electric charge should be much smaller than the electron electric charge. Otherwise, they will emit photons and will not remain dark. The dark matter lifetime should be much longer than the age of the Universe. The observation of bullet clusters constrain the self-interaction of the dark matter particles indicating that they cannot have strong interactions. Last but not least, at the onset of structure formation, the dark matter should be non-relativistic. None of the Standard Model (SM) elementary particles satisfies these conditions. As a result, the observational hints for dark matter opens a window towards new physics beyond the standard model by introducing a new neutral (meta)stable massive particle that can play the role of dark matter.\footnote{The primordial blackholes as dark matter candidates have received popularity after observation of gravitational waves by LIGO and VIRGO. We will not however pursue this idea in this letter.} The models for dark matter candidates are quite diverse. Depending on the details of the model, different strategies for dark matter detection have been developed. In a wide class of models known as WIMP, where dark matter has a mass in the wide range of MeV to 100 TeV and it weakly interacts with the Standard Model (SM) particles\textsuperscript{1}, there are three approaches to dark matter detection: (1) direct dark matter detection, (2) indirect dark matter detection and (3) dark matter production in colliders. No conclusive evidence for dark matter detection has been reported from any of these experiments so far but every now and then, a signal is reported by indirect search experiments which may be interpreted as dark matter. Some of these signals disappear by collecting more data or are eventually proved to be artifacts. There are however signals that pass the test of time but after some period of time simply go out of fashion because the simplistic dark matter models developed to explain them become ruled out by independent observations. However, the dark sector does not need to be simplistic. Indeed, judging from the rich and dazzling structure of the SM, simplicity and minimalism is not the taste of the physics of the elementary particles. Going beyond simplistic dark matter models, it may be possible to resurrect dark matter explanation for the signal. One example of the signals that has survived the test of time is the 511 keV line observed from galactic center which must come from the positronium (the $e^- e^+$ atom) decay: $e^- + e^+ \rightarrow \gamma + \gamma$. The intensity of the line indicates an excess of $e^+$ in the galactic center. Although the excess may come from the SM sources such as pulsars but entertaining a DM origin is too tantalizing to dismiss, especially that the morphology of the intensity of the line follows a $\rho_{DM}^2$ distribution, further pointing towards DM pair annihilation as the origin of the excess. In Ref.\textsuperscript{1}, it had been shown that dark matter with mass of a few MeV can explain the excess of the 511 keV line observed from the galactic center by pair annihilation into pairs of electron positron with a cross section of $\langle \sigma[DM + DM \rightarrow e^- e^+]v \rangle \sim O(10^{-38})$ cm$^2$. This model implies huge energy pump into the plasma at the era of recombination when the DM density is high, through annihilation into $e^- e^+$. Such entropy pump can distort the CMB fluctuation pattern. The Planck data therefore rules this solution out\textsuperscript{2}. Moreover, this model is disfavored by the lack of observation of the 511 keV line from dwarf galaxies\textsuperscript{3}. Ref.\textsuperscript{4} proposes a dark matter model for the 511 keV line from the center of the Milky Way (MW) that can avoid the bounds from CMB and explain the lack of the 511 keV line from satellite dwarf galaxies. In this model, dark matter is composed of a scalar particle, $X$, with a mass of a few MeV which can be identified with the SLIM particle\textsuperscript{5,6}. $X$ decays into a pair of pico-charged particle ($C + \bar{C}$) with a lifetime much longer than the age of Universe: $\Gamma(X \rightarrow C + \bar{C})t_0 \ll 1$ in which $t_0$ is the age of the Universe. The magnetic field of the MW accumulates the $C$ and $\bar{C}$ particles. The $C\bar{C}$ pairs eventually annihilate with each other producing $e^- e^+$ pairs. At the recombination and during the dark ages, the density of $C\bar{C}$ would be too small to pump energy to the plasma so the CMB bound can be avoided. \textsuperscript{1}Here, by weak interactions, we do not necessarily mean nuclear weak interaction, characterized by the SM $W$ and $Z$ gauge boson exchange. Moreover, the magnetic field in the dwarf galaxies may be too weak to accumulate $C\bar{C}$ so the lack of the 511 keV line from them can be explained. Ref.\textsuperscript{4} predicted a signal at direct dark matter search experiments with low energy threshold. The relativistic $C$ particles wandering around can impart a recoil energy of a few keV via Coulomb interaction to the electrons. In 2020, XENON1T reported an excess of electrons with recoil energy of 2–4 keV\textsuperscript{7}. In Ref.\textsuperscript{8}, we discussed the parameter space where the model can explain both signals. Refs.\textsuperscript{9} have also attempted to explain the XENON1T signal via DM. This paper is organized as follows. In Sect. 2, we describe the model and the bounds that constrain its parameter space. In Sect. 6, we show how the model can explain the 511 keV line from the galactic bulge. In Sect. 4, we describe how the model can explain the XENON1T data. We show that the XENON1T electron recoil spectrum can also set the strongest upper bound on $f = \Gamma(X \rightarrow C\bar{C})t_0$. A summary is given in Sect. 5. In Sect. 5, we also discuss how the model can be tested with further collection of the 511 keV data from satellite galaxies and studying the correlation of the intensity with the magnetic field of the host [dwarf] galaxy. We also discuss the signatures in diffuse gamma ray background from the inflight annihilation of positron. We also comment on the positron flux to be detected by Voyager. 2. The model and its constraints As mentioned in the introduction, the model is based on a scalar dark matter, $X$ with a mass of a few MeV which decays into a pair of pico-charged particles, $C\bar{C}$. The $X$ particle can be identified with the SLIM particle introduced in Ref.\textsuperscript{3}. In the SLIM scenario, the DM is thermally produced via interaction with neutrinos and its abundance is set \textit{a la} freeze-out scenario through lepton number violating annihilation into neutrino or antineutrino pair, $\langle \sigma (X + X \rightarrow \nu + \nu) v \rangle = \langle \sigma (X + X \rightarrow \bar{\nu} + \bar{\nu}) v \rangle \sim pb$. Within the SLIM scenario, the annihilation to neutrino or antineutrino pair takes place via the $t$-channel exchange of a new Majorana neutrino which has a Yukawa coupling with the SM neutrino and $X$. A $Z_2$ symmetry, under which only the new Majorana neutrino and $X$ are odd, guarantees the stability of $X$ which is the lightest among the $Z_2$ odd particles. The same $Z_2$ symmetry forbids Yukawa coupling between the leptons, the Higgs and the new Majorana fermions and as a result, neutrinos cannot obtain a Dirac mass at the tree level. The SM neutrinos obtain a Majorana mass at one loop level within the SLIM scenario. The smallness of the neutrino mass is explained, thanks to the loop suppression despite a relatively large Yukawa coupling between, $X$ and neutrinos of order of $10^{-3}$. The model is testable by searching for the three body decay of Kaon and pion where along with the charged leptons, these new particles are emitted and appear as missing energy\textsuperscript{10}. Ref.\textsuperscript{6} shows how the SLIM scenario can be embedded within a UV complete model. Combining the condition $\langle \sigma (X + X)v \rangle \sim pb$ (to obtain the required DM relic abundance via the freeze-out mechanism) with the one-loop contribution to the neutrino mass to be equal to $\sim \sqrt{\Delta m^2_{atm}}$ sets an upper bound of $\sim 10$ MeV on the dark matter mass. On the other hand, $X$, being in thermal equilibrium with neutrinos at the time of the neutrino decoupling, cannot be lighter than a few MeV\cite{11}; otherwise, it will lead to too many extra relativistic degrees of freedom. If $X$ is not the SLIM and its production is not thermal, these lower and upper bounds on its mass can be relaxed. Taking the pico-charged particles $C$ and $\bar{C}$ to be scalars, the decay $X \rightarrow CC$ can take place via a soft $Z_2$ symmetry breaking trilinear term of form $AXCC$ with a very small $A$ making $X$ metastable. Let us now discuss how $C$ and $\bar{C}$ acquire their minuscule electric charge. For this purpose, we need a new Abelian gauge symmetry such that the $C$ and $\bar{C}$ are charged under the new $U_X(1)$ gauge symmetry. We denote the new gauge boson with $A_\mu$. The kinetic and mass mixings between $A_\mu$ and the SM hypercharge gauge boson $B_\mu$ lead to a small electric charge for the $C$ and $\bar{C}$ particles. We follow the notation of Ref.\cite{12} and write $$-\frac{A_{\mu\nu}A^{\mu\nu}}{4} - \frac{B_{\mu\nu}B^{\mu\nu}}{4} = \frac{\delta}{2} A_{\mu\nu} B^{\mu\nu} - \frac{1}{2} (\partial_\mu \sigma + M_1 A_\mu + \epsilon M_1 B_\mu)^2,$$ where $\delta, \epsilon \ll 1$. In the kinetic and mass basis, the three neutral gauge bosons are the SM $\gamma, Z$ bosons and a new gauge boson $A'_\mu$ which we shall call dark photon. To the first approximation in $\epsilon$ and $\delta$, the mass of $A'$ is decoupled from $m_Z$ and is equal to $M_1$. Up to $O(\epsilon^2, \delta^2)$, the coupling between the SM charged fermions, $f$ and $A'$ can be written as the following: $$q' f^\dagger \gamma^\mu f A'_\mu \quad \text{where} \quad q' = e \cos \theta_W (\epsilon - \delta) Q_f,$$ in which $\theta_W$ is the weak (Weinberg) mixing angle. The coupling of $C\bar{C}$ to $A'$ is given by $g_X J_C^\mu A'_\mu$ where $J_C^\mu$ is the current of the $C$ particles. Then, the $C$ particle obtains an electric charge of $$q_C = -g_X \epsilon \cos \theta_W.$$ From Eq. (2), we observe that in the limit $\epsilon \rightarrow \delta$, the SM fermions do not couple to $A'$. In this limit, if $A'$ is lighter than the $C\bar{C}$ pair with a mass of order of a few keV, it will become metastable with a lifetime much greater than the age of the Universe\cite{4}. This means the $A'$ particles produced in the early universe remain as relics today and contribute as a subdominant DM component. In Ref.\cite{8}, we do not however consider this limit and allow for a faster decay of $A'$. As shown in Ref.\cite{8}, the $U_X(1)$ symmetry can be identified with the $L_\mu - L_e$ symmetry with a gauge coupling of $g_{\tau - \mu}$ which opens up the possibility of fast decay $A' \rightarrow \nu_\mu \nu_\mu, \bar{\nu}_\tau \nu_\tau$. The decay can relax the bound from supernova cooling, opening up the window $m_{A'} \sim 100$ keV, $q' \sim 10^{-10} - 10^{-9}$ and $g_{\tau - \mu} \sim 10^{-3}$. Such values of coupling are too small to produce $A'$ with significant abundance in the early universe so the bounds on the extra relativistic degrees of freedom from CMB and BBN can be relaxed. On the other hand, the $A'$ particles produced in the supernova core decay into neutrino and antineutrino pairs in the outer layers of supernova. This process can severely change the supernova evolution. It can even facilitate the shock revival. Thus, in future with better observation of supernovae and more advanced simulations, this part of the parameter range of the model can be tested. For the time being, however, this range of the parameter space is accepted within the uncertainties. The $A'$ particles escaping the core decay into $\nu\bar{\nu}$ pairs so similarly to the standard picture, the binding energy of the star will be emitted in the form of neutrinos to be observed by detectors on the Earth such as Super-Kamiokande. The $C$ particles produced by the $X$ decay will be relativistic at the production with a velocity far exceeding the escape velocity from the galaxy. They are kept in the galactic disc by the magnetic field. To accumulate the $C$ and $\bar{C}$ particles, their Larmour radius, $$r_c = (100 \text{ pc}) \frac{3 \times 10^{-11}}{q_C} \frac{m_X}{10 \text{ MeV}} \frac{1 \mu\text{Gauss}}{B}$$ should be smaller than the thickness of the galactic disc. This requirement sets a lower bound on $q_C$: $q_C > 10^{-11}$, see Refs. 4,8. The strongest upper bound on $q_C$ in this mass range comes from supernova cooling consideration and is $q_C < 10^{-9}$, see Ref. 13. As shown in Ref. 14, expanding supernova remnants can pump energy to the charged particles, enhancing their Larmor radius and expelling them from the galactic disc with a rate of $(100 \text{ Myr})^{-1}$ unless there is an efficient mechanism for energy loss for the charged particles. As discussed in Ref. 4, in the limit $\delta \to \epsilon$ where $A'$ is (meta)stable, the relic background of $A'$ can behave as a coolant. That is the $C$ and $\bar{C}$ particles can scatter off the background $A'$, losing energy. As discussed in Ref. 8, in general case $\delta \neq \epsilon$ where there is no $A'$ background, another coolant should be introduced. Following Ref. 8, we denote the coolant with $Y$ and take $10 \text{ eV} < m_Y < 10 \text{ keV}$, $n_Y\|_{local} = (\rho_{DM}\|_{local}/\langle \rho_{DM} \rangle)/n_Y)$ with a coupling of $g_Y$ to the $C$ particles leading to the scattering cross section of $\sigma_S \sim g_Y^2/(4\pi E_C^2)$. The energy loss at each collision can be written as $\Delta E_C = m_Y v^2(E_C/m_C)^2$. The energy of the $C$ particle at the production from $X \to C\bar{C}$ is equal to $m_X/2$. In the absence of energy pump from supernova, the time scale for the energy loss from $m_X/2$ to $m_C(1 + v_f^2/2)$ is $$\tau_E = \int_{m_C(1+v_f^2/2)}^{m_X/2} \frac{dE_C}{\Delta E_C} \frac{1}{\sigma_S v n_Y} \sim \frac{4\pi m_C^3}{g_Y^2 n_Y m_Y} \left(\frac{1}{v_f} - \frac{1}{v_i}\right). \quad (3)$$ Equating $\tau_E$ with the time scale of the energy gain from expanding supernova remnants ($i.e., 100 \text{ Myr}$), we find $$v_f = 0.08 \left(\frac{0.25}{g_Y}\right)^4 \left(\frac{m_C}{3\text{MeV}}\right)^3 \frac{0.1 \times \rho_{DM}\|_{local}}{n_Y m_Y}. \quad (4)$$ Let us now discuss the bounds on $\Gamma_X = \Gamma(X \to C\bar{C})$ or equivalently on $f = \Gamma_X t_0$. A rather strong bound comes from the requirement that annihilation $C\bar{C} \to $A'A'$ does not deplete the $C$ abundance of the galactic disc: That is $\sigma(C\bar{C} \rightarrow A'A')n_Cv_f t_0 < 1$ which leads to $$f < 10^{-2} \frac{m_X}{10 \text{ MeV}} \left( \frac{0.15}{g_X} \right)^4 \left( \frac{m_C}{5 \text{ MeV}} \right)^2 \frac{0.06}{v_f}.$$ A stronger bound comes from the requirement that the process of cooling during the history of the galaxy does not repel all the coolants. We should bear in mind that the coolant is a subdominant DM component which contributes less than 10% to the whole dark matter content. Ref. shows that this requirement implies $$f < 10^{-3} \frac{m_X}{10 \text{ MeV}} \frac{m_Y}{10 \text{ keV}} \frac{10\rho_Y}{\rho_{DM}}.$$ As we shall see in sect. 4, the bound from XENON1T detector on $f$ is much stronger. In order to produce $e^- e^+$ from $C\bar{C}$ annihilation yet another interaction is required. We will discuss this coupling in Sect. 6. ### 3. Explaining the 511 keV line The 511 keV photon line has been detected from the galactic bulge of the MW for more than 40 years. The INTErnational Gamma-Ray Astrophysics Laboratory (INTEGRAL) is a space telescope launched in 2002. The data from the SPI spectrometer at INTEGRAL over the years has confirmed the 511 keV signal at 56 $\sigma$ C.L. The narrow photon line peaked at 511 keV, whose energy coincides with the rest mass of electron or positron, should come from the annihilation of the electron-positron pairs at rest which form positronium atoms. The source of such large positron abundance in the bulge of MW is subject to debate. An intriguing possibility can be annihilation of DM leading to the $e^- e^+$ production. Astrophysical sources such as pulsars or X-ray binaries may also contribute to the positron flux but their characteristic spectral shapes do not fit the observations. Moreover, the morphology of the observed 511 keV line is better compatible with the $\rho_{DM}^2$ distribution than what is predicted from these point sources, hinting further toward a DM annihilation origin. Ref. proposes DM with mass of few MeV annihilating to $e^- e^+$ as a solution to the 511 keV line. With an annihilation $\langle \sigma(DM + DM \rightarrow e^- + e^+)v \rangle \sim 10^{-2}$ pb, MeVish DM can account for the intensity of the 511 keV line. However, if the annihilation to $e^- e^+$ is a $S$-wave process, the reionizing energy dump at recombination will cause delayed recombination which is in tension with the CMB data. If the annihilation is $P$-wave, at the freeze-out temperatures, the annihilation cross section will be $\sim 10$ pb, washing out the relic density of DM. Thus, we should go beyond this minimal scenario. In the following, we show that the model described in sect. 2 can provide a reliable explanation. The $X$ DM particles in the galaxy decay into $C\bar{C}$ with a rate of $\Gamma_X$. If these $C$ particles are accumulated close to their production point, their number density after $t^0$ will be $$n_C = \frac{\rho_X}{m_X} f \quad \text{with} \quad f = \Gamma_X t^0$$ \hspace{1cm} (5) in which $f$ is the fraction of $X$ particles that have decayed. According to the INTEGRAL findings$^{21}$, the intensity of 511 keV line from galactic center is $\Phi_{511} = (0.96 \pm 0.07) \times 10^{-3} \text{ph cm}^{-2} \text{sec}^{-1}$. If each $C\bar{C}$ annihilation produces $N_{pair}$ pairs of $e^- e^+$ and subsequently $N_{pair}$ pairs of photons, the flux from the galaxy bulge will be $$\Phi_{511} \simeq 2 N_{pair} \int_0^{r_b} (n_X(r))^2 f^2 (\sigma_{CC} v) 4\pi r^2 dr \frac{1}{4\pi r_{sol}^2}$$ \hspace{1cm} (6) where $r_{sol} \simeq 8$ kpc is the distance between us and the center of MW, $r_b \simeq 1$ kpc is the galactic bulge radius and $n_X = \rho_X(r)/m_X$. In the minimalistic dark matter scenario in which the dark matter pair annihilates directly into $e^- e^+$, the flux is given by the formula in Eq. (6), replacing $N_{pair} f^2$ with 1 and $\sigma_{CC}$ with $\sigma(\text{DM + DM} \rightarrow e^- e^+)$. In this scenario, the cross section should be of order of $10^{-39} - 10^{-38}$ cm$^2$ to account for the 511 keV line intensity which implies that in our model: $$\sigma_{CC} \sim 2 \text{ nb} \left( \frac{1.5 \times 10^{-4}}{f} \right)^2 \left( \frac{m_X}{5 \text{ MeV}} \right)^2 \frac{1}{N_{pair}}.$$ \hspace{1cm} (7) Let us now discuss the couplings that leads to the $e^- e^+$ production from the $C\bar{C}$ annihilation. Taking effective coupling $(\bar{e}e)(C\bar{C})/\Lambda$, the cross section of 2 nb implies $\Lambda < 100$ GeV which would be within the reach of the LEP collider and other accelerator experiments. We therefore invoke the possibility of intermediate light neutral particles, $\phi$, to avoid the bounds from the LEP. Then, the production of $e^- e^+$ takes place in two steps: First, $C\bar{C} \rightarrow \phi \phi$ and subsequently $\phi \rightarrow e^- e^+$. As a result, $N_{pair} = 2$. The $\phi \rightarrow e^- e^+$ decay can take place through $$g_\phi \phi \bar{e}e.$$ \hspace{1cm} (8) As long as $g_\phi < 10^{-11}$, the $\phi$ production in the supernova core will be negligible. We need $g_\phi > 10^{-15}$ in order for the $\phi$ decay length to be shorter than 10 pc. If $\phi$ travels less than $\sim 100$ pc in the bulge (i.e., of order of the resolution of the INTEGRAL observatory) before decay, the $n_{511}^2$ dependence of the 511 keV line morphology will be maintained, as favored by observation. Thus, $0.3 \times 10^{-15} < g_\phi < 10^{-11}$. $\phi$ is mainly an electroweak singlet mixed with the SM Higgs via $$a_\phi \phi \|H\|^2$$ Then, $$g_\phi = \sqrt{2} \frac{a_\phi v}{m_h^2} Y_e$$ \hspace{1cm} (9) in which $Y_e$ is the Yukawa coupling of the SM Higgs to the electron. Fortunately, the hierarchy between the $H$ and $\phi$ masses does not require unnatural fine tuned cancellation at tree level because $a_\phi \sim m_\phi$. In the early universe, $\phi$ can be produced via $e^- e^+ \rightarrow \phi \gamma$ with a number density $n_\phi \sim (\sigma_\phi v) n_e^2 H^{-1} \|_{T = m_\phi}$ where $\sigma_\phi = e^2 g_\phi^2 / 8\pi m_\phi^3$. Thus, during the period between $T = m_\phi$ and $T^{-1}$, we can write $n_\phi / n_\gamma = 8 \times 10^{-4} (g_\phi / 10^{-11})^2$. As a result, $n_\phi$ at $T \sim 1$ MeV would be small enough to satisfy the BBN bounds. In the range of $g_\phi$ in which we are interested, the decay of the $\phi$ particles takes place long before the recombination. For more details, see Ref. 4. 4. Direct detection of the $C$ particles in the galaxy The $C$ particles, having a tiny electric charge, can have Coulomb interaction with the electrons and protons. It can also interact with matter fields by the $t$-channel exchange of dark photon, $A'$. The velocity of $C$ arriving at the detector is larger than the typical velocity of the DM particles (i.e., velocity of $C \gg 10^{-3}$). As a result, the recoil energy imparted by the $C$ collision can be much larger than that expected from the collision of a DM particle of mass of MeV. This rises the hope to search for $C$ particles by the direct dark matter search experiments, despite $C$ being light. In this section, we shall show how the electron excess with recoil energy of $2 - 4$ keV at XENON1T can be explained within our model. We will then show that XENON1T sets strongest upper bound on the fraction of the $X$ particle decaying into $CC$, $f$. According to Eq. (4), the velocity of $C$ particles around us is $v_f \sim 0.08$. The Larmor radius in the Earth magnetic field is $5 \times 10^8$ km$(10^{-11}/g_C)(m_C/3$ MeV). Since the Larmor radius is much larger than the Earth radius, the Earth magnetic field cannot accumulate the $C$ particles so their density around the Earth will be similar to anywhere in the solar system (i.e., anywhere in a distance of 8 kpc from the MW galactic center): $(\rho_{DM}\|_{local}/m_X)f$ in which $\rho_{DM}\|_{local} \sim 0.4$ GeV/cm$^3$. Consider a $C$ particle with a velocity of $v_f$ scattering off a non-relativistic particle of mass, $m$. Up to corrections of $O(v_f^4)$, the maximum recoil energy, which corresponds to the backward scattering, can be written as $$E_{max} = 2mv_f^2 \frac{m_C^2}{(m + m_C)^2}. \quad (10)$$ For $m_C \sim 1 - 5$ MeV, $v_f = 0.08$ and $m = m_e$, $E_{max} = 3 - 5.5$ keV which is tantalizingly within the range of electron recoil excess observed by XENON1T. Taking $m$ equal to the mass of Xenon or of Argon, $E_{max}$ turns out to be much below the detection energy threshold of the experiments such as XENON1T, DarkSide and even the CRESST detector. The spectrum of recoiled electrons per detector mass will be $$\frac{dN}{dE_r} = \frac{Z_{out}}{m_N} (2f \frac{\rho_X}{m_X}) \int f_C(v) \frac{d\sigma}{dE_r} dv$$ where $m_N$ is the mass of the nuclei inside the detector; i.e., $m_N = 131$ GeV for the Xenon. $Z_{out}$ denotes the number of the electrons in the outer orbitals of the detector atoms whose binding energies are smaller than $E_r$. Ref. 20 shows that to a reasonable approximation, the binding energy can be treated with a step function in computing the scattering cross section. For Xenon, $Z_{out} = 44$ which is the number of the electrons in the $n = 3, 4, 5$ orbitals. These outer electrons have velocities smaller than $v_f$ so we neglect their velocity in our analysis. The velocity distributions of the $C$ and $\bar{C}$ particles are given by $f_C(v)$. In our computation, we take $f_C(v) = \delta(v - v_f)$ for simplicity. As far as $\|\epsilon\|/2m_e E_r \gg \|\epsilon - \delta\|/(2m_e E_r + m^2_{A'})$, the $t$-channel photon exchange gives the dominant contribution to the $C$ particles scattering off the electrons. The differential cross section can be written as 24 $$\frac{d\sigma}{dE_r} = \frac{e^2 q^2_{C}}{8\pi m_e v^2 E^2_r} \cdot \frac{1}{E^2_r} \left(1 + \frac{2m_e E_r}{2m_e E_r + m^2_{A'}} \cdot \frac{\delta - \epsilon}{\epsilon}\right)^2$$ where $E_r < E_{max} = 2m_e v^2 \left(\frac{m_C}{m_C + m_e}\right)^2$. The first (second) term in the parenthesis comes from the contribution from $t$ channel photon (dark photon, $A'$) exchange. Analyzing the XENON1T data, we take $v_f = 0.08$, $q_C = 10^{-11}$ and $m_X = 10$ MeV and use the data presented in Fig. 4 of Ref. 7. Let us define $\chi^2$ as follows $$\chi^2 = \sum_{bins} \frac{[N^{obs}_i - N^{pred}_i]^2}{\sigma^2_i}$$ where $N^{obs}_i$ is the observed number of events at bin $i$ and $N^{pred}_i$ is the prediction for the sum of the background and the signal from the $C$ scattering in the $i$th bin. The values of the uncertainty ($\sigma_i$), the background and $N^{obs}_i$ are obtained from Fig. 4 of Ref. 7. We only focus on the first seven bins because the excess appears only at $E_r < 8$ keV. The variation of $\chi^2$ with $m_C$ seems to be only mild. In sect. 2, we discussed that for $100$ keV $< m_{A'} <$ few MeV, identifying the $U_X(1)$ with the $L_\mu - L_\tau$ gauge symmetry, the bounds on $q' \propto (\delta - \epsilon)$ from supernova cooling can be relaxed because $A'$ can decay into neutrino pairs relatively fast. For $m_{A'} \sim$few MeV, $2m_e E_r \ll m^2_{A'}$ and we can treat $(\delta - \epsilon)/(em^2_{A'})$ and $f$ as two free parameters to be fitted with the data. Taking $m_C = 5$ MeV, we find that the minimum of $\chi^2$ lies at $f = 10^{-7}$ and $(\delta - \epsilon)/\epsilon = -2 \times 10^3(m_{A'}/\text{MeV})^2$. For $\epsilon = -q_C/(g_X \cos \theta_W) \sim 10^{-11}/(g_X \cos \theta_W)$, this corresponds to $q' \sim (10^{-8}/g_X)(m_{A'}/\text{MeV})^2$. At this point, $\chi^2 = 4.7$ with $5 = 7 - 2$ degrees of freedom which corresponds to a one-sided goodness of fit of 45 %. The results are shown in Fig. 1 with a red curve. In the first bin, contribution from the dark photon and the SM photon cancel each other. In case of the green curve shown in Fig. 1, we have taken $2m_e E_r(\delta/\epsilon - 1)/m^2_{A'} \to 0$ so there is no cancellation between dark and SM photons. As seen in the figure, in this case the spectrum diverges for small $E_r$. For $m_{A'} \sim 100$ keV, $2m_e E_r$ is comparable to $m^2_{A'}$ and we should use the whole formula to fit the data. Taking $m_C = 5$ MeV and $m_{A'} = 100$ keV ($m_{A'} = 200$ keV), the best fit will correspond to $(\delta - \epsilon)/\epsilon = -13.9$ and $f = 6.6 \times 10^{-7}$ ($( \delta - \epsilon)/\epsilon = -76$ and $f = 2 \times 10^{-7}$) and to the minimum $\chi^2 = 4.11$ ($\chi^2 = 4.5$). From Fig. 1, we Fig. 1. XENON1T datapoints superimposed on our predictions. We have set $m_C = 5$ MeV corresponding to $E_{\text{min}} = 3.3$ keV. The red curve shows the prediction for $m_{A'} > 1$ MeV, $f = 10^{-7}$ and $(\delta - e)/e = -46$, $g' = 10^{2.5} m_{A'}/(\text{MeV})^2$. The blue curve depicts the prediction for $m_{A'} = 0.1$ MeV, $(\delta - e)/e = 13.9$ and $f = 6.6 \times 10^{-7}$. The black curve illustrates the prediction for $m_{A'} = 0.2$ MeV, $(\delta - e)/e = -76$ and $f = 2 \times 10^{-7}$. The green curve shows the SM photon dominant regime with $f = 5 \times 10^{-6}$. The SM background is shown by the cyan dashed curve. observe that both fits result in very close prediction. At this fit, $g'$ is right above the supernova bound ($g' \sim 10^{-10} - 10^{-9}$) which can be ruled in by opening the possibility of decay of $A'$ into neutrino pairs in the outer layers of the supernova as discussed in sect. 2. The effects of such couplings can be tested by studying the flavor composition, the neutrino energy spectrum and the duration of the neutrino emission. The $g'$ coupling can lead to the production of the $A'$ particles in the early universe, contributing to the extra relativistic degrees of freedom, $\delta N_{eff} < 1$. More accurate $\delta N_{eff}$ determination by future CMB and BBN studies can test this model with $g' \sim 10^{-10} - 10^{-9}$. Let us now discuss the bound that can be derived on $f$ from the XENON1T data. To do so, we consider the bins with energy less than 8 keV. Fig. 2 shows $\chi^2$ as a function of $f$. Since the first bin is close to the detection threshold and it suffers from large uncertainty in the evaluation of the background from $^{214}\text{Pb}$, we present results both including and excluding the first bin. We have taken $m_C = 4$ MeV, but the results are rather insensitive to the value of $m_C$. To draw the blue and green lines, we have set $\delta = e$ which means the dark photon decouples and does not contribute to the scattering. Drawing the orange line, we have chosen the value of $2n_e(\delta - e)/(m_{A'}^2 e)$ such that the contributions from SM photon and dark photon cancel each other at $E_r = 1.5$ keV. From the figure we find an upper bound of $1.5 \times 10^{-6}$ on $f$ at 3 $\sigma$, taking all the bins with $E_r < 8$ keV: $$f < 1.5 \times 10^{-6} \quad \text{at } 3\sigma.$$ Excluding the first energy bin relaxes the bound to $10^{-5}$. Notice that these bounds are much stronger than the constraints discussed in Sect. 2. Since $f = \Gamma_X t_0$, the upper bound on $f$ is equivalent to a lower bound on the dark matter lifetime. The lifetime of $X$ should be larger than $10^{15} - 10^{16}$ years. Let us revisit the explanation for the 511 keV line taking into account this very stringent upper bound on $f$. To explain the intensity of the 511 keV line from the galactic bulge, the annihilation cross section of $C\bar{C} \rightarrow \phi\bar{\phi}$ should be $\text{mb}(10^{-7}/f)^2(\omega_{\chi X}/5 \text{ MeV})^2$. Taking a quartic coupling between these scalars of form $\lambda_{\phi C}\|\phi\|^2\|C\|^2$, we need $\lambda_{\phi C} \sim 0.02$ to obtain an annihilation cross section of mb. 5. Conclusion and discussion We have presented a dark matter model that can simultaneously explain the 511 keV line from the galactic bulge and the electron excess with recoil energy of $2-4$ keV reported by XENON1T. The model is based on introducing a dark matter particle with a mass of $O(1-10)$ MeV that decays into a pair of pico-charged particles. Although the velocities of the $C$ and $\bar{C}$ particles are much larger than the escape velocity from the galaxy, the galactic magnetic field can accumulate them inside the galactic disk. The $C$ particles with velocities of order of 0.08$c$ can scatter off the electrons of the direct dark matter search experiments, imparting a recoil energy of a few keV which can explain the electron excess at the low energy bins reported by XENON1T. Moreover, the $CC$ accumulated in the galactic bulge can annihilate with each other, giving rise to the excess of the 511 keV line. We have shown that the XENON1T data also sets strong upper bound of $10^{-6}$ on the fraction of $X$ particles that have decayed into $C\bar{C}$ since early universe until today, $f < 10^{-6}$. The density of $CC$ at the time of recombination and during the dark ages would be too low to pump any significant energy and ionize the atoms so the bounds from the CMB on the minimalistic DM explanation for the 511 keV line do not apply here. The model enjoys a rich phenomenology and can be tested by a myriad of methods. Most obvious test is further data from upcoming direct dark matter search experiments, especially the XENONnT detector which is an upgrade of XENON1T. The XENONnT detector is in the construction phase. The $C$ particles obtain their electric charge through the mixing of the photon and the dark photon, $A'$ which is the gauge boson of a new $U_X(1)$ symmetry under which $C$ and $\bar{C}$ are charged. Simultaneous explanation for the excess of events with recoil energy between $2 - 4$ keV and for the absence of an excess at the 1 keV bin requires a cancellation between the contributions from the $t$-channel exchanges of the photon and $A'$. Such a cancellation is possible with a light $A'$ with a mass of 100 keV–1 MeV. We have shown that by identifying the $U_X(1)$ symmetry with the $L_\mu - L_\tau$ symmetry and therefore allowing for $A' \rightarrow \nu_\mu \nu_\mu, \nu_\tau \nu_\tau$, it is possible to avoid the present bounds from the supernova cooling consideration. The $A'$ decaying to $\nu \bar{\nu}$ in the outer layers of supernova can however alter the evolution of supernova, leaving its imprint on the duration of neutrino emission, flavor composition, energy spectrum and shock revival. All these signatures provide alternative methods to test the model. The intensity of the 511 keV line from dwarf galaxies is proportional to the product of the number of $C$ and $\bar{C}$ particles. As a result, there should be a correlation between the magnitude of the magnetic field in dwarf galaxies and the intensity of the 511 keV photon line emitted from them which can be tested in the future. The positrons from the $C\bar{C}$ annihilation at the production are relativistic. A fraction of these positrons may annihilate in flight giving rise to a continuous $\gamma$ ray spectrum on which there is a bound, implying that the $C$ particles should be lighter than $6 - 15$ MeV depending on the models for ionization fraction in the galaxy\cite{21,22}. In future with more data, lighter $C$ can be probed by this consideration. Moreover, there will be an excess cosmic positron at few MeV range. At these energies, the solar wind renders measurements of cosmic positron flux in the MeV range obsolete within the solar system. Fortunately, there are detectors on board the Voyager spacecraft that have set an upper bound on the cosmic positrons outside the solar system. The bound can be interpreted as $m_C < 20$ MeV. Further data from Voyager can test smaller values of $m_C$ or hopefully detect a positron excess which will be a hint in favor of the present model. Acknowledgments* The article is prepared for the proceedings of the sixteenth Marcel Grossmann meeting (MG16). The author would like to thank the organizers and conveners of this meeting for the kind invitation. She is also grateful to ICRANET whose support made this participation possible. This project has received funding/support from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 860881-HIDDeN. The author has also received partial financial support from Saramadan under contract No. ISEF/M/98223, No. ISEF/M/9916 and No. ISEF/M/400188. References 1. C. Boehm, D. Hooper, J. Silk, M. Casse and J. Paul, Phys. Rev. Lett. 92 (2004) 101301 doi:10.1103/PhysRevLett.92.101301 [astro-ph/0309686]. 2. R. J. Wilkinson, A. C. Vincent, C. Boehm and C. McCabe, Phys. Rev. D 94 (2016) no.10, 103525 doi:10.1103/PhysRevD.94.103525 [arXiv:1602.01114 [astro-ph.CO]]. 3. T. Siegert, R. Diehl, A. C. Vincent, F. Guglielmetti, M. G. H. Krause and C. Boehm, Astron. Astrophys. 595 (2016) A25 doi:10.1051/0004-6361/201629136 [arXiv:1608.00393 [astro-ph.HE]]. 4. Y. Farzan and M. Rajase, JHEP 12 (2017), 083 doi:10.1007/JHEP12(2017)083 [arXiv:1708.01137 [hep-ph]]. 5. C. Boehm, Y. Farzan, T. Hambye, S. Palomares-Ruiz and S. Pascoli, Phys. Rev. D 77 (2008) 043516 doi:10.1103/PhysRevD.77.043516 [hep-ph/0612228]. 6. Y. Farzan, Phys. Rev. D 80 (2009) 073009 doi:10.1103/PhysRevD.80.073009 [arXiv:0908.3729 [hep-ph]]. 7. E. E. Aprile et al. [XENON], Phys. Rev. D 102 (2020) no.7, 072004 doi:10.1103/PhysRevD.102.072004 [arXiv:2006.09721 [hep-ex]]. 8. Y. Farzan and M. Rajase, Phys. Rev. D 102 (2020) no.10, 103532 doi:10.1103/PhysRevD.102.103532 [arXiv:2007.14421 [hep-ph]]. 9. G. Choi, T. T. Yanagida and N. Yokozaki, [arXiv:2007.04278 [hep-ph]]; J. Hisano, A. Ibarra and R. Nagai, [arXiv:2007.03216 [hep-ph]]; D. Aristizabal Sierra, V. De Romeri, L. J. Flores and D. K. Papoulias, [arXiv:2006.12457 [hep-ph]]; G. Choi, T. T. Yanagida and N. Yokozaki, [arXiv:2007.04278 [hep-ph]]; S. Karmakar and S. Pandey, [arXiv:2007.11892 [hep-ph]]; L. A. Anchordoqui, I. Antoniadis, K. Benakli and D. Lust, [arXiv:2007.11697 [hep-th]]; J. Cao, X. Du, Z. Li, F. Wang and Y. Zhang, [arXiv:2007.09981 [hep-ph]]; C. W. Chiang and B. Q. Lu, [arXiv:2007.06401 [hep-ph]]; N. Okada, S. Okada, D. Rau and Q. Shafi, [arXiv:2007.02908 [hep-ph]]; W. DeRocco, P. W. Graham and S. Rajendran, [arXiv:2006.15112 [hep-ph]]; M. Lindner, Y. Mambrini, T. B. d. Melo and F. S. Queiroz, [arXiv:2006.14590 [hep-ph]]; K. Nakayama and Y. Tang, [arXiv:2006.13159 [hep-ph]]; G. Alonso-Álvarez, F. Ertaş, J. Jaeckel, F. Kahlohofer and L. J. Thormaehlen, [arXiv:2006.11243 [hep-ph]]; K. Kannike, M. Raidal, H. Veermäe, A. Strumia and D. Teresi, [arXiv:2006.10735 [hep-ph]]. P. Ko and Y. Tang, [arXiv:2006.15822 [hep-ph]]; L. Delle Rose, G. Hüttsi, C. Marzo and L. Marzola, [arXiv:2006.16078 [hep-ph]]; U. K. Dey, T. N. Maity and T. S. Ray, [arXiv:2006.12529 [hep-ph]]; H. Alhazmi, D. Kim, K. Kong, G. Mohlabeng, J. C. Park and S. Shin, [arXiv:2006.16252 [hep-ph]]; P. Ko and Y. Tang, [arXiv:2006.15822 [hep-ph]]; L. Delle Rose, G. Hüttsi, C. Marzo and L. Marzola, [arXiv:2006.16078 [hep-ph]]; Y. Jho, J. C. Park, S. C. Park and P. Y. Tseng, [arXiv:2006.13910 [hep-ph]]; B. Fornal, P. Sandick, J. Shu, M. Su and Y. Zhao, [arXiv:2006.11264 [hep-ph]]; Q. H. Cao, R. Ding and Q. F. Xiang, [arXiv:2006.12767 [hep-ph]]; Y. Chen, J. Shu, X. Xue, G. Yuan and Q. Yuan, [arXiv:2006.12447 [hep-ph]]; W. Cho, K. Y. Choi and S. M. Yoo, [arXiv:2007.04555 [hep-ph]]; H. J. He, Y. C. Wang and J. Zheng, [arXiv:2007.04963 [hep-ph]]; H. Davoudiasl, P. B. Denton and J. Gehrmann, [arXiv:2007.04989 [hep-ph]]; H. An and D. Yang, [arXiv:2006.15672 [hep-ph]]; J. Smirnov and J. F. Beacom, [arXiv:2002.04038 [hep-ph]]; N. F. Bell, J. B. Dent, B. Dutta, S. Ghosh, J. Kumar and J. L. Newstead, [arXiv:2006.12461 [hep-ph]]; W. Chao, Y. Gao and M. j. Jin, [arXiv:2006.16145 [hep-ph]]; S. Baek, J. Kim and P. Ko, [arXiv:2006.16876 [hep-ph]]; I. M. Bloch, A. Caputo, R. Essig, D. Redigolo, M. Sholapurkar and T. Volansky, [arXiv:2006.14521 [hep-ph]]; J. Bramante and N. Wang, [arXiv:2006.14083 [hep-ph]]; J. M. Lee, [arXiv:2006.13978 [hep-ph]]; M. Bayarakhbar, A. Berikbaev, J. Jin and N. Weiler, [arXiv:2006.13918 [hep-ph]]; D. Choudhury, S. Mahanta, D. Sarkar and V. Sadhu, [arXiv:2007.08205 [hep-ph]]; H. An and D. Yang, [arXiv:2006.15672 [hep-ph]]; Y. Ema, F. Sala and R. Sato, [arXiv:2007.09105 [hep-ph]]; S. Chigusa, M. Endo and K. Kohri, [arXiv:2007.06163 [hep-ph]]; A. Das and M. Sen, [arXiv:2104.00027 [hep-ph]]; S. Vagnozzi, L. Visinelli, P. Brax, A. C. Davis and J. Sakstein, [arXiv:2103.15834 [hep-ph]]; H. J. He, Y. C. Wang and J. Zheng, [arXiv:2012.05891 [hep-ph]]; A. Aboubrahim, M. Klasen and P. Nath, JHEP 02 (2021), 229 doi:10.1007/JHEP02(2021)229 [arXiv:2011.08053 [hep-ph]]; R. Harnik, R. Plestid, M. Pospelov and H. Ramani, Phys. Rev. D 103 (2021), no.7, 075029 doi:10.1103/PhysRevD.103.075029 [arXiv:2010.11190 [hep-ph]]; B. Salehian, M. A. Gorji, H. Firouzjahi and S. Mukohyama, Phys. Rev. D 103 (2021), no.6, 063526 doi:10.1103/PhysRevD.103.063526 [arXiv:2010.04491 [hep-ph]]; A. N. Khan, Phys. Lett. B 819 (2021), 136415 doi:10.1016/j.physletb.2021.136415 [arXiv:2008.10279 [hep-ph]]; A. Karozaas, S. F. King, G. K. Leontaris and D. K. Papoulias, Phys. Rev. D 103 (2021), no.3, 035019 doi:10.1103/PhysRevD.103.035019 [arXiv:2008.03295 [hep-ph]]; V. Brdar, A. Greljo, J. Kopp and T. Opferkuch, JCAP 01 (2021), 039 doi:10.1088/1475-7516/2021/01/039 [arXiv:2007.15563 [hep-ph]]. 10. Y. Farzan, Mod. Phys. Lett. A 25 (2010), 2111–2120 doi:10.1142/S0217732310034018 [arXiv:1009.1234 [hep-ph]]. 11. N. Sabti, J. Alvey, M. Escudero, M. Fairbairn and D. Blas, JCAP 01 (2020), 004 doi:10.1088/1475-7516/2020/01/004 [arXiv:1910.01649 [hep-ph]]. 12. D. Feldman, Z. Liu and P. Nath, Phys. Rev. D 75 (2007), 115001 doi:10.1103/PhysRevD.75.115001 [arXiv:hep-ph/0702123 [hep-ph]]. 13. S. Davidson, S. Hannestad and G. Raffelt, JHEP 05 (2000), 005 doi:10.1088/1126-6708/2000/05/005 [arXiv:hep-ph/0001179 [hep-ph]]. 14. L. Chuzhoy and E. W. Kolb, JCAP 07 (2009), 014 doi:10.1088/1475-7516/2009/07/014 [arXiv:0809.0436 [astro-ph]]. 15. R. M. Bandyopadhyay, J. Silk, J. E. Taylor and T. J. Maccarone, Mon. Not. Roy. Astron. Soc. 392 (2009) 1115 doi:10.1111/j.1365-2966.2008.14133.x [arXiv:0810.3674 [astro-ph]]. 16. R. Diehl, D. Hartmann, P. Hoppe and N. Prantzos. (2002). Astronomy with Radioactivities, Springer. 17. E. Aprile et al. [XENON], Phys. Rev. Lett. 121 (2018) no.11, 111302 doi:10.1103/PhysRevLett.121.111302 [arXiv:1805.12562 [astro-ph.CO]]. 18. P. Agnes te al. (DarkSide-50), Phys. Rev. Lett. 121 (2018) 081307. 19. J. Schieck, G. Angloher, A. Bento, C. Bucci, L. Canonica, X. Defay, A. Erb, F. v. Feilitzsch, N. Ferreiro Iachellini, P. Gorla, A. Gütlein, D. Hauff, J. Jochem, M. Kiefer, H. Kluck, H. Kraus, J. C. Lanfranchi, J. Loebell, M. Mancuso, A. Münster, C. Pagliarone, F. Petricca, W. Potzel, F. Pröbst, R. Puig, F. Reindl, K. Schäffner, S. Schönert, W. Seidel, M. Stahlberg, L. Stodolsky, C. Strandhagen, R. Strauss, A. Tanzke, H. H. T. Thi, C. Türkoglu, M. Uffinger, A. Ulrich, I. Usherov, S. Wawozny, M. Willers, M. Wüstrich and A. Zöller, PoS ICHEP2016 (2016), 217 doi:10.22323/1.282.0217 [arXiv:1611.02113 [astro-ph.CO]]. 20. C. C. Hsieh, L. Singh, C. P. Wu, J. W. Chen, H. C. Chi, C. P. Liu, M. K. Pandey and H. T. Wong, Phys. Rev. D 100 (2019) no.7, 073001 doi:10.1103/PhysRevD.100.073001 [arXiv:1903.06085 [hep-ph]]. 21. T. Siegert, R. Diehl, G. Khachatryan, M. G. H. Krause, F. Guglielmetti, J. Greiner, A. W. Strong and X. Zhang, Astron. Astrophys. 586 (2016), A84 doi:10.1051/0004-6361/201527510 [arXiv:1512.00325 [astro-ph.HE]]. 22. J. Beacom and H. Yuksel, Phys. Rev. Lett. 97 (2006), 071102 doi:10.1103/PhysRevLett.97.071102 [arXiv:astro-ph/0512411 [astro-ph]]. 23. P. Sierolski, M. Casse and S. Schanne, Phys. Rev. D 74 (2006), 063514 doi:10.1103/PhysRevD.74.063514 [arXiv:astro-ph/0607374 [astro-ph]]. 24. R. Harnik, Z. Liu and O. Palamara, JHEP 07 (2019), 170 doi:10.1007/JHEP07(2019)170 [arXiv:1902.03246 [hep-ph]].
Sodium nitrite as corrosion inhibitor for mild steel in simulated concrete pore solution M. Pandiarajan¹, S. Rajendran¹,² J. Sathiya Bama¹ and R. Joseph Rathish³ ¹Corrosion Research Centre, PG and Research, Department of Chemistry, GTN Arts College, Dindigul, Tamil Nadu, India ²Corrosion Research Centre, Department of Chemistry, RVS School of Engineering and Technology, Dindigul, Tamil Nadu, India ³PSNA College of Engineering and Technology, Dindigul, India ABSTRACT The corrosion inhibition effect of mild steel in simulated concrete pore solution (SCPS) by Sodium Nitrite and Zn²⁺ has been investigated using weight loss method and cyclic voltammetry. The results show that 93% of inhibition efficiency is achieved with binary system consisting of 100 ppm of sodium nitrite – 50 ppm of Zn²⁺. Surface evaluation technique like FTIR is used to determine the nature of the protective film formed on the metal surface. The protective film consists of Fe²⁺ - SN complex, Zn²⁺- SN complex and Zn(OH)₂. Cyclic voltammetry study reveals that the protective film is more compact and stable even in 3.5% NaCl environment. Key words: Concrete Corrosion, Simulated Concrete Pore Solution, Mild Steel, Sodium nitrite, Well water. INTRODUCTION Effect of organic inhibitors on chloride corrosion of steel rebars in alkaline pore solution has been investigated by Marina Cabrini et al.. The inhibition properties of aspartic and lactic acid salts are compared with nitrite ions with regard to their effect on critical chloride concentration. The tests were carried out on carbon steel specimens in simulated pore solutions with initial pH in the range of 12.6 to 13.8. The results confirm that 0.1 M aspartate exhibits an inhibiting effect comparable with nitrite ions of the same concentration[1]. The inhibition of corrosion of reinforcing steel in simulated concrete pore solution (SCPS) has been studied using mass loss, gasometric measurements, potentiodynamic polarization and impedance studies using Mezlocillin (MZN) as a green inhibitor. The studies clearly revealed that MZN acted as cathodic inhibitor. Diffused reflectance spectra confirmed the formation of adsorbed film of inhibitor on reinforcing steel in SCPS[2]. The corrosion inhibition effect and mechanism of D-sodium gluconate for reinforcing steel in the simulated concrete pore solution containing Cl⁻ were studied by electrochemical techniques, including corrosion potential, potentiodynamic polarization, and electrochemical impedance spectroscopy measurements. The results indicate that 0.01 M D-sodium gluconate showed a good corrosion inhibition effect on reinforcing steel in the simulated concrete pore solution containing 0.1 M NaCl because it strongly hindered the anodic reactions, by forming a compact adsorptive film by strong chelation and effectively inhibit the initiation of reinforcing steel corrosion[3]. The Effect of NO₃⁻ on the early stages of pitting corrosion of Q235 carbon steel in a simulated concrete pore solution (pH=11) was studied by means of polarization curves, Mott-Schottky curves and XPS methods. It was found that, NO₃⁻ inhibits both the stable pitting corrosion and the meta stable pitting corrosion [4]. The newly synthesized polymer, azo methine-based polyester, has been employed as inhibitor against the corrosion of rebar in artificially simulated concrete pore solution with chloride contamination (blank) by means of Tafel polarization and electrochemical impedance measurements. Polarization studies exhibited the maximum inhibition efficiency of 98% at 1000 ppm concentration. Electrochemical studies revealed the mixed-type nature of the inhibitors [5]. EXPERIMENTAL SECTION 2.1. Preparation of the specimens: Mild steel specimen was used in the present study. (Composition (wt %): 0.026 S, 0.06 P, 0.4Mn, 0.1C and balance iron. The dimension of the specimen was 1 x 4 x 0.2 cm were polished to a mirror finish and degreased with trichloroethylene and used for the weight-loss method and surface examination studies. The environment chosen is well water and the physico-chemical parameter of well water is given in table 1. 2.2. Preparation of Simulated Concrete pore solution (SCPS): Simulated concrete pore solution is mainly consisted of saturated calcium hydroxide Ca(OH)$_2$, sodium hydroxide (NaOH) and potassium hydroxide (KOH) with the pH ~ 13.5. However in numerous studies of rebar corrosion, saturated Ca(OH)$_2$ has been used as a substitute for pore solution. A saturated calcium hydroxide solution is used in present study, as SCP solution with the pH ~ 12.5. Table – 1 Physico-chemical parameters of well water \| Parameters \| value \| \|----------------\|----------------\| \| pH \| 8.38 \| \| Conductivity \| 1770 $\mu\Omega^{-1}\text{cm}^{-1}$ \| \| Chloride \| 665 ppm \| \| Sulphate \| 214 ppm \| \| TDS \| 1100 ppm \| \| Total hardness \| 402 ppm \| \| Total Alkalinity \| 390 ppm \| \| Magnesium \| 83 ppm \| \| Potassium \| 55 ppm \| \| Sodium \| 172 ppm \| \| Calcium \| 88 ppm \| 2.3. Weight loss method Weighing the specimens before and after Corrosion: All the weighing of the mild steel specimens before and after corrosion was carried out using Shimadzu balance, model AY62. Determination of Corrosion Rate: The weighed specimens in triplicate were suspended by means of glass hooks in 100 ml SCPS prepared in well water containing various concentration of potassium chromate in the presence and absence of Zn$^{2+}$ for one day, the specimen were taken out, washed in running water, dried, and weighed. From the change in weights of the specimens, corrosion rates were calculated using the following relationship: $$\text{CR} = \left[\frac{\text{Weight loss in mg}}{\text{Area of the specimens in dm}^2 \times \text{Immersion periods in days}}\right] \text{ mdd}$$ \hspace{1cm} (1) Corrosion inhibition efficiency (IE, %) was then calculated using the equation: $$I.E = 100\left[1 - \left(\frac{W_2}{W_1}\right)\right] \%$$ \hspace{1cm} (2) Where, $W_1 =$ corrosion rate in the absence of the inhibitor, and $W_2 =$ corrosion rate in the presence of the inhibitor, RESULTS AND DISCUSSION Analysis of results of the weight loss method The calculated inhibition efficiencies (IE) and corrosion rates (CR) of sodium nitrite in controlling corrosion of mild steel immersed in simulated concrete pore solution (SCPS) for a period of one day in the absence of and presence of Zn$^{2+}$ ion are given in Table 2. The inhibition efficiency is also given in this Table 2. The corrosion rates of the potassium chromate systems at various concentrations are shown in Fig 1. It is observed from Table 2 that sodium nitrite (SN) shows some inhibition efficiency, 50 ppm of SN has 85% IE. As the concentration of SN increases, the IE increases, and corrosion rate decreases. 50 ppm of SN has 85% IE. 100 ppm of SN has 93% IE. When 50 ppm of Zn$^{2+}$ is added, the IE shifting increases in each case. That is addition of SN and Zn$^{2+}$ increases the corrosion protection of mild steel immersed in SCPS prepared in well water. Table 2: Inhibition efficiencies (IE%) and corrosion rates (CR) obtained from SN-Zn$^{2+}$ system in controlling corrosion of mild steel immersed in SCPS prepared in well water Inhibitor system: SN – Zn$^{2+}$ Immersion period: 1 day* \| SN ppm \| Zn$^{2+}$ ppm \| IE % \| CR mdd \| \|--------\|-----------------\|------\|--------\| \| 0 \| 0 \| 0 \| 25 \| \| 0 \| 50 \| 20 \| 20 \| \| 50 \| 0 \| 85 \| 3.7 \| \| 100 \| 0 \| 87 \| 3.2 \| \| 50 \| 50 \| 90 \| 2.5 \| \| 100 \| 50 \| 93 \| 1.7 \| Influence of Zn$^{2+}$ on the corrosion inhibition efficiency: It is observed that when 50 ppm Zn$^{2+}$ is added, the inhibition efficiency increases in both cases. The formulation consisting of SCPS + 100 ppm of sodium Nitrite and 50 ppm of Zn$^{2+}$ has 90% corrosion inhibition efficiency. In presence of Zn$^{2+}$, more amount sodium Nitrite is transported towards the mild steel surface. On the mild steel Fe$^{2+}$-Nitrite complex is formed on the anodic sites of the mild steel surface. Thus, the anodic reaction is controlled. The cathodic reaction, the generation of OH$^-$ is controlled by the formation of Zn(OH)$_2$ on the cathodic sites of the mild steel surface. Thus, the anodic reaction and cathodic reaction are controlled effectively. \[ \text{Fe} \rightarrow \text{Fe}^{2+} + 2e^- \] (anodic reaction), \[ \text{Fe}^{2+} + \text{Zn}^{2+} - \text{Nitrite} \rightarrow \text{Fe}^{2+} - \text{Nitrite} + \text{Zn}^{2+} \] \[ \text{O}_2 + 2\text{H}_2\text{O} + 4e^- \rightarrow 4\text{OH}^- \] (cathodic reaction), \[ \text{Zn}^{2+} + \text{OH}^- \rightarrow \text{Zn(OH)}_2 \downarrow \] ![Graph showing inhibition efficiency (IE) of SN – Zn$^{2+}$ system in the corrosion of mild steel immersed in SCPS prepared in well water](image) Analysis of FTIR spectra FTIR offers quantitative and qualitative analysis for organic and inorganic samples. Fourier Transform Infrared spectroscopy (FTIR) identifies chemical bonds in a molecule by producing an infrared absorption spectrum. The spectra produce a profile of the sample, a distinctive molecular fingerprint that can be used to screen and scan samples for many different components. FTIR is an effective analytical instrument for detecting functional groups and characterizing covalent bonding information. FTIR spectra were used to analyze the protective film formed on metal surface [6]. The FTIR (KBr) spectrum of pure sodium nitrite is given in figure 2a. The NO$_2^-$ stretching frequency appears at 1268 cm$^{-1}$. The FTIR spectrum of the film formed on the metal surface after immersion in SCPS prepared in well water containing 100 ppm of NaNO$_2$ and 50 ppm of Zn$^{2+}$ is shown in figure 2b. The NO$_2^-$ stretching frequency of NaNO$_2$ shifted from 1268 to 1205.67... cm$^{-1}$. This confirms that the oxygen atom of the nitrite has coordinated with Fe$^{2+}$ resulting in the formation Fe$^{2+}$-nitrite complex on the metal surface. Also there is possibility of anchoring of nitrite on the layer consisting CaO, Ca(OH)$_2$, CaCO$_3$. ![Fig. 2a FTIR spectrum of pure sodium nitrite](image) ![Fig. 2b FTIR spectrum of film formed on the metal surface after immersion in aqueous solution containing SCPS+ 100 ppm SN+ 50 ppm of Zn$^{2+}$](image) Peak appears at 1398 cm$^{-1}$ is due to Zn–O stretching. The - OH stretching frequency appears at 3439.79 cm$^{-1}$. These observations indicate the presence of Zn (OH)$_2$ formed on the metal surface. Peak appears at 1515, 746 and 1398 cm$^{-1}$. These peaks confirm the presence of calcium carbonate, calcium oxide, calcium hydroxide and on the metal surface[7-8]. 2. CYCLIC VOLTAMMETRY Cyclic voltammograms have been used to investigate the corrosion behaviour of metals[9-12]. Deyab and Keera[9] have analysed the influence of sulphide, sulphate, and bicarbonate anions on the pitting corrosion behaviour of mild steel in formation water containing chloride ions by means of cyclic voltammetry technique. The cyclic voltammograms were recorded in the presence of increasing amounts (0.1 to 0.3M) of NaCl at a scan rate of 10 mV s$^{-1}$. The anodic response exhibits a well defined anodic peak followed by a passive region. The anodic peak is due to active metal dissolution and formation of ferrous hydroxide [13]. The cathodic sweep shows two cathodic peaks. The appearance of cathodic peak around -1.1 V is due to reduction of corrosion product, namely iron oxide to iron. The appearance of cathodic peak around -0.7 V is due to the reduction of pitting corrosion products precipitate on the electrode surface. In the present study, cyclic voltammograms were recorded by measuring the working electrode, mild steel, in 3.5% NaCl solution. The cyclic voltammogram of mild steel immersed in 3.5% NaCl is shown in Fig 3a. It is observed that during anodic scan, no peak is observed but a passive state is noticed. This can be explained as follows: When the metal dissolves, ferrous hydroxide is formed. When the concentration of ferrous oxide at the anodic surface exceeds its solubility product, precipitation of solid oxide occurs on the electrode surface. When the surface is entirely covered with oxide passive film, anodic current density does not increase indicating onset of passivation. In the passive state, the Cl$^-$ ion can be adsorbed on the bare metal surface in competition with OH$^-$ ions. As a result of high polarizability of the Cl$^-$ ions, the Cl$^-$ ions may adsorb preferentially [14]. The cathodic sweep shows only one peak at -0.652 V. This is due to the reduction of corrosion product, iron oxide to iron. The peak due to reduction of pitting corrosion product is absent. (This reveals that pitting corrosion does not take place under the given experimental conditions). The cyclic voltammogram of mild steel, which has been immersed in SCPS for one day and dried is shown in Fig 3b. (brown iron oxide was observed on the mild steel electrode). It is observed that during anodic sweep, no peak appears, but a passive region is observed. During the cathodic sweep, the peak due to reduction of pitting corrosion product appears at -955 mV indicating that pitting corrosion takes place. However, the peak due to reduction of corrosion product, iron oxide, appears at -2.7 V. The current density increases from $-2.5 \times 10^{-7}$ A to $-4.6 \times 10^{-5}$ A. This indicates that when mild steel electrode is immersed in SCPS for one day, a protective film of iron oxide is formed on the electrode surface. It is stable in 3.5% NaCl solution. The increase in current density is explained as follows: Chloride ion is adsorbed on the passive film. The adsorbed chloride ion penetrates the oxide film especially at the flaws and defects in the oxide film [15]. When the penetrated chloride ion reaches the metal surface, they promote local corrosion. Figure 3a. Cyclic voltammogram of mild steel electrode immersed in 3.5% NaCl When the mild steel electrode is immersed in SCPS water containing 100 ppm of SN and 50 ppm of Zn$^{2+}$ for one day, a protective film is formed. It consists of Fe$^{2+}$–SN complex, Zn$^{2+}$–SN complex, and Zn(OH)$_2$ as revealed by FTIR spectroscopy. The cyclic voltammogram of mild steel electrode deposited with the above protective film is shown in Fig.3c. It is observed that during anodic sweep, dissolution of metal does not take place. This indicates that the protective film is stable and compact. Electrons are not transferred from the metal surface, and a passive region is observed. During cathodic sweep, the peak corresponding to reduction of pitting corrosion product appears at -850 mV. However, the peak due to reduction of iron oxide to iron appears at -1.31 V. The current density increases from $-2.5 \times 10^{-3}$ A to $-1.5 \times 10^{-3}$ A. The decrease in current density suggest that the adsorbed inhibitor molecule on the electrode surface not only effectively retard the anodic dissolution iron, but also enchanced the stability of ferrous species to be further oxiesed(Fe$^{2+}$) into ferric iron. It is observed from the Fig 2a, 2b, 2c that the pitting potentials for the three systems are at -656 mV, -970 mV, and -645 mV respectively. That is when mild steel electrode is immersed in the SCPS medium; the pitting potential is shifted to more negative side (active side, i.e., -970 mV). It accelerates corrosion because the protective film formed is porous and amorphous. When the electrode is immersed in the inhibitor medium, the pitting potential is shifted to the noble side, i.e., -645 mV. This indicates that the passive film found on the metal surface in the presence of inhibitors is compact and stable. It can withstand the attack of chloride ion present in 3.5 NaCl. REFERENCES [1] Marina Cabrini, Francesca Fontana, Sergio Lorenzi, Tommaso Pastore, and Simone Pellegrini, J. Chem, 2015, 2015, Article number 521507. [2] ELHarish; S Karthikeyan; SK Sekar; Int. J. Chem Tech Research, Volume, 2015, 7(4), 2003-2006. [3] JH Li ; B Zhao ; J Hu, ; H Zhang ; S G Dong ; R G Du ; C J Lin. Int. J.Electrochem.I Sci, 2015, 10(1), 956-968. [4] Y Tang; B Niu; B Lin X Zhao; Y Zuo; Corrosion Science and Protection Technology, 2015, 27(2), 123-128. [5] B Bhuvaneshwari; A Selvaraj; NR Iyer; L Ravikumar. Materials and Corrosion, 2015, 66(4), 387-395. [6] S Agnesia Kanimozhi; S Rajendran; Arab. J. Sci. Engg, 2010, 35(A), 41-52. [7] A Richard Nyquist; O Ronald Kagel, Infrared Spectra of Inorganic Compounds, Academic press, New York and London, 1971 p.553. [8] A Richard Nyquist; O Ronald Kagel, Infrared Spectra of Inorganic Compounds, Academic press, New York and London, 1971 318. [9] CAM. Dutra; EN Codaro; RZ Nakazato. Mater. Sci.Appl., 2012, 3, 348. [10] L Feng; H Yang; F Wang. Electrochim. Acta, 2011, 58, 427. [11] Y Li, P Kumar; X Shi; TA Nguyen; Z.Xiao; J Wu. Int J Electrochem Sci, 2012, 7, 8151. [12] MA Deyab; ST Keera. Egypti J Petrol, 2012, 21, 31. [13] VK Gouda; Proc. 12th Int.Corrosion Cong., Houston, TX, USA, 1993, 19. [14] S AM Refaey; SS; Abed El-Rehim, F.Taha, M.B. Saleh, R.A. Ahmed. Appl Surf Sci. 158, 2000, 190. [15] DCW Kannagara; BE Conway; J Electrochem Soc., 134, 1987, 894.
Bridging across cognitive training and brain plasticity: a neurally inspired computational model of interactive skill learning Wai-Tat Fu,1* Hyunkyu Lee,1 Walter R. Boot2 and Arthur F. Kramer1 This article reviews recent empirical and brain imaging data on effects of cognitive training methods on complex interactive skill learning, and presents a neurally inspired computational model that characterizes the effects of these training methods. In particular, the article focuses on research that shows that variable priority training (VPT), which requires learners to shift their priorities to different task components during training, often leads to better acquisition and retention of skills than fixed priority training (FPT). However, there is only weak evidence that shows that VPT can enhance transfer of complex interactive skills to untrained situations. Brain imaging studies show that VPT leads to significantly lower activations and a higher reduction of activities in attentional control areas after training than FPT. Research also shows that the volume of the striatum predicts the learning effects, but only in VPT. The computational model, developed based on learning mechanisms at the neural level, bridges across the empirical and the brain imaging results by explaining the effects of VPT and FPT at both the behavioral and neural levels. The results were discussed in the context of previous findings on cognitive training. © 2012 John Wiley & Sons, Ltd. How to cite this article: WIREs Cogn Sci 2013, 4:225–236. doi: 10.1002/wcs.1214 INTRODUCTION Skill learning in complex, interactive environments has long been an important cognitive science research topic.1–4 Decades of behavioral research has shown that although practice invariably improves performance, explicit training methods can more effectively enhance learning, retention, and transfer of skill.5–8 Recent research shows some evidence that cognitive training using simple laboratory tasks can improve specific basic cognitive operations and abilities, such as visual selective attention,9 working memory,10 or fluid intelligence.11 On the other hand, research on cognitive training using more complex interactive tasks (e.g., video games) shows mixed results—while training performance consistently increases, transfer of complex skills to untrained situations is limited.12,13 To further understand the effects of cognitive training, recent brain-imaging studies have focused on associating training outcomes to changes in brain activities.14,15 In particular, researchers are interested in understanding how different training methods may lead to changes in patterns of activities in different brain regions, and to identify biomarkers for effective acquisition, retention, and transfer of complex interactive skill. Although there has been an increasingly large volume of empirical and brain imaging data, there is still a lack of an integrated theory that bridges across theories of cognitive training and brain plasticity to better predict the effectiveness of training strategies on interactive skill learning. The current focus article is... a first step toward such an integrated theory. First, we provide a brief review of recent advances in behavioral and brain imaging research on training of interactive skills that show how different training methods lead to changes in brain structure and function and behavioral performance. Second, we introduce a neurally inspired computational model that explains and predicts effectiveness of training methods based on mechanisms and representations at the neural level. COGNITIVE TRAINING USING COMPUTER GAMES The Space Fortress Game One of the first attempts to systematically study effects of training methods on complex interactive skills was the learning strategies initiative.\textsuperscript{1,16} Under this initiative, a series of training strategies were outlined to enhance complex interactive skill learning using a multifaceted video game called space fortress (SF). Given that there has been a large volume of research using the same SF game, we briefly describe the general design of the game. The SF game was designed such that learners need to master a large number of basic cognitive operations (as opposed to a single specific cognitive function, such as working memory, see e.g., Jaeggi et al.\textsuperscript{11}) by practicing when and how different cognitive operations should be combined and deployed. Learning to play the SF game, therefore, involves not only training on a specific cognitive function, but also the strategies for interleaving different operations, combining micro operations into macro operations, as well as the decision on when and how different operations should be used in dynamic situations. Training methods for this kind of complex interactive skills are, therefore, also more complex, and their effects may also differ from cognitive training methods that focus on a single cognitive function. The main objective of the SF game is to destroy the SF, while maintaining a spaceship within a certain velocity limit and prespecified boundaries on the screen (Figure 1). These two subtasks are reflected by the velocity and control scores, respectively, which are continuously updated on the screen. Learners use a joystick to control the spaceship, which flies in a frictionless environment. Missiles are fired from the spaceship. In addition to destroying the fortress, the learner has to protect his/her spaceship against damage from the fortress and mines. Learners can also shoot the mines to gain points. In addition to tasks related to the fortress and mines, there is a constant monitoring task embedded in the game. Symbols appear periodically below the fortress and whenever a dollar sign appears for the second time, players can use the mouse to either select bonus points or bonus missiles (which are a limited resource). However, if learners incorrectly identify the first dollar symbol as the second, they miss their opportunity to obtain a bonus when the second dollar sign does appear. Thus, learners are always encouraged to monitor this information. The four subscores: points ($P$), control ($C$), velocity ($V$), and speed ($S$) add up to the total scores, which are also continuously displayed on the bottom of the screen (see Figure 1). Learners are asked to keep their ship within the two hexagons on the screen, and doing so increases the Control subscore. Flying the ship outside of the large hexagon or leaving the screen entirely (referred to as ‘going into hyperspace’) subtracts from the Control subscore. The velocity subscore rewards participants for going slowly and punishes participants for flying at high speeds. The speed subscore rewards/punishes participants for how quickly they deal with mines, and the points subscore rewards participants for shooting and destroying the fortress, but subtracts points for damage and destruction of the player’s ship. Typically, in fixed priority training (FPT), learners are instructed to give equal weight to the subscores (or just the total score) throughout the sessions. In variable priority training (VPT), learners are instructed to emphasize one of the four subscores in each game, and the emphasis changes throughout the sessions. FIGURE 1 \| The space fortress game display. See text for details of the game. Given the complexity of the task, learning to effectively perform and integrate the various cognitive, perceptual, and motor operations requires at least 10–20 h of training. Because each of the components involves relatively different cognitive operations, analysis of changes in behavior across time not only can shed light on the effectiveness of training methods on improving performance in these components, but also the extent to which training can enhance basic cognitive operations and lead to transfer of these complex interactive skills to untrained situations. A wide range of training strategies have been studied (see Ref 1). Part-task training involves principled decomposition of the complex task into functional components, which can be individually mastered by the learner. Whole-task training with VPT involves training on the integrated whole task with changing emphasis on the subcomponents of the task; while whole-task training with FPT involves equal emphasis of the subcomponents throughout the training. Finally, hybrid variable priority training (HVT) combines part-task training with VPT. Learners in HVT begin with a small number of components, but as they gradually master them, more components are added until the whole task is used (see Refs 17 and 20 for a comparison). Behavioral Findings Using the SF game, Gopher et al. showed that VPT could lead to better learning and performance in interactive tasks. They argued that VPT-enabled participants to explore different strategies and thus develop a better match between the requirements of the tasks and the efficiency of their efforts. Compared to FPT, VPT makes people better able to strategically allocate attention to multiple components of the task to comply with the change in emphases during training. Although benefits of VPT on global performance and some general transfer of skills to a new environment have been demonstrated through a number of studies, there is still a lack of understanding on the specifics of whether and how it promotes transfer to untrained situations. Recent research on cognitive training has led to a reexamination of various training strategies. VPT was found to be beneficial for dual-task performance, faster learning and higher level of mastery of complex skills, and better working memory performance in older adults. Some studies found that VPT could lead to better transfer to untrained tasks. A recent study by Lee et al. found that HVT (a variation of VPT) led to higher level of mastery of skills in the SF game than FPT, and the benefits of HVT remained even after 7 months. HVT was also found to be most useful for participants who started training with lower performance, presumably because HVT allowed them to focus more on their weaknesses early on, such that their performance improved more in HVT than in FPT. Contrary to previous claims that VPT could lead to transfer of skills to untrained tasks (often called ‘far transfer’, as opposed to ‘near transfer’, in which the trained and untrained tasks are similar or have functionally overlapping components), Lee et al. found that performance gained from VPT did not transfer to untrained tasks. The result was also in contrast to recent studies on cognitive training using simple laboratory tasks, in which general increase in ability was observed after training for certain participants. On the other hand, the results were consistent with previous studies using complex interactive tasks, in which only near transfer was observed but not far transfer. The results not only suggest that far transfer is difficult to achieve, but also the fact that more research is needed to understand why certain training strategies using certain tasks lead to far transfer for some participants and not others. As we explain later, one possibility is to study the cognitive representations (a set of features that the person uses to encode the stimuli and responses in the environment) of the different task components as individuals are trained, as the same tasks could induce different cognitive representations and may influence skill learning and transfer. We elaborate on the role of cognitive representations when we introduce our computational model. First, we review recent neuroscience research on skill learning, which guides the construction of the computational model. Brain Changes Induced by Training Strategies In addition to behavioral research, recent brain imaging studies have shown how changes in brain activities induced by different training strategies differ. Using the SF task, Anderson et al. compared brain changes during part-task and whole-task training. They found that individual learning gains could be predicted by critical regions of the brain involved when individual task components were learned. Most importantly, individual differences could be explained by the extent to which additional activation was observed when these regions already had a high level of activation. This finding was consistent with the cognitive load hypothesis, which claims that learning effectiveness tends to decrease as brain activation reaches saturation (such that additional activation is limited). In other words, learners with lower capacity may benefit more from part-task training as it requires less attentional demand, whereas learners with high capacity may benefit more from whole-task training as it provides more training opportunities per unit time. Prakash et al.\textsuperscript{15} and Lee et al.\textsuperscript{14} compared brain activation changes induced by HVT and FPT as individual learned to play the SF game over a 30-h period. Consistent with previous results, they found that individuals trained by HVT reached higher performance and showed less brain activation in areas related to visuospatial attention\textsuperscript{14} and top-down attentional control\textsuperscript{15} than those trained by FPT. In addition, HVT learners showed a significantly higher reduction of activation in these areas after training than FPT leaners, along with reduced activation in the motor and sensory cortices and the posteromedial cortex. These findings suggest that VPT can lead to better mastery of skills by reducing the reliance on top-down attentional networks in the prefrontal cortices. Given that the striatum plays a critical role in skill learning, Erickson et al.\textsuperscript{28} imaged the striatum with high-resolution MRI before the SF game training using an automated segmentation algorithm, after which the volume of each region was calculated based on voxel dimensions. Analysis was conducted on performance across the entire 20-h training period, collapsed across both training groups, to determine whether striatal volumes were predictive of performance improvements in the SF game. Multiple regression analyses were conducted between changes in performance and the volume of each region of the striatum, while including initial performance as covariates in the model. No correlation was found between striatum volumes and initial total scores in both groups, suggesting that initial performance of the task was not correlated with striatum volumes. However, striatum volumes significantly predicted change in performance when the groups were collapsed for the left and right caudate nucleus. The volumes of the dorsal striatal regions were positively correlated with training-induced performance improvements, and accounted for 23% of the variance in learning amounts across training. Analysis including the two training groups showed significant positive training group $\times$ volume interactions for the points, velocity, and speed subscores. The dorsal striatal volumes were predictive of performance only for subjects in the VPT group, but not for the FPT group. In sum, results showed that the preexisting volumes of the striatum predicted learning improvements \textit{only} in the VPT group, but not in the FPT group. In other words, individuals with a larger striatum learned more effectively in VPT, but this benefit was not found in FPT. No such correlation was found between performance and other brain regions, such as the putamen and hippocampus, supporting the critical role of the striatum in skill learning. Summary of Results As described above, results from behavioral and brain research provide support for the benefit of VPT on learning complex interactive skills. Brain imaging shows that VPT lead to higher reduction of activation in attentional control regions [the right middle frontal gyrus, right superior frontal gyrus, and the ventral medial prefrontal cortex (PFC)] compared to FPT,\textsuperscript{14,15} suggesting that VPT learners are able to perform better with less reliance on top-down attentional control. In addition, learners who have low initial performance tend to benefit more from part-task training and VPT,\textsuperscript{13} and individual learning gains were found to be limited by the cognitive load imposed by whole-task training.\textsuperscript{26} To summarize, convergent evidence supports the following general characteristics of VPT. First, VPT is a more effective method for reaching the \textit{automatic} stage of skill acquisition, in which stimulus–response associations (or \textit{habits}) are selected and executed with less reliance on top-down attentional control. Second, one reason why VPT is more effective is because they tend to impose a lower cognitive load during training, such that individual components can be focused on as training emphasis switches. A NEURALLY INSPIRED COMPUTATIONAL MODEL OF FLEXIBLE SKILL LEARNING Researchers interested in developing integrated theories often rely on computational models to provide mechanistic accounts of behavior. To increase the biological plausibility of these models, researchers often rely on the large body of literature on computational properties of different brain structures to provide constraints on the models. These models are often referred to as \textit{neurally inspired computational models}, as they attempt to provide multilevel explanations by simulating how computations in the brain can lead to specific behavioral patterns.\textsuperscript{29–31} We will adapt a similar approach to provide an integrated theory that bridges what we know about brain plasticity to effects of cognitive training by different training schedules. Decades of research showed that skill learning emerges as a result of the experience-dependent plasticity in the basal-ganglia-cortical neural circuits.\textsuperscript{29,32,33} Two major forms of learning are observed in these circuits. \textit{Hebbian stimulus–response (S–R) learning} allows direct association between stimuli and responses. This form of learning requires little executive processing, but often requires extensive training. \textit{Discrimination learning} allows recognition of patterns of stimuli and selection of correct responses. This form of learning requires executive processing of information at the attentional control areas including the PFC that guides the selection of actions and is mediated by external feedback. Theories of skill acquisition often assumes that learning shifts from slow goal-directed behavior requiring executive processing to discriminate patterns of stimuli and the selection of correct responses to fast execution of S–R behavioral rules requiring less executive processing.\textsuperscript{34,35} The basal ganglia is often believed to be involved in the acquisition of these behavioral rules, in which activities at the caudate tend to increase during initial training and decline as the rules become ‘habitual’ (see Ref 36 for a review). We discuss how the co-occurrences and shift between these two learning processes may explain the benefits of VPT at both the neural and behavioral levels. However, we first briefly describe the neural components of these two processes, which inform the major components of the model. \textbf{Hebbian (Habit) Learning} Hebbian learning between the frontal cortex and the premotor cortex allows fast selection of responses tied to an environmental stimulus. Unlike learning in the striatum, Hebbian learning is often independent of the outcome of the responses, that is, association is strengthened whenever the response is selected when the stimulus is perceived. During initial learning, because the correct S–R rules have not yet been learned, none of the responses will be activated. Instead, goal-directed behavior will guide the selection of the right response. With practice, the correct S–R rules are strengthened, which allow correct responses to be activated when the stimuli that tied to them are perceived. \textbf{Discrimination Learning} During discrimination learning, goal-directed actions that require attentional function at various brain regions including the PFC are selected based on behavioral rules acquired through the declarative system (i.e., by following instructions in an experiment to associate a stimulus to a response). This form of learning involves the connections between the PFC to the diverse set of spiny neurons in the striatum for pattern recognition computations\textsuperscript{37} and the existence of relatively ‘private’ feedback loops of connectivity from diverse cortical regions that converge onto those striatal spiny cells, via the pallidum and thalamus, and lead back to the frontal cortex.\textsuperscript{38,39} Unlike neurons that learn through a Hebbian-like mechanism, spiny neurons are found to receive specialized inputs that appear to contain training signals from dopamine (DA) neurons.\textsuperscript{40,41} \textbf{The Role of the Striatum in Skill Learning} Research shows that the striatum is activated while performing tasks that require cognitive flexibility such as task switching and transfer to untrained tasks.\textsuperscript{33,42–44} PET studies in humans have shown that DA release and binding are increased in both these striatal regions when subjects play a video game, and that greater DA binding is associated with better performance.\textsuperscript{45} As described above, Erickson et al.\textsuperscript{38} shows that the differential size of the striatal regions predicts learning on an unfamiliar video game. They found that individual structural differences in the striatum were effective predictors of procedural learning and cognitive flexibility and were sensitive indicators of ventral-to-dorsal differences in striatal recruitment during learning. These findings suggest that changes in the striatum are predictive of learning effects observed during the video game. As we show later, our model shows that discrimination learning at the striatum induced by different training methods can explain differences of their effectiveness in skill transfer. \textbf{The General Structure of the Model} The general structure of the model is shown in Figure 2. The activations of the neurons in the attentional control regions including the PFC (represented collectively in the figure as PFC) represent the different stimulus patterns perceived by the corresponding sensory cortical units (I). Neurons in the PFC are fully connected to the neurons in the association striatum (S), and the connection strength is changed through a DA-moderated discrimination learning mechanism (discussed next). The activated neurons in the striatum then send inhibitory signals to the globus pallidus (G), which send inhibitory signals to the thalamus (T). Neurons in the premotor cortex (M) are connected to both the thalamus and the PFC. Goal-directed behavior is modeled by the connections from the PFC to the striatum, through the globus pallidus and thalamus, and eventually score) or incorrect (negative), the dopaminergic signal ($D$) would moderate learning of the connections (Eqs (A8)–(A11)). This form of DA-mediated learning is shown to resemble the reinforcement learning process that is extensively studied in machine learning.\textsuperscript{40,46} With practice and directed feedback, the model learns to select the correct responses when external stimuli are perceived (and interpreted by the PFC). The repeated firing of the correct S–R pairs strengthens the connections between the PFC and the premotor area through Hebbian learning. With enough practice, the connections become strong enough that the correct responses can be directly selected using a random walk model\textsuperscript{47} when the associated stimuli are perceived at the PFC, by-passing the slower subcortical path through the basal ganglia. The model, therefore, characterizes skill learning through the shift from discrimination learning through the basal ganglia to direct activation of the S–R rules at the cortex (a similar process on perceptual learning was modeled by Ashby et al.\textsuperscript{29}). The computational model was implemented by differential equations (see Appendix) that simulate the activations of neurons in each brain structure shown in Figure 2. The Simulations To illustrate computationally, the general effects of VPT and FPT on performance and brain activation changes, it suffices to use an abstract representation of the SF game and neuron activations in each brain region. Specifically, there were 1000 possible stimuli and 1000 possible responses, each represented by a vector with length 100. The vectors represented by stimuli and responses were randomly generated by assigning either $-1$ or $1$ to each element of the vectors (but no two stimuli or responses were identical), and the correct response vectors were randomly paired with the stimulus vectors. These numbers were chosen such that learning the right responses required a significant amount of learning, reflecting the amount of learning required by participants to master the task. On the other hand, we verified that increasing these numbers only prolonged the training periods of the model without impacting the patterns of results reported in this article. In other words, the general computational characteristics of the model are not sensitive to the specific sizes of the stimuli and responses. In the simulation, each of the four subscores (velocity, speed, control, and points) was considered a task component. Each response was randomly assigned a score in each task component (ranged from $-4$ to $4$), such that the maximum total point for a response was 16 ($4 \times 4$) and the minimum is $-16$ ($-4 \times 4$). Each stimulus vector was directly fed to 100 neurons ($P$) in the PFC, which were fully connected to 100 neurons at the premotor cortex ($M$). The size of the striatum ($S$) varied from 20 to 100 neurons, each of which was connected to other structures ($G$ and $T$) as shown in Figure 2. On the basis of the diffusion model of response decision,\textsuperscript{47} when the integral of the difference between any two responses exceeded a threshold, the response with the largest activation would be selected (see Eq. (A6)). In addition to inputs from different regions, activations in each neuron were also decreased by two mechanisms (indicated by the negative terms, see Eqs (A1)–(A5) in Appendix): (a) lateral inhibition from neighboring neurons and (b) decay of activation over time. Discrimination learning occurred at the connections ($W$) between PFC ($P$) and the striatum ($S$) (Eq. (A7)). Discrimination learning depended on the strength of $P$, $S$, and the reward signal received ($D$). When the response was correct, the value of $D$ would be positive (Eq. (A10)); when the response was incorrect, the value of $D$ would be negative (Eq. (A11)). The connections that led to the correct response would then be reinforced based on the weighted sum of the task components (Eq. (A9)). In VP training, the weights would change across blocks; in FP training the weights were all set to 0.25. This process was based on the reinforcement learning process that was shown to reflect the reward-based learning process at the basal ganglia.\textsuperscript{40} Connections ($V$) between $P$ and $M$ were updated based on Hebbian learning mechanism (Eq. (A8)), in which the strength of the connection is strengthened by an amount proportional to $P \times M$. Training and Testing of the Model We randomly selected 500 stimuli for training in each session, and repeated the training for 20 sessions. In VP training, the weight for one task component was set to 0.85 and the rest set to 0.05 every 100 stimuli. We changed the parameter $\alpha$ to simulate the low ($\alpha = 0.01$) and high ($\alpha = 0.05$) performance groups, which controlled how fast it learns to select the correct responses. These values were chosen to best fit the general patterns of the data we collected from the experiment. Note that there was no significant difference in the wide set of cognitive abilities that were tested between the two groups. Thus, the use of different learning rates to simulate the two groups was based on the finding that only the initial training performance was different between the two groups. We discuss alternative possibilities of the difference of the two groups in the discussion section. Separate simulations were conducted for a low- and high-performance group in VPT and FPT, respectively. Figure 4 shows the mean proportion correct of the model during testing in each performance group in the VPT and FPT across 20 sessions. The simulation results show close resemblance to the empirical results by Boot et al.\textsuperscript{12} shown on the left. In particular, for the high-performance groups, there was virtually no difference between VPT and FPT. However, for the low-performance groups, VPT led to much better learning than FPT. In the model, the main reason why VPT was better than FPT in the low-performance group was that discrimination learning was more effective in VPT than FPT. Given that the learning rate ($\alpha$) was lower in the low-performance group, initial learning was equally slow in both VPT and FPT. However, ![Figure 4](image-url) FIGURE 4 \| Performance scores by human participants (left) and mean scores of 100 simulations of the model (right) across the 20 sessions. The model scores were normalized to 0 to 1. FP, fixed priority training; VP, varied priority training; L, low-performance group; H, high-performance group. because the weight given to the emphasized task component was higher in VPT, feedback received was more effective because the value of $D$ would be more distinctive when encoding the correctness of the response with respect to the emphasized task component (Eq. (A9)). In particular, when responses that were more important for a specific task component (e.g., a score of 4 in one component and 1 in the rest) and were emphasized in VPT, these responses would be reinforced more strongly in VPT than FPT (another way to look at it is that in FPT, learning from feedback was diluted across components). In VPT, feedback encoded by the dopaminergic signal is therefore more effective than in FPT to guide the model to strengthen the correct S–R pair through the feedback-driven reinforcement of the connections that activate the right set of neurons at the striatum, which eventually activate the correct responses at the premotor area. This subcortical pathway for response selection was then more effectively transferred to the cortical-cortical S–R rules through the Hebbian learning mechanism (Eq. (A8)) in VPT than FPT. The results were consistent with the findings that brain activation in the attentional control regions (i.e., PFC in the model) reduced significantly more in VPT than FPT. In the high-performance group, because the higher learning rate ($\alpha$) compensated for the diluted feedback received in FPT, the difference between VPT and FPT was reduced. As both groups achieved asymptotic performance, the difference between the two groups was not significant. However, further analysis did show that even in the high-performance group, participants in VPT seemed more effective in learning sophisticated strategies than FPT, suggesting that VPT would more likely induce optimization of strategies with respect to each task component, while in FPT, responses that were generally good across components were learned. Due to the space limitation, these analyses could not be included here. To simulate effects of size of striatum on learning, the number of neurons in the striatum was increased from 20 to 100. Figure 5 shows the correlations between the percentage increase in performance of the model and the size of the striatum in the VP and FP training conditions. Consistent with empirical results, the model shows that the size of the striatum is positively correlated with performance improvement in VPT more than in FPT. In the model, the main reason why the size of the striatum predicted performance improvement more in VPT was because discrimination learning was more efficient with more neurons in the striatum. This effect was apparent from the model’s perspective, considering the fact that the number of connections between the PFC and striatum would increase as the size of the striatum increased. A higher number of connections would naturally expand the capacity of the network to encode more S–R patterns. The interesting question is why the model showed a stronger correlation in VPT than FPT. This was again because discrimination learning was more effective in VPT than in FPT. When different task components were emphasized, learning responses that led to higher score to the task component would be learned more efficiently in VPT. On the other hand, in FPT, learning of actions would more likely be based on their total scores, rather than their specific effects on each task component (Eq. (A9)). The overall effect was that responses specific to certain task components were more likely encoded to different striatal neurons in the model in VPT (which would more likely lead to distinct responses selected), while FPT would more likely learn to select generally good responses. Because the granularity of the discrimination was higher in VPT, learning would more likely be limited by the number of neurons at the striatum than in FPT (i.e., mapping between stimuli and responses was more sensitive to whether it was correct with respect to each task component in VPT, thus practically creating another dimension in the mapping). This explained why performance improvement was more highly correlated with the size of the striatum in VPT than FPT. CONCLUSIONS AND DISCUSSION Consistent with previous results, VPT is found to be more effective than FPT for the acquisition of skills. Although there is some evidence that VPT enhances skill transfer to tasks that are functionally similar to some components of the trained tasks, broad transfer from trained to untrained tasks was not generally observed. Learners with low level of initial performance before training tend to benefit more from VPT. Results are consistent with the saturation hypothesis in learning, in which the relatively high level of cognitive load imposed by FPT or whole-task training may create extra obstacles for learning of complex skills, while VPT may allow learners to focus on specific components to improve individual components of the skills. To explain the effects of different training strategies, we developed a computational model of complex skill learning at the level of neural computations between the attentional control areas in the PFC, the basal ganglia, and the premotor cortex. The model successfully explained how VPT and FPT induced different discrimination learning at the converging connections between the PFC and the striatum, and how they eventually led to different effectiveness in overall learning. The model provided novel explanations to two major phenomena: (1) VPT benefits low-performance participants more than FPT, and (2) the size of the striatum is highly correlated with performance improvement in VPT but not in FPT. In VPT, experiences of how different subcomponents were dynamically related to each other were learned more effectively than in FPT. Under FPT, participants received feedback based on the total score that represented the sum of subcomponents; while under VPT, participants received feedback that emphasized individual subcomponents. This difference led to more distributed and effective encoding of cognitive representations of skills at the striatum, which led to better overall training effectiveness. In our model, we simulated the low and high-performance groups based on initial training performance (i.e., the learning rate). One may argue that differences in initial performance could be attributed to differences in preexisting ability in playing the game. However, we did not find significant differences in their general ability measures between the two groups, suggesting that the difference between the two groups lied at how well they learned during initial training. This is consistent with previous results\textsuperscript{11} and prior meta-analysis,\textsuperscript{48} which showed that performance gained during initial training predicted later training benefits. It is interesting to note, however, that, the current model predicts that initial training performance interacts with the training method \textit{and} preexisting capacity—i.e., a lower learning rate combined with VPT leads to better overall training, especially when the size of the striatum is large. Perhaps, because we are using a complex interactive task (as opposed to simple laboratory task such as the n-back memory task), learning multiple representations is more important for predicting training gain, and thus the size of the striatum is an important factor. On the other hand, a recent study has also found that reinforcement learning in the striatum can also explain why bilinguals have better executive function.\textsuperscript{49} Thus, it is also possible that a larger striatum may also lead to broader improvement in general cognitive functions, although we did not observe this difference in the experiments using the SF game. The model demonstrated that the preexisting volumes of the striatum predicted performance improvement as subjects learning a complex video game, and the predictive power of the size of the striatum was much stronger in VPT. The model also captured these relations by showing that a larger striatum could accommodate more distributed cognitive representations of skills experienced in VPT. In contrast, in FPT, discrimination learning would more likely select actions that were generally good across all task components, and did not require as many neurons to encode the mapping. Thus, the size of the striatum was not a limiting factor in FPT. The model thus provided an explanation based on the interactions between training procedures and the computational characteristics of brain structures. The explanation was consistent with previous hypothesis that VPT could enhance coordination and integration of cognitive, motor, and perceptual operations, and allow more development of more flexible cognitive representations of the task environment. If VPT is more effective than FPT for learning by capitalizing on the computational characteristics of basal ganglia-based circuits as a consequence, then this type of training could prove more useful for enhancing cognitive function in a number of applied settings. NOTE \textsuperscript{a}Although Gopher et al. found that training in SF game transferred to new tasks, the similarities of the SF game and the new task were not well controlled in the study. See discussions in Ref\textsuperscript{13}. REFERENCES 1. Donchin E, Fabiani M, Sanders A. The Learning Strategies Program: An Examination of the Strategies in Skill Acquisition. Amsterdam: Elsevier; 1989. 2. Fu W-T, Anderson JR. Dual learning processes in interactive skill acquisition. J Exp Psychol Appl 2008, 14: 179–191. 3. Fu W-T, Chin J, Dong W, Liao V. Interactive skills and dual learning processes. In: Seel NM, ed. Encyclopedia of the Sciences of Learning. Springer-Verlag; 2011, 1612–1615. 4. Salvucci D, Taatgen NA. The Multi-Tasking Mind. New York, NY: Oxford University Press; 2010. 5. Carrier CA, Davidson G, Williams M. Selection of instructional options in a computer-based coordinate concept lesson. Educ Commun Technol J 1985, 33: 199–212. 6. Gopher D. Emphasis change as a training protocol for high-demand tasks. In: Kramer AF, Wiegmann DA, Kirlik A, eds. Attention: From Theory to Practice. New York, NY: Oxford University Press; 2007, 209–224. 7. Hannafin MJ. Guidelines for using locus of instructional control in the design of computer-assisted instruction. J Instr Dev 1984, 7:6–10. 8. Schmidt RA, Bjork RA. New conceptualizations of practice: common principles in three paradigms suggest new concepts for training. Psychol Sci 1992, 3:207–217. 9. Green CS, Bavelier D. Action video game modifies visual selective attention. Nature 2003, 423:534–537. 10. Klingberg T. Training and plasticity of working memory. Trends Cogn Sci 2010, 14:317–324. 11. Jaeggi SM, Buschkuehl M, Jonides J, Shah P. Short- and long-term benefits of cognitive training. Proc Natl Acad Sci U S A 2011, 108: 10081–10086. 12. Boot WR, Basak C, Erickson KI, Neider M, Simons DJ, Fabiani M, Gratton G, Voss MW, Prakash R, Lee H, et al. Transfer of skill engendered by complex task training under conditions of variable priority. Acta Psychol 2010, 135:349–357. 13. Lee H, Boot W, Basak C, Voss M, Prakash R, Neider M, Erickson K, Simons D, Fabiani M, Gratton G, et al. Performance gains from directed training do not transfer to untrained tasks. Acta Psychol 2012, 139:146–158. 14. Lee H, Voss M, Prakash R, Boot W, Vo L, Basak C, VanPatter M, Gratton G, Fabiani M, Kramer AF. Videogame training strategy-induced change in brain function during a complex visuomotor task. Behav Brain Res 2012, 232:348–357. 15. Prakash RS, De Leon AA, Mourany I, Lee H, Voss MW, Boot WR, Basak C, Fabiani M, Gratton G, Kramer AF. Examining neural correlates of skill acquisition in a complex videogame training program. Front Hum Neurosci 2012, 6:115. 16. Mane A, Donchin, E. The space fortress game. Acta Psychol 1989, 71:17–22. 17. Gopher D, Weil M, Siegel D. Practice under changing priorities: an approach to the training of complex skills. Acta Psychol 1989, 71: 147–177. 18. Kramer AF, Larish JF, Strayer DL. Training for attentional control in dual task settings: a comparison of young and old adults. J Exp Psychol Appl 1995, 1:50–76. 19. Kramer AF, Larish JL, Weber TA, Bardell L. Training for executive control: task coordination strategies and aging. In: Gopher D, Koriat A, eds. Attention and performance XVII: Cognitive Regulation of Performance: Interaction of Theory and Application. Cambridge, MA: The MIT Press; 1999, 617–652. 20. Fabiani M, Buckley J, Gratton G, Coles MGH, Donchin E, Logie R. The training of complex task performance. Acta Psychol 1989, 71:259–299. 21. Stern Y, Blumen HM, Rich LW, Richards A, Herzberg G, Gopher D. Space Fortress game training and executive control in older adults: a pilot intervention. Aging Neuropsychol Cogn 2011, 18: 653–677. 22. Boor WR, Kramer AF, Simons DJ, Fabiani M, Gratton G. The effects of video game playing on attention, memory, and executive control. Acta Psychol 2008, 129:387–398. 23. Owen AM, Hampshire A, Grahn JA, Stenton R, Dajani S, Burns AS. Putting brain training to the test. Nature 2010, 465:775–778. 24. Kotovsky K, Hayes JR, Simon HA. Why are some problems hard? Evidence from the Tower of Hanoi. Cogn Psychol 1985, 17:248–294. 25. Rumelhart DE, Norman DA. Representation in memory. In: Atkinson RC, Heimstein RJ, Lindsey G, Luce RD, eds. Stevens’s Handbook of Experimental Psychology. 2nd ed. New York, NY: John Wiley & Sons; 1988, 511–587. 26. Anderson JR, Bothell D, Fincham J, Anderson AR, Poole B, Qin Y. Brain regions engaged by part- and whole-task performance in a video game: a model-based test of the decomposition hypothesis. J Cogn Neurosci 2011, 23:3983–3997. 27. Sweller J, van Merrienboer J, Paas F. Cognitive architecture and instructional design. Educ Psychol Rev 1998, 10:251–296. 28. Erickson KI, Boot WR, Basak C, Neider MB, Prakash RS, Voss MW, Graybiel AM, Simons DJ, Fabiani M, Gratton G, et al. Striatal volume predicts level of video game skill acquisition. Cereb Cortex 2010, 29:3:1–9. 29. Ashby FG, Ennis JM, Spiering BJ. A neurobiological theory of automaticity in perceptual categorization. Psychol Rev 2007, 114:632–636. 30. Wang H, Fan J. Human attentional networks: a connectionist model. J Cogn Neurosci 2007, 16: 1678–1689. 31. Wang H, Liu X, Fan J. Cognitive control in majority search: a computational modeling approach. Front Hum Neurosci 2011, 5:16. 32. Graybiel AM. Habits, rituals and the evaluative brain. Ann Rev Neurosci 2008, 31: 359–387. 33. Stocco A, Lebiere C, Anderson JR. Conditional routing of information to the cortex: a model of the basal ganglia’s role in cognitive coordination. Psychol Rev 2010, 117: 540–557. 34. Fitts PM, Posner MI. Human Performance. Oxford: Brooks/Cole; 1967. 35. Schneider W, Shiffrin RM. Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory. Psychol Rev 1977, 84:127–190. 36. Seger CA, Spiering BJ. A critical review of habit learning and the basal ganglia. Front Syst Neurosci 2011, 5:1–9. 37. Houk JC, Wise SP. Distributed modular architectures linking basal ganglia, cerebellum, and cerebral cortex: their role in planning and controlling action. Cereb Cortex 1995, 2:95–110. 38. Amos A. A computational model of information processing in the frontal cortex and basal ganglia. J Cogn Neurosci 2000, 12:505–519. 39. Kelly RM, Strick PL. Macro-architecture of basal ganglia loops with the cerebral cortex: use of rabies virus to reveal multisynaptic circuits. Prog Brain Res 2004, 143:449–459. 40. Fu W-T, Anderson JR. From recurrent choice to skill learning: a reinforcement learning model. J Exp Psychol Gen 2006, 135:184–206. 41. Schultz W, Dayan P, Montague PR. A neural substrate of prediction and reward. Science 1997, 275:1593–1599. 42. Ragozzino ME, Jih J, Tzavos A. Involvement of the dorsomedial striatum in behavioral flexibility: role of muscarinic cholinergic receptors. Brain Res 2002, 953: 205–214. 43. Meiran N, Friedman G, Yehene E. Parkinson’s disease is associated with goal setting deficits during task switching. Brain Cogn 2004, 54:260–262. 44. Dahlin E, Neely AS, Larsson A, Backman L, Nyberg L. Transfer of learning after updating training mediated by the striatum. Science 2008, 320: 1510–1512. 45. Koepf MJ, Gunn RN, Lawrence AD, Cunningham VJ, Dagher A, Jones T, Brooks DJ, Bench CJ, Grasby PM. Evidence for striatal dopamine release during a video game. Nature 1998, 393:266–268. 46. Sutton RS, Barto AG. Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press; 1998. 47. Ratcliff R. A theory of memory retrieval. Psychol Rev 1978, 85:59–108. 48. Colquitt JA, LePine JA, Noe RA. Toward an integrative theory of training motivation: a meta-analytic path analysis of 20 years of research. J Appl Psychol 2000, 44:117–186. 49. Stocco A, Yamasaki B, Natalenko R, Prat C. Bilingual brain training: a neurobiological framework of how bilingual experience improves executive function. Int J Bilingual 2012, 1–26. APPENDIX The model was implemented as a set of differential equations shown below. The equation for one neuron at each brain structure was shown below. $P$, prefrontal cortex; $S$, striatum; $G$, globus pallidus; $T$, thalamus; $M$, premotor; $D$, dopaminergic signal; $w$, weights between prefrontal and striatum; $v$, weights between prefrontal and premotor; $i$, input stimuli. $\alpha$, $\beta$, and $\gamma$ are free parameters that control the learning, and they were chosen to be equal (0.05) in each structure to minimize the number of parameters. The value of $D$ depends on whether the response is correct (positive score) or incorrect (negative score). See text for details. In Eq. (A1), the time derivative of the activation of a neuron in the PFC (left side of the equation) increases with input stimuli from the sensory cortex ($i$), its own activation ($P$), and activation from the thalamus ($T$). The $(1 - P_K)$ term is to account for saturation, such that the activation cannot exceed 1. The negative terms represent inhibition, which include self-inhibition ($P_K$) and lateral inhibition ($P_I$). In Eq. (A2), the time derivative of the activation of a striatum neuron increases as it receives excitatory inputs from the PFC ($P$), which are weighted by the massive number of convergent connections ($w$). Discriminatory learning occurs through the updating of these weights (Eq. (A7)). It has the similar saturation, self-inhibition, and lateral inhibition terms as in Eq. (A1). In Eq. (A3), the time derivative of the activation of the globus pallidus neuron receives inhibitory inputs from the striatum. It also has the self-inhibition and lateral inhibition terms. In Eq. (A4), the time derivative of the activation of the thalamus neuron receives excitatory inputs from the profrontal cortex ($P$) as well as self-excitation. It also receives inhibitory inputs from the striatum. It also has the self-inhibition and lateral inhibition terms. In Eq. (A5), the time derivative of the activation of a premotor neuron receives self-excitation ($M$), as well as the massive connections from the PFC ($v$). Hebbian learning occurs in these weights (Eq. (A8)). It has the same saturation, self-inhibition, and lateral inhibition terms. In Eq. (A6), the integral represents the accumulation of the differences between the activations of different premotor neurons. As soon as this integral exceeds a threshold, the action represented by the premotor neuron with the highest activation will be selected. In Eq. (A7), each weight is updated by the sum of activations in the prefrontal and striatal neurons. However, the update is also moderated by the DA signals, which depends on feedback in the environment (Eq. (A9)). It also has a saturation term $(1 - w)$ that prevents its value from exceeding 1. Learning rate is controlled by the parameter $\alpha$. In Eq. (A8), each weight is updated by the sum of activations in the prefrontal and premotor neurons. It also has a saturation term $(1 - v)$ that prevents its value from exceeding 1. Learning rate is controlled by the parameter $\alpha$. In Eqs (A9)–(A11), the value of the DA signal is assumed to varied depending on the feedback received from the task component. In fixed-priority training, the weight for each component will be equal, but in varied-priority training, one of them will be higher than the others (see text for details). \[ \frac{dP_K}{dt} = (i_K + \alpha \cdot P_K)(1 - P_K) \cdot T_K - \beta(P_K) - \gamma \cdot P_J \] (A1) \[ \frac{dS_K}{dt} = \left[ \sum_m w_{mK} P_l \right] (1 - S_K) - \beta(S_K) - \gamma \cdot S_J \] (A2) \[ \frac{dG_K}{dt} = -\alpha(S_K G_K) - \beta(G_K) \] (A3) \[ \frac{dT_K}{dt} = -\alpha(S_K T_K) - \beta(T_K) + \alpha(T_K P_K) \] (A4) \[ \frac{dM_K}{dt} = -\alpha(M_K) + \left[ \sum_m v_{mK} P_K \right] (1 - M_K) - \beta(M_K) - \gamma \cdot (M_J) \] (A5) \[ \Delta = \int (M_J - M_K) dt \] (A6) \[ w_{mK}(n + 1) = w_{mK}(n) + \alpha \sum_l P_m \sum_l S_K \cdot D_K \cdot (1 - w_{mK}(n)) \] (A7) \[ v_{mK}(n + 1) = v_{mK}(n) + \alpha \sum_l P_m \sum_l M_K \cdot (1 - v_{mK}(n)) \] (A8) \[ D_K = \sum_i \text{weight (task component)} \cdot D_K(\text{task component}) \] (A9) When response is correct: \[ D_K(\text{task component}) = 1 - \frac{e^{\frac{S_K}{T}}}{\sum_i e^{\frac{S_i}{T}}} \] (A10) \[ D_K(\text{task component}) = -\frac{e^{\frac{S_K}{T}}}{\sum_i e^{\frac{S_i}{T}}} \] (A11)
Wireless Weekly Friday, February 5, 1937 Vol. 29 No. 6 Registered at the G.P.O., Sydney, for Transmission by post as a Newspaper. John Dunne Manager of 2SM and "Hail-Fellow-Well-Met" of Sydney Broadcasting. MISS BERNARDINE KING - Los Angeles - Flew upside down 25 minutes, and 26 seconds... CAESAR WAS NEVER EMPEROR OF ROME... HE RULED AS DICTATOR MARY-ANN Hollywood movie elephant wears false teeth! Her tusks are artificial and worn only on the set. JOHN PAUL JONES WAS A NAVAL OFFICER FOR 4 DIFFERENT NATIONS AND REFUSED ADMIRALSHIP FROM ANOTHER... "STRANGE AS IT SEEMS" Oddities and Curiosities in Nature and Mankind Mondays to Fridays 8.45 p.m. 2GB "The Favourite Station" QUEEN WILHELMINA - Of Holland - Europe's only ruling Queen, has been on the throne longer than any living ruler... Her mother ruled before her and her daughter is sole heir to the throne! LAWRENCE TIBBETT STANDS ON HIS HEAD FOR SEVERAL MINUTES BEFORE EVERY APPEARANCE! CZAR PAUL - of Russia - Ordered that no one in the capital use the word for "snub-nosed" because they might be referring to him! SPRAGUE THE FINEST CONDENSERS AT ANY PRICE For many years, SPRAGUE Condensers have been the quality standard for the leading Radio Manufacturers and Servicemen. THEY ARE MADE RIGHT! THEY ARE PRICED RIGHT—and they are FULLY GUARANTEED by one of the World's largest Condenser Manufacturers. They are your assurance of long, trouble-free, radio receiver performance. SPRAGUE Condensers are made in the famous WET ELECTROLYTIC and TUBULAR types. Write for free Catalogue "March of Time." INTERNATIONAL RADIO CO. LTD. CHAS. E. FORREST, Managing Director 254 Castlereagh Street, Sydney 103 Bourke Street, Melbourne Grey Buildings, Courthouse Lane, Auckland, N.Z. 178 Rundle Street, Adelaide C.T.A. Building, Elizabeth Street, Brisbane YEARS OF FAITHFUL SERVICE The Fisk Radiola has been recognised as Australia's Finest Broadcast Receiver since the commencement of broadcasting in this country. The Radiola is designed, manufactured and guaranteed by Australia's national Wireless Organisation, and will give years of satisfactory service. High quality is combined with price economy and models for Australian reception can be obtained from 15 guineas, and for overseas' reception from 19 guineas. Distributors everywhere will gladly demonstrate and explain a convenient deferred payment system. THE FISK RADIOLA MANUFACTURED AND GUARANTEED BY AMALGAMATED WIRELESS (A/SIA) LIMITED Yours Leonard Bennet Alan Reavey It happens to Rion Voigt exactly as you see it pictured across the top of these pages. Much too often, Rion says, for the general comfort of ringsiders at the Leichhardt Stadium, where he has been describing wrestling and boxing bouts for several years. Sometimes, the wrestlers, as a result of holds like those Tony Lomaro and Mike McGill illustrate on the lower portion of this page, will spin out of the ring and land their fifteen or twenty stone on top of the announcer. On several occasions wrestlers who thought Rion wasn't giving them a fair go did what you see Tony doing in a friendly way at the bottom of the opposite page. TO ANNOUNCER AT THE STADIUM Leonard Bennett and the "Compagnie de Quinze" LISTENERS to the B.S.A. Players who have heard Leonard Bennett in many outstanding parts, would not guess that this fluent actor is equally at home in French drama. In fact, Leonard Bennett is a member of the Compagnie de Quinze, so named because there were fifteen players in the original cast, all of them pupils of Jacques Copeau, founder of the famous French theatre which is known as "The Theatre de Vieux Colombier." The theatre was founded in 1913, but a year later with the coming of the War, it was closed down. However, it was reformed in America under the auspices of the French Government for propaganda purposes round about 1916, when it was financed by Otto Kahn, the millionaire art-lover. "Incidentally," says Leonard Bennett, "it was the inspiration of this famous Company that led to the formation of the even more famous New York Theatre Guild." After the war, however, Monsieur Copeau decided that the routine of a repertory theatre was cramping his style, so with twenty of his favorite pupils, he moved to a village in Burgundy with the unpronounceable name of Pernand-Vergelesses. It was at this period that Leonard Bennett joined the Company and he says he always feels proud that he has the distinction of being one of the very few foreign pupils accepted by the master. "It meant hard work," continues Leonard Bennett, "We were made to work 12 hours a day so that we might learn the rudiments of the comedian's art. We took a course in gymnastics, had lessons in elementary ballet, music, mask-making and pantomime (that is to say, the art of expressing oneself by action instead of words.)" "Our first material was drawn from the folk-songs and stories of the Cote d'Or. This is that celebrated part of Burgundy which produces such famous wines as Clos-Vougeot, Hospice de Baune and Pommard, and our songs and stories were those sung and told and handed down from generation to generation by the vignerons who worked in those famous vineyards." The cottage where Leonard Bennett stayed during this period was close to the "Vineyard of Charlemagne." It is famous for a very fine white wine and it is said that the vineyard was originally planted by Charlemagne himself. "I cannot answer for the truth of this," says Leonard Bennett, "I only know that the wine was very good." The training continued for 12 months when Monsieur Copeau decided that the company was fit to tour, Holland, Switzerland and Belgium were visited with a presentation called "La Danse de Ville et des Champs." "After this tour," continues this talented young actor, "we were considered by our master to be far enough advanced to work by ourselves and the following year in Versailles and Geneva, we presented the two plays for which "La Compagnie de Quinze" became famous, "Le Viol de Lucrece" based on Shakespeare's famous poem, and "Noe," a brilliant portrayal of the famous flood story from the Bible. Both these plays were later presented by La Compagnie de Quinze in London, but, unfortunately, for Leonard Bennett, a law was brought in by the French Government during the depression forbidding the payment of salaries to foreign artists on French soil. This prevented him from continuing with the company to Paris, and he very sadly returned to England. "It was good experience," he says, and the most thorough training any dramatic artist is likely to get. Furthermore, I met many celebrated people such as Andre Gide, the great French novelist, Jean Cocteau, the famous playwright and Stravinsky, whose ballet 'Petrouchka' was recently heard in Sydney. Receiving an offer from Sybil Thorndike I joined her Company at Port Said for an Australian tour." But bad luck still dogged Leonard Bennett. Taken ill in Sydney, he was forced to abandon the tour and return to England. The doctor, however, ordered him to a warm climate, so back he came to Australia, this time to join 2GB and the B.S.A. Players. Speaking of Jean Cocteau, Leonard Bennett, and one of his colleagues on 2GB, hope during 1937 to collaborate in two stage productions. "That is if our work in radio permits," adds Leonard Bennett. "We shall probably choose 'Orphee' and 'La Machine a Ferille,' two brilliant works by this famous French dramatist." The GIBB Family THE odds are ten-to-nothing that the topic of conversation around the Brunton Gibb meal table is BROADCASTING, for every member of the family down to 9-year-old Barbara, is an old hand at facing the mike. Peter, 12, and Wendy, 11, are the star players of "Tales Told to Peter and Pam," 2GB, Monday and Wednesday at 7.45 p.m. and Saturday at 7.55 p.m. Peter has to his credit title roles in "Oliver Twist" with the A.B.C. and "David Copperfield" with the George Edwards Players (parts taken by Freddie Bartholomew in the screen versions), and with sister Wendy, played in the A.B.C. Children's serial, "Rosie, the Babe of the Circus." But Peter's private ambitions are definitely apart from broadcasting fame. He is mechanically inclined, has built his own radio set, and repaired with complete success the sets of neighbors. Wendy made her radio debut at 4 years of age in a song scene written by Wilfred Thomas. She spoke a dozen lines sitting on her father's knee. Which brings us to Barbara, who, at the age of 6, assisted in the 2FC Children's session and since has had minor roles. Mrs. Brunton Gibb and her children, Peter and Barbara. Mr. Brunton Gibb Broadcasters The trio laugh at the very idea of mike fright, but confess to being somewhat bored by rehearsals. Mrs. Brunton Gibb is best-known as Ethel Laing of 2SM (Hazel in “One Man’s Family”), and 2GB fame. Brunton Gibb himself, apparently deciding that you can have too much of a good thing, has practically forsaken broadcasting to concentrate on imparting the art of elocution. Wendy and Peter pose for the camera in unorthodox guises. No wonder Sir Harry Lauder is so pleased with the microphone. A report says the A.B.C. engaged him at a fee of £500 to broadcast from the national stations on Monday night, January 25. This figure has not been confirmed either by Mr. Cleary or the canny Sir Harry. However, it was good publicity for the programme which was looked forward to by thousands of Scots throughout Australia. Sir Harry Lauder, who is 66, is here on a private visit to his brother, who lives at Stockton, Newcastle. Above are THE RAGGLE TAGGLE GYPSIES, broadcasting weekly through 2FC, commencing February 12. On the right is THE ELIZABETHAN QUARTET, which has been broadcasting from Hobart and Sydney. The four girls, Jean Beddoe (pianist), Joy Tulloh (violinist), Marion Ogilvie (viola player) and Greta Vincent (cellist) are Melbourne musicians. They Called Him: Few people realise the most popular play dealing with the Great War, "Journey's End," was produced in nearly every capital city in Europe by an Australian—Tom Hudson, who is one of the newest additions to 2UE’s announcing staff. Tom Hudson’s father, one of South Australia’s theatrical pioneers, built the Tivoli Theatre in Adelaide. To-day Tom Hudson is back in Australia after twenty-five years. In that time, he has worked as stage director and producer to most of London’s leading managers, including Lewis Casson; (Dame Sybil Thorndike’s husband); Leon M. Lion, Archie de Bear, Basil Dean and in latter years, Maurice Browne of “Journey’s End” fame, “one of the greatest bosses and pals a stage producer ever had.” During his association with Dame Sybil Thorndike, Tom Hudson stage-managed Bernard Shaw’s famous “St. Joan” at the New Theatre, London, and he has vivid recollections of the great G.B.S., sitting in the dress circle of that theatre during rehearsals, with his famous leather-bound and massive note-book, brooking interference from no-one, and carefully recording every move and gesture of the actors on the stage. Nothing escaped the notice of this vegetarian octogenarian. ● TOM HUDSON, latest addition to 2UE’s announcing staff, is by his own account the off-stage hero of “Journey’s End,” the record-breaking play by R.C. Sheriff. “TOMMY with the BOMBS” R. C. Sheriff’s famous war drama, “Journey’s End” was originally produced by James Whale for the Incorporated Stage Society at the Appollo Theatre in London for a trial Sunday night performance. But the night “Journey’s End” was produced theatrical history was made. Hollywood promptly secured Jimmy Whale and very few Sydney folk realise that it is thanks to “Journey’s End” that the Liberty Theatre looks like breaking all records with “Show Boat.” For this is a James Whale production. Many fortunes were made (and lost again) from “Journey’s End.” Its author, Bob Sheriff, must be an exceedingly wealthy man to-day. Yet, success did not turn the head of this £6 per week ex-insurance clerk one iota. At the height of the London run, Tom Hudson had a map of the world hung outside the theatre and tiny lights showed all the cities at which “Journey’s End” curtain would rise at approximately 8 o’clock that night. At its peak period, thirty-three lights dotted the map, including Sydney, and when one realises that Bob Sheriff’s cheque each week for nearly two years from the Savoy Theatre alone was over £300 in royalties, well work the figures out for yourself. “When the continental productions began,” says Tom Hudson, “my one anxiety was to see that the correct portrayal of a British soldier in uniform was given. No easy task, when the continental habit of over-gesticulation is re-called. I made so many trips between London and the Continent in 1929, that I became known, at all the frontiers, as ‘Tommy with the Bombs,’ for on each occasion I left with a complete ‘Journey’s End’ outfit, including rifles, revolvers, gas masks and electrical timing bombs for the explosion effects in the play. These bombs would put the “wind-up” the customs officers. They were perfectly harmless until electrically connected.” “My first trip was anything but smooth going, however, for in spite of being laden with all the necessary documents from foreign consuls to get me through with my precious cargo, I might easily have been detained as an “agent provocateur.” I rushed from one capital to another, visiting Paris, Stockholm, Copenhagen, Berlin, Budapest and even as far north as Helsingfors in the Arctic Circle.” “Wherever I went I was received with the utmost hospitality. At the Grand Hotel Stockholm, I was given a most palatial suite, and like all hotels in Sweden was excessively overheated by steam. I contracted the usual visitor’s chill the next morning. On retiring to bed that night, I found a bottle of whisky and two lemons by my bed-side. Yet in that year, Sweden was a prohibition country! “In spite of Sweden and Finland being neutral countries during the Great War, “Journey’s End” appealed to them just as strongly as any combatant country.” Tom Hudson recalls with amusement, a photograph that was given quite an amount of publicity in London at the time “Journey’s End” was being played. In Berlin, the German actor, who played the school-master officer Osborne, insisted on being photographed with an eye-glass to which was attached a one-inch-wide piece of ribbon fastened to his tunic button. He was quite hurt when informed that not even an English school-master would wear such a thing, particularly in the front line. But to pacify him a negative was taken although he did not appear on the stage with this diplomatic appendage. Tom Hudson considers that of all the Continental people he visited, the Swedes are easily the most hospitable and charming race to be found anywhere. LIKES THE MUSIC CRITICISMS Dear Sir,—I am renewing my subscription for the half-year, and am including a note of appreciation of your articles. I have been taking your paper for well over 10 years, and have followed your change in size from the little first ones to the large copy, and now your rotogravure numbers. I have appreciated the following: Keith Barry, Discobolus—Robert McCall, Prerauer, whose criticisms are a source of pleasure to me. I enjoy James, and was very sorry when his articles stopped. The technical articles by Hull and Moyle are, of course, first class. I built many sets in the early days of 2BL and 2FC in Sydney, and I had many thrills from early interstate reception at Toowoomba. I had also had many headaches from the earphones. I ceased constructing for the last few years, but hope to take it up again. One item of your paper I have missed, and that is "the best music of the week." "Curves Ahead" gave much pleasure, but it is enough to say that the paper has been a pleasure to read for the years that I have got it. Yours, etc., F. B. PERRY. Gympie, Qld. (10/1/37). RECEIVERS WRONG NOT TRANSMITTERS Dear Sir,—I note with great interest Mr. P. Lumsden's meanderings (W.W., 15/1/37) on the subject of recorded commercial announcements. Evidently our technical friend believes that the said meanderings are calculated to bring forth humility on the part of the recording engineers, whose inability (?) to cope with the Stroboscope causes much acceleration on the play-back. As a matter of fact, very few engineers rely on a Stroboscope, as it is generally as unreliable as Mr. Lumsden's statements. Again, it is very seldom that recorded advertisements are used, except on transcriptions which are circulated from station to station, thereby being badly damaged. Might I suggest to this gentleman that the cut frequencies are a result of his receiver's inability to cope with any high fidelity output. In conclusion I would mention that a Stylus is an exceedingly less obtuse article than Mr. Lumsden's self and that he limit his remarks to subjects of which he knows a great deal more than of recording.—Yours, etc... "RECORDER." Sydney (13/1/37). CRICKET AT CANBERRA The Canberra commercial station 2CA will give a ball-to-ball description of the match England v. New South Wales Country Eleven on February 10 and 11 at the Manuka Oval. The hut given by the Olympic Games Committee to the Commonwealth Government, and placed by them on the Manuka Oval, will serve a useful purpose on this occasion, the P.M.G. Department having converted it for use as a telegraph despatch station during the match. A battery of instruments is to be installed for the use of the large number of pressmen who will be in attendance at the match. Overseas journalists will make use of the facilities to transmit news as to the progress of the match to a very large group of metropolitan and provincial newspapers in the United Kingdom. ARE YOU A MUSIC LOVER? YOUR OPINION Dear Sir,—In recent months I have been somewhat disappointed in "W.W." My "complaint" is due to the absence of such musical items as Mr. McCall used to be responsible for, viz., the summary of the week's best music and the criticism of the monthly record issues. I should think, that there is no doubt that music is the most important part of radio programmes, and those listeners interested in the best that music has to offer us are undoubtedly increasing very steadily, perhaps rapidly, but they are only too poorly catered for by Australian publications. "W.W." did show signs of developing along lines which would have helped to fill that vacancy, but, with the passing to another sphere of Mr. McCall, all that it now retains is Mr. Prerauer's weekly comment. The latter is an admirable item in your publication, and forms the most interesting and stimulating commentary available to us. It always claims my prompt attention. The suggestion I wish to make is that "W.W." should again let us have features similar to Mr. McCall's previous efforts. Mr. Prerauer is, refreshingly, not afraid to praise or to criticise severely, at times, and this is what is wanted. If he (for preference) could deal with the monthly record lists in his usual critical manner, I am sure readers would give his comments a warm welcome. If, in addition, he could give the "once-over" and a few notations to a summary of the week's best music, recorded and otherwise, I feel that readers would double their welcome. Music-lovers who are listeners—and who among them is not?—must, if they are to get what they want, study programmes carefully, and, therefore, this class must form a substantial part of "W.W.'s" readers, and regular readers, too. Are they not, therefore, deserving of some further attention and encouragement, to confirm those among them who are already readers and to attract others. Music-lovers would then be fairly well catered for, with the programmes of the best of both recorded and studio music, and Mr. Prerauer's (I hope) criticism of the studio performances as they are broadcast (which he already gives us), and of recorded music as it is issued. I feel sure that very many of your readers would agree with my suggestion, in its general import at least, and I hope that you can see your way to give this letter favorable consideration. Yours, etc., J. E. M. DIXON. Wahroonga (15/1/37). "—and the ball is going straight to the fence" Seething with interest and full of surprises, cricket is like the Radioplayer which brings it into your home — the apex of exciting entertainment. Made by the makers of the famous Philips Lamps, each Radioplayer represents true value for every penny spent. Ask your Accredited Radioplayer retailer to demonstrate these amazing musical instruments. BROADCAST MODELS A.C. OPERATED Model 6506, 5 Valves 15 gns. Model 6507, 5 Valves 19 gns. Model 6500, 5 Valves 23 gns. WORLD-WIDE MODELS A.C. OPERATED Model 6608, 6 Valves 27 gns. Model 6709, 7 Valves 34 gns. BATTERY-OPERATED RECEIVERS Model 6517, for Broadcast Reception, 5 Valves — 27 gns. Model 6612, for World-wide Reception, 6 Valves — 37 gns. TERMS AVAILABLE PHILIPS radioplayers MADE BY THE MANUFACTURERS OF THE FAMOUS PHILIPS LAMPS THE MUSIC CRITIC WHAT WE SHOULD LIKE TO HAVE ON various occasions in these columns I have attacked artists for their interpretations. I think it is about time that I gave my readers an idea of what I mean. The other day I had an interesting experience in listening to a record of a world-famous artist. Two friends and I were also listening, told me they did not care much for the record, and I pointed out to them that when they felt vaguely, I could express distinctly by saying that in the interpretation there were neither more nor shorter than one crotchet; which is wrong. Why does an audience prefer one artist to another? It is because they have faith in the interpretation of the lesser artist which makes the layman weary. The professional will be able to point out why. There are many people in Australia who know nothing of them professional, and also some ardent listeners who are living in a young country, their opinion may not get so strong avialt that it may render impossible the task of putting over certain "readings" which would be common down under the older countries. For example, I found out that in a performance of the fifth symphony of Beethoven, some years ago, a famous Australian conductor was beating the first bar in four broad beats instead of one quick beat. The first bar is an old rhythm. It is not easy to get all the instruments to play together. The solution is to do exactly what Stravinsky did in his "Rite of Spring," give one beat but in advance, with such a movement that the audience does not notice it. At the same time the orchestra gets the definite idea that the tempo must be played in. Not to know this is, as far as the fifth symphony is concerned, an omission which ought not to happen without being challenged, even 11,000 miles away from London. The tempi which singers take generally in Strauss lieder, especially in "Morgen," are by far too slow. It is necessary to know that the language of Mozart and Beethoven is comparable to an Andante or Allegretto of other composers. To take a breath in lieder by Wagner or German language seems to require it (un which people wrongly believe that they must sing these pieces) is wrong. Brahms' declamation goes very often against the language. Brahms' according to his language, therefore, damages the musical phrase which is still the most important thing to be observed. I have heard from Miss Burt (voi la sappe!) where he did not preserve that to sacrilege. On the other hand, I notice sometimes that artists are inclined to give simply the written music without any feeling at all. This is as bad as the mistake which I pointed out above. One gets the wrong interpretation when one neglects the inner meaning of a phrase, according to the expression which the composer has accorded to the laws of dynamics, in parallel to other works of the same composer, to the historical growing of style, and so on. This is much more difficult than simple enunciation. For example, I noticed that no performance of the Preussische-Overture I have heard yet in Australia, except that of the Berlin Philharmonic, discoveries of Wagner, who wrote at length about it. I am aware that the things I am saying here cannot be accepted in a minute. However, I hope that my readers will agree with me that it is necessary to mention them. And I shall do so again. SYMPHONY HOUR 2FC, January 24 THE A.B.C. Orchestra, under Mr. Percy Code, sounded very good to-night. The programme started with Schumann's fourth symphony. The annotations did not have anything to be desired. The symphony hour is always now at 8.0, and I must say I appreciated that the old mistake has been made. One is freer to appreciate music at 8.0 than at 9.30. The symphony was played very well. The interpretation was good, though the characteristic jump of one octave in the scherzo might have been heavier, and the second subject, the last movement, lighter and more amiable, but otherwise I enjoyed the performance. Listening to Constance Burt left me again wondering why this sweet, charming voice should not be controlled better in its technical requirements. One felt that it sounded often rather amateurly, indicating a lack of breath control, and sometimes bordering on becoming flat, especially in the longer aria. I am under the impression that Miss Burt does not support her voice sufficiently with her diaphragm. Vo! she sapped it was not singing with an expression of joyousness but who wanted to make love to everybody, but sounded a trifle peevish. The ritardando at the end of each verse was added to Miss Burt's Miss Burt. If it would have been smaller, nothing much could be said against it. The aria was always extended over the bars. "Laimen" was not what we are used to consider as Mozart style. There was hardly anything but the music. I waited always for some more expression, but there was only the sweetness of the voice. The section in sicario ought to be livelier than the other sections, according to the style of the opera seria. And what about Miss Burt to sing the two dotted semiquavers before the end? There was a little in the aria? If so, let us a major sin. Miss Dawson played the solo violin with greatest expression. The final item was the second "Wand der Wunder" suite, which was well played by the quintette, though sometimes wanting for more transparent lightness and amiability of expression. The basses in "The Bears" could have been more prominent. RECITAL, ARVED KURTZ 2BL, January 25 HAVING talked a lot about style now, I am glad to say that the style in which Mr. Kurtz played the Vivaldi Concerto in G Minor was of flawless perfection. Old music asks for a sense of restraint, and the part of the performer evokes the right amount of it. It was noticeable to-night, and yet there was a lot of feeling in it. SYDNEY STRING QUARTET 2BL, January 25 RUTH PEARCE-JONES sang two brackets of songs. She has not yet overcome a certain timidity of colour, and her top notes have to be developed further. But this is the accompanist knew what to do with Strauss's "Chaconne." It calls for more expansion and more legato. Williamson would be an excellent song writer if he would discover his own personality instead of simply surrendering to the influence of German music. Singing by Rubbra was interesting. However good the singer's German may be, she ought to sing in English. The programme of the quartet had to be cut short. The items cut out were supposed to be Beethoven's string quartets. The latter seems to be much more important than that the A.B.C. could allow a string quartet to build up by playing a number of variations standards. The Bach Quartet, written 1912, is an amazing work. The slow movement of the work, which is decidedly Bachian, shows Elgar's influence. The last movement, which is an essentially Gaelic folk songs and dances, is simply wonderful in its combination its temperament, and its humour. The performance was good, except in this last movement, which was not always clear. Messrs. White and Miller played with a better tone than Messrs. Krasnik and Ashton. SHORT NOTES 2BL January 27 MR. MURISON BOURN, conducting the A.B.C. String orchestra, confirmed again the impression that he is not a polished and competent musician, though not a strong personality. The Vivaldi concerto in G Minor was played by Miss Joyce Leighton. I suppose, the name was not given. The variations by Hardee Wilson are an excellent and interesting work, with help in the way Max Regger used to write variations. Miss Constance Cattafio, failed completely to impress me, either by the quality of her voice or by her interpretation. The second recital of contemporary compositions, among works by Scriabine, who is not contemporary, because he died in 1915, 22 years ago. Paul Vinogradov was the excellent pianist who proved thorough understanding. Ha Turnbull sang a bracket of British songs, one of them, "Inthus," not being contemporary music. Her voice was extremely interesting, and she sang with her accustomed intelligence and feeling, though the voice itself did not strike me as being particularly attractive. —CURT PRERAUER "ALL IS NOT GOLD THAT GLITTERS" It sounds attractive to be able to buy a radio set for country use that does not require B batteries to operate. But all is not gold that glitters. You may save the cost of dry batteries. BUT . . . • • • The initial cost of such a set is higher. • • • You have other units to replace, just as you replace batteries. • • • You have to purchase a bulky six-volt accumulator instead of an economical two-volt. • • • And this has to be recharged more frequently. In other words, you may save with one hand but throw away money with the other. Investigate fully any claims you may hear before you buy your new radio, and you'll find that there is NO PROVEN SUBSTITUTE for the true, tried and tested B-BATTERY EQUIPPED SET! An advertisement issued by Ever-Ready Co. (Aust.) Ltd., Sydney, in the interests of more economical 'out-back' radio. E.X.7 EVERY SET IS A BETTER SET POWERED WITH EVER READY RADIO BATTERIES BETWEEN YOU AND ME AND THE MICROPHONE Troubadour I was not surprised, last Monday night, to find myself listening to Sir Harry Lauder's £625 hour. For one thing, being a good deal Scotch myself, I didn't like to see all that money going to waste; and then there had been so much brilliantly worked publicity over how much they were paying him. One rumor I heard was that the Commission was not paying him £625 to do the hour, but as a bribe not to do any more than an hour. But one does not have to believe everything one hears, does one? And then, his voice was one of the first voices I ever heard on a gramophone, and not even a gramophone, but a phonograph which had been stowed away in a Scotch pantry: the records were most of Harry Lauder's then extant, and "Flow Gently, Sweet Afton," and "Rock of Ages Cleft for Me"; and it all came back to me, the home of these Scotch friends of ours, the pantry full of details, in the discard, or, rather, I think it was a linen room; the garden full of flowers, the orchard full of plums and apples; the Scotch cooking imported by the lady of the house, the plum sauce, the apple chutney, the shortbread, etc., etc., etc. They was very nice people. Well, I heard this Harry Lauder session; one song I missed was "Courting Jennie Lindsay"; I think that was the title; and I said to myself, now what makes the old boy worth £625? First, of course, it was because he was Harry Lauder, the war-time entertainer, and, second, because he would appear just the once, and third because he is an artist; but, altogether, I thought it was because he was a troubadour; which is making a certain statement about the state at which the art of the troubadour has arrived. Yet such arts are like religion and witchcraft and sculpture; they never snuff out, but persist in strange and unexpected forms. Observe, therefore, that in the days of the troubadour in Provence, on all of which I am pretty hazy, but you can look it up in the Essay by Ezra Pound, on whom be W. B. Yeats's blessing and peace, a troubadour used to write his words out, and sometimes he sang them himself, and other times he had a jongleur to sing them for him; from place to place as he went he was entertained in state by the lords—and the ladies, and sometimes he was a wealthy man, and sometimes he wasn't and sometimes he was better entertained than other times: sometimes he got £625 an hour, and other times he didn't. Then they used to have Tournaments of Song, in which the troubadours competed, and part of the competition was for one man to make up a set of rhythms and rhymes for the other to follow immediately impromptu; but this has nothing to do with Sir Harry, except that he writes his own songs and sings them himself, is very much in demand because of them, and if you want to be smart and say that he uses the gramophone for a jongleur, I won't make any objections. The other day I saw another essay of Ezra Pound's in "The Listener," the B.B.C. paper, in which he pointed out that the art of matching words to music so that an ordinary man could enjoy the song had gone from the lighkliars writers and composers to the music-halls, cabarets and music-modes; which is no doubt a reason why the lovely American critics like Kascoe don't like Ezra Pound any more; but I contend you can always tell a boob by hearing him quote from "Esquire," which advertises itself as a habit-forming magazine; so if you want to be a Hart Schaffner and Marx writer in a yellow shirt with a pink tie, you know where to go for opinions. So it is not so far out to say that Sir Harry Lauder (they used to knight them in the old days, too) is an example of the troubadour in our times; that is, he is not of these very times, but pre-war; and his songs, like "Roaming in the Glomming" and "Jubilee," are really very good songs, because you can understand them, and they have something in them that is peculiar to Harry Lauder, and the words are properly matched to the music, and he sings them very well, and he is not like the vaudeville lady who said to one of Australia's foremost ballad writers, "Why don't you write me a song? I'll pay you for it. I sometimes give as much as thirty shillings for a good song." She was doing him the honor, you see; which amounts to uppishness in jongleurs; and it is some satisfaction to see a man who is proud of writing his own songs getting away with £625; you feel that that is as it should be. It pleases me to see the troubadours getting away with it. It's something about the same in Hungary, Sally says. The gypsies come into a village and start playing, and the people all come out of their houses and dance and sing, and the other gipsies sneak in the houses and take all the valuables; or, says Sally, "They see a man and girl in a cafe, and they come over to your table and play to you, and you just feel you could give them anything—" "Anything," Sally says, looking out the window, "Everything you've got." "Absolutely anything at all." Times when she tells me this, which is almost every time the wireless plays Liszt's Hungarian rhapsodies or any Hungarian gipsy music, I sort of congratulate myself she has never been in Hungary, and make a resolution that she must never go there. Any thing. On the other hand, it is not so good when the troubadour begins to see himself as The Light of the World, or The Licht o' the Wurruld; and the fifteen or twenty minutes we spent on Monday night hearing how it was Sir Harry's mission in life to cheer the little children and comfort the distressed, interspersed with illustrations of how Sir Harry could sing Irish songs—unaccompanied—how do they get such notions? Do they get them from touring the same circuits as evangelists and reformers? Or is it just one of the results of growing old and wealthy? Anyhow, the fifteen or twenty minutes we spent listening through these inartistic noises were perhaps the greatest tribute we could pay to a very great artist. HOWLERS & HANDCLAPS WEEK'S BEST Miss Constance Duncan, 2FC (9.22 p.m., January 12): As you know, the Chinese have always been closely associated with China. Sent in by Ulick F. S. King, 19 Moseley Street, Strathfield, who wins this week's prize of 10/-. Cricket announcer, 4QG (5.12 p.m., Jan. 18): And now Pearson comes in blowing—er—bowling to Taitton. "The Captain," 2GB (5.30 p.m., Jan. 18): It is noticed that cuckoos which are hatched in other birds' necks—er—nests. Speaker, after fight, 2BL (9.45 p.m., Jan. 19): The issue was never in—er—was always in doubt. Mr. Huck Finlay, 2BL (9.5 p.m., Jan. 19), at the Sports Ground during boxing match: Leads with his body to Blatch's body. Howlers and Handclaps are humorous mistakes, slips, mispronunciations, juxtapositions, wisecracks, witticisms, word tangles, and anything heard over the air which shocks or amuses the listener by its spontaneous or impromptu quality. Howlers should be written down in exactly the same way as they are shown on this page. The name of the station, the time, the date, and, where possible, the name of the artist or announcer are essential. The sender's name and address should be neatly written below the howler or handclap. One shilling is paid for every one published. There is no limit to the number which may be sent in. Address your entries to "Howlers," Box 3366PP, Sydney. Announcer, 2BL (6.40 p.m., Jan. 19), on the Spanish front: Bullets were fired over the heads and hoses were played on bread queues because they refused to disburse. Harry Solomons, 2UE racing session (2.40 p.m., Jan. 20): Cigarette should be a good thing if you have enough to back 'er. Harry Solomons, 2UE (4.10 p.m., Jan. 20), Victoria Park: The last race starts at 4.25 and finishes a little earlier. Bert Lewis, 2GN Goulburn (9.20 p.m., Jan. 20), advertising duco-sprayer: Even if you only want your mudguards or some parts of your body touched up with duco, go to —. Harry Dearth, 2GB (9.15 p.m., Jan. 20): The appearance of your car may be marred by spots; but Y—— cleanser will make it disappear right away. Mr. Lyle, 2KO (10.15 a.m., Jan. 20), giving information about famous actor: David Garrick, married in 1799, of course that was before he died. Announcer, 2KY (9.5 a.m., Jan. 21): —and I hope that to-day will be very much brighter and happier than what it was yesterday. Announcer, 2UW (8.45 p.m., Jan. 21): These ice chests of C—— have been before the public for twenty years. Aunty May, 2UE (10.5 a.m., Jan. 22), speaking about afternoon to be held: Mrs. X—— has quite a lot of little things in her head. Announcer, 2WG (8.30 p.m., Jan. 22), advertising chain stores: Come in and have a look around, and you won't be tempted to buy. Announcer, 2KO (8.45 p.m., Jan. 22): Now the time from 2KO isn't fifteen minutes to nine. Cricket commentator, 5CL (5.47 p.m., Jan. 23): The crowd here is very concerned to see Bradman is not himself. Gwen Laws, 2KY (11.10 p.m., Jan. 23): The sheep are very tame in Hyde Park; they will even let you eat out of their hand. Cricket commentator, 2BL (12.55 p.m., Jan. 23): Beattie stood straight up and took that ball off his chest. Cricket commentator (1.14 p.m., Jan. 23), 2BL Sheffield Shield match: The umpire definitely gave him out l.b.w., then called him back and changed his name—I mean his mind. The most popular of all Wireless Weekly Circuits Dual Wave Pentagrid Six Complete Kit of Parts . . . . . £16/10 Including Valves, Speaker and all Batteries This Battery-operated Dual Wave is offered at an astounding low price for such quality! This set has everything you could wish for in a high-grade radio—exceptional selectivity, a tonal purity that's second to none, worldwide reception. Features of construction include full automatic volume control. When the set has been wired you can have it aligned on the oscillator free of charge! Complete Kit of Parts is £16/10/. Should you prefer to install the set in your own cabinet, we will supply it ready wired and tested—with all batteries and equipment for £18. The same Set ready assembled in Handsome Cabinet for £21/10/- The lowest price we've ever asked for a 6-valve dual wave battery set! Complete with all batteries, equipment, aerial, earth, tested and in a very attractive Cabinet, for £21/10/- EASY TERMS Only £3 deposit and small weekly payments will secure this splendid radio for you! TWO FREE OFFERS DURING FEBRUARY! - Special Offer! We will pay the freight on any Kit Set purchased during February, to any part of the State! Save on freight charges and get your Kit Set now! - Free 30-amp. hour accumulator will be given free with every Kit Set bought during February! With a spare battery you can still have all the radio entertainment you want, while your other accumulator is being charged. Order before February 28th to share in this splendid offer! DAVID JONES' RADIO HALL MEN'S MODERN STORE, GEORGE ST., Opp. G.P.O., SYDNEY To DAVID JONES', G.P.O. Box 503A, SYDNEY Please send me further particulars of the Kit Set advertised in "Wireless Weekly" of February 5th. (If you require details of any of the following sets, please mark those concerned with an X.) Pentagrid Battery 6 . . . . A.V.C. Dual Wave 4 . . . . 5-v All Electric . . . . A.V.C. Battery 4 . . . . Name Address FRIDAY 2FC 7.40 A PROGRAMME OF BALLADS By MAISIE RAMSAY, Soprano The Piper from Over the Way May Bracke in My Garden Idabelle Fergusons Gossiping Dodge The Singer Elsa Maxwell Red Rose and Dead Rose Louis Lortie The Thrush Edith Hartry What's the Time, Blackbird? Oliver 2FC 8.0 BLUE BONNETS O'ER THE BORDER—No. 1 FAST CASTLE By RUSSELL SCOTT Featuring SIDNEY MACEWAN CHARACTERS: MALCOLM, a Young Borderer. DONALD, his Friend. MADGE, his Fiancée. SERGEANT AT ARMS. SANDY. Production: CHARLES WHEELER. 2FC 8.35 ROY SMEDLEY'S MANDOLIN ENSEMBLE Bridal Dance from Ferma Cosmopolitan Traditional Airs of Russia— About a River. An Old Evening. Russian Dance Traditional Airs of Russia— At the Gate. Kazaky—My Volinka. Gopak Trepak Bright Shines the Moon 2FC SYDNEY, 610 K'cycles (Aust. B'casting Commission) EARLY SESSION 7.0: Big Ben, Opening Musical Item (r.). 7.10: This morning's news from the "Sydney Morning Herald" Cables (copyright) from the Australian Associated Press. 7.20: A Musical Interlude (r.). 7.25: Market Report—Fruit and Produce Report supplied by the State Marketing Bureau; Wool and Wool Textiles; Summary of the Homebush Stock Markets, Wool and Shipping. 7.45: Sectional Weather (repeated) and River Heights. 7.48: A Musical Interlude (r.). 8.0: Close. MORNING SESSION 9.30: Marching Music for School Children (r.). 9.35: Railway Train Arrivals, Musical Miniatures (r.). 10.0: British Official Wireless News Service. 10.05: Musical Interlude (r.). 10.35: Early Overseas Wheat Quotations. 10.56: A Musical Item (r.). 10.58: A Woman's Comments on Overseas Topics. 1.55: Stock Exchange, second call. 1.57: Programme Announcements. 2.0: Close. AFTERNOON SESSION 3.0: EDUCATIONAL SESSION—BROADCAST TO SCHOOLS—SECONDARY. Mrs. D. State, B.Sc., Dom., formerly Lecturer Sydney Teachers' College, will speak on: George Westinghouse. 3.15: Mr. G. A. Sussemlch, F.G.S., will speak on: The Extinct Volcanoes of Australia. 3.30: Musical Items (r.). 4.0: THINGS I'VE NOTICED ROUND THE WORLD. TALK BY MARCIA GREGORIA. 4.15: Close. EARLY EVENING SESSION 5.30: Young People's Session, conducted by Miss F. P. Foster. Elle Price and Co. present the serial: The Young Royalist. 5.40: Kevin Maegraith will give a Sketching Talk. 5.50: Pat. 6.0: Dinner Music (r.). 7.0: Features in To-night's Programme. 7.2: Dinner Music (r.). 7.40: A PROGRAMME OF BALLADS MAISIE RAMSAY, Soprano. (See Panel) EVENING SESSION 2FC 9.0 PROGRAMME BY NATIONAL MILITARY BAND Conducted by STEPHEN YORKE Assisted by COURTNEY BRAY, Baritone BAND— Overture: William Tell Spanish Dances Nos. 2 and 6 BARITONE— The Soliloquy, from Boris Godounoff La Cucumina, from The Barber of Seville BAND— Selection from La Gioconda 2FC 9.50 IN A PERSIAN GARDEN A SONG CYCLE Words Selected from Rubaiyat of Omar Khayyam. Music by LIZA LEHMANN Presented by A.B.C. (SYDNEY) CONCERT ORCHESTRA AND WIRELESS CHORUS Conducted by PERCY CODE CHORUS: Wake, for the Sun Who Scattered into Flight. SOLO: Before the Phantom of False Morning Dies, BASS: For the New, Devouring Oil Dances (For Bass.) TENOR: Iran, Indeed is Filled with All His Rose. CHORUS: Come, Fill the Cup, and in the Fire of Spring BASS: Whether at Nainshapur or Babylon. CONTRALTO RECIT.: Ah, Not a Drop, as from Our Cups We Throve. CONTRALTO: I Sometimes Think That Never Blows So Red. SOPRANO and TENOR: A Book of Verse Underneath the Bougainvillea. BASS (Solo): Myself When Young, Did Eagerly Frequent. BASS (Solo): Ah, Make the Most of What We Yet May Spend. CONTRALTO: When You and I Behind the Veil are Far. SOPRANO RECIT.: But if the Soul can Fling the Dust Aside. SOPRANO (Solo): I Sent My Soul through the Invisible. TENOR: Alas, that Spring should vanish with the Rose. CONTRALTO: The Worldly Hope Men set Their Hearts upon. SOPRANO: Each Morn a Thousand Roses Bring—You Say. CHORUS: They say the Lion and the Lizard Keep. TENOR (Recit.): Ah, Fill the Cup—What Boots it to Remain? TENOR: Ah, Moon of My Delight, That Knows No Waning. BASS: As Then the Tulip for Her Morning Sup. CHORUS: Alas, That Spring should vanish with the Rose. FRIDAY, FEBRUARY 5 . . . CONTINUED 8:30: Interlude (r.). 8:35: ROY SMEDLEY'S MANDOLIN ENSEMBLE (See Panel) 9:0: PROGRAMME OF NATIONAL MILITARY BAND Assisted by COURTNEY BRAY, Baritone. (See Panel) 9:45: Interlude (r.). 9:50: IN A PERSIAN GARDEN—A SONG CYCLE. (See Panel) 10:30: Late Casting Weather Forecast and Late News from "The Sun." 10:40: PHILHARMONIC SYMPHONY ORCH.—OF NEW YORK—Andante—Menuetto—Allegretto. (Clock Symphony) ... (Haydn) 10:56: ROSE TRIO—Wiegenduft ... (Schubert); Moment Musicale ... (Schubert) 11:4: JOHN MCCORMACK, Tenor—Baby Angel—When the Children Say Their Prayers 11:10: EDITH LORAND ORCHESTRA—La Lettre a Manon ... (Gillet); Serenade ... (Drida) Eurotis ... (Jarniceloff) 11:22: FRED HARTLEY'S QUINTET—Song of the Nightingale (Allabout Music) ... (Peter) 11:30: Close. 2BL SYDNEY, 740 K'cycles [Aust. B'casting Commission] OPENING SESSION 7:0: G.P.O. Chimes and Announcements—Morning Music (r.). 7:15: Call-Up Notice from the Government Labor Exchange. 7:20: Morning Music (r.). 7:35: Morning News from the "Sydney Morning Herald" and Cables Copyright from the Australian Associated Press. 8:0: Morning Music (r.). 8:15: British Official Wireless News Service. 8:20: Morning Music (r.). 9:0: TAL ORDELL—TELLS A STORY. 9:30: Close. MIDDAY SESSION 11:30: G.P.O. Chimes and Announcements—An Orchestral Interlude (r.). 12:0: Stock Exchange, First Call. 12:7: FROM THE SYDNEY TOWN HALL—COMMUNITY SINGING CONCERT (Arranged by The Australian Broadcasting Commission). Associate Artists: THE MUSICAL CARLSONS; MAR JUNE, AND JULIE, Vocal Trio; LES CONEY, Comedian. 2:0: FROM THE STUDIO—Afternoon News from "The Sun" Cables (Copyright from the Australian Associated Press Metropolitan Section). Forecasts and Weather Synopsis. What's on the Air This Afternoon? 2:10: Music of the Moment (r.). 2:30: Musical Interlude (r.). RADIO MATINEE 3:0: Recital by BINA ADDY, Distinguished Indian Soprano, Soloist, Violinist DAISY RICHARDS, SOPRANO— Sundari Radha (with 'Cello Accompaniment) ... (Tagore) Al Sabahan Purwina (with 'Cello and Flute Accompaniment) ... (Tagore) VILDADE— Andante from Concerto in E Minor ... (Mendelssohn) Horses Melody ... (Achron) Mohobella ... (Grieg-Hartmann) SOPRANO— Shakhi Baahi Baje (with Flute Accompaniment) ... (Tagore) Qamar Shitar Dey (with 'Cello Accompaniment) ... (Tagore) 2BL 8.0 RECORDED CELEBRITY RECITAL Arranged and Annotated by ROBERT McCALL THE CONCERTS COLONNE ORCHESTRA, under the direction of Gabriel Pierre— Marche Jeunesse ... Chabrier NINON VALLIN, Soprano— L'Automne ... Faure Clair de Lune ... Faure L'Air ... Faure D'Une Prison ... Hahn Marguerite Long, Pianist, with the Colonne Orchestra— Symphony for Piano and Orchestra; On a French Mountaineer's Song ... D'Indy HEDDLE NASH, Tenor— To the Queen of My Heart ... Delius Love's Philosophy ... Delius THE LONDON PHILHARMONIC ORCHESTRA, conducted by Sir Thomas Beecham— Hassan—Interlude and Serenade ... Delius TSCHERPENPIN, Pianist— Bagatellen ... Tscherpennin 2BL 9.0 JUDITH ANNE AND HER UNCLE JOHN 2BL 9.10 RECORDED CELEBRITY RECITAL (Continued) NEW YORK PHILHARMONIC ORCHESTRA— Ein Heidenleben ... Strauss GEORGES BUZILL, Tenor, and GERMAINE MARTIN-NIEMAN, Soprano— The Mastersingers—Excerpts from Acts 2 and 3 ADOLF BUSCH, Violin, and RUDOLF SERKIN, Pianist— Sonata in C Minor, Op. 30, No. 2 ... Beethoven and Flute Accompaniment) (Tagore) 3:30: MUSICALS (r.). Marek Weber and His Orchestra—Chocolate Soldier Selection (O. Strauss) Jack Buchanan and Gerald's Orchestra— Brewster's Millions (Noble-Furber) Patricia Routledge, Piano— Kilgallen's Blackbird ... (Balfe) Herman Finck and His Orchestra— Gaity Echoes (Caryll-Monckton) E.E.C. Dance Orchestra, Directed by Henry Hall— Noel Coward Medley— Jeannette Macdonald, Soprano— Beyond the Blue Horizon (Harling) Isn't It Romantic ... (Rodgers) Always ... All My Days ... (Harling) Rudy Wiedeott, Saxophone— La Cinquantaine (Poirier, arr. Wiedeott) New Mayfair Orchestra— Big Broadcast of 1936 Selection. 4:30: CHAT OVER THE TEACUPS—BY PAT COTTON. 4:35: Trade Demonstration Music (r.). 5:0: Stock Exchange, Third Call. 5:5: Trade Demonstration Music (r.). 5:23: Weather Information. 5:30: Close. EARLY EVENING SESSION 6:0: Musical Item (r.). 6:10: The Sporting Editor will discuss Current Sporting Results. 6:20: Stock Exchange Report. 6:30: The A.B.C. Racing Commentator. 7:0: CURRENT BOOKS WORTH READING. 7:15: Interlude (r.). 7:20: NATIONAL NEWS BULLETIN. 7:30: Local News. 2NR GRAFTON, 700 K'cycles [Regional Station] 7:0 to 7:25: Relayed from 2FC. 7:25 to 9:30: Relayed from 2BL 9:30 to 10:00: From 2FC. 10:0 to 12:20: Relayed from 2FC. 12:20 to 1:20: Relayed from 2FC. 1:20 to 1:30: Local News from the Grafton Exchange. 1:30 to 2:00: Relayed from 2FC. 2:0 to 4:15: Relayed from 2FC. 4:15 to 6:10: Relayed from 2FC. 6:10 to 6:30: Local News, Sporting Session, from 2NR Studio. 6:25 to 8:0: Relayed from 2BL. 8:0 to 11:30: Relayed from 2FC. 2UE SYDNEY, 950 K'cycles [Commercial Station] 6:0: Wake Up Section—Lewis Bennett. 6:30: Weather, General Markets. 6:35: Meditation Music. 6:45: Lewis Bennett at the Piano. 7:00: News from London. 7:10: Bright Breakfast Music. 8:0: Randwick Track Highlights. 8:30: Famous Selections. 9:02: News from "The Daily Telegraph." 9:45: Songs for Mother. 9:50: Deep Harmony Lullaby. 9:48: Racing Tips. 10:32: Music. 10:0: Women's Interests — 10:0: Office Hours and Club Announcements. Mrs. Filmer 10:30: Morning Tea Session. 10:45: Sacred Music. 11:00: Music—2UE Radio Matron; 11:30: Household Hints and Recipes; 11:45: Miss Julie Russell. 12:0: Serial Story—St Meredith. 12:30: Music. 12:45: Garden Chat—Mr. S. H. Hunt. 1:10: Luncheon Music. 1:30: Close. 2:0: Between Ourselves—An Afternoon Hour, Conducted by Frank Sturge Hardy. 2:15: Music. Music. 2:45: The Radio Adviser—Life's Problems. 3:15: Pot Pourri. 3:45: Songs relayed by Frank Hartly. 4:15: Musical Memories. 5:0: Dance Rhythms of To-day. 5:30: With the Light Orchestras. 5:45: New Record Releases. 6:0: Dinner Music. 40: Racing Preview by Harry Solomons. 6:45: Dance Music. THE SIGN OF THE PURPLE SPUR—A B.S.A. PRODUCTION. 7:15: Bright Music. 8:0: THE MUSICIANS' LIBRARY. 8:30: Quick Stepping Rhythm. 8:45: Music. 9:0: POLITICAL AFFAIRS—By Mr. A. H. HAUFFTANN. 9:15: Radio News and Announcements. 9:30: SPORTING PREVIEW AND SUMMARY—By Mr. JACK ALLISON. 10:0: Music—To-day. 10:30: Froth and Bubble. 11:0: Words and Music. 11:30: Close. 2NC NEWCASTLE, 1230 K'cycles [Aust. B'casting Commission] 7:0 to 9:30: Relayed from 2BL. 9:30 to 4:00: Relayed from 2FC. 10:0 to 10:30: From the Newcastle Cathedral Organ Recital by T. H. Munger. 10:30 to 10:40: From the Newcastle Studio—News Service by courtesy of the "Newcastle Morning Herald." 10:40 to 11:30: Relayed from 2FC. 12:0 to 1:15: Relayed from 2FC. 1:15 to 2:00: Relayed from 2FC. 3:0 to 3:40: Relayed from 2FC. 3:40 to 3:45: Chat Over the Teacups by Celso Starbird. 3:45 to 15: Relayed from 2FC. 5:30 to 6:0: Relayed from 2FC. 6:0 to 6:25: Women's Sporting Session, by Cecily Sullivan. General Sporting Session, by Halford Hooker. 6:25 to 8:0: Relayed from 2BL. 8:0 to 11:30: Relayed from 2FC. 2CH SYDNEY, 1190 K'cycles [Commercial Station] 4:5: Good-morning Man. 6:0: Early Morning Entertainment. 7:15: Syncope—Spotlight. 7:30: Beach Forecast. 8.0: Bird Calls from Tasmania. 8:30: "Why's Off to School." 8:45: Mothers' Session. 9:30: Australian Woman's College Session—Janet Austin. 9:45: Music. 10:0: British Official Wireless News. 10:2: Morning Devotion—Rev. E. Butterworth. 10:25: Music. 10:30: The Quiet Quest. 10:45: Microphone Melody. 11:45: Look Up and Laugh. 11:50: Weather Report. 12:0: Hope Suttor conducts the Housewife Service Session. 12:15: "Windows Open." 1:0: FROM PARRAMATTA TOWN HALL—COMMUNITY SINGING, conducted by Bob Strether. FRIDAY, FEBRUARY 5 . . . OVERSEAS STATIONS WHAT'S ON THE AIR TO-DAY 5.45 a.m. (GSD 25.5, GSB 31.5, GST 19.6, with GSC 31.3 at 7): "Ghosts of London." Musical Memories by the B.B.C. Theatre Orchestra. 6.45: Interval. 7.00 News. 7.20: B.B.C. Theatre Orchestra. Overture, The Bronze Horse. 8.00 Music. Synopseation (Kreckler). Slavonic Rhapsody, No. 2. 6.0 a.m. (TPA 32.19): Variety Programme from Italian Stations. 9.0: News in English. 9.15: Music. 9.30: (LMD 2.0) DML 19.8, DJG 49.8): A Military Concert. 7.5: News. 7.20: Close. 5.30 p.m. (DJB 19.7, DJA 31.3, DUN 11.4): Pendergast of North, South, East, West. Amusing Stories from All Quarters. 6.0: The Spirit of the New German University. 6.45: The German Songs. 7.30: News in German. 7.45: Concert of Light Music. 8.0: Close. 6.0 p.m. (GSD 19.5, GSO 17.6, GSG 16.8): "John Londoner at Home." An Evening with His Varied Acquaintances. 6.31: Beethoven String Quartets. 7.0: Suggestions for the Bookshelf. 7.15: Programme of New Gramophone Records. 7.40: News. 8.0: Close. 6.30 p.m. (CLR 31.3): Recorded Music. Sporting and Commercial News. 7.0: News Bulletin. 7.15: Collaborative News. Music. 8.0: National Evening Musical Programme. 10.30: News Bulletin. 10.45: Music. 11.30: Close. 7.0 p.m. (TPA 32.19): News from Radio Paris. 7.40: News in English. 7.50: Music. 8.0: Close. 7.0 p.m. (SME 31.5): Recorded Music. 8.0: Weather and News. 9.0: Music. 10.0: Close. 7.0 p.m. (EDW 31.0): News. Music. 9.0: Programme from Hongkong Studio. 10.0: News. 10.15: Music. 8.30 p.m. (VPD 31.45, YDD 31.4, YDC 19.8, VPD 27.2, FMN 2.0): Evening Musical Programme. 10.0: VPD2 News. 9.0 p.m. (OSB 31.4, GSO 16.8, OSH 13.9): Big Ben. Organ Recital from Hits Cinema, Belfast. 9.30: "From Sail to Steam." Story of Ships. 10.10: News. 10.15: Music. 7.0: The English Symphony by Herbert Holdsworth. M.P. 10.30: The B.B.C. Dance Orchestra (Henry Hall). 10.55: News. 11.15: News. 11.45: Descriptive Commentary on a Modern Tweed Mill. 11.45: Close. 9.0 p.m. (DJB 19.7, DJA 31.3, DUN 31.4, DJQ 19.6): Concerts. 9.15: News. 10.0: News in English from Berlin. 10.15: Light Music. 10.30: News in German. 11.15: Eternal Monuments of German Culture. V. Cathedral, Castle and Haunts of Kant in Konigsberg. Midnight: News in German. 10.0: News. DJA 19.6, DJQ. 9.0 p.m. (TPA 19.6): Concert Programme from Paris. 10.0: News in English. 10.15: Music. 9.30: Musical Scrapbook. Pearls on Velvet. I Love You Truly. Red River Valley, Furiani. 9.45: ENTERTAINING AND HUMOROUS PALESTINIANS 10.15: N.R.M.A. Service to Motorists—Mr. Mitchell. 10.30: Willem van der Meer and his Orchestra. 10.45: The London Symphony Orchestra—Perpetuum Mobile (Strauss). Richard Crooks, tenor. 11.00: The Magic of Your Mother. Colombo and His Telugue Orchestra—A Russian Night at the Hungaria. Lucienne Boyer. 11.30: The Magic of Romance (Parish-Delettre). The Band of H.M. Coldstream Guards—The Quaker Gift. Selections. 12.00: Frank Deane, baritone. On the Road to Mandalay (Speaks). 11.0: Slimmer Music. 11.30: 2GB Goodnight Song. 11.30: Close. 10.30: The 2UW Guest Artist. 10.40: Musical Comedy Selections. 11.0: Katy and John—Hilda Morse and Clifford Arnold. 11.20: Mr. Piggle. Fun and Humour. 11.30: Light Lyrics. 11.45: Dr. Frederick. 12.0: Vocal and Instrumental Music. 2.00: Musical Matinee. 3.0: Matti Moore. Musical Highlights. 3.30: Prelude to Afternoon Tea. 4.30: Melody Girl. 4.0: United Association Session. 4.45: Pablo Paredes Newman. 5.15: Bobby and Betty Bluegum. 6.0: George Edwards—"David and Goliath Under the Southern Cross." 6.15: Cricket Stumps—Drawn Scores. 6.18: Dinner Music. 6.30: The Talk of the Town. 6.37: NORMAN LYONS—MODEL AEROPLANES. 6.50: Sporting Session—Charles Lawrence, Adelphians, Moorfield and Williamson. 7.0: Trade Demonstration Music. 7.10: Speedway Topics. 7.30: Something for Everyone. 11.45: ANGLERS' WEEKLY GUIDE SESSION — O. J. LEIGHTON. 8.0: The Household of Notre Dame—A George Edwards Production. 3.30: Personality Series, No. 1. Richard Tucker, Tenor. 4.30: Festival Recordings. 4.45: Eric Coates, Composer, Conductor. 9.0: From the Old Music Cabinet. 9.30: The Light Orchestra. 10.0: Famous Composers. 10.15: A Christmas Recital—Elleen Joyce, Pianist. 10.30: Orchestral Highlights. 10.45: Radio Gnomish. 11.0: Musical Miscellany. 12.0: Vitality in Rhythm—Henry Gregory, Percussionist. 12.30: Keep It in the Party. Going 1.0. Clarinet Calls. 1.30: Winding Up the Clock. 2.0: News Findings. 2.30: Hospital Nurse's Singing Session. 3.0: Relay. 3.30: Hospital Children's Bath Tub Session. 6.0: Angelus. 6.1: Count McCormack Tenor—Just for To-day. 6.5: Breakfast Session. 6.10: News. 6.30: The Listeners' Choice Session. 6.50: Music. 7.30: The Listeners' Choice Session. 8.0: Music. 9.0: Close. 1.0: Lunchtime Evening Afternoon Sessions—John Tuttle. 1.30: Jim Sampson, Jazz Pianist. 1.45: Music. 2.0: Story Time, with John Tuttle. 3.0: Afternoon Session—Miss Doreen McKay. 3.5: The Music Box. 4.0: The Variety and Notorieties. 4.15: Lovely Ladies. 5.0: Uncle Tom and his Gang. 6.0: Afternoon. 6.1: Dinner Session—John Tuttle. 7.0: Preview of to-morrow's Races in England. 7.35: Preview of to-morrow's Grade Cricket Matches—S. Nagle. 7.45: The Balladaires. 8.1: Programme of Dance Music. 8.15: CAMDEN MORRISBY—Bookmarks. 8.30: "The Grocer and Madame." 8.52: "Thermo-Ray." 9.0: THE AMATEUR HOUR, Accompanied by John Dunn. 10.0: Marie Ormiston, Pianist. 10.25: When You Come to the End of the Day—Paul Oliver. 10.30: Close. 2SM SYDNEY, 1270 K'cycles [Commercial Station] 6.0: Angelus. 6.1: Count McCormack Tenor—Just for To-day. 6.5: Breakfast Session. 6.10: News. 6.30: The Listeners' Choice Session. 6.50: Music. 7.30: The Listeners' Choice Session. 8.0: Music. 9.0: Close. 1.0: Lunchtime Evening Afternoon Sessions—John Tuttle. 1.30: Jim Sampson, Jazz Pianist. 1.45: Music. 2.0: Story Time, with John Tuttle. 3.0: Afternoon Session—Miss Doreen McKay. 3.5: The Music Box. 4.0: The Variety and Notorieties. 4.15: Lovely Ladies. 5.0: Uncle Tom and his Gang. 6.0: Afternoon. 6.1: Dinner Session—John Tuttle. 7.0: Preview of to-morrow's Races in England. 7.35: Preview of to-morrow's Grade Cricket Matches—S. Nagle. 7.45: The Balladaires. 8.1: Programme of Dance Music. 8.15: CAMDEN MORRISBY—Bookmarks. 8.30: "The Grocer and Madame." 8.52: "Thermo-Ray." 9.0: THE AMATEUR HOUR, Accompanied by John Dunn. 10.0: Marie Ormiston, Pianist. 10.25: When You Come to the End of the Day—Paul Oliver. 10.30: Close. 2KY SYDNEY, 1020 K'cycles [Commercial Station] 6.45: Weather Report. News, and Commentary—Robert Butler. 7.0: Happy Hour. 8.0: With the Children. 8.15: Music. 9.0: Diggers Session. 9.15: Sports. 10.0: Collins. 9.30: The Stay-at-Home. 10.0: Musical Programme. 10.15: Morning Story—Captain Bainhalmair. 10.30: Women's Session—Mrs. Gray. 11.0: Let Us Look Through These Books. 11.15: Music. 12.0: Close. 1.0: Special "Argus" Broadcast. 1.15: Labor News Commentary. 2.0: Musical Programme. 2.30: Suburban Session—Fred Garland. 3.30: Close. 4.0: Melodies Session—John Harper. 5.15: Alley, Rion, and Essex—Children's Session. 5.30: Musical Divertissements. 6.0: Stable Spy. 6.30: Dinner Entertainment—Goodo. 6.45: Spot of Humor. 6.45: Radio Snapshots. 7.0: Racing Revelations. 7.15: Labor News Commentary. 7.30: News Topics—A. W. Davies. 7.30: Music. 7.45: Unemployed Question. 7.50: Music. 8.0: Quiz. 8.15: Music. 8.15: "THE CRIME SHEET" —RADIO SERIAL 8.30: Sports Sketch—Alf Flanagan. 8.45: Music. 8.50: PREVIEW OF GREYHOUND RACES FROM HARROD PARK—DAVE ALEXANDER 9.0: Music. 9.15: Radio Radio—E. C. Turner. 9.40: Feature. 10.0: Radio Rhythms. 10.30: Happy Hour. 12.0: Close. 2GZ Central N.S.W. 990 K'cycles [Commercial Station] 6.30: Music. 6.35: Weather. 6.40: News. 6.45: Programmers' Service Information—W. T. 7.0: Programmers' Calendar. 7.30: News and Weather. 8.15: News. 8.30: Show's Morning Session. 9.0: Programs Projects. 9.15: Popular Recordings. 9.30: British Official Wireless News. 9.30: Music. 9.30 to 11.0: Close. 11.5: The Kitchen of the Air. 12.0: News. 12.30: News Break. 11.35: News Through the Eyes of a Woman. 11.45: Far West. FRIDAY, FEBRUARY 5 . . . CONTINUED 2KA KATOOMBA, 1160 K'cycles [Commercial Station] 7:30 Happiness Breakfast Session. 7:36 Weather Report and Stock Sales. 7:38 Music. 9:00 The Happy Hour. 9:05 Golf Hints. 9:30 Weather Report. 10:00 Close. 2CA CANBERRA, 1050 K'cycles [Commercial Station] 12:30 Luncheon Music. 12:45 Horoscope Reading. 12:50 Music. 1:00 Candid Comments on Current Releases. 1:15; Music. 1:30 Close. 5:30 Children's Session. 5:45 Children's Story. 6:00 Adventures of Ben, Snow Gypsy, and Tom. 6:15 What's on in Canberra? 6:35 Dinner Music. 7:00 Vocal Hall Hours. 7:30 Vagabonds of the Prairies. 7:45 Famous Dance Bands. 8:00 Famous Orchestras. 8:15 Globetrotters Talk. 8:30 Favourites Old and New. 8:45 Instrumental Quarter Hour. 9:00 Amateur Hour. 10:00 Announcements and Close. 8:50 What's on the Air. 9:00 Close. 12:00 Programme Review. Lunch Music. 2:00 Close. 6:00 Dinner Music and Birthday Calls. 7:15 The Man on the Land. 8:00 Irish Session. 8:30 Dance Music. 9:30 News; Market and Weather Reports. 10:30 Close. 2GN GOULBURN, 1390 K'cycles [Commercial Station] 7:00 Bright and Popular Music. 7:30 Competition. 8:00 Bright and Popular Music. 9:00 Close. 12:00 Bright Lunch Time Musical Programme. 1:15 Musical Box. 1:40 Close. 5:30 Entertainment for the Tiny Tots. 5:45 Special Story for Children. 6:00 Dinner Music. 6:10 Tunes from Talkies. 7:15 Screen News from the Empire Theatre. 7:30 Favourites Old and New. 8:00 Musical Varieties. 9:45 New Releases. 10:15; Slumber Music. 10:30 Close. 2GF GRAFTON, 1210 K'cycles [Commercial Station] 7:30 to 8:30: Merry Morning Session. 12:00 Weather and Market Reports for the Man on the Land. 12:30 Music. 1:15 Musical Box. 1:30 Church Service. 2:00 Music. 2:30 Close. 5:30 to 6:00: Annie and Uncle Fred will entertain the Children. 6:00 to 6:30: Dinner Music. 6:30 to 7:30: Musical Potpourri. 7:30 Church Notices. 7:40 Music. 10:30 Close. 2LM LISMORE, 900 K'cycles [Commercial Station] 7:00 Jim Sharp, the Early Bird. 7:30 Latest News. 7:45 Short News. 7:50 3LM Market Place. 8:15 Richard Tauber. 8:30 News. 8:30 Music. 9:00 Close. 10:00 Jim Sharp Entertains You. 12:00 Musical Box. 2:00 Recorded Rhythm. 2:35 Let's Get Together. 3:00 Close. 5:30 Children's Corner conducted by Aunt Mollie. 6:00 Dinner Music. 6:30 Musical Comedy Montage. 6:45 Dramatic Interlude. 7:15 Sporting Session. 7:30 Gilbert and Sullivan Gross. 7:45 Heroes of Civilisation. 8:00 Fifteen Minutes with the Girls. 8:15 N.R.A. News. 8:30 Mr W. Mason. 9:00 New Releases. 9:15 Spot of Comedy. 9:30 Let's Go Dancing. 9:45 News. 9:50 Birthday Greetings. 10:00 Close. 2NGR GRIFFITH, 1470 K'cycles [Commercial Station] 6:00 Children's Session. 6:30 Music Lovers' Potpourri. 7:00 Songs at the Piano. 7:30 Cyril James. 7:45 Australian Artists in Person. 7:50 Market Reports. 8:15 Cultural and Horticultural News. 8:30 Musical Box. 9:30 One Hour Dance Programme. 10:30 Goodnight Song. 2MO GUNNEAH, 1360 K'cycles [Commercial Station] 7:00 Opening March. 7:30 Weather Report and Early News Service. 7:15 Bright Music. 8:30 Weather, News, and Market and Sales. 8:30 Music. 8:45 Kiddie Capers. 10:30 Close. 2TM TAMWORTH, 1300 K'cycles [Commercial Station] 7:00 Jim Sharp, the Early Bird. 7:30 Latest News. 7:45 Short News. 7:50 3LM Market Place. 8:15 Richard Tauber. 8:30 News. 8:30 Music. 9:00 Close. 10:00 Jim Sharp Entertains You. 12:00 Musical Box. 2:00 Recorded Rhythm. 2:35 Let's Get Together. 3:00 Close. 5:30 Children's Corner conducted by Aunt Mollie. 6:00 Dinner Music. 6:30 Musical Comedy Montage. 6:45 Dramatic Interlude. 7:15 Sporting Session. 7:30 Gilbert and Sullivan Gross. 7:45 Heroes of Civilisation. 8:00 Fifteen Minutes with the Girls. 8:15 N.R.A. News. 8:30 Mr W. Mason. 9:00 New Releases. 9:15 Spot of Comedy. 9:30 Let's Go Dancing. 9:45 News. 9:50 Birthday Greetings. 10:00 Close. 2AY ALBURY, 1480 K'cycles [Commercial Station] 12:15 Open. 12:20 News. 1:15 Musical Box. 2:00 Close. 5:00 Children's Community Singing. 6:00 Dinner Music. 6:30 Amalgamated Gramophones Programme. 7:30 Sporting Session. 8:00 Gem of the Evening. 8:30 Record Hour. 9:15 The House of Peter MacGregor. 9:30 Country Man's News. 9:45 Music of the Moment. 10:30 Close. 3LO MELBOURNE, 770 K'cycles [Aust. B'casting Commission] 7:00 Time. Early News. Shipping. 7:15 Weather. Aviation Forecasts. 7:15 Music. 7:20 Sydney Market Reports. 7:30 Music. 7:45 Melbourne Market Reports. 8:00 Close. 9:30 A Recent Recording. 9:35 Irish Folk Stories—"Innisfail." Read by Miss Moya Carey. 9:35 Music. 10:00 Express Train Information. Aviation Forecasts. British Official Wireless News. London Edition. Questions to Australian Rates and Foreign Exchange. 10:10 Current Happenings in Sport—By A.B.C. Commentator. 10:20 Musical Interlude. 10:25 Women's Session, conducted by "Judy." 10:30 "Our Weekly London Letter," from Miss Margaret Neville. Music. "Philosophy of Modern India"—Mrs. Gertrude Sherrill. 10:50 Music. 11:00 Aviation Forecasts. 11:30 Musical Interlude. 12:00 Close. 12:30 Music—Light and Bright. 12:15 Midday News Bulletin—Corn Exchange. Midday Report. Interstate Market Report. Cattle Prices. Police Messages. 12:35 Music. 12:45 From Sydney—"At Home and Abroad" with the Collector. 1:00 From Melbourne—Interstate Weather Forecasts. Central Banker's Interest Ratings. Rainfall Bulletins. River Gaugings. Weather Conditions in Border Areas. 1:00 Sydney-Melbourne Air Route. 1:10 Stock Exchange Report for London. Exchange. London Metal Prices. Shipping Information. 1:15 Music. 2:00 Close. 3:30 Recital by Raymond Lambert, Pianist. 4:00 Children's Corner Session (Claude Debussy). 3:30 The Symphony Orchestra, under Sir Edward Elgar—"Romantic Elgar"; Serenade Lyrique (Poulenc); Agnête (Maurice Ravel), Mezzo-Soprano. 4:00 Unreliable, Pagen Hoji I Taarinet (Jacobsen-Nielsen). 4:00 Farken. Else Ammertorp. Sophie von Schleicher (Holstein - Nielsen). Georges Thill. Tenor—"Liebestraume (Liszt-Hensel). 4:00 "The Blue Willow"—Gavotte in F Major (Beethoven). Menuet (Bach). 3:30 Classic Music. 4:00 Close. 3:30 Young People's Session—Overtures. Zoo News by David Fleay. 4:00 "How Would You Like to be an Artist?" Fun and Music, with "Mandy Lou." Finish. 6:00 Dinned Music. 7:10 A Programme of Ballads. (See 2PC.) 8:00 The Sundowners Male Quartette. (See 2PC.) 8:30 The National Dance Orchestra. (See 2PC.) 9:00 National Military Band. (See 2PC.) 9:15 Interlude. 9:50 "In a Persian Garden." (See 2PC.) 10:00 A Summary of the Day's News. FRIDAY, FEBRUARY 5 . . . CONTINUED 10:49: William Backhaus, Pianoforte Solo—Sonata in C Sharp Minor, Op. 27, No. 2 (The Moonlight Sonata) (Beethoven). Sir Thomas Beecham conducting the Royal Philharmonic Orchestra—On Hearing the First Cuckoo in Spring (Debussy), Maurice Marsche, 'Cello Solo—Gigle (Faure). London Symphony Orchestra—Symphony by Sir Hamilton Harty—Passacaglia (Handel Trans.); Arietta (H. Harty). Wilhelm Backhaus, Pianoforte Solo—Passacaglia (Bach), Budapest String Quartet—Andante con Tenabile (Tchaikowsky). 11:30: Close. 3AR MELBOURNE, 580 K'cycles [Aust. B'casting Commission] 7.0: Melbourne Central Post Office Chimes & Shipping, Express, Trains Information. 8.1: Relayed from 2BL, Sydney—Morning Music. 8.50: Morning News Bulletin, Police Messages. 9.0: Melbourne Weather. 9.1: Music. 9.30: Close. 11.30: Interstate Air Mails, Music. 11.45: Stock Exchange Reports (Morning Change), London Metal Prices. 12.0: Luncheon Music. 2.1: Music. 5.0: Dinner Music. 5.30: Close. 6.0: Music. 6.45: Sporting Session, conducted by A.B.C. Commentator. 7.0: From Sydney—National Talk—Current Books Worth Reading—Denial Batchelor. 7.15: Music. 7.30: National News Bulletin. 7.30: Victorian News Bulletin. 7.35: Countryman's Session; Melbourne Stock Exchange Daily Report (including London Metal Prices), Advance Notices of Country Stock Sales, Official Quotations of Antwerp Wool Futures, Carriage Meals, Sheep Cow Markets, Beef Hearts, Wholesale Fruit and Citrus Fruit Market Reports, London Butter Quotations, Weather Forecasts for Victoria, New South Wales, and South Australia, All States' Rainfall Bulletins and River Gaugings. 8.0: For Farmers Only—Review and Forecast of Grain Stocks. Marketer—Mr. M. McNamara. Milkage Dowling. A Primary Producers' Problem—Mr. J. Leith Gillespie. 8.25: Close. 8.30: "Dawn, Judith and Joan" in Harmonised Numbers. The Swiss Tap (Rogan and Reinhart), Day and Day (Con Porter), Weather Man (Castan and Chase), Old Folks at Home (Robert Foster), Nagasaki (Dixon and Warren). 8.45: From Hall of Honor, Geelong—Community Singing, conducted by J. Finney. 9.15: A Comedy Interlude by "Moss and Mandy"—Episode No. 3: "Going to Their Country Home." 9.30: From Hall of Honor, Geelong—Community Singing. 10.0: Interlude. 10.10: News, Road and Touring Notes. 10.30: Close. 3GI SALE, 830 K'cycles [Regional Station] 7.0: See 3AR. 7.37: See 3LO. 8.0: See 3AR. 9.30: See 3LO. 11.30: Close. 12.0: See 3LO. 1.10: See 3AR. 2.0: Close. 3.0: See 3LO. 4.15: Close. 5.20: See 3LO. 6.0: See 3AR. 10.10: Musical Interlude from 3GI. 10.30: See 3LO. 11.30: Close. 11.15: The Women's Magazine, conducted by Miss Rita Humphress. Talk by J. F. Bailey—Home Gardening. 11.40: Talk by a Woman Doctor—Mothercraft: The Tired Child, 12.0: Close. 12.15: Broadcast to Schools by D. S. A. Smith, M.A., Inspector of Schools. Geography—1. Where Man has settled and why. 12.30: Market Reports. British Official Wireless News, Weather Data. 12.45: For the Man on the Land—Talk arranged by the Department of Agriculture and Stock: The Influence of Type in Queensland Merino Sheep, by Mr. G. L. Hodge, Instructor in Sheep and Wool. 1.0: Music. 2.0: Close. 3.0: Matinee Recital by Jill Mannering (soprano) and Arthur Sherman (piano). Piano—Prelude, Fugue, and Variations (Chopin, Franck). Violin—The Answer (Terry). Poor Butterfly (Hubbell). The Cuckoo (Brahms). Piano—Autumn Idyll (Hull). Rossiniade (Schumann). Penouhoff). Nocturne in A Major (Fiori). Serenade—The Bird and the Babe (Lisztance). The Nightingale (Kjerulf). 2.30: Yehudi Menuhin (violin) and Orchestre Symphonique de Paris—Concerto in E minor for Violin—Elisabeth Schumann (soprano)—Bist du bei mir (Bach). Colonne Symphony Orchestra—A Night on a Bare Mountain (Moussorgsky). 4.15: Dance Music. 4.30: Close. 3.30: Children's Corner, conducted by Uncle Max. 6.0: Dinner Music. INTRODUCING The Latest BREVILLE Model 89. A "B" Batteryless Mantel Set with 5 valves for Australian Broadcast reception. No B or C Batteries required. Now country residents may enjoy a Mantel set without cumbersome B Batteries—a set that gives the performance of its Big Brothers which have already become so popular. The Cabinet is the Breville Jewel constructed of choice Walnut Veneers and hand polished to a delightful finish. The large Permagnetic Speaker, Automatic Volume Control, Tone Control and large Aero Dial lift the Model 89 into a class by itself. And yet the price has been kept extremely low for such a high-grade Radio Receiver. The care taken in selecting component parts and strict supervision during manufacture, in addition to a series of most rigid tests, assures the purchaser more than usual satisfaction in being the possessor of one of these fine sets. The Radio Frequency Stage, a 3-gang tuning condenser, plus the Breville Ferrodine Intermediate Frequency transformers at 175KC, gives extreme selectivity and freedom from background noise. It's economical, too — Maximum current drain .9 of an amp. DEALERS: Write for particulars of the Breville Franchise for your district. • FULLY GUARANTEED: Warranty card accompanies each receiver. BETTER RADIO VALUE is not obtainable anywhere—it's a BREVILLE—therefore the LOGICAL CHOICE. Write for illustrated leaflet. Manufactured and fully guaranteed by— BREVILLE RADIO—also—BREVILLE RADIO Pty., Ltd. 486 Elizabeth Street, SYDNEY. Phones ME391 (2 lines). 191 Queen Street, Melbourne. Phone MU1548. DISTRIBUTORS IN ALL STATES: QUEENSLAND REPRESENTATIVE: O. Thomas, King House, Queen Street, Brisbane. Phone B371L PRICE: In N.S.W., Vic. and Tas. £25'15'- In Q'ld., W.A. & S.A. £26'15'- BREVILLE RADIO—also—BREVILLE RADIO Pty., Ltd. 486 Elizabeth Street, SYDNEY. Phones ME391 (2 lines). 191 Queen Street, Melbourne. Phone MU1548. DISTRIBUTORS IN ALL STATES: QUEENSLAND REPRESENTATIVE: O. Thomas, King House, Queen Street, Brisbane. Phone B371L FRIDAY...CONTINUED 6:45: Sporting Notes. Racing prospects reviewed by Keith Smith. 7:0: From 2BC—National Talk by Denzil Batchelor B.A. (Oxon.). Current Books Worth Reading. 7:15: Interlude. 7:20: From 2BL—National News Bulletin. 7:30: News. 7:35: Movements of Air-mail Flights. Weather, Daily and Market Reports. General Commercial News. Programme Announcements. 8:0: From "Dobrowolsky" and his Ballets featuring the A.B.C. (Brisbane) Augmented Orchestra. The Programme arranged and conducted by W. Nelson Burton. Presented by Dion Wheeler. 8:45: Legal Criticly. A visit to the London Law Courts. Narrated by Ian Hay (B.R.C.) Recording. 9:0: From 2PC—The National Military Band. 9:45: The Lyrical Male Choir, directed by J. T. O'Connor, assisted by Max Muller Jun., (violin). Choir in the March "Heavenly." Violin—Sonata in D Minor (Vera- cini), Largo, Allegro con fuoco. Choral Hymn—"O Messias, Jesu, Herrh!" Led Kindt Light arr. Purdy; The Little Dustman (Brahms). Violin—Romance in E Major (Mendelssohn). Siciliano and Riusadon (Francoeur-Kremer). Guit.-Hope will Battle. Suite—(Spanish air); The Mulligan Musketters (Atkinson); Southern Memories. 10:25: Interlude. 10:30: Weather, late Sporting Summary. 10:40: Dance Music. 11:30: Close. 4RK RH'TON, Q. 910 K'cycles [Aust. B'casting Commission] 7:0 Relayed from 4QG. 9:0: Close. 10:30: Relayed from 4QG. 10:45: News. 11:0: Relayed from 4QG. 2:0: Close. 3:0: Relayed from 4QG. 4:30: Close. 5:30: Relayed from 4QG. 6:0: News. 6:15: Relayed from 4QG. 11:30: Close. 4QN Nth. Regional 600 K'cycles 7:0: Relayed from 4QG. 5:0: Close. 10:30: Relayed from 4QG. 10:45: News. 11:0: Relayed from 4QG. 2:0: Close. 3:0: Relayed from 4QG. 4:30: Close. 5:30: Relayed from 4QG. 6:0: News. 6:15: Relayed from 4QG. 11:30: Close. 5CL ADELAIDE, 730 K'cycles [Aust. B'casting Commission] Main Features 3:15: Educational Broadcast—Nature Knowledge, by Mr. Inspector A. G. Edquist. 3:30: Afternoon Tea Matinee, with Marian Hardy, Mezzo-Soprano. 4:0: A Short Talk by Mary Trent. 4:15: News. 4:27: News Flashers. 4:30: Close. 5:30: The Children's Session, conducted by Marie Louise and Olga Pal. 6:0: Music. 6:35: What of the Roads?—A Talk under the Auspices of the Royal Automobile Association, by Mr. S. T. Facey. 6:50: From Sydney—National Talk: Current Books Worth Reading. 6:50: From Sydney—National News. 7:0: From Adelaide—Evening News. 7:5: Official Stock Exchange Information and Market Reports. 10:0: From Melbourne—A Programme of Ballads by Maisie Ramsay, soprano. 10:30: From Adelaide—Regional Programme. 10:45: From Sydney—Minstrels—A Feature Production by the British Broadcasting Corporation. 10:50: From Sydney—National Programme. The National Military Band, conducted by Stephen York, with Colin Chapman, Baritone. Alternative Programmes. 9:15: Station 5CK—From Adelaide—Regional Programme. The A.B.C. (Adelaide) Theatre Orchestra, conducted by William G. Cade. 9:15: Station 5CK—Regional Programme. Wool Sales Resume. 10:0: Stations 5CL and 5CK—Announcements for To-day's Racing and Tennis Fixtures. 10:3: World Radio—A Weekly Talk by Sydney Shute. 10:15: News and Weather Notes. 10:25: Dance Music by the Painis Royal Dance Band, conducted by Harry Bonke Smith. 11:25: Signature Melody. 11:30: Close. 7NT Launceston, 710 K'cycles [Aust. B'casting Commission] 7:0: Time. Weather. 7:3: News. 7:5: Shipping Mails. 7:10: Music. 7:15: Sydney Prices of Tasmanian Products. 7:30: Music. 9:0: Close. 10:30: Time. Weather. Music. 10:37: Introduction to Daily Broadcast Service. 10:40: Daily Broadcast Service. 10:55: Music. 11:0: Women's Session, conducted by Elizabeth Rowlands. 11:30: Music. 11:45: Melbourne Stock Exchange. 12:9: Time. Music. 12:10: Corn Reports. 12:30: News. 12:35: Music. 12:40: Hobart Stock Exchange. 12:45: "Home and Abroad," by The Spectator. 1:0: Weather. Music. 1:10: Melbourne Stock Exchange. 1:15: Musical Reproductions. 1:30: Weather Information. 4:0: Close. 5:0: Music. 4:30: Close. 5:30: Young People's Session. 6:0: Dinner Music. 6:45: Geoff Stannard's Racing Selections, for tomorrow's Races. 7:0: National Talk—"Current Books Worth Reading." 7:15: Music. 7:20: National News Bulletin. 7:30: News, Markets, Stock Exchange. 7:40: From Launceston—Recital by Madge Elliott, Soprano. 8:0: From Northern Churches of Christ, A Programme of Characteristic Music and Humor from Northern Ireland. Compiled and Produced by Henry McMillan. Presented from the Studios of the British Broadcasting Corporation. 9:0: From Sydney—National Military Band (See 2FC). 9:45: From Hobart—The Hobart Repertory Theatre Society presents "Wo-wo-wo," by the Great Bell of Chiao-Shing-Sun. By James and Sung Ko. The principal characters in the play are as follows: Wen-li, the bell founder to the Emperor Yung-lo. Yen-lin (also known as Liang), daughter of the bell founder. Chang Feng, the young and clever court maker and bell-founder. Guo-tzu-tzu, Li Sheh, the Controller of the Imperial Household. The Lady Pao, the wife of the bell-farer and Way-larn, personal attendants of the Lady Pan Amah, the nurse of Golden Genius and now companion. Hsieh-tsin (known to most as Chin), the well-known poet who lived in the reign of Yung-lo. He was petted by Poon-Wong, the Controller. Secretary, a mean fellow. Nameless people of the sacred mountain Hwa Shan in Shenxi. Guardsmen of the Controller. Heroine of the Controller. Fountain men. Yen-li. Craftsmen of Chang-Feng, the sword maker. Producer: James Pratt. 10:30: News. Weather. 10:40: British Symphony Orchestra (Sir Henry Wood)—Brandenburg Concerto, No. 2, and G. Bach's The Philadelphia Orchestra, Op. 93, Symphony No. 3, in E Minor (Dvorak). 11:30: Close. ENTERTAINMENT ON EVERY PAGE FEBRUARY NUMBER NOW ON SALE Better and Brighter Than Ever BEAUTIFULLY PRODUCED, PRINTED IN COLOURS, PACKED WITH PICTURES, FEATURING: THE ART OF CONDUCTING COMMUNITY SINGING.—Bob Strother reveals the secret of his success. "MY FIGHT FOR SUCCESS"—Robert Waldron, author of "The Flying Doctor," now attached to the staff of 2UW, tells of his struggle to gain recognition. PRINCE OF DRIVERS AND DRIVER OF PRINCES—Story of a well-known radio star who was chauffeur to Duke of Windsor. LIFE STORY OF FRANK STURGE HARTY (9th Instalment). RADIO COPYWRITING OR A HUSBAND—A true story of a girl in a radio advertising agency who had to decide this question. MEN I FALL FOR—"Susie" of Broadway Pie reveals her likes and dislikes. STORY OF A WELL-KNOWN RADIO BACHELOR who keeps nine cats. ANOTHER OF ABBEY'S ADVENTURES IN RADIOLAND. OFFICE BOY TO STATION MANAGER—How a well-known broadcasting executive rose to the top of his profession. NEWCASTLE PAGES, with beach pictures of 2EO Personalities. INTIMATE INTERVIEWS, behind the scenes, studio news, latest gossip about your radio favorites, splendid feature articles, and a host of other features of interest to every listener. DON'T MISS IT FREE ASTROLOGICAL SERVICE. A question from your Horoscope answered free of charge. Full details, page 30. Also, Jack Davey's Horoscope and Birthday Definitions of Radio Stars. ANOTHER JUMBLED LETTER RADIO COMPETITION. Here is one of the six groups of letters. Can you make the name of a well-known personality from it? L A W C H I E L A N C E E R Get Radio Pictorial and test your skill in this fascinating competition. Prizes for the winners. FEATURE PROGRAMMES IN ALPHABETICAL ORDER A MONTH AHEAD. SPECIAL CHILDREN'S SECTION SPECIAL REPRODUCTION OF GOODIE REEVES' Poem "SLIPPER," by courtesy "Woman." If your newsagent can't supply, send 6d in stamps with coupon, which they will be posted by return mail. The Manager, Radio Pictorial, 64 State Building, 48 Market St., Sydney. Dear Sir, Enclosed please find 6d in stamps. Please forward the February Issue of Radio Pictorial. Name ........................................... Address ........................................... W.W.J. 5DN Adelaide, 960 k.c. 7:15 a.m.: Countryman's Session. 7:30: Bright Breakfast Music. 8:45: Business Man's Session. 9:40: Morning 1100 Sponsored Session. 11:15: Morning Tea Music. 12:45: Weather Report. 1:45: Portionari. 1:0: Close. 2:0: Dinner Music. 7:0: Sports Resume, by "The Stroller." 7:30: "Home, Sweet Home." 7:35: Musical Pot-pourri. 8:0: Running description of events at the Anzac Highway Motordrome. 9:0: Variety Interlude. 11:0: Sponsored Hour. 12:0: Close. 5KA Adelaide, 1200 k.c. 5:30 a.m.: Bright Morning Music. 6:45: Review of To-day's Races. 7:30: Weather Report and Programme resume. 8:0: Children's Session. 8:45: Melody Session. 9:0: Classified Ad.'s Session. 10:0: Ladies' Session. 10:45: Housewives' Competition and Correspondence. 11:15: Social Story Time. 11:35: Mystery Competition and Correspondence. 12:0: Noon: Classified Ad.'s Session. 12:30 p.m.: Turf Talk and 'Spearefts' Selections and Latest Sprinks results from Melbourne. 1:0: Racing Session, with Eric Welsh. Progressive results of Adelaide events. 3:0: Children's Session. 5:45: "Screwy School." 6:0: Sporting Session. 6:35: Classified Ad. Session. 6:30: Dinner Music. 6:50: "S.A. Worker's" News Service. 10:0: Dance music. 12:0: Close. Sunday, Feb. 7 5CL Adelaide, 730 k.c. 5CK Crystal Brook, 640 k.c. 10:30: Cathedral Bells. 10:37: Morning Hymn—"Brightest and Best of the Sons of the Morning." 11:0: Morning Divine Service, from Maughan Methodist Church. 12:15: From Studios—Interlude. 12:30: From Studios—News Notes. 12:45: "The Adelaide" Feature, with Cameos, 1:0: "This Week's Good Cause." 1:5: Classical Cameos. 1:30: Close. 1:31: Pleasant Sunday Afternoon Service, from Maughan Church. Chairman: Rev. S. Forsyth. 4:30: From Studios—Light Classical Incidental Music. 5:0: Close. 5:30: The Children's Session, conducted by "The Announcer." 5:45: Wednesday Night Story Session, conducted by Mrs. Lindsay Sewton. 5:55: Interlude. 6:0: From Melbourne—National Programme: National Talk: A Sermon for Those who may not like Sermons—"A Thing of Naught." 6:15: From Sydney—National Programme: "Random Pages," a Piano Recital, by Isidor Goodman. 6:30: "Gems From Favorite Operas"—"Carmen." Alternative Programmes. 7:0: 5CL—From Adelaide—Regional Programme: "The Crucifixion of Cremona," a Special Recorded Presentation. 7:30: From Sydney—National Programme: "Symphony Concerto," with the A.B.C. (Sydney) Symphony Orchestra. 7:0: 5CK—From Adelaide—Regional Programme: Morning Divine Service, from St. Peter's Cathedral. 8:15: From Studios—Interlude. 8:30: "Cape and Cove" from Sydney—National Programme: "A Talk on 'International Affairs'." 8:55: From Broadcast—"Under Your Spell," with Lawrence Tibbett. 9:45: From Adelaide—Regional Programme—A Piano Recital, by Spruhan Kennedy—(a) Prelude and Fugue in A Minor (Bach-Liszt), (b) Sonatina (Ravel). 10:5: News and Weather Notes. 10:10: Arnold Myers (Holtone), a Special Recital, re-broadcast from the B.B.C., London. 10:20: Musical Epilogue. 10:30: Close. 5AD Adelaide, 1310 k.c. 5PI Crystal Brook, 1040 k.c. 5MU Murray Bridge, 1450 k.c. 5:30: Chimes. Relay to 5MU and 5PI begins. Building: Choral Services, under the direction of the Council of Religious Education, conducted by the Church of Christ. 6:0: Chimes. Light Orchestral Numbers. 6:30: Celebrity. 6:50: Chimes for the Sick, conducted by Mr. T. H. Prince. 6:45: Memorial Interlude. 7:0: Hour of Melody. 7:0: Survey of Feature Programme from 5 AD-PI-MU for To-morrow. News, Tides, Weather Forecast. 9:30: Songs by famous Singers. 9:50: Epilogue. 10:0: Close. 5DN Adelaide, 960 k.c. 5RM Murray Heights, 850 k.c. 6:0 p.m.: Studio Music. 6:45: S.A. Theatre. 7:0: "Time Marches On" feature session. 7:30: Devotional Music. 8:0: Studio Feature. 8:30: Studio Music. 9:0: Mayfair Theatre. 9:15: Studio Music. 10:15: Close.* 5KA Adelaide, 1200 k.c. 9:0 a.m.: Selection of Operatic Gems, played by the Palladium Orchestra. 9:30: Golden Age Session. 10:15: Historical Session. 11:15: Melody Lane. 11:30: Jack Dale's Variety. 12:00 noon: To-day with Asson. 12:30: From Studios—News resume and close. 1:30: Lyric Suite, played by the London Philharmonic Orchestra. 2:30: Watch Room Programme. 4:0: Wizards of the Wurlitzer. 4:45: Children's Session. 5:0: Dinner Music. 6:30: Sponsored Sessions. 7:0: Hymns of All Churches. 7:30: Variety Feature Session. 2:30: Melody Session. 7:50: Strange As It Seems. 8:0: "Gems of Life." 8:15: Aspro presents George Edwards and his Players. 8:30: Brisbane's Radio Sketch. 9:0: Sponsored Session. 9:15: Hilar of the Mountains. 9:45: Regent Wurlitzer Organ Recital. 11:0: Weather Report and Close. Monday, Feb. 8 5CL Adelaide, 730 k.c. 5CK Crystal Brook, 640 k.c. 7:0: News and Weather. 7:10: Bright Morning Music. 7:30: News and Notes of Mary Tootle—Shipping, Mail, Train, and Market Reports. What's On To-day? The Sporting Diary. 7:30: From Sydney—Bright Morning Music and News Notes. 8:30: From Adelaide—Items of Interest. 8:45: Studio Music. 8:50: Close. 10:31: Daily Broadcast Service—Under the auspices of the Council of Churches. Preacher: Capt. R. Dellow for Salvation Army. 10:45: Housewives' Morning Tea Session. 11:15: "Health and Happiness," a Talk by Walter Jones. 11:30: Boston Orchestra — "Caucasian Sketches" (Ippolitov-Ivanov). 12:0: "What's Yours?"—Fifteen minutes of conviviality. 12:10: From Sydney—"At Home and Abroad," a New Commentary "Spectator." 12:30: From Adelaide—News Service. 12:35: Weather, River, and Grain Reports. 12:45: Official Stock Exchange Information. 12:50: "Caucasian" music. 1:0: Weather Report and Grain Reports. 2:0: Close. 3:15: Melody Lane Memories. 3:15: Educational Broadcast, historical play, by Mr. Brian Elliot and Company. 3:30: "Basses of Renown"—a recorded presentation. 4:0: "Piano-time Melodies" and Fleetsam and Jetsam. 4:30: News Flashes. 4:30: Close. 5:30: The Children's Session, conducted by "Minnie Louise," with "Auntie Peggy" and "Old Pal." 6:0: Dinner music. 6:25: Sporting Results. 6:30: From Melbourne—National Talk, "The Problem of Urban Slum Clearance—Housing in Cities," by Mr. W. O. Butt. 6:50: From Adelaide—Interlude. 6:55: From Melbourne—National News Bulletin. 7:0: From Adelaide—Morning News Service. 7:5: Official Stock Exchange Information. 7:10: Interlude. 7:22: Cattle Market Reports. 7:30: From Sydney—National Programme—Every Monday Night at 7:30—"As It Sow." An Australian comedy, written for Radio by Edmund Barclay. Production: Lawrence H. Cerny. 8:0: From Adelaide—Regional Programme—"Moments of Melody," with Maud Clayton. Contralto: Dave Howard, Saxophonist, and Tom Kinnaird. 8:30: From Melbourne—National Programme—The National Dance Orchestra, conducted by Al Haynes. Presents a programme of latest dance successes. 9:0: From Adelaide—Regional Programme—Vocal Interlude. 9:10: From Torch Theatre—The A.B.C. (Adelaide) Theatre Orchestra, conducted by William R. Cade, with Thea Phillips, operatic soprano. 10:10: From Studios—Interlude. 10:15: News and Weather Notes. 10:23: Modern Dance Music. 11:0: "Swing Music." 11:28: Signature Melody. 11:30: Close. 5AD Adelaide, 1310 k.c. 5PI Crystal Brook, 1040 k.c. 5MU Murray Bridge, 1450 k.c. 6:0: Relay to 5PI begins. What is on at 5 AD-PI-MU? Morning Melodies. 7:0: Wheat Markets, Tides, River Levels, Mails, Shipping, Music. 7:15: Music. 8:0: Variety. 8:10: Musical Pot-pourri. 11:0: "Nannie" and the Home Circle Club. 11:30: Music. 12:30: Weather Forecast and Grain Report. 1:0: News Summary, including Weather Forecast Programme Survey. 1:30: Orchestral Selections. 2:0: Closed. Studio closes down until 5 p.m. 2:0: Holiday Hour. 3:15: Dancing Feet. 4:0: Variety. 5:0: Studio resumed. 5MU Relay begins 5 AD-PI-MU Kangaroo Club. 6:0: Dinner Music. 6:30: Magazine of the Air. A Talk by "Observer." 7:0: Musical Medley. 7:45: News, Wheat Markets, Tides, River Levels, Weather Forecast. Programme Survey. 8:0: Evening Handley in a Comedy Session. 3:15: Melody and Song. 8:30: Variety. 9:0: Survey of Feature Programme for To-morrow. From 5 AD-PI-MU News Summary, Wheat Market, Abattoirs Stock Market (Independent report), Stock Exchange, Mails and Shipping, Tides, River Levels, Weather Forecast. 9:15: Accordeonists. 9:30: Melody Lane. 10:0: News Summary. 10:10: Variety. 11:0: Close. 5DN Adelaide, 960 k.c. 5RM Murray Heights, 850 k.c. 7:15 a.m.: Countryman's Session. 7:30: Morning Medley. 7:45: Brighter Breakfast Music. 8:45: Business Man's Session. 9:0: Studio Music. 10:15: Sponsored Session. 10:30: Music. 11:0: Sponsored Sessions. 11:30: Dinner Music. 12:0 p.m.: G.P.O. Chimes and Music. 12:30: Sponsored Session. 12:45: Weather Report. 12:45: Pot-pourri. 2:0: Musical Hour. 2:15: Matinee. 4:30: Tailgazers' Club. 4:45: Matinee continued. 4:0: Children's Session. 5:30: Variety Music. 6:30: Dinner Music. 7:0: Feature Session. 7:0: "Home Sweet Home." 7:20: Wide Range Interlude. 7:30: Musical Interlude. 7:30: "Tarzan of the Apes" Feature Session. 7:45: "Cavalcade of Music." 8:0: Travel Interlude—Victorian Government Tourist Bureau. Friday, February 5, 1937 Wireless Weekly Programme Supplement for South Australia 8:20: Musical Variety. 8:45: "Hunchback of Notre Dame," featuring George Edwards. 9:0: Wide Range. 9:15: Gems from Wide Range. 9:30: Radio Play, Arizona. 9:45: "Radio Charades." 10:15: Wide Range Celebrity Recital. 10:30: Popular Variety. 11:15: Close. 5KA Adelaide, 1200 k.c. 5:20 a.m.: Early Morning Session. 7:0: Weather Report and Programme Review. 8:0: Children's Session. 9:0: Classified Ad's Session. 10:0: Ladies' Session. 10:45: Housewives' Competition and Correspondence. 11:15: Serial Story Time. 11:35: Mystery Competition. 12:0: noon: Classified Ad's Session. 12:45: The Bookman. 5:15: Children's Session. 5:45: Dinner Music. 6:25: Classified Ad's Session. 6:30: York Theatre Feature Session. 6:50: "S.A. Worker's" News. 7:0: Sponsored Sessions. 8:15: Black Magic. 8:30: Dan, the Used Car Man. 9:10: Radio Players. 9:30: Physical Culture, with Reg. Shorthose. 9:45: Kolinsky and Clancy. 11:0: Weather Report and Close. Tuesday, Feb. 9 5CL Adelaide, 730 k.c. 5CK Crystal Brook, 640 k.c. 7:0: News and Weather. 7:10: Bright Morning Music. 7:20: News and Notes of Many Topics—Shipping, Mail, Train, and Market Reports. What's on To-day? The Sporting Diary. 7:30: From Sydney—Bright Morning Music and News Notes. 8:30: From Adelaide—Items of Interest. 8:40: Music. 9:0: Close. 10:31: Daily Broadcast Service, under the auspices of the Council of Churches. Preacher: For Congregational. 10:45: Women's Morning Tea Session—"Diet and Nutrition," a Talk by Sister Pines. 11:30: John Barbicelli's Orchestra and Essie Achabian, Contralto. 12:0: Educational Broadcast. "Your Home and Mine," by Mr. Norman L. Haines. 12:15: From Sydney—"At Home and Abroad," a News Commentary by "Spectator." 12:30: From Adelaide—News Service. 12:45: Weather, River, and Grain Reports. 12:50: Official Stock Exchange Information. 12:55: Luncheon Musicale. 1:57: Weather, River, and Grain Reports. 2:0: Close Down. 3:1: "In the Sudan"—a Descriptive Sketch, by the Super-Cinema Orchestra. 3:15: Educational Broadcast—Science Talks—"The Story of Sound," by Mr. E. S. H. Gibson, M.Sc. 3:30: Recital by Lotte Lehmann, Joseph Szigeti, and Helen Joyce. 4:0: The A.B.C. Bridge Club, conducted by "No Trump." 4:15: Norman Long expresses some of his opinions supported by the London Palladium Orchestra. 4:27: News flashes. 4:30: Close. 3:30: From Children's Session, conducted by "Marie Louise," with "The Wattle Lady" and "Old Pal." 6:0: Dinner Music. 6:25: Sporting Results. 6:30: Interlude. 6:50: From Melbourne—National News Bulletin. 7:0: From Adelaide: Evening News Service. 7:5: Official Stock Exchange Information. 7:15: News Reports. 7:20: Interlude. 7:30: From Melbourne—National Programme, conducted by Radio Play, "W. W. Beale's Characters," Nick Cusley, Fisher, and Peter. The Action takes place during a train journey in England. Production: John Cairns. 8:15: Our Heritage at Work. Fourth of the Series, presented by Harold Browning. 8:30: National Talk, "A Novelist and a Minstrel," by Mr. William G. Smith. 8:45: From Adelaide—Regional programme. A Ballad Recital, by Margaret Pirie, mezzo-soprano. 9:10: Interlude. 9:15: Dance Music, by the Palais Royal Dance Band, conducted by Harry Roake Smith. 10:0: News Service and Weather Notes. 10:23: Dance Music, by Palais Band (edtl.). 11:0: "Meditation" Music. 11:28: Signature Melody. 11:30: Close. 5KA Adelaide, 1200 k.c. 5:20 a.m.: Early Morning Music. 7:0: Weather Report and Programme Review. 7:15: "Retreat," Tearoom's Session. 8:45: Melody Session. 9:0: Classified Ad's Session. 10:0: Ladies' Session. 10:45: Housewives' Competition and Correspondence. 11:15: Story Reader. 11:35: The Dale Entertainers. 11:35: Mystery Competition. 11:40: Truby King League Talk to Mothers. 12:0: noon: Classified Ad's Session. 3:0 p.m.: The Bookman. 3:15: Children's Session. 6:30: Classified Ad's Session. 6:45: The Parliamentarian Speaks. 6:50: "S.A. Worker's" News Service. 7:0: Sponsored Sessions. 8:0: Sponsored Feature. 9:0: "Eb. and Zeb." 11:0: Weather Report and Close. Wednesday, Feb. 10 5CL Adelaide, 730 k.c. 5CK Crystal Brook, 640 k.c. 7:0: News and Weather. 7:10: Bright Morning Music. 7:20: News and Notes of Many Topics—Shipping, Mail, Train, and Market Reports. What's on To-day? The Sporting Diary. 7:30: From Sydney—Bright Morning Music and News Notes. 8:30: From Adelaide—Items of Interest. 8:40: Music. 9:0: Close. 10:31: Daily Broadcast Service, under the auspices of the Council of Churches. Preacher: Rev. S. T. Earl, for Baptist. 10:45: Housewives' Morning Tea Session. 11:0: A Talk by a Member of the Housewives' Association. 11:15: "Hall Vienna"—a Potpourri of Viennese Grace and Charm. 12:15: From Sydney—"At Home and Abroad," a News Commentary by "Spectator." 12:30: From Adelaide—News Service. 12:35: Weather, River, and Grain Reports. 12:45: Official Stock Exchange Information. During the afternoon, morning services from Strathalbyn and Geelong will be provided, with music from the Studios. 5:30: The Children's Session, conducted by "Marie Louise," with "Jeanette" and "Old Pal." 6:0: Dinner Music. 6:15: The 5CL Bluebird Club, conducted by The Bird Lady. 6:30: From Perth—National Talk, Three Modern Arts—"The Art of Listening," by Professor Walter Murdoch. 6:45: From Adelaide—Sporting Results. 6:50: From Sydney—National News Bulletin. 7:0: From Adelaide—Evening News Service. 7:5: Official Stock Exchange Information. 7:15: Interlude. 7:22: Cable Market Reports. 7:30: From Sydney—National Programme—The National Military Band, conducted by Stephen Yorke. 8:0: "Those Were Hits," a feature presentation, by Jim Davidson's A.B.C. Dance Band. 8:30: From Adelaide—Regional Programme—Recital, by Paul Vinogradov (distinguished Russian Pianist), "Sonata Appassionata in F Minor," Opus 57 (Beethoven). 8:50: Interlude. 9:0: From Torch Theatre—The A.B.C. (Adelaide) Theatre Orchestra, conducted by William R. Cade, with Thea Phillips—Operatic Repertoire. 10:0: From Studios—"Songs of the Sea," a Recorded Interlude. 10:15: News Service and Weather Notes. 10:23: Dance Music. 11:28: Signature Melody. 11:30: Close. Thursday, Feb. 11 5CL Adelaide, 730 k.c. 7:0 News and Weather. 7:10 Bright Morning Music. 7:20 News and Items of Interest—Shipping, Mail, Traffic and Market Reports. What's on Today? The Sporting Diary. 7:30 From Sydney—Bright Morning Music and News Notes. 8:30 From Adelaide—Items of Interest. 8:40 Music. 9:0 Close. 10:31 Daily Broadcast Service under the auspices of the Congregational Churches. Preacher: Rev. S. Martin for Presbyterian. 10:45 Housewives' Morning Tea Session. 11:0 Topical Gleanings," by "Lady Tulliver." 11:15 Choral Interlude, by the B.B.C. Wireless Chorus. 11:30 From Melbourne—Community Singing. 12:0 From Adelaide—Educational Broadcast, "Steps for Little Folk," by Miss M. Wauchope, M.A., Dip. Ed. 12:15 From Sydney—"At Home and Abroad," a News Commentary by "Spectator." 12:30 From Adelaide—News Service. 12:38 Weather, River, and Grain Reports. 12:45 Official Stock Exchange Information. 12:55 Luncheon Musicale. 1:57 Weather, River, and Grain Reports. 2:0 Children's Session. 2:11 Four Ways Suite" (Eric Coates). 3:15 Educational Broadcast, "English Speech," by Mr. D. A. David, M.A. 3:30 Brahmsiana. 4:0 "Aboard the Windjammer," with Stuart Robertson and Male Chorus. 4:27 News Flashes. 4:30 Close. 5:0 From Children's Session, conducted by "Marie Louise," assisted by "Dilly" and "Rosie." 5:0 Dinner Music. 6:15 Boy Scouts' Corner. 6:25 Sporting Results. 6:30 Music. 6:30 From Sydney—National News Bulletin. 7:0 From Adelaide—Evening News Service. 7:15 Official Stock Exchange Information. 7:15 Market Reports. 7:30 Interlude. 7:30 From Sydney—National Programme: A Light Orchestra Programme, by the B.B.C. String Orchestra, conducted by Murison Bourn. 8:0 From Adelaide—Regional Programme—Interlude. 8:30 Recital, by John Robertson, Trumpet Virtuoso. 8:55 Interlude. 8:55 From Adelaide—National Programme: National Talk. 9:00 Salon Recital, by Frederick Williamson, Tenor; Constance Pether, Flautist, and Vina Harden, Pianiste. Piano—Scherzo in C minor, Mendelssohn; Tempest, Arrangement of Songs by Roger Quilter. 1. Go Lovely Rose; 2. O Mistress Mine; 3. To Daisies. 4. Concertino for Flute and Piano (Chaminade); 2. Pastorelle and Bourée (M. German). 9:40 Regional Programme—A Feature Record, "The Nightingale," by Victor Arden in Retrospect, including Elisabeth Bertheng, Eileen Joyce, Ezio Pinza, Dr. Malcolm Sargent, Dorothy Heimrich, and Percy Grainger. 10:15 News Service and Weather Notes. 10:23 Modern Dance Music. 11:38 Signature Melody. 11:30 Close. 5CK Crystal Brook, 640 k.c. 7:0 News and Weather. 7:10 Bright Morning Music. 7:20 News and Items of Interest—Shipping, Mail, Traffic and Market Reports. What's on Today? The Sporting Diary. 7:30 From Sydney—Bright Morning Music and News Notes. 8:30 From Adelaide—Items of Interest. 8:40 Music. 9:0 Close. 10:31 Daily Broadcast Service under the auspices of the Congregational Churches. Preacher: Rev. S. Martin for Presbyterian. 10:45 Housewives' Morning Tea Session. 11:0 Topical Gleanings," by "Lady Tulliver." 11:15 Choral Interlude, by the B.B.C. Wireless Chorus. 11:30 From Melbourne—Community Singing. 12:0 From Adelaide—Educational Broadcast, "Steps for Little Folk," by Miss M. Wauchope, M.A., Dip. Ed. 12:15 From Sydney—"At Home and Abroad," a News Commentary by "Spectator." 12:30 From Adelaide—News Service. 12:38 Weather, River, and Grain Reports. 12:45 Official Stock Exchange Information. 12:55 Luncheon Musicale. 1:57 Weather, River, and Grain Reports. 2:0 Children's Session. 2:11 Four Ways Suite" (Eric Coates). 3:15 Educational Broadcast, "English Speech," by Mr. D. A. David, M.A. 3:30 Brahmsiana. 4:0 "Aboard the Windjammer," with Stuart Robertson and Male Chorus. 4:27 News Flashes. 4:30 Close. 5:0 From Children's Session, conducted by "Marie Louise," assisted by "Dilly" and "Rosie." 5:0 Dinner Music. 6:15 Boy Scouts' Corner. 6:25 Sporting Results. 6:30 Music. 6:30 From Sydney—National News Bulletin. 7:0 From Adelaide—Evening News Service. 7:15 Official Stock Exchange Information. 7:15 Market Reports. 7:30 Interlude. 7:30 From Sydney—National Programme: A Light Orchestra Programme, by the B.B.C. String Orchestra, conducted by Murison Bourn. 8:0 From Adelaide—Regional Programme—Interlude. 8:30 Recital, by John Robertson, Trumpet Virtuoso. 8:55 Interlude. 8:55 From Adelaide—National Programme: National Talk. 9:00 Salon Recital, by Frederick Williamson, Tenor; Constance Pether, Flautist, and Vina Harden, Pianiste. Piano—Scherzo in C minor, Mendelssohn; Tempest, Arrangement of Songs by Roger Quilter. 1. Go Lovely Rose; 2. O Mistress Mine; 3. To Daisies. 4. Concertino for Flute and Piano (Chaminade); 2. Pastorelle and Bourée (M. German). 9:40 Regional Programme—A Feature Record, "The Nightingale," by Victor Arden in Retrospect, including Elisabeth Bertheng, Eileen Joyce, Ezio Pinza, Dr. Malcolm Sargent, Dorothy Heimrich, and Percy Grainger. 10:15 News Service and Weather Notes. 10:23 Modern Dance Music. 11:38 Signature Melody. 11:30 Close. 2FC 7.40 A.B.C. (Sydney) WIRELESS CHORUS Conducted by JOHN ANTILL Little Miss Muffet ........................................... Alfred Wheeler Gipsy Love Song ................................................ Victor Herbert Sylvia ............................................................... Oley Speaks Hungarian Dance No. 6 ........................................... Brahms-Walts Tales from the Vienna Woods ................................. Strauss-Lavater 2FC 8.0 HOTCH POTCH FOR THOSE WHO LIKE IT A Musical and Dramatic Miscellany SOMETHING FOR— HIGHBROW, LOWBROW, MIDDLEBROW and NOBROW Compered and Produced by RUSSELL SCOTT 2FC 8.30 PROGRAMME By THE RAGGLE TAGGLE GYPSIES Tschirchacha .................................................. Arr. by Leader of Band Tango Argentinia Mendoza .................................. Llosas Czardas .......................................................... Kempner Santuzza .......................................................... Berco Natalia ............................................................ Hahn-Mc Hublitzki ......................................................... Arr. Leader of Band Two Guitars ..................................................... Arr. Leader of Band 2FC 9.0 FROM THE HILL BILLY CABIN Fifteen Minutes of Song and Story from the Mountains of Virginia Featuring COLIN CRANE With Assisting Harmonists Arranged by REG. STONEHAM 12.0: Stock Exchange, First Call. 12.7: A Pianoforte Recital (r.) by RAIE DA COSTA— Rigoletto Paraphrase (Verdi-List) Sing Brothers ... (Tunbridge) VIVIAN ELLIS—Public Nuisance No. 1. 12.30: Vocal and Instrumental Music for the Dance Hour (r.). 1.30: Radio Rhythm (r.). 2.10: Music of the Moment (r.). RADIO MATINEE: 2.30 (Approx.): FROM THE NEPEAN RIVER— Descriptions of the Eight-Oar Championships of New South Wales. (Interspersed with Recorded Musical Items from the Sydney Opera House.) 5.0 (Approx.): FROM THE STUDIO— Music and Interlude (r.). 5.15: Complete Racing Resume. 5.33: Weather Information. 5.30: Close. EARLY EVENING SESSION 6.0: Musical Items (r.). 6.20: WORKING SESSION—Results and Comments. 7.25: NATIONAL NEWS BULLETIN 7.30: Local News. 7.35: Weather Report, Stock Exchange Report and Tasmanian Police Report. 7.45: David G. Stead will speak on "World Peace Congress at Brussels." 7.55: What's on the Air To-night? 2NC NEWCASTLE, 1230 K'cycles [Aust. B'casting Commission] 7.0 to 9.0: Relayed from 2BL 10.0 to 10.30: Relayed from 2FC 10.30 to 10.40: News Service, by courtesy of the "Newcastle Morning Herald." 10.40 to 11.30: Relayed from 2FC. 11.30 to 12.00: Relayed from 2FC. 5.30 to 6.0: Relayed from 2FC. 6.0 to 7.45: Relayed from 2BL. 7.45 to 7.58: Local Sporting Results. 7.58 to 11.30: Relayed from 2BL 2NR GRAFTON, 700 K'cycles [Regional Station] 7.0 to 7.20: Relayed from 2FC. 7.25 to 9.0: Relayed from 2BL 10.0 to 11.30: Relayed from 2FC. 2FC 9.15 FATHER BROWN By G. K. CHESTERTON THE FIFTH CASE THE ABSENCE OF MR. GLASS Adapted for Radio by MAX AFFORD CHARACTERS: DR. ORION HOOD, Criminologist and Specialist on Moral Disorders. MRS. JANET McNAB. MAGGIE McNAB, Her Daughter. JAMES TODHUNTER, Fiance of Maggie. MR. GLASS, a Creature of Evil. FATHER BROWN. Scene: Scarborough, England. Production: JOHN CAIRNS. 2FC 9.50 RECITAL BY ALICE PROWSE, Contralto. PHYLLIS McDONALD, Violinist. MARJORIE HESSE, Pianist A PROGRAMME OF MUSIC BY AUSTRALIAN COMPOSERS PIANOFORTE— Reverie ......................................................... Frank Hutchens Scotsch Air (Was's Me for Charlie) .................. Alfred Hill A May Day ..................................................... Roy Agnew Twilight ......................................................... Marjorie Hesse The Piper. VOCAL— The Isle of Innisfree ........................................ Ramsey Peacock In Early Green Summer ..................................... Marjorie Hesse O Men from the Fields ...................................... James Bresh Moorland (with Violin Obbligato, played by Phyllis MacDonald). VIOLIN— Always Afternoon ........................................... Frank Hutchens Caravan ......................................................... Alfred Hill Molly on the Shore .......................................... Percy Grainger 2UE SYDNEY, 950 K'cycles [Commercial Station] Day Sessions as Friday, Except: 8:00: Randwick Track Highlights 8:30: Famous Selections 8:30: News from "The Daily Telegraph." 8:45: Songs for Mother 9:00: Down Home Music 9:15: Racing Topics 10:00: Greetings and Club Messages—Mrs. Filmer. 11:00: Close. 12:00: DESCRIPTION OF THE MOOREFIELD RACES BY HAROLD PARKINSON. In Scripting of the Races WILIAMSTOWN BY FRED FREEDMAN. Replied from 3WF (MELBOURNE). 5:00: Dance Rhythms of Today 5:30: With the Light Orchestras 5:45: Rhythms of Yesterday—Footlight Parade 6:00: The Virginian Gentleman 6:15: Kiddie Kapers 6:30: Dinner Music 6:40: COMMENTARY ON THE DAY'S RACING—H. C. COLLINS. 7:00: Sporting Review and Summary by Sydney Allison 7:12: Box Positions from Harold Park 7:30: DESCRIPTION OF THE GREYHOUND RACING FROM HAROLD PARK. 7:33: Music 7:45: Second Race 7:50: Home Sweet Home 7:55: The Singing Troubadour 8:00: Third Race 8:4: Dance Moods 8:16: Fourth Race 8:19: Music of Today 8:32: Fifth Race 8:35: Music that Cheers 8:46: Sixth Race 9:00: Bright Music—Specially Arranged for Dance Lovers 9:04: Seventh Race 9:08: Dance Music 9:20: Eighth Race 9:23: Dance Music 9:26: Ninth Race 9:30: Dance Music 9:32: Tenth Race 9:35: Dance Music 10:0: Eleventh Race 10:12: Dance Music 10:30: Twelfth Race 10:30: Froth and Bubble 10:45: Continuous Dance Music 12:00: Close. 2GB SYDNEY, 870 K'cycles [Commercial Station] 6:00: Morning Session, conducted by Jack Lawrence and Noel Jude 6:45: News 7:00: Albert Russell and Reg Morgan 8:00: Around the Breakfast Table 8:15: Bluebird Session by Auntie Val 9:00: UNCLE FRANK'S CHEER-UP SESSION, CONDUCTED BY UNCLE FRANK. 10:00: Musical Promenade, conducted by Jack Davey. Overture 11:00: Hall of Song 11:45: Music 1:00: Music 1:15: Tunes from the Radio Library 1:30: Music 1:45: THE GARDEN BEAUTIFUL—MR. S. H. HUNT. 2:00: New Tunes for Old—Serenade; What a Difference a Day Made; Maybe On in the Old Again; Spanish Serenade 2:15: Special Session of New Recordings 2:30: George Somers conducts the Afternoon Session 2:45: ARTISTS AT HOME 3:00: FRED WARING AND HIS PENNSYLVANIANS 3:30: Jewel Box—Traviata Prelude; Pourquoi Valsante; La Danse des Demoiselles 3:45: Dance Rhythms 4:00: FOR MEN ONLY—AIE YOUR GROUCH 4:30: Music 4:45: Annie Goodie—Songs to the Memory 5:00: Children's Sextet, conducted by Uncle George 5:15: Uncle Jerry 5:30: King's College 5:45: Grand Finale 6:00: Special Session 6:15: Feature Session, presenting the Music Box 6:30: Final Sporting Resume—Mr. Oscar Lawson 6:55: Tales Told to Peter and Pam 7:00: A B.B.C. Production 7:15: Debroy Somers Band—Evergreen, Schubert, Humoresque 7:30: Schmidt Tranz—One Life, One Love (May-Neunacht), Don Miguel and his Cuban Music—El Sombrero de Copas Azules, coupled with Lorand and her Viennese Orchestra—Dancing Demoiselle (Leo Fall) 7:30: Bill Mack and Jimmie 7:45: World Revue—Flight of the Humble Bee; Spring Song; Zigeuner; You're Everywhere; Sunny South 8:00: Sir Henry J. Wood and Orchestra—Prelude to the Dance, John Charles Thomas, Baritone—Mah Lindy Lou (Strickland), Great Britain Mascot—Song of the Islands (King Murgatroyd) and Winterbottom—Music (Crickhambly—Bunkau). New Symphony Orchestra, conducted by Dr. Malcolm Sarzant—Merrymaker's Dance, from Nell Gwyn Dance (Grimm) 8:30: The McCook-Mendel Agency 8:35: Debroy Somers Band—Out of the Bottle, Sel (Levant-Ellias), I've Got a Quilt—Flemenco; What a Little Moonlight Can Do; Castillian Moonlight; Sugar Plum; Lady of Spain 9:00: THE HEART OF JACK—A FUNNY STORY SHOW 9:30: Feature Session, presenting Grand Hotel Orchestra, with Forces Langford, Vocalist, Grand Hotel Orchestra, conducted by Leslie Jeffreys—Dance of the Icicles (Kennedy-Russell), Frances Langford, Vocalist—I Don't Want to Make History (Haider), Grand Hotel Orchestra, conducted by Lesley Jeffreys—Musical Comedy Gems, Frances Langford, Vocalist—Speaking Confidentially (McHugh-Fields) 9:45: MERRY-MAID Moods—Children's March (Wolff), Blue My Bubblichi; My Heart at Thy Sweet Voice; Breakin' in a Pair of Shoes; Pittsburg Medley 10:00: Frivolities 10:15: The Bohemians, Light Orchestra—Won't You Buy Me? (Kern-Warren), Fraser Garage, Baritone—Shower On, My Little Gypsy Sweetheart 10:45: Special Session of New Recordings 11:00: DANCE SESSION, CONDUCTED BY JACK DAVEY 12:30: Close 2SM SYDNEY, 1270 K'cycles [Commercial Station] Day Sessions as Friday, except: 8:45: Preliminary Talk on To-day's Races at Moorefield 12:00: RACING SESSION—During the afternoon, a description in running will be given of the Races from Moorefield, with Music. Replied to 2GZ 9:00: Uncle Tom and his Gang 9:15: Angelus 9:30: Uncle Tom's Serenade 10:00: Re-broadcast of to-day's Races, with Correct Starting Prices 10:15: Music 10:30: Jim Sampson, Jazz Pianist 10:45: Resume of to-day's Field Sports—S. Nagle 11:00: Our Domestic Talks 11:45: Patterns in Harmonies 12:00: Selected Musical Items 12:30: "The Grocer Entertains" 9:00: RADIO DANCE NIGHT 10:00: Reception of the Week—A Review 10:25: When the Crowd is Off the End of the Day—Paul Oliver 10:30: Close 2UW SYDNEY, 1110 K'cycles [Commercial Station] Day Sessions as Friday, except: 4:45: Junior Farmers' Session—J. Crawford 5:00: Sporting Preview by 2UW Sporting Editor, with suggestions for Moorefield and Williamstown 5:30: Greyhound Form and Selections—Cyril Angles 10:00: Track Gallops 10:20: Around the Band Stand 11:15: Dance Music 11:20: Light Orchestral Music During the afternoon, 2UW Surf Patrol 2KY SYDNEY, 1020 K'cycles [Commercial Station] 6:45: Weather Report, News, and Commentary—Bert Beaver 7:00: Musical Programme 7:15: The World Today in Boxing 7:30: Music SATURDAY, FEBRUARY 6 10:00 FROM EMPIRE THEATRE—2KY RADIO PARTY WITH ALGY, RION AND KATHY SNARES, KIDDIE KIDS AND A BIG PROGRAMME OF NOVELTIES. 11:00 Around the English Inns with Gwen Lark. 11:35: Close. 12:00 BROADCAST OF RACES AT MOOREFIELD—IAN GARDEN. 1:15: Labor News Commentary. 5:00 Algy Entertains. 5:30 Dinner Divertisements. 6:00: Stable Spy. 6:15: Boxing, with Charles Lucas. 6:30: Dinner Entertainment—Goode and Uncle Bert. 6:37: Spot of Humor. 6:45: THE DANCE MUSIC OF JACK SPOONER'S BAND. 7:15: Labor News Commentary. 7:20: Resume of Day's Sports—Andy Flanagan. 7:30: News. 7:45: COMMENDING AT 7:45 AND CONTINUING THROUGHOUT THE EVENING, A RUNNING DESCRIPTION OF GREYHOUND EVENTS FROM HAROLD PARKER'S GARDEN. 5:00: Dinner Music. 9:00: BROADCAST OF WRESTLING FROM SYDNEY STADIUM—DESCRIBED BY RION VOIGHT. 10:30: Happy Hour. 12:00: Close. 2GZ Central N.S.W. 990 K'cycles [Commercial Station] 6:30: Music. 6:35: Weather. 6:40: Market Resume. 6:45: Producers' Service Session—Poultry, Eggs, etc. 7:00: Producers' Calendar. 7:15: News and Weather. 7:15: Mr. Stop's Morning Service. 8:00: Programme Projects. 8:15: Poppy's Tea-Party. 8:30: British Official Wireless News. 9:30: Close. 11:00: Women's Morning Session. 12:55: Programme Projects. 1:00: MOOREFIELD RACE DESCRIPTIONS. RESULTS OF MELBOURNE RACES. Music. 1:15: British Official Wireless News. 1:15: Weather. 5:0: Popular Recordings. 6:00: Dinner Music. 7:0: Sporting Resume. Results of day's races. Country Sporting News. 7:30: Weather. 7:30: Popular Recordings. 8:00: Music. 8:45: Dance Music. 9:15: DESCRIPTION OF WRESTLING FROM RUSHCUTTER'S BAY STADIUM. Music. 10:30: Close. 2HD Newcastle 1140 K'cycles [Commercial Station] 6:00: Musical Clock. 6:15: Early Morning Melodies. 6:25: Morning News. 6:45: Morning Meditations. 7:10: More News. 8:00: Joy Club Session. 8:15: Variety Music. 1:00: BROADCAST DESCRIPTION OF MOOREFIELD RACES. 5:15: Variety Entertainment for the Kiddies. 6:00: The Dinner Music Overture. 6:35: News and Theatre. 6:30: Lechritette. 6:45: Joyster Note and News Flashes. 6:40: Mr. and Mrs. Everybody. 6:55: Peter and Pam. 7:10: Alan Rippon Talk. 8:00: People's Programme. 10:00: News. 10:10: 2HD Radio Vaudeville and Old Time Dance Music. 12:30: Close. 2KO Newcastle, 1410 K'cycles [Commercial Station] 10:00: Weather Report. Studio News and Open Number. 10:30: Radio Trade Demonstration. 10:30: The Flame of Service. 11:00: Fifteen Minutes of Brightness. 11:30: Music. 12:00: Close. 2:00: Matinée, with Benny and Bubbs. 2:30: The Peabody's written and read by Vernon Little with Mora Downie. 2:45: Rhythm Is Our Business. 3:00: Vaudeville of the Air. 3:30: Matinee, Group 4. The Big Parade. 4:45: Songs My Mother Taught Me. 5:00: Early Evening Programme. 5:45: Birthday Calls. 6:45: Sports Resume, by Allan Webber. 7:15: Music Lessons. 8:00: Melodious Moments. 8:15: Songs of Yesteryear. 8:30: DESCRIPTION OF THE MAIN CONTESTS AT THE NEWCASTLE STADIUM, BY STAN HEANEY. 11:00: Close. 2CA Canberra, 1050 K'cycles [Commercial Station] 6:00: Music. 6:10: Children's Birthday Cheerio. 6:15: Popular Recordings. 6:25: What's On in Canberra? 6:30: District Sports Results. 7:00: Vocal Half Hour. 7:30: Famous Dance Bands. 7:45: Hot and Hotter. 8:00: Famous Orchestras. 8:30: Popular Old and New. 8:45: Instrumental Quarter Hour. 9:00: New Record Releases. 9:15: Dance Music. 10:00: Announcements and Close. 2LM Lismore, 900 K'cycles [Commercial Station] 7:00: Jim Sharp, the Early Bird. 7:30: Wake Up Session. 7:30: A Charivari, Mirth, Melody, and Music. 8:00: The Lismore Week's Happenings. 8:45: Music. 9:00: Close. 1:00: Molly's Shunny Session. 1:30: Music As You Like It. 2:00: Gracie Fields. 2:30: Radio Matinees. 3:00: Close. 2MO Gunnedah, 1360 K'cycles [Commercial Station] Day Sessions as Friday. 6:00: Musical Presentation. Serial Story. 6:15: Birthday Calls and Cheerios. 6:30: Entertainment, News and Bright Music. 7:00: Outing Music. 7:15: Dance Music. 7:30: Sporting Results. 7:35: Stock Sales. 7:40: Old-Time and Jazz Dance Music (contd.). 9:00: Happy Hour. 9:30: Weather Report. 10:00: Close. 2TM Tamworth, 1300 K'cycles [Commercial Station] Morning Session as Friday. 11:30: Musical Presentation. 12:00: Luncheon Music. 1:15: What's On. 1:27: Radio March. 1:30: Close. 5:15: Pussum Club conducted by Uncle Patsy. Uncle Bud and Rustin. 6:00: Dinner Music. 6:30: Sporting Commentator. 7:00: Popular Music. 7:15: Miniature Variety Programme. 7:45: Uncle Bud's Radio Session. 2:00: Radio Rhythms. 8:15: Piano Popularities. 8:45: Weather Reports. 9:00: Entertainment for Everyone. 9:15: Dance Programme. 11:00: Close. 2KA Katoomba, 1160 K'cycles [Commercial Station] Morning Sessions as Usual. 6:00: Children's Studio Concert. 6:30: Sponsor's Session. 6:45: Dinner Music. 9:00: Rhythm Highlights. 7:30: Vocal and Instrumental Music. 8:30: A Spot of Humor. 8:45: A Show. 9:00: 2KA Radio Dance Night. 10:28: Weather. 10:30: Close. 2AD Armidale, 1080 K'cycles [Commercial Station] 11:30: Programme Review. 11:35: Race News. 11:45: News. 11:15: In Town To-night. 1:30: Close. 12:00: Dinner Music and Birthday Calls. 6:55: In Town To-night. 7:00: Talk by Mr. Ferguson, Fruit Inspector. 8:00: Listening-in on Hollywood. 8:45: Favorites, Old and New. 9:00: The Music Lovers' Session. 9:15: Dance Music. 9:30: Weather; Markets, News, Cabaret Cameo. 10:30: Close. 2GF Grafton, 1210 K'cycles [Commercial Station] 7:30 to 8:30: Merry Morning Session, with Community Singers. 12:00: Weather and Markets. 12:30: Music. 12:45: Musical Box. 1:30: Music. 2:00: Close. 5:30: Kitty Uncle Cot and the Golliwoggs' Orchestra will entertain the Children. 6:30: Music. 7:15: Sporting Observations. 8:30: Listening-in on Hollywood. 8:40: Music. 9:30: Mentors Lane with Historical Talk. 10:00: Music. 10:30: Close. 2GN Goulburn, 1390 K'cycles [Commercial Station] 7:00: Jim Sharp, the Early Bird. 7:30: Wake Up Session. 7:30: A Charivari, Mirth, Melody, and Music. 8:00: The Goulburn Week's Happenings. 8:45: Music. 9:00: Close. 1:00: Molly's Shunny Session. 1:30: Music As You Like It. 2:00: Gracie Fields. 2:30: Radio Matinees. 3:00: Close. 2RG Griffith, 1470 K'cycles [Commercial Station] 6:00: Children's Session. 6:30: Music Lovers' Potpourri. 7:00: Songs at the Piano, by Cyril James. 7:15: Suspense of Comedy. 7:30: Market Reports. 7:45: Pianoforte Recital. 8:30: One Hour Dance Programme. 10:30: Goodnight Song. SATURDAY, FEBRUARY 6 2AY ALBURY, 1480 K'cycles [Commercial Station] Mid-day Sessions as Friday. 5:30: Children's Session. 6:0: Dinner Music. 7:0: Station Features. 7:20: Sporting Session. 8:0: Gem of the Evening. 8:30: Old Time Dance Programme. 10:30: Interlude. 10:45: Late Sporting Session. 10:50: Old Time Dance Programme. 12:0: Close. 3LO MELBOURNE, 770 K'cycles [Aust. B'casting Commission] Day sessions as Friday, except: 10:10: Current Happenings in Sport by A.B.C. Commentator. 1:20: Radio Matinee of Specially-selected Recorded Music. 5:0: Close. 5:30: Young People's Session. Overture. Sporting Results from Simon Kenton "and Corkin's Birthday"—"The Pilgrims," by Irene Skatchiok, Music, "Wing Tips," by the Group Captain Wigley, of the R.A.F. Final. 6:0: Dinner Music. 7:40: A.B.C. (Sydney) Wireless Chorus (see 2FC). 8:0: "Hotch Potch" (see 2FC). 8:30: Interlude. 9:0: Paul Vingadoff (see 2FC). 9:0: "From the Hill Billy Cabin" (See 2FC). 9:15: "Father Brown" (see 2FC). 9:45: Interlude. 6:0: Lismore Choral Choristers, conducted by Edward Hocking. Solists: Warwick Kitson, piano; Allan Richards, Boy Soprano; Eddy and Bunny Hocking, piano duets; Dorothy Robinson, soprano. Choir—Tales from the Vienna Woods arr. Louis LaRue. Country Places (Margaret Sutherland), piano. The Girl in the Bathing Pool (Frank Hutchison). Ballad—Soprano—Hedie Roses (Schubert). Little Boy Blue (Eth. Nevin). Clouds—Quartet for the May John Taite); Gentle Maiden (Louis van Water). Piano Duet—Morning Song (Maud Green). Soprano Solo—The Thrush (Edith Haynes). When the Time, Blackbird? (Herbert Oliver). Choir—To the Night (A. E. Floyd); Cradle Song (J. A. Steel). 10:30: Cricket—Detailed Scores of To-day's Play. News Brevities, Summary of the Day's News. 10:40: Interlude. 10:55: Dance Music by Jim Davidson's A.B.C. Dance Band. 11:30: Close. 3GI SALE, 830 K'cycles [Regional Station] 7:0: See 3AR. 7:37: See 3LO. 8:0: See 3AR. 9:0: Close. 10:0: See 3LO. 11:30: Close. 12:0: See 3AR. 12:30: See 3AR. 5:0 (approx.): Close. 5:30: See 3LO. 6:0: See 3AR. 6:15: Australian Sporting News. 6:30: See 3AR. 8:0: See 3LO. 11:30: Close. 2CO COROWA, 670 K'cycles [Aust. B'casting Commission] 7:0: Music. 7:30: Weather, River, and Grain Reports. During the afternoon, Racing Services from Morphettville and Warrnambool will be broadcast, with Music from the Studios. 5:30: The Children's Session, conducted by Marie Louise, with Austin Ellis. 6:0: Dinner Music. 6:40: To-day's Tennis, by Ken Beriman. 6:45: Bowling Results. 6:55: From Sydney—National News Bulletin. 7:0: From Adelaide—Evening News Service and Race Results. 7:10: To-day's District Cricket. 7:15: The SCL Boys' Club Session, interspersed with Trotting Descriptions from Wayville. 8:0: Trotting Descriptions from Wayville, interspersed with Light Orchestral Selections. 9:15: Grand Band Concert by the Hobart Municipal Band, conducted by David Hopkins, with Trotting Descriptions interspersed. 10:15: News Service and Weather Notes. 10:23: Dance Music and Novelty Numbers, with Trotting interspersed. 11:28: Signature Melody. 11:30: Close. 4QG BRISBANE, 800 K'cycles [Aust. B'casting Commission] 7:0: Music. 7:30: Weather, Cable, Racing Anticipations, movements of Air-mail planes. 7:30: Music. 8:0: From 2BL—Morning Music. 8:30: Close. 9:0: Close. 10:0: The Daily Broadcast Service, conducted by Pastor D. R. Stirling. 11:15: Music. 12:0: Close. 12:30: News. British Official Wireless News, Weather Data, Music. During the afternoon descriptions will be given of Moorefield Races, Sydney. Albion Park and Melbourne Race Results. 5:30: Children's Corner, conducted by Cap and Babe. 6:0: Dinner Music. 6:45: Race Results by Keith Neld. General Sports Notes. 7:45: From 2BL—National News Bulletin. 8:30: News. 9:30: Movement of Air Mail Planes. Weather Data, Programme Announcements. 8:0: From 2FC—Hotch Potch. (See 2FC). 8:30: Coffee Stall Chorus, featuring Ian Maxwell and June Carter. 9:0: From 3LO—From the Hill Billy Cabin. (See 2FC). 9:15: From 3LO—Father Brown. (See 2FC). 9:30: The National Eisteddfod of Wales, 1936—A Programme of some of the Successful Competitors. (B.B.C. Recording). 10:0: Interlude. Late Sporting Summary. 10:40: From 2BL—Dance Music by Jim Davidson's A.B.C. Dance Band. 11:30: Close. 4RK RH'TON, Q., 910 K'cycles [Aust. B'casting Commission] 7:0: Relayed from 4QG. 8:30: Close. 11:0: Relayed from 4QG. 11:15: News. 11:30: Relayed from 4QG. 6:0: News. 6:15: Relayed from 4QG. 11:30: Close. 4QN Nth. Regional 600 K'cycles 7:0: Relayed from 4QG. 8:30: Close. 11:0: Relayed from 4QG. 11:15: News. 11:30: Relayed from 4QG. 6:0: News. 6:15: Relayed from 4QG. 11:30: Close. 5CL ADELAIDE, 730 K'cycles [Aust. B'casting Commission] Main Features: 11:55: Williamstown Acceptances. 12:45: Weather, River, and Grain Reports. During the afternoon, Racing Services from Morphettville and Warrnambool will be broadcast, with Music from the Studios. 5:30: The Children's Session, conducted by Marie Louise, with Austin Ellis. 6:0: Dinner Music. 6:40: To-day's Tennis, by Ken Beriman. 6:45: Bowling Results. 6:55: From Sydney—National News Bulletin. 7:0: From Hobart—Vocal Numbers, by Miss Ellen, Carne, etc. A Soft Day (Stanford). Here in the Quiet Hills (Carne). Dawn and Dusk (Forster). 8:0: From Sydney—"Hotch Potch" (See 2FC). 8:30: From Hobart—Light Orchestral Session, by the A.B.C. (Murrayville) Orchestra, conducted by Clive Douglas. Muss Bat. Overture: Vanity Fair (Fletcher). Reverie (MacDowell). Chant Et Danse (Thomson). Suite, from the Ballet, "Prince Ador" (by Cornelius Rybner). Dance of the Flowers, Danse de la Gipsy (Mikurka). Pas de Deux (Polka). Ballabile (Grand Waltz). 9:0: From Melbourne—"From the Hill Billy Cabin." (See 2FC). 9:15: "Father Brown" (See 2FC). 9:45: From Hobart—The Royal National Eisteddfod of Wales. (See 4QG). A Programme of some of the successful competitors, presented from the Studios of the British Broadcasting Corporation. 10:30: Weather. Our Launceston Letter. 10:40: Dance Session. 11:30: Close. --- MAN CAN NOW USE STRANGE INNER POWER The story of a fascinating and amazing teaching which tells of a strange inner power so dynamic and forceful that it may carry man to great heights of happiness, success and achievement is told in the 5,000 word booklet "Triumphant Accomplishments Through Inner Wisdom" recently compiled by Yogi Alpha, noted American philosopher. He tells of his discovery and his belief that all of the laws of the Universe may be controlled because the laws themselves depend upon the great Universal Mind, or God Mind, for their existence. He further teaches that this Divine Power is not limited to a fortunate few, but is latent in every normal human being regardless of training, education or environment. He believes that man is made in the image of God and that the kingdom of heaven, of health, happiness and success is contained within man to be enjoyed today in our present world. The author offers to send this amazing booklet to the readers of this paper free of cost or obligation. It tells of his discovery of the secret which he believes may enable you within the next few months to unlock the reservoir of vast riches within you and explains how you may receive this unusual teaching in your own home. For your free copy, address Inner Wisdom Fellowship, Dept. B, Box 98, San Diego, Calif. Copyright 1936 by Yogi Alpha. 2FC 6.30 NATIONAL TALK A SERMON FOR THOSE WHO MAY NOT LIKE THEM—A THING OF NOUGHT 2FC 6.45 RANDOM PAGES A Recital by ISIDOR GOODMAN, Pianist. The Butterfly Grieg In the Hall of the Mountain Kings (from Peer Gynt) Grieg Arabesque Leschetizky Mazurka in A Minor Chopin Walz in E Flat Chopin Dance of the Gnomes Liszt 2FC 7.0 GEMS FROM FAVORITE OPERAS "CARMEN" Presented by A.B.C. (Sydney) SYMPHONY ORCHESTRA Conducted by PERCY CODE SOLOISTS: EILEEN BOYD (Contralto) WALTER KINGSLEY (Bassoon) LANCE JEFFREE (Tenor) EVELYN LYNCH (Soprano). Chorus of Street Boys. CONTRALTO: Habanera. TENOR and BARITONE—Duet: Speak to Me of My Mother. CONTRALTO: Seguidilla. BARITONE: Torrador's Song. TENOR: Flower Song. 2FC 7.30 CHAMBER MUSIC RECITAL By SYDNEY INSTRUMENTAL TRIO Trio in E Flat, Op. 1 Beethoven Allegro. Adagio Cantabile. Scherzo—Allegro Assai. Finale—Presto. 2FC 8.0 A.B.C. SYMPHONY ORCHESTRA Conducted by PERCY CODE Suite: Les Faunnes de Bourre Humeur Domenico-Scarlatti (The Good Humored Ladies) Presto. Allegro. Andante. Non Presto in Tempo di Ballo. Presto. ORCHESTRA— Song Cycle: Songs of a Wayfaring Man, Gustav-Mahler Soloist: Heather Kinnaird. Symphonie Espagnole, Op. 21 Lalo Allegro Non Tropppo. Scherzando. Intermezzo. Andante. Rondo. For Violin and Orchestra—Soloist: Lionel Lawson. 2FC 9.0 INTERNATIONAL AFFAIRS Professor A. H. CHARTERIS 2FC 9.20 A MUSICAL FILM BROADCAST Presenting LAWRENCE TIBBETT In UNDER YOUR SPELL With WENDY BARRIE, GREGORY RATOFF, ARTHUR TREACHER Based on the Story by BERNICE MASON and SY BARTLETT With Music and Lyrics by ARTHUR SCHWARTZ and HOWARD DEITZ By Arrangement with 20th Century-Fox Film Corp. CAST OF CHARACTERS: ANTHONY ALLEN LAWRENCE TIBBETT CYNTHIA WRIXEL WENDY BARRIE PETE OFF GREGORY RATOFF BOTTS ARTHUR TREACHER COUNT RAUL DU REINNE GREGORY GAYE THE JUDGE BERTON CHURCHILL MR. WREN CLAUDE PRETTY MRS. TWERP CLAUDIA HOLMAN UNCLE BOB CHARLES RICHMAN Arranged and devised by THE AUSTRALIAN BROADCASTING COMMISSION'S HOLLYWOOD REPORTER. SUNDAY, FEBRUARY 7 . . . CONTINUED 6:00 Morning Service—Manly Presbyterian Church. 12:30 Close. 2:55 Temperance Sermon—J. Court. 2:59 Willem Mengelberg and Orchestra—Leonore Overture Beethoven. 3:25 FROM LYCEUM: Pleasant Sunday Afternoon Programme. Conducted by Mr. G. R. DRUMMOND, M.L.A. 4:30 Book Reviews by Rev. W. A. Marsh. 5:00 Children's Session by Aunty Margaret. 5:40 Service for the Sick. 5:58 C.R.E. Session for Young Workers. 6:15 Close. 7:00 Evening Service—St. Mathurin-street Baptist Church. 8:15: Sacred Music. 8:30: Marc Weber and Orchestra—Mazurka (Schumannstich). 8:34: Joseph Schmidt—Tenor—Ah So Pious (Mozart). 8:37: Fritz Kreisler, Violinist—Slavonic Dance No. 1 in G Minor (Dvorak-Kreisler). 8:40: Richard Czarka, tenor—Berceuse de Jourdy (Ganne). 8:43 London Philharmonic Orchestra—Mozart—Piano Suite (Berlioz). March 147. Herbert Gröb and Emyh Meyerbeer—Opera Songs. 8:55 Symphony Orchestra—Concert from London Again Suite (Eric Coates). 9:00 The Bible Man—Rev. A. W. Stuart. 9:18: Richard Tauber, Tenor—Exsult (Ganne). 9:18: Glenn Dyer and the London Philharmonic Orchestra—Symphonic Variations (Caesar Franck). 9:34: Queen Elizabeth Terry, mezzo-soprano—Hand (Puccini). 9:48: Hall Orchestra, conducted by Sir Hamilton Harty—Overture, Rosamunde Schumann. 9:46 Epilogue. 10:00: Close. 2GB SYDNEY, 870 K'cycles [Commercial Station] 8:00 London Philharmonic Orchestra, conducted by Sir L. Ronald-Fuller—4th Movement, Chopiniana Suite (Glazounov). Peter Dawson, bass-baritone—The Anvil House—Airings. Alfredo Campoli and his Orchestra—Columbina (Fontenelle), arr. Crooki. Richard Crooks, tenor—Only My Love (Lefort). 8:15 Masters of Music. 8:30: The Minneapolis Symphony Orchestra, with Amelia Galli-Curci, soprano—Suite in Minnesota. Minneapolis Symphony Orchestra, conducted by E. Ormandy—Scarf Dance from La Source—Cavatine (Delibes). Amelia Galli-Curci, soprano—Paria Valse (Arditi). The Minneapolis Symphony Orchestra, conducted by Eugene Ormandy—Tambourin Chinois (Kreisler). Amelia Galli-Curci, soprano—The Gipsy and the Bird (Benedetti). 8:45 Divertissement: Why Stars come out at Night. 8:50: Rosita, Lazy River, All I Need. 9:00: Keith Collins, pianist. 9:15: Melody Palette. Sleeping Beauty Waltz. 9:20: Laurie Song of Easter. Romany Rhapsody. 9:30 True Confessions. 9:45: Orange Reavers. 10:00: The Hawaiian Club. 10:15 Mr. W. D. Fitzsimmons of the Wentworth District Council of the Parents and Citizens' Association. 10:30: Liberal Catholic Church of St. Alban's Service. 11:45 Questions and Answers by Rev. W. Burt. 12:00: Music. 12:15 Miniature Ballad Concert. 12:45 Luncheon Session conducted by Harry Dearing. New Zealand. 1:15 Sydney Theatre Orchestra—Concerto—Piece selection (Coward). Lillian O'Leary Company—High Society. Vocal Gems. 1:30: Petite Musicals. 1:45: Highlights from Opera. 3:00 The Gleeson Quartet presented by Uncle Frank. 3:15: Castles in Music. Hike, Hike; Poem; White Analysis; On the Road to Mandalay. 3:30 RICHARD WANT, M.A.—The Psychologist's Notebook. 3:30: Islamey, Oriental Fantasy (Balakireff, arr. Casella)—New Symphony Orchestra. 3:39 A Trip to the Zoo—Samuel Biber. 4:00: Massed Military Band—1091 March; Distanti Grenadiers. 4:00 Harold Williams, Baritone—My Sheepdog and I (Thayer); Fire in My Heart; When the Harp's in the Clouds. 4:15: Solo from the Console—Faust, Ballet Music (Gounod, Finlandia (Sibelius). 4:30 Harold Williams, Baritone—My Sheepdog and I (Thayer); Fire in My Heart (Thayer); When the Harp's in the Clouds. 4:35 Music from the Console—Faust—Ballet Music (Gounod). Finlandia (Sibelius). 4:35 Medley of Chappell Band—Money and Boosy Ballads—Jack Flynn and his Orchestra. 4:45: Feature Recordings. 5:00 Bobby and Betty Bluegum. 5:30 The Destiny of the British Empire—The Cameo of the Camera. 6:30 Lightfoot Music. 7:00 Festival in Seville (Albeniz—arr. Rzewuski) — Philadelphia Symphony Orchestra. 7:30: Happenings Ahead. 7:30: TIME MARCHES ON—roduced by C. R. HICKLING. 8:00 The Joy of Living. 8:15 Musical Romeo—Third Dimension Sound. 8:45 GEORGE EDWARDS IN "THE PHANTOM PLANET." 9:15 Theatrical Revue. 9:45: Film Star Parade. 10:00 FOREIGN AFFAIRS—J. M. PRENTICE. 10:10: The Cathedral Close. 10:30: Midnight Lovers. Half Hour. 11:00: Music for Everyone from Everywhere. 12:00 midnight: Vitality—Rhythm—O. Leigh—Interpretation. 12:30: Keeping the Party Going. 1:00: California. 1:30: Winding Up the Clock. 2:00: The Clock. 2:30: Hospital Nurses' Supper Session. 3:00: British and Continental Relays. 3:30: Hospital Children's Bath Tub Session. 2SM SYDNEY, 1270 K'cycles [Commercial Station] 11:00 HIGH MASS—St. Mary's Basilica. 11:20 Sermon—St. Mary's Basilica. 12:13: "MELODIANA," Compered by Dominic Harnett. 1:00 A Composer—His Life and Music—The Musical. 2:00: "Pageant of the Air-waves." 3:30: Famous Symphonies. ★ ★ ★ CERTITUDE RECITAL—George Weir. 6:00: Angelus. 6:1: Uncle Tom and Joan. 6:30 Court Symphony Orchestra—Mikado (Sullivan). 6:35: FATHER WALSH, S.J.—"Night of the World"—Lecturette. 7:00: Father Meaney—Lecturette. 8:30: FOLLIES OF 1937. 9:00 The Saucy Quartette—John Dunne, Fred Webber, Fred Witt, Walter King. 9:15: "THE MARCH OF TIME," produced for Radio by Reg. Hawthorne. 9:30: "The Red Room," featuring Ethel Holden and her Violin. 10:00: "The Dream Ship," piloted by John Dunne. 10:30: Close. 2KY SYDNEY, 1020 K'cycles [Commercial Station] 7:30: Film Fashions. 8:00: Wake Up Session. 8:30: Music. 9:00: Radio Review. 9:30: Humor for the Menfolk. 10:00: Comedy Capers. 10:30: At the Piano with the Melody Girl. 11:00 Carnival Criers. 11:15: Maori Music. 12:00: Music. 12:30: Traders' Topics—John Harper. 1:30: Music for Everybody. 7.0: From 2FC—Gems from Opera: Carmen. (See 2FC.) 7.30 Evening Service from St. John's Anglican Cathedral, Brisbane. 8.45: Weather; A.B.C. News Service. 9.0: From 2FC—Talk by Professor A. H. Charteris: International Affairs. 9.30: From 2FC — Blossom Time. (See 2FC.) 10.15: Poetry Reading. 10.25: Epilogue. 10.30: Close. 4RK R'H'TON, Q., 910 K'cycles [Aust. B'casting Commission] 10.0: Relayed from 4QG. 1.30: Close. 3.0: Relayed from 4QG. 5.0: Close. 5.45: Relayed from 4QG. 10.30: Close. 4QN Nth. Regional 600 K'cycles 10.0: Relayed from 4QG. 1.30: Close. 3.0: Relayed from 4QG. 5.0: Close. 5.45: Relayed from 4QG. 7.30: Evening Service, from St. James' Cathedral, Townsville. 8.45: Relayed from 4QG. 10.30: Close. 5CL ADELAIDE, 730 K'cycles [Aust. B'casting Commission] 10.30: Cathedral Bells. 10.33: Morning Hymn—Brightest and Best of the Sons of the Morning. 11.0: Morning Divine Service, from Maughan Methodist Church. 12.15: From Studios—Interlude. 12.20: From Sydney—News Notes. 12.40: From Adelaide—Classical Carols. 1.0: This Week's Good Cause. 1.5: Classical Cameos. 1.30: Close. 2.1: Pleasant Sunday Afternoon Service from Maughan Church, Chairman: Rev. S. Forsyth. 3.30: From Studios—Light Classical Instrumental Music. 5.0: Close. 5.30: The Children's Session, conducted by The Announcer. 5.45: Sunday Night Story Session, conducted by Mrs. Lindsay Sowton. 5.55: Interlude. 6.0: From Melbourne—National Programme: The Abolition Talk—A Sermon for Those Who say not like Sermons—A Thing of Naught. 6.15: From Sydney—National Programme: Recital from Paris—A Piano Recital by Isidore Goodman. 6.30: Gems From Favorite Operas—Carmen. Alternative Programmes: 7.0: From Adelaide—Regional Programme: The Craftsman of Crewe—A Special Recorded Presentation. 7.30: From Sydney—National Programme: Symphony Hour, with the A.B.C. (Sydney) Symphony Orchestra. 7.5: 5CK—From Adelaide—Regional Programme: Evening Divine Services, From St. Peter's Cathedral. 8.15: From Studios—Interlude. 8.30: 5CL and 5CK—From Sydney—National Programme: A Talk on International Affairs. 8.45: From Adelaide—Land of Song, with Lawrence Tibbett. 9.45: From Adelaide—Regional Programme: A Piano Recital by Spruhan Kennard—(a) Prelude and Fugue in A Minor (Bach-Liszt), (b) Sonatina (Ravel). 10.0: News and Weather Notes. 10.10: Arnold Walters (Hallerton)—A Special Recital, Re-broadcast from the B.B.C., London. 10.20: Musical Epilogue. 10.30: Close. AUSTRALIAN GENERAL ELECTRIC LIMITED Sydney Melbourne Brisbane Adelaide Hobart AMALGAMATED WIRELESS (AUSTRALASIA) LTD. 47 York Street, Sydney 167-169 Queen Street, Melbourne (Advertisement of Amalgamated Wireless Valve Co. Ltd.) In To-night's Programme (r.): MILITARY BAND (see Panel) "JOW" (see Panel) PORTLAND AND LAND (see Panel) NATIONAL DANCE BAND (see Panel) "VARIETY" (see Panel) "INN" (see Panel) Weather Forecast from The Music (r.): ORCHESTRA— Luna (Becca) Piano (Moszkowski) MAN, Violin)— Soleil (Schubert) PHILHARMONIC (Schubert) CASALS (Cello) ... (Haydn) ... (Cassado) EL STRING OR- CHER (Mozart) SYDNEY, 740 K'cycles g Commission] 12:7: Pianoforte Recital (r.) by ARTHUR RUBENSTEIN, with the London Symphony Orchestra—Concerto in A Major . . . (Mozart) Allegro. Andante. Presto. THE LUNCH HOUR: 12:30: Vocal and Instrumental Music for the Lunch Hour (r.): 1:0: MILESTONES AND MELODY (r.): 1:30: Radio Rhythm (r.): 3:0: Programme Music: THE HAYDOCK DUO (Violin and Piano). In association with A. E. Y. BENHAM, Basso. DUO— Legende . . . . (Wieniawski) BASSO— Kurawiwak . . . . (Wieniawski) BASSO— Two Two Grenadiers (Schumann) Myself When Young (Lehmann) DUO— Chanson Hindou . . . (Kriesler) Perpetuum Mobile . . . (Ricci) 3:30: Interlude (r.). 3:35: HALF AN HOUR OF WIT AND HUMOR. Presented by WIN. W. WINDLE. 4:3: Interlude (r.): 4:39: CHAT OVER THE TEACUPS BY NOELLE BRENNAN. 4:35: Trade Demonstration Music (r.): 5:0: Stock Exchange—Third Call. 5:5: Trade Demonstration Music (r.): 5:50: Weather Information. 5:30: Close. 2FC 9.0 NATIONAL DANCE ORCHESTRA Conducted by AL. HAMMETT GRAMME OF THE LATEST DANCE SUCCESSES 2FC 9.30 NON-STOP VARIETY K FIRE VAUDEVILLE PROGRAMME Arranged by SYDNEY HOLLISTER 2FC 10.0 LISTENERS' INN A WAYSIDE WASSAIL Introducing RHYTHMIC REFRESHMENT by the NATIONAL DANCE ORCHESTRA Conducted by AL. HAMMETT A strange tale this—a peep into the parlor of Listeners' Inn, where 'tis said that listeners hear nought but good of themselves; and every man goes out better than he came in. Here we find a simple Landlord, an eloping couple, and a dance orchestra. The Story written by GEORGE MATTHEWS. PRODUCTION: SYDNEY HOLLISTER. MONDAY, FEBRUARY 8 . . . CONTINUED EARLY EVENING SESSION 6.0: Musical Interlude (r.) 6.3: GIRL GUIDES' SESSION—"Brownies." 6.12: THE GUNNER will give a Talk entitled, "Three Men in a Dugout." 6.23: Stock Exchange—Report. 6.25: SPORTING SESSION—Results and Comments. 7.0: "THE PROBLEM OF SLUM CLEARANCE AND HOUSING IN CITIES." MR. W. O. BURT. 7.20: International News. 7.25: NATIONAL NEWS BULLETIN. 7.30: Local News. 7.35: Weather Report; General Market; Produce and Fruit Report and Late Sussex Street Prices, supplied by the State Marketing Bureau; Wool Sales; Truck Booking. 7.58: What's on the Air To-night. EVENING SESSION 8.0: Recital by LIONEL LAWSON, Violinist. (See Panel) 8.15: THE NATIONAL MILITARY BAND. (See Panel) 8.45: OUR FILM REVIEW, by E. J. FRANCIS. 9.0: Interlude (r.). 9.5: Trio Recital. (See Panel) 10.5: Musical Items (r.). 10.29: Weather Information. 10.30: Close. 2BL 8.0 RECITAL BY LIONEL LAWSON, Violinist Nocturne in E Flat ........................................... [Chopin] Allegro ......................................................... [Ficco] Estrellita ....................................................... [Ponce-Heifetz] 2BL 8.15 THE NATIONAL MILITARY BAND Conducted by STEPHEN YORKE Associate Artist: WILLIAM GREEN, Tenor BAND— Overture: Der Freischutz .................................. Weber TENOR— Ah, Moon of My Delight .................................... Lehmann Ay! Ay! Ay! .................................................. Freire The Star ....................................................... Rogers BAND— Ballet Music: Le Prophete ................................. Meyerbeer Valse Pas de la Redowa. Quadrille des Patineurs. Galop. 2BL 8.45 OUR FILM REVIEW By E. J. FRANCIS 2BL 9.5 TRIO RECITAL Arranged by THE BRITISH AND INTERNATIONAL MUSIC SOCIETY EVELYN BLANCHE, Piano DAGMAR THOMSON, Violin FLORENCE FORSHAW, Viola And RAYMOND BEATTY, Bass TRIO— Trio for Piano, Violin, and Viola (In One Movement) Baz BASS— Come to Me in My Dreams .................................. Frank Bridge So Perverse ..................................................... Frank Bridge Wood Magic .................................................... Martin Shaw Wishin' as the Rye ............................................ Peter Warlock VIOLIN— Reve d'enfant .................................................. Ysaye Burleska ......................................................... Suk TRIO— Trio for Piano, Violin and Viola (In One Movement) Jongen Prelude. Variations. Finale. 2UE SYDNEY, 950 K'cycles [Commercial Station] Day Sessions as Friday, except: 12.30: Poultry—Talk by Mr. C. H. Hutton. 2.30: Between Ourselves—An Afternoon Session, conducted by Frank Sturge Hardy. 2.30: Relaxation Music. 2.35: The Radio Adviser on Life's Problems. 3.15: Liling Melodies. 3.30: Story, read by Frank Hardy. 4.0: Selections from Opera. 4.15: Studio Party—Guest of Honor, Miss Bertha Clarke—Japan From Inside. 4.45: New Record Release. 5.0: Dance Rhythms of Today. 5.30: With the Light Orchestra. 5.45: Films of Yesterday—Love in Bloom. 6.0: Dinner Music. 6.15: Trotting Results. 6.20: Dinner Music. 6.35: Sue and Buddy Entertain. 6.45: Dinner Music. 11.30: Our Legally Qualified Medical Practitioner will deliver a Talk on Mothcraft. 4.15: The Friendly Session, featuring The Picture Gallery, conducted by Margaret Herd. 4.55: Children's Session by the Fairy Godmother. 5.30: Peter and Patty. 5.45: The Hello Man's Children's Session. 6.0: Dad and the Girls. 6.15: The Happy Family. 6.25: 2CH Sporting Session—Amateur Athletics Association, conducted by G. C. Crutcher. 6.30: KNIGHT BARNETT at the BURWOOD PALATIAL CINEMA. 6.45: Dinner Music. 6.50: Weather Report. 7.0: The Voice of Henrietta. 7.5: Ramon Novarro, Baritones—El Relicario (Padilla). 7.8: Norman Martin and Accordeon Band—They Came that Morning (Biala) and Sterni. 7.11: Ruby Newman and his Orchestra—Me and the Moon (Hirsch) and Heywood. 7.15: Jimmy Green and Orchestra—Swing Time (Fields and Kern). 7.20: The Crocodile. 7.25: Grand Cinema Organ Solo—St. Petersburg Sledge Drive. 7.30: Further Air Adventures of Jimmy Allen. 7.45: THE CHOCOLATE MINSTRELS, produced by BOB BROTHIER and featuring LES WARTON AND GEORGE JENNINGS. 8.0: Praise and Praise Everybody—featuring Edward Howell and Therese Desmond. 8.12: Orchestra Mascotte—Sicles of Blue (Kutatel). 8.15: Stars of the Network. 8.20: Melody from the Sky. Introduced by Bob Brothier—Out of Heaven, I'm Shootin' High, Moonburn, So This is Heaven. 8.45: THE IN-LAWS. 9.0: Tuneful Tales, presenting My Sweetie, Come Away. 9.15: The Cavalier You Started Me Dreaming, One Rainy Afternoon. 9.15: Music by Candlelight, presenting The White Rose (Rachmaninoff), Lurline Fleming, Soprano; Willa Hindman, Soprano. 9.35: Selection—Mother of Pearl. 9.45: Masters of Music. 10.0: 2CH Sporting Service—Bowling Association News. 10.10: W. Jongs. 10.20: Please Yourself. 10.50: Meditation and Music. 11.0: Close. 2GB SYDNEY, 870 K'cycles [Commercial Station] Day Sessions as Friday, except: 12.15: Housewives' Association Session. Mrs. Ruby Duncan. 4.45: Auld Goodie—Songs to the Time Taps. 5.0: Children's Session, conducted by Uncle George. 5.15: Uncle Jerry. 5.20: Kitty Kottage. 5.35: Music. 5.45: Growing Up. 6.0: The Children's Newspaper. 6.15: The Australian Football League—Under the Sea, by Jules Verne. 6.30: The Stage Orchestra—Waltzes from Vienna Selection (Strauss). Peter Dawson, bass-baritone. 6.45: The Good Green Fields of Home (Kahal-Fain). Gerry Moore, pianist—Fox Trot Medley. 7.0: Philharmonic Orchestra—Turkish March from Ruins of Athens (Beethoven). 6.45: TALES TOLD to PETER AND WENDY—A B.S.A. PRODUCTION. 7.0: Music. 7.15: The Tale Drums. 7.30: Bill Mack and Jimmie. 7.45: Personalities at the Piano. 8.0: Music. 8.5: A Spot of Humor. 8.10: Gems that Thrill You. 8.20: McCOY MENDEL AGENCY. 8.35: Bring a Little Home. 8.45: STRANGE AS IT SEEMS. 9.0: Teach Bearers of History. 9.0: Chimes of the Orient. 9.30: Dramatic Presentation. 9.45: A PAIR OF JACKS—A FUNNY STORY SESSION. MONDAY, FEBRUARY 8 . . . CONTINUED OVERSEAS STATIONS WHAT'S ON THE AIR TO-DAY 5.15 a.m. (DJD 25.4, DJC 49.8, DIL 19.8): Masquerade and Carnival Music and Poetry. 6.0: Little Evening Entertainment. 6.45: Sporting Review. 7.0: News. 7.20: Close. 6 a.m. (2RO 31.19): Variety Programme from Italian Stations. 7.0: News. 7.15: English. 6 a.m. (GSS 31.5, GSD 25.8, GSI 19.6, with GSC 31.3 at 7.0): Religious Service (Methodist) from Bondi Park. 6.45: Inter-vals. 7.0: News. 7.20: The Hunchback String Quartet. 8.5: The Royal Bridgwater Quintet. 8.30: Sports. 8.45: Close. 5.30 p.m. (DJB 19.7, DJN 31.4, DJA 31.3): Orchestral Concert. 6.30: Dance Music. 7.15: Sports Review. 7.30: News in German. 7.45: Children's Session. 8.15: Close. 6 a.m. (GSS 31.5, GSO 19.7): "Talk—This England." 6.40: BBC Empire Orchestra. 7 p.m. (TPAS 25.23): Musical Programme. 7.40: News in English from Paris. 8.0: Close. 7 p.m. (TPAS 31.19): Recreational Music. 9.0: Weather and News. 9.10: Music. 10.0: Close. 7 p.m. (ZBW 31.49): Native Music. 9.0: Programme from Hongkong Studio. 10.0: Weather and Announcements. 10.15: Music. 8.30 p.m. (VPD2 31.45, YDB 31.1, PLP 27.2, PNM 29.2, YDC 19.8): Evening Music Programme. 10.0: VPD2 closes. 9 a.m. (TPAS 19.4): Concert from Radio Paris. 10.0: News in English. 10.15: Music. 9 p.m. (DJB 19.7, DJA 31.3, DJS 31.4, DJE 16.8, DJQ 19.6): Concert from Radio Paris. 10.0: News from Berlin. 10.15: More Light Music. 11.0: News in German. 11.15: Little Evening Entertainment. Midnight: News in English from Paris. 11.45: DJQ closes. 9 p.m. (GSS 31.5, GSO 19.7, GSH 13.9): Big Ben. A Chopin Recital by Poussaint (Piano). 9.30: Points of View by Travellers from the Dominion. 9.45: Haydn Orchestra from Westminster Cathedral, Birmingham. 9.55: "London Pie"—a Radio Revue. 10.25: Memories of the London Theatre. 10.55: News. 11.15: Organ Recital. 11.45: Close. 10 a.m. (WTK 25.19): News of the World from Boston. 10.15: French Morning Music. 11.30 p.m. (PHI 25.57): Dutch National Anthem. Music. 11.40: News from Holland. 11.50: Music. Midnight (2RO 35.4): English Hour. News of the World from Rome. 12.15: Music. Midnight (GSS 31.5, GSP 19.8, GSH 13.9): "John Londeron at Home"—An Evening with His Varied Acquaintances. 2MO GUNNEDAH, 1360 K'cycles [Commercial Station] Day Sessions as Friday. 6.0: Children's Music. 6.3: Serial Story. 6.15: Birthday Calls and Cheerio Session. 6.30: Entertainments and Bright Music. 7.0: Buddies—Thrilling Drama. 7.15: Light Music. 7.30: Market Report and Stock Market. 7.25: Light Music. 8.0: The Happy Hour. 8.30: Weather Report. 9.31: Hollywood Review. 10.0: Close. 2CA CANBERRA, 1050 K'cycles [Commercial Station] 12.30: Lunchroom Music. 12.45: Horoscope Reading. 1.0: Music. 1.30: Close. 3.30: Tiny Tots' Session. 5.45: Children's Story. 6.0: Adventures of Ben, Sam, Gorkey, and Tom. 6.20: Children's Birthday Cheerios. 6.30: What's On in Canberra. 6.35: Dinner Music. 7.0: Vocal Half Hour. 7.45: Famous Dance Music. 8.0: Serenade Mamma. Women conducted by Wyn. Glimmer. 9.0: Famous Orchestras. 9.15: Instrumental Quarter Hour. 9.30: Popular Old Time Music. 9.45: Relaxation Music. 10.0: Announcements and Close. 2GN GOULBURN, 1390 K'cycles [Commercial Station] Day Sessions as Friday. 5.30: Entertainment for the Tiny Tots. 5.45: Serial Story for Children. 6.0: Dinner Music. 7.0: Selections from Talkies. 7.15: Variety Music. 7.30: Popular Releases. 7.45: Radio Rhythm. 8.0: Down Memory Lane. 8.30: Heroes of Medical Science. 8.45: Selected Music. 10.30: Close. 2TM TAMWORTH, 1300 K'cycles [Commercial Station] Day Sessions as Friday, except: 3.30: Afternoon Story. 4.0: Afternoon Selections. 4.45: Medley of All Kinds of Things. 5.0: What's On. 5.15: Possum Club conducted by Auntie Patsy, Uncle Bud and Rambler. 6.0: Dinner Music. 6.30: Popular Music. 7.0: Racing Talk. 7.15: Miniature Variety Programme. 7.45: Uncle Peter's Radio Session. 8.0: Radio Rhythm. 8.15: Kosinsky and Claney. 8.30: Poetics of 1937. 9.0: Weather Reports and Markets. 9.15: Popular Music. 10.0: What's On. 10.27: Spin a Web of Dreams. 10.30: Close. 2KA KATOOMBA, 1160 K'cycles [Commercial Station] 7.25: Breakfast Session. 7.28: Weather Report. 7.30: Close. 6.0: Children's Session. 6.30: Dinner Music. 7.0: Snappy Syncopations. 7.30: Vocal and Instrumental Music. 8.30: Service to Motorists. 8.45: Special Session. 9.0: Feature Record Readings. 9.15: The 2KA Story Man. 9.30: Recent Record Releases. 9.45: Dance Music. 10.10: Slumber Music. 10.26: Weather. 10.30: Close. 2AD ARMIDALE, 1080 K'cycles [Commercial Station] Day Sessions as Friday. 6:05: Dinner Music and Birthday Calms. 7:15: News. 7:30: To-night. 7:0: Rhythm Highlights. 8:0: Heroes of Civilisation. 8:10: Music and Songs of the Masters. 8:30: Post of Call. 9:0: Dance Music. 9:30: News, Markets, Weather. 9:45: Music, Love Songs. 10:0: Slumber Music. 10:27: Tranquillity Tales. 10:30: Close. 2GF GRAFTON, 1210 K'cycles [Commercial Station] 7:30 to 8:30: Merry Morning Session. 12:0: Weather and Markets. 12:30: Music. 1:15: Musical Box. 5:30: Aunt Judy and Uncle Fred will entertain the Children. 6:0: Dinner Music. 7:0: Current Programme. 7:15: Musical Variety. 7:30: Radio Studio Mystery. 8:0: Strange Adventures in Strange Lands. 8:15: Music. 8:30: Mysteries, Crooner (Repeat Guest). 9:13: Your Telling Me. 9:30: Music. 10:30: Close. 2LM LISMORE, 900 K'cycles [Commercial Station] 7:0: Jim Sharp, the Early Bird. 7:5: Op 'Nuff to Get Up In the Morning. 7:30: What's on Tonight? 7:45: The Rambling Pianist. 8:0: News. 8:10: Touring the World. 8:15: Charlie's Comedy. 8:45: More Music. 9:0: Close. 7:0: Jim Sharp Entertains You. 1:30: Songs and Their Singers. 2:0: Helpful Hints. 2:30: Musical Parodies. 3:0: Close. 5:30: Children's Story Book, conducted by Aunt Mollie. 5:45: Cup Session, conducted by Grey Wolf. 6:0: Dinner Music. 6:30: Beau Brummel's Music Box. 7:0: Africa. 7:15: Sporting Session. 7:30: Touring. 8:0: Square Celeste Oct. 9:0: Radio Dance Hour. 9:45: News. 9:50: Birthday Greetings. 10:0: Close. 2RG GRIFFITH, 1470 K'cycles [Commercial Station] 6:0: Children's Session. 6:30: Music. 6:45: Poiroturri. 7:0: Songs at the Piano by Cyril James. 7:15: Snapshots of Comedy. 7:30: Market Reports. 7:45: Local News. 8:30: One Hour Dance Programme. 10:30: Good-night Song. 2AY ALBURY, 1480 K'cycles [Commercial Station] Mid-day Sessions as Friday. 5:30: Children's Session. 6:0: Dinner Music. 6:45: George Osborne on the Radio. 7:0: Stairs. 7:30: Stair Features. 7:6: Rale Langley on Beauty Culture. 7:30: Sporting Session. 8:0: Gens of the Evening. 8:45: The House of Peter MacGregor. 9:0: Heroes of Christmas. 9:15: Country Man's News. 9:30: Tuneful Melodies. 10:30: Close. 3LO MELBOURNE, 770 K'cycles [Aust. B'casting Commission] Day Sessions, as Friday, except: 2:0: The Radio Serenade "Cattle Camp" by J. J. Hasler. Read by Scrible. 10:0: Current Headlines in Sport by A.B.C. Commentator. 10:25: Women in the News. The Week's Health Talk—"Influenza Epidemic." Dr. John Dale. 11:3: Home Exercises. Mr. F. W. Pearce. 12:0: Health Talk—"How the Dutchman made little things look big—the Microscope." By Hipparia. 12:45: From Sydney—At Home and Abroad. By the Contributor. 1:0: From the Constitutional Club—Luncheon Speeches. 3:0: Lectures on Modern History—Mildred Anne's Stripes. 3:15: Philharmonic Symphony Orchestra—Toscanini—Longinuri, Puccini. Chevalier of New York introduced to Act Three (Wagner): Götterdämmerung. Siegfried's Journey to the Well of Wishing. Wagner's Festival Chorus and the London Philharmonic Orchestra—Israel in Egypt—The Land is a Man of War (Handel). Gaetan Cassado, 'cello—Koi Nidrei (Bruch). 3:45: Boston Promenade Orchestra, conducted by Arthur Fiedler. Thais—Meditation (Massenet). Xerxes—Large (Handel). Wan-Lee Larkins, soprano. Harlequin Goldmark Variations. 1 to 12 Bach. London Philharmonic Orchestra, conducted by Sir Hamilton Harry. Punter, March for the last scene of "Hamlet." Op. 14, No. 1. 11:0: Close. 3:30: Young People's Session. Overture. Story Corner, conducted by the Storyteller. Music. Tragedy News from London. Ann. Noel Mewton Wood will play some of his own compositions. Finale. 7:0: Dinner Music. 7:45: Recording Room Music. 7:30: From 3AR—National Talk by W. S. Stuart, Member of the Victorian Housing and Rent Adjustment Board—The Problem of Slum Clearance. Housing in Cities. 7:20: Introductions. 7:25: From 2BL—National News Bulletin. 7:30: Moments of the Day. Main Place, Melbourne. Date, Late Market Reports; General Commercial News; Copra Prices; Irish Storms; Investments; Programmes; Announcements. 8:0: From 2FO. (See 2FO.) 7:0: From The Sales Ensemble. Ensemble Demoule Chic (Pletcher): Chansonne (Primi). Violin Sonatina (Media). Cinderella's Birthday. Procesion (Dicker): Poeme (Fibich): Symphonie (Kracal): Piano: Shepherd Pastoral. Dance du Dineri. Ensemble Sorfly. Unawari (Lincke). In the Middle of a Kiss (Schoo) (Herrmann). 9:0: From 3LO—The National Dance Orchestra, conducted by Al Hammert. 9:30: The Ipswich Vice Regal Band, conducted by J. W. Burmeister. Assisted by Patrick McGonig (Soprano) and Harry Morrallde (Entertainer). 10:0: Dinner Music. Late Sporting. 10:40: Recorded Programme. 11:30: Close. 3AR MELBOURNE, 580 K'cycles [Aust. B'casting Commission] Day Sessions as usual. 8:0: Alice Prisse, Contralto. 8:15: Raymond Dombell, Violinist. Marjorie Hease, Pianist. 8:30: Special Reproductions. 8:45: Interlude. 10:0: News Session. 10:30: Close. 3GI SALE, 830 K'cycles [Regional Station] 7:0: See 3AR. 7:37: See 3LO. 8:0: See 3AR. 9:35: See 3LO. 11:30: Close. 12:0: See 3LO. 1:10: See 3AR. 2:0: See 3LO. 4:15: Close. 5:30: See 3LO. 6:15: Local News Service, including Gippsland Market Reports. 8:30: See 3AR. 8:0: See 3LO. 11:30: Close. 2CO COROWA, 670 K'cycles [Aust. B'casting Commission] 7:0: See 2FC. 7:25: See 3AR. 9:35: See 3LO. 11:30: Close. 12:0: See 3LO. 12:30: Local News. 12:45: See 3LO. 1:10: See 3AR. 2:0: Close. 3:0: See 3LO. 4:15: Close. 5:30: See 3LO. 6:15: Local News Service, including Market Reports from Albury. 8:30: See 3AR. 8:0: See 3LO. 11:30: Close. 4QG BRISBANE, 800 K'cycles [Aust. B'casting Commission] Day Sessions as Friday, except: 11:0: The Daily Broadcast Service, conducted by Canon D. J. Garland. 11:15: The Weekly Book Review, by Mrs. Margaret Mackay. 11:40: Talk by Mrs. Shand Findlay—"Self Expression in Animals." 12:15: Broadcast to Schools by D. S. Drury, B.A., Inspector of School Literary Appreciation: "Persophone." 2:0: Broadcast to Schools, by B. J. McKenna — Current News Bulletin. 3:15: A Programme for the Diggers. 3:30: Story Corner, conducted by Cup and Bebe. 5:30: The Knowledge Man. 6:0: Dinner Music. 6:45: Recording Room Music. 7:0: From 3AR—National Talk by W. S. Stuart, Member of the Victorian Housing and Rent Adjustment Board—The Problem of Slum Clearance. Housing in Cities. 7:20: Introductions. 7:25: From 2BL—National News Bulletin. 7:30: Moments of the Day. Main Place, Melbourne. Date, Late Market Reports; General Commercial News; Copra Prices; Irish Storms; Investments; Programmes; Announcements. 8:0: From 2FO. (See 2FO.) 7:0: From The Sales Ensemble. Ensemble Demoule Chic (Pletcher): Chansonne (Primi). Violin Sonatina (Media). Cinderella's Birthday. Procesion (Dicker): Poeme (Fibich): Symphonie (Kracal): Piano: Shepherd Pastoral. Dance du Dineri. Ensemble Sorfly. Unawari (Lincke). In the Middle of a Kiss (Schoo) (Herrmann). 9:0: From Adelaide—Regional Programme—Vera Interlude. 9:30: From Torrens University—The A.B.C. (Adelaide) Theatre Orchestra, conducted by William R. Cade, with Thea Phillips, Operatic Soprano. 10:0: News and News Notes. 10:30: Modern Dance Music. 11:30: Swing Music. 11:30: Signature Melody. 11:30: Close. 4RK R'H'TON, Q., 910 K'cycles [Aust. B'casting Commission] 7:0: Relayed from 4QG. 9:0: Close. 10:30: News. Music. 11:0: Relayed from 4QG. 2:0: Close. 3:0: Relayed from 4QG. 4:30: Close. 3:30: Relayed from 4QG. News. 6:15: Relayed from 4QG. 11:30: Close. 4QN Nth. Regional 600 K'cycles [Aust. B'casting Commission] Main Features— 1:0: "Drury Lane Memories"—Educational Broadcast, historical play, by Mr. Brian Elliot and Company. 3:0: The Children's Session. 6:0: Dinner Music. 7:30: Swing Results. 8:0: From Melbourne—National Talk—The Problem of Slum Clearance. Housing in Cities. 7:0: From Sydney—National Programme—Vera Interlude. 7:30: From Torrens University—The A.B.C. (Adelaide) Theatre Orchestra, conducted by William R. Cade, with Thea Phillips, Operatic Soprano. 10:0: News and News Notes. 10:30: Modern Dance Music. 11:30: Swing Music. 11:30: Signature Melody. 11:30: Close. ELIMINATE "B" and "C" BATTERIES FROM YOUR SET ESSEX RADIO "B" Batteryless FOR COUNTRY USE No "B" or "C" Batteries are necessary, and drain from the accumulator is considerably less than any other set of similar size. CASH PRICE, £32/10/- And may be purchased on extended terms. NEW ELECTRIC RADIO Models from £12/19/6 New Airzone—Radiola. Philips—Ultimate. Breville—Tasma. Trade your old Set—generous allowance. Easy Terms from 5/- weekly Open Friday Nights. Write, Call or Phone. Phone: B1623. WINKWORTH'S 51 YORK ST., SYDNEY (Top of Wynyard St. G.P.O. tram stop) TUESDAY, FEBRUARY 9 . . . CONTINUED We're Toss on Saturday Night—(Kennedy-Carr) There's a New World—(Kennedy-Carr) O.K. for Sound—(Kennedy-Carr) 4:10: RAMON NOVARRO, Baritone—Long Ago in Alcala (Messenger) 4:13: THE GLEN GRAY ORCHESTRA—Champagne Waltz—(Drake) Moments of Music—(Schnell) 4:19: LOUIS LEVY and his Gaumont British Symphony—Swingtime Selection (Fields-Kern) My, what a different Night—(Gordon-Reely) 4:30: CHAT OVER THE TEACUPS—by NOELLE BRENNAN 4:36: Trade Demonstration Music (r.) 5:0: Stock Exchange, Third Call. 5:5: Trade Demonstration Music (r.) 5:23: Weather Information. 5:30: Close. EARLY EVENING SESSION 6:0: A FINANCIAL SESSION—conducted by Dr. L. D. Woodward. 6:25: Stock Exchange Report. 6:35: SPORTING SESSION—Results and Comments. 6:45: The A.B.C. Racing Commentator. 6:55: SPORTING SESSION (Continued). 7:0: Prof. Henry L. Brose will speak on "Electricity in Our Bodies." 7:15 Interlude (r.). 7:30: NATIONAL NEWS BULLETIN. 7:35: Weather Report, General Market; Produce and Fruit Report and late Sydney Street Prices, compiled by the State Marketing Bureau. Wool Sales and Truck Business. 7:36: What's on the air To-night? 7:36: FROM THE AUBURN TOWN HALL. COMMUNITY SINGING CONCERT. (See Panel) 9:0: Interlude (r.). 9:15: FLORA'S HOLIDAY. (See Panel) 10:0: Interlude (r.). 10:5: THREE GREAT TENORS (See Panel) 10:29: Weather Information. 10:30: Close. 2BL 8.0 FROM THE AUBURN TOWN HALL COMMUNITY SINGING CONCERT Associate Artists: ROBERTS and BROOMHEAD, Entertainers in Harmony and Humor THE BLUE FIVE, Noveltly Instrumentalists And THE SISTERS THREE, Vocal Ensemble 2BL 9.15 FLORA'S HOLIDAY A Cycle of Old English Melodies, Words Written and Adapted and Music Arranged and Composed by H. LANE-WILSON Presented by THE A.B.C. (SYDNEY) ORCHESTRAL AND WIRELESS CHORUS Conducted by PERCY CODE Come, All Ye Lads and Lasses, Lover's Greeting, Tell Me Charming Creature, Gentle Dawn, The Country Dance, Maiden, Beware Ye, Sound Argument, The Pedlar, The Commotion of Love. 2BL 10.5 THREE GREAT TENORS (r.) ANDRE D'ARKOR— Berceuse de Jocelyn . . . Godard Ave Maria . . . . . . . . Gounod JOHN MCMORMACK— Who is Sylvia? . . . . Schubert The Last Hour . . . . Kramer CHARLES HACKETT— Wherever You Walk . . . Handel I Heard a Thrush at Eve . . Eberhart Sturge Harty. 2:30: Meditation Music. 2:45: The Radio Adviser on Life's Problems. 3:10: Pourri. 3:45: Story, read by Frank Harty. 4:15: Musical Impressions. 5:30: Dance Rhythms of To-day. 5:50: With the Light Orchestras. 6:15: News of Yesterday—George White's Scandal. 6:0: Dinner Music. 6:18: Racing Results. 6:30: Dinner Music. 6:45: Racing Preview—Harry Solomon. 7:0: The Sign of the Purple Spider—A B.S.A. Production. 7:15: A Shot of Humor. 7:30: Dance Moods. 7:45: Home Sweet Home. 7:50: The Singing Troubadour. 8:0: Modern Composers. 8:15: Build a Little Home. 8:20: Children's Favorites. 8:45: Racing Talk. 9:0: To-day's New Record Releases. 9:20: A TALK BY REV. CANON HAMMOND. 9:30: Cities of Music—Buenos Aires. 9:45: Fun Rhythm. 10:0: Music of Today. 10:15: KENNETH FOWLES WARRING STORY. 10:30: Out of the Band Box. 11:0: Words and Music. 11:30: Close. Selection, Carmen (Bizet). 9:45: Masters' Music. 10:0: 2CH Sporting Service—Surf Association News by E. C. Sier. 10:5: Mr. Wren and His World, featuring Edward Howell and Les Warion. 10:20: Please Yourself. 10:50: Meditation and Music. 11:0: Close. 2GB SYDNEY, 870 K'cycles [Commercial Station] Day Sessions as Friday, except: 10:30: Dorothy Jordan—"Beauty Through the Ages." 11:0: Dorothea Vautier—"The London Letter." 11:40: Dorothea Vautier—"Overseas News." 4:45: Little Goodie—Songs to the Tiny Tot. 5:0: Children's Session, conducted by Uncle George. 5:15: Uncle Jerry. 5:30: Miss Koffe. 5:35: Music. 5:45: Growing Up. 6:30: Dinner Music. 6:45: Marek Webe and His Orchestra—For You, Rita (Sanginini); John Hendrik, Tender-Me; Dingle, My Blue Heaven; Mello 'Cello. Hal Robinson and His Orchestra—Me and the Moon; Alfred Hauseman; Debroy Somers Bond, with Vocal by Debroy Somers—Nippy Sill, Pt. 2 (Eyton-Mason). 7:30: Music. 7:15: Echoes of the Stars and Scenes—We'll Make Hay While the Sun Shines; Then You'll Remember Me; I'm in the Mood for Love. 7:30: Bill Mack, Jimmie. 7:45: Musical Moments—Featuring Wide Range Orchestras and Vocalists. 8:0: A Spot of Humor. 8:5: Harmony Time. 8:20: McCoy Mendei Agency. 8:35: Build a Little Home. 8:45: Strangers as It Seems. 9:0: Susie and Jack. 9:15: Let's Go Places—Federmann Winkler; Call of the Desert; Dogie's Lullaby; Ainin' the Scotch. 9:30: FREED WARING AND HIS PHILLYVANIANS. 10:0: The World Moves On. 10:15: Frivolities. 10:30: The B.S.A. Dance Orchestra—Music in the Air, Selections. Pt. 2 (Kern), Paul Robeson, Bar—Mah Song (Strickland), Harry Vernon (Harp) and His Music Lovers—Conchita Supervia, Soprano—Canzona Del Sole (Sertorio). 10:45: New Records. 11:0: Slumber Music. 11:25: 2GB Goodnight Song. 11:30: Close. 2UW SYDNEY, 1110 K'cycles [Commercial Station] Day Sessions as Friday, except: 9:20: Trudy Galloway. 10:30: Woman's Magazine of the Air—A Trans-Radio Production. 12: NOON: FROM SYDNEY TO NEW YORK COMMUNITY SINGING, FRANK HATHERLEY, CONDUCTOR. 4:0: Young Citizens' Association—Mr. McNamara. 4:40: Radio Pictures Newman. 5:15: Bobby and Betty Bluegum. 5:30: George Edwards in "David and Dawn Under the Southern Cross." 6:0: Cricket: Stumps — Drawn Scores. 6:20: Dinner Music. 6:30: The Town and the Town. 6:37: Tea Time Tunes. 6:55: Anticipations for Rosebery. 7:0: Orchestral Selections. 7:15: Mrs. Arris and Mrs. Igza. 7:30: Orchestral Novelties. 7:45: Singing Salesmen. 7:50: Highlights of Harmony. 8:0: THE HUNCHBACK OF NOTRE DAME — A GEORGE EDWARDS PRODUCTION. 6:15: The Crockett Raymond. 8:30: George Edwards and Neil Stirling in "Notable British Trials." 8:45: Music Parade. 9:45: TIER GAS. TUESDAY, FEBRUARY 9 . . . CONTINUED 2SM SYDNEY, 1270 K'cycles [Commercial Station] Day Sessions as Friday, except: Note: During the afternoon Race Results will be announced. 4:15 Draw for Saturday's Greyhound Meeting. 4:15 Beauty Talk—"Lovely Ladies." 5:00 Uncle Tom and his Gang, 6:0: Angelus. 6:1 Dinner Session—Dominic Harbord. 7:0 Preview of To-morrow's Races—H. Millard. 7:30 "ONE MAN'S FAMILY." ★ Book 8, Chapter 5—"Paul Meets Lindy"—produced for Radio by John Brown. 8:00 "THE GIRL IN THE FLAT NEXT DOOR," featuring Scott Treher, Hilda Scurt, Victor Godfrey. 8:15 "HATS OFF!"—Names in the News. 8:30 "The Order and Madame." 8:52 "Thermo-Ray." 9:0 "YOUR RADIO"—Radio Drama. 9:30 Revue, 10:0 Ten O'clock Music, 10:25 When You Come to the End of the Day—Paul Oliver, 10:30: Close. 2KY SYDNEY, 1020 K'cycles [Commercial Station] Day Sessions as Friday, except: 11:0 BOOK TALK—DR. LLOYD JOHNS. 1:0 Special "Arts" Broadcast. 4:23 Melody Session—John Harper. 5:15 Aley, Rion, and Esme—Children's Session. 5:30 Radio Divertissements. 6:00 Stable Spy. 6:30 Dinner Entertainment—Gado and Noble Her, 6:37 Spot of Humor, 6:45: Radio Snapshots. 7:0 Racing Revelations. 7:15 Latest News Commentary. 7:20 Racing Topics—A. W. Davies. 7:30 Music, 7:45: Unemployed Questions, 8:00 Music, 8:00 Overseas Topics, 9:5 Music, 8:30 Musical Highlights, 9:45: The Story Behind the Stamp—To-day with 2KY Phanero, 9:0 Music. 9:15: "LABOR MARCHES ON"—Radio Play with 2KY Players. 9:30 Melbourne Turf Topics—Mr. Francis, 9:35 Music. 9:45 Holiday Bureau: Secondo Landi with His Piano-Accordion. 10:15 Radio Rhumba, 10:30 Happy Hour, 12:0: Close. 2GZ Central N.S.W., 990 K'cycles [Commercial Station] Day Sessions as Friday, except: 8:45 Producers' Service Session—Mutton, Sheep, and Beef Cattle. 11:35 News Through the Eyes of a Woman. 11:45 Let's Talk About Our Children. 3:30 Music, 3:40 All and Sunny, 3:55 Over the Teacups, 4:00 Dreams Out Beyond, 4:10 The Men Who Made Music, 4:28 The Pictorial of the Air, 4:38 Music of Yesterday. 2HD NEWCASTLE, 1140 K'cycles [Commercial Station] Day Sessions as Friday, except: 5:15 Children's Session. 6:00 Dinner Music, Overture, 6:15 Armadoneddon, 6:20 Joyster Notes and News Flash, 6:30 Entertainment, News, and Bright Music, 7:0 Overture, 7:15 2MO Race Broadcaster, 7:30 Stock Sales, 7:35 Light Music. 8:00 Strange Adventure, 8:13 Light Music, 9:00 The Happy Hour, 9:30 Weather Report, 9:31 Listening-in on Hollywood, 10:00 News, 10:30: Close. 2MO SUNNEDAH, 1360 K'cycles [Commercial Station] Day Sessions as Friday. 5:0 Children's Music, 6:3: Serial Story, 6:45 Birthday Calls and Cheerio Session, 6:30 Entertainment, News, and Bright Music, 7:0 Overture, 7:15 2MO Race Broadcaster, 7:30 Stock Sales, 7:35 Light Music. 8:00 Strange Adventure, 8:13 Light Music, 9:00 The Happy Hour, 9:30 Weather Report, 9:31 Listening-in on Hollywood, 10:00 News, 10:30: Close. 2CA CANBERRA, 1050 K'cycles [Commercial Station] 12:30 Luncheon Music, 12:45 Horoscope Reading, 12:50 Film Star Competition, 1:0 Music, 1:15 Close. 5:30 Children's Session, 5:45 Children's Story, 6:0 Adven-tures of Ben, Sam, Gurkey, and Tom, 6:20 Children's Birthday Cheerios, 6:30 News from Canberra, 6:35 Dinner Music, 7:0 Vocal Half Hour. 7:00 Vagabonds of the Prairies, 7:45 Famous Dance Bands, 8:0 Famous Orchestra, 8:15 Album of Love, 8:30: Music Conducted by Theatre New Mar, 9:0 Famous Overtures, 9:18: Instru-mental Quartet, 9:30: Pro-fetes, 9:45 News and Song, 9:45 Relaxation Music, 10:0 An-nouncements and Close. 2GN GOULBURN, 1390 K'cycles [Commercial Station] Day Sessions as Friday. 5:30 Entertainment for the Tiny Tots, 5:45 Serial Story for Children, 6:0 Dinner Music, 7:0 Selections from Talkies, 7:15 Screen News from Empire Theatre, 7:30 Radio Play—One Man's Family. 8:0 Musical Selections, 8:30 Musical Variety, 8:45 Famous Airs by Famous Composers, 8:15 Bright Musical Selections, 10:18 Slumber Music, 10:30: Close. 2TM TAMWORTH, 1300 K'cycles [Commercial Station] Day Sessions as Friday, except: 3:30 Afternoon Story, 4:0 Afternoon Selections, 4:45 Medley of All Kinds of Things, 5:0 What's On. 5:15 Possum Club conducted by Auntie Faisy, Uncle Bud and Bessie, 6:0 Dinner Music, 6:30 Sporting Commentator, 7:15 In Town Tonight, 7:15: Miniature Variety Programme, 7:30 Relay from 2SM, Sydney, 7:45 One Man's Family, 8:0 Radio Birthday. 8:30 Entertainment for Everyone, 8:45 Weather, 9:0 Market Reports, 9:15 Musical Presentation, 9:15 Heroes of Civilisation, 9:45 Talk on Douglas Social Credit, 10:0 What's On, 10:27: Spin a Web of Dreams, 10:30: Close. 2KA KATOOMBA, 1160 K'cycles [Commercial Station] 7:0 Breakfast Session, 7:28 Weather, 8:35: Close. 6:0 Children's Session, 6:30 Dinner Music, 7:0 Popular Medley, 7:15 Frank and Annie, 7:30 Vocal and Instrumental Music. 8:0 Heroes of Civilisation, 8:30 A Shot of Numbat, 8:45 Musical Appreciation, 9:0 Feature Recording, 9:15 The UKA Army Man, 9:30 Dance Music, 10:10 Slumber Music, 10:28: Weather, 10:30: Close. TUESDAY, FEBRUARY 9 . . . CONTINUED 2GF GRAFTON, 1210 K'cycles [Commercial Station] 7:30 to 8:30: Merry Morning Session. 8:30: Weather and Market Reports. 12:30: Music. 1:15: Musical Box. 1:30: Radio Rhythm. 2:0: Close. 5:30: Aunt Judy, Uncle Col, and Wendy will entertain the children. 6:0: Dinner Music. 6:30: Musical Programme. 7:30: Religious Lecture. 7:45: Musical Programme. 8:0: Heroes of Civilisation. 9:30: Music. 10:30: Close. 2LM LISMORE, 900 K'cycles [Commercial Station] 7:0: Jim Sharp, The Early Bird. 7:5: Merry Melodies. 7:50: Sunshine Songs. 8:0: News. 8:30: To-day's Programme. 9:0: Close. 1:0: Mollie's Sunday Session. 1:30: Meet the Band Leaders. 2:0: Scandal of the Stars, with Jim Smith. 2:30: Quiet Chat with the Girls. 3:0: News. 5:30: The Children's Corner, conducted by Mollie. 6:0: Dinner Minstrel. 6:30: Keith Spencer. 6:30: Farewell Rhythm. 7:0: Close. 7:0: Songs Without Words. 7:15: Sporting Session. 8:0: Raiput. 8:15: Radio Theatre, compered by Eric Harrison. 8:45: Richard Grooves. 9:15: Let's Go Dancing. 9:45: News. 9:50: Birthday Greetings. 10:0: Close. 2RG GRIFFITH, 1470 K'cycles [Commercial Station] 6:5: Children's Session. 6:30: Music Lovers' Program. 7:0: Songs at the Piano, by Cyril James. 7:15: Snapshots of Comedy. 7:30: Musical Recital. 7:45: Pianoforte Recital. 8:2: Heroes of Civilisation. 9:30: One Hour Dance Programme. 10:30: Good-Bye Song. 2AY ALBURY, 1480 K'cycles [Commercial Station] Mid-day Sessions as Friday. 5:30: Children's Session. 6:0: Dinner Music. 7:0: Station Features. 7:30: Sporting Session. 8:0: Gem of the Evening. 8:30: Around the Bandstand. 8:45: Watch Tower Bible and Tract Society. 10:15: Slumber Music. 10:30: Close. 3AR MELBOURNE, 580 K'cycles [Aust. B'casting Commission] Day Sessions as usual. 7:30: Victorian News Bulletin. 8:0: Rehearsing the Act—Austrian Harpist, Miss Quinlan. 8:20: Swing's the Thing (Mr. Matthews). 8:30: Interlude (r.). 8:35: There's Nothing Half So Sweet in Life (Mr. Grant). 9:30: Alabama Moonshine (Mr. Colin Crane). 10:0: Interlude (r.). 10:15: News. 10:30: Close. 3GI SALE, 830 K'cycles [Regional Station] 7:0: See 3AR. 7:37: See 3LO. 8:0: See 3AR. 9:30: See 3LO. 11:30: Close. 12:0: See 3LO. 10:10: See 3AR. 2:0: Close. 3:0: See 3LO. 4:15: Close. 5:15: See 3AR. 5:30: See 3LO. 6:15: Local News Service including Market Reports from Albury. 6:30: See 3AR. 8:0: See 3LO. 11:30: Close. 4RK R'H'TON, Q., 910 K'cycles [Aust. B'casting Commission] 7:0: Relayed from 4QO. 9:0: Close. 10:30: News. Music. 11:0: Relayed from 4QO. 2:0: Close. 3:0: Relayed from 4QO. 4:30: Close. 5:30: Relayed from 4QO. 6:0: See 4QO. 6:15: Relayed from 4QO. 11:30: Close. 4QG BRISBANE, 800 K'cycles [Aust. B'casting Commission] Day Sessions as Friday, except: 11:0: The Daily Broadcast Service, conducted by Rev. H. M. Wheller. 11:15: Talk by James Devaney—Australia. 11:45: Talk by S. D. England — The Health Way of Slimming. 12:15: Broadcast to Schools by C. E. W. Jones, Inspector of Schools, Queensland—Epic Voyages: (1) Tasman. 3:0: Recital by Effie Howell (Piano) and Vido Dupas (Harpone). 3:30: Piano Romance in E (Sharp) (Schumann). 3:45: Widmung (Schumann). Baritone O Star of Love (Schumann). Debussy, piano sextet (Don Giovanni) (Mozart). Piano: Joy of Autumn (Mendelssohn). Minuet from Sonatas (Ravel). Baritone La Paloma (Vraderi). Lolita (Buzzac-Pecca). 3:33: Broadcast to Schools by R. P. Cummings, A.R.I.B.A. — The Renaissance in Architecture. 5:30: Children's Corner, conducted by Uncle Max. 6:0: Dinner Music. 6:45: Sporting Notes; Music. 7:0: Talk by Miss Meta Maclean—Four Ways Son Do You Know How? 7:15: Interlude (r.). 7:20: From 2BL — National News Bulletin. 7:30: News. 7:45: Movement of the Market. 8:0: Weather Data; Late Market Reports; General Commercial News; Programme Announcements. 8:0: From 3LO—Quits. (See 2TC.) 8:45: Our Heritage of Song—One of the Series presented by Harold Browning. 9:0: Talk by William Tainsh—A Novelist and a Minstrel. 9:15: From A.B.C. (Brisbane) Concert Orchestra, conducted by W. Nelson Burton, with Clement Q. Williams (Baritone). At the Palace Ethel Conley. Orchestra: Overture (Herold). Baritone: Songs of Love (Yrjo Kilpinen). Home Over a Thousand Mountains. A Little Song. Playful Bargains. Orchestra: The Merchant of Venice—Suite (Rossini). Baritone: Songs of Death (Yrjo Kilpinen). Black Bird. Death and the Lone Drum. Winter Night. The Sower. Orchestra: Ballet Egyptien (Lunigini). 10:15: Interlude. 10:30: Weather; Late Sporting Summary. 10:40: From 3LO—Dance Music by the National Dance Orchestra, conducted by Al Hammett. 1:30: Close. 5CL ADELAIDE, 730 K'cycles [Aust. B'casting Commission] Main Features: 12:0: From National Broadcast—"Your Home and Mine," by Mr. Norman L. Haines. 12:15: "The Bush"—A Descriptive Sketch, by the Super-Cinema Orchestra. 12:30: Educational Broadcast—Science Talks—"The Story of Sound," by Mr. E. S. H. Gibson, M.Sc. 3:0: Recital by Lotte Lehmann, Joseph Siegel, and Eileen Joyce. 3:30: From A.B.C. (Adelaide) Club, conducted by "No Trump." 4:15: Norman Long expresses some of his opinions, supported by the London Philharmonic Orchestra. 4:27: News Flashes. 4:30: Close. 5:30: The Children's Session, conducted by Marie McLeod, with "The Wattle Lady" and "Old papa." 6:0: Dinner Music. 6:25: Sporting Results. 6:50: From Sydney—National News. 7:0: From Adelaide—Local News. 7:15: Market Reports. 7:20: Interlude. 7:30: From Melbourne—National Programme—"Quits," a Radio Drama by F. W. Bentley. Characters: Nick Curley, Fisher, a porter. The Action takes place during a train journey in England. Produced by Charles Cullen. 8:15: "Our Heritage of Song," fourth of the Series, presented by Harold Browning. 8:30: National Talk—"A Novelist and a Minstrel," by Mr. William Tainsh. 8:45: From Adelaide—Regional Programme. A Ballad Recital, by Margery Pirie, mezzo-soprano. 9:0: Interlude. 9:15: Dance Music, by the Palais Dance Band, conducted by Harry Burke-Smith. 10:15: News Service and Weather Notes. 10:25: Dance Music, by Palais Band (cd.). 11:0: "Meditation" Music. 11:15: Signature Melody. 11:30: Close. 2FC 7.40 ON THE BEACH AT BALI BALI Presented by JIM DAVIDSON'S A.B.C. DANCE BAND 2FC 8.0 NATIONAL MILITARY BAND Conducted by STEPHEN YORKE March: Fantasia, Colonel Bogey on Parade ... Alford Suite In Days Of Old ... Ball Music Dance Danse Caracteristique Satyr's Revel. Reminiscences of the Plantations Chambers 2FC 8.30 THESE WERE HITS Presented by JIM DAVIDSON'S A.B.C. DANCE BAND 2FC 9.5 LA LUNE BLEU An Operetta in Two Acts Book by PAUL FURNESS Lyrics by ALTHEA LAMBERT-GLASBY and PAUL FURNESS Music by HORACE KEATS IN ASSOCIATION WITH THE A.B.C. (Sydney) STRING ORCHESTRA Conducted by the Composer, HORACE KEATS Antony Cler ran a Parisian Beauty Salon, but was practically bankrupt when his American aunt offered a large sum for the best perfume in Paris, and Antony, with the help of a Professor of Perfumes in the South of France, set out to win the money. CHARACTERS: ANTONY CLER (TONY), of La Lune Bleu. ALOUETTE, a Dancer. LOUIS BLANCHET, their Friend. MRS. DYSON-EVERETTE-CONDOR, wealthy American Woman. M'SIEUR BILLARD, a Creditor. ARMAND, an Soldier. FREDERIC, an Actor. PHILIPPE MONT-VALLET, a Friend. PROF. MONT-VALLET, his Brother. SENORA GOMEZ, a Spanish Beauty Specialist. EMILIO, her Brother. 2FC 10.0 KOSLOV TRIO Serenade Espagnol ... Athens Patellia ... Ponce Rokero ... Moszkowski Andante ... Mendelssohn Gavotte ... Bach Interlude Orientale ... Korsakov Chanson Meditation ... Cottenet Entre Act La Valse ... Helmsberger 2BL 7.40 A Pianoforte Recital (r.) by MISCHA LEVITZKI—Hungarian Rhapsody No. 6 in D Flat Major (Liszt) La Campanella (Paganini-Liszt) Staccato Etude ... (Rubenstein) Hungarian Rhapsody, No. 13 (Liszt) 12.70: Vocal and Instrumental Music for the Lunch Hour (r.). 2NR 7.40 MILESTONES AND MELODY 1.0: Radio Rhythm (r.) 2.0: CHAT OVER THE TEACUPS, Neville Venman 3.0: GRAMOPHONE PIE (r.) 4.30: Trade Demonstration Music (r.) 2UE 7.40 DESCRIPTION OF THE RACES FROM KILBERY, BY HARRY SALOMONS 5.0: With the Light Orchestras. 5.20: With the Light Orchestras. 5.40: Local News from Sydney. 5.45: Keith Collins at the Piano. 6.0: Dinner Music. 6.45: RECORDED DESCRIPTION OF TO-DAY'S ROSEBERRY RACES, PRESENTED BY H. C. COLLINS. WEDNESDAY, FEBRUARY 10 CONTINUED 6:55 Music. 7:00 THE SIGN OF THE PURPLE SPIDER—A B.S.A. PRODUCTION. 7:15 A Spark of Humor. 7:30 Dance Moods. 7:55: The Singing Troubadour. 8:00 In the Key of Happiness. 8:15 From Our Standard Radio Library. 8:30: Bijou Programme. 8:45 Marek Weber and His Orchestra. 9:00 STRANGE ADVENTURES IN STONE LAKES. 9:15 Film: Rhythm. 9:30 Childhood Memories. 9:45 Peter Dawsons and Eddie Lorand. 10:00 Music for the War. 10:15 Kenneth Fowles Reads a Story. 10:30 Out of the Band Box. 11:00 Words and Music. 11:30 Close. 2BL 8.0 ORGAN MUSIC RONALD MARSTON, Organist In Association with STANLEY CLARKSON, Bass ORGANIST— Precesional March ........................................... Cousted A Morning Song .................................................. Meade Bells and Carillon ............................................... Meade O, Star of Eve (Tannhauser) .................................. Wagner BASS— Sons of the Sea .................................................. Coleridge Taylor The White Owl .................................................. Lauster Ma Little Banjo .................................................. Dichtman ORGANIST— Largo from The New World Symphony ..................... Dvořák Minuet from Suite Gothique .................................. Boëllmann Everlong .......................................................... Martin Easter Hymn, from Cavalleria Rusticana .................... Mascagni BASS— A Sailor's Prayer .............................................. Keel Bed in Summer .................................................. John Ireland Turn Ye to Me (Hebridean Melody) ......................... Traditional 2BL 8.40 RECITAL BY VETA WAREHAM, Violinist And PAUL KAIN, Tenor VIOLIN— Air ............................................................... Purcell Serenade ......................................................... d'Ambrasio Hora Slaccaeto (Rumanian Dance) .......................... Dinicu-Heifetz TENOR— The Cloths of Heaven ......................................... Dunhill Spelvin .......................................................... Sinding Star Vichno ..................................................... Salzator-Rosa 2BL 9.0 THE NATIONAL MILITARY BAND Conducted by STEPHEN YORKE Overture: Rosamunde ........................................... Schubert Suite: The Ballet, The Moon Pay ............................. Moreton The Waltzed Wood at Moonrise Dance of Phidias and the Moon Pay Tarantella. CLARINET SOLO— Caprice in A Flat ................................................. Mayeur (Soloist: Douglas Williamson) Three Dances from the Suite: The Bavarian Highlands Elgar 2BL 9.50 MUSICAL GIANTS OF YESTERYEAR (r.) A Programme Arranged by BEN SULLIVAN and GEORGE JOHNSTON Great Homes of Opera, Part 3 THE PARIS OPERA HOUSE Artists to be heard:— MADAME BROHLY. MAURICE RENAUD. CLEMENCE. PLANCON. MARY GARDEN. MARCEL JOURNET. 2SM 1270 K'cycles [Commercial Station] Day Sessions as Friday, except: 8:45: Preliminary Talk on to-day's Race—Mr. Rubens. 12:00: RACING SESSION— During the Afternoon a Description in Running will be Given of the Races from Rosebery, with Music. Re- layed to 2GZ. 5:00: Under Tom and his Gang. 6:00: Angels. 6:11: Dinner Session—Dominic Har- rington. 6:30: Re-broadcast of to-day's Races, with Correct Starting Prices. 7:00: 2SM's Skin Specialist—Talk. 7:15: Music. 7:30: Ashton Jarry—"The Passer." 8:00: Wednesday Night at 5. 8:15: "UNCLE PETA—THE STAMP ADVENTURER," fea- turing Reg. Hawthorne. 8:30: The Great British Medicine. 9:00: "ALL OVER SYDNEY"—An Dramatic Presentation, by Ray Hawbinder. 9:30: "The Hatton Twins" Variety Programme. 10:00: Ten O'clock Musicale. 10:25: When You Come to the End of the Day—Paul Oliver. 10:30: Close. 2KY SYDNEY, 1020 K'cycles [Commercial Station] Day Sessions as Friday, except: 10:15: Talk—Miss Mary Gibbons. 11:30: Suburban Session — Fred Gardner. 12:30: BROADCAST OF RACES FROM ROSEBERY — Ian Garden. 2:45: Moody Sermons—John Harper. 5:15: Egyg, Rion, and Emeil—Childe Hasselbeck. 5:30: Dinner Divertissements. 6:0: Stable Spy. 6:30: Dinner Entertainment—Gooey and Uncle Bert. 6:37: Spot of Humour. 8:45: Radio Snapshots. 7:0: Anything Goes. 7:15: News. 7:20: Commentary. 7:20: FREED WITT — WITH SONGS AT THE PIANO. 7:30: News. 7:45: Unemployed Music. 8:0: Overture. 8:5: Music. 8:15: THE COWBOY SHEET — RADIO SERIAL WITH 2KY PLAYERS. 8:30: Musical Highlights. 8:45: Musical Comedy Musical Review. 9:0: CORONATION PLAYERS RACING TRIALS — FROM ASSEMBLY HALL — SEMI-FINALS. 10:0: Radio Rhythm. 10:35: Happy Hour. 12:0: Close. 2GZ Central N.S.W., 990 K'cycles [Commercial Station] Day Sessions as Friday, except: 6:45: Promenade Session — Wires and Joints Cereals. 11:45: "Hospital Development in Victoria" Talk by Mr. K. K. Oates. 12:0: Midday Session. 12:55: Programme Projects. 1:0: Rosebery Race Despatches. Music. 1:5: English Pacific Wireless News. 1:15: Weather Report. 5:0: Children's Session. 6:0: Dinner Music. 7:0: Sports Resume—Racing Results. 7:15: Wool Report. 7:24: Produce. 7:30: Weather. 7:50: The Man on the Land. 7:50 Music. 8:18: "Hunchback of Notre Dame." 8:30: CAVALCADE OF MUSIC. 9:0: Keyboard Kings. 9:0: "Mr. and Mrs. Everybody." 9:30: Ballads We Love. 9:48: Racing Talk—George Francis. 10:0: Novelty Notions. 10:15: Happy Harmony. 10:30: Close. 2HD NEWCASTLE, 1140 K'cycles [Commercial Station] Day Sessions as Friday, except: 1:0: RACES FROM ROSEBERY. 5:15: Children's Session. 5:20: Uncle Rex and Horace. 6:0: Dinner Music. 6:15: Nuggets from the Golden Age. 6:20: Joyites, Games and News Flash. 6:45: Peter and Pam—A B.S.A. Production. 7:5: Alan Kippax Talks. 7:50: Mrs. Arris and the Togs. 8:30: Edmundo Breeze in Memories. 8:5: Eb and Zeb, Mirth Makers. 9:20: Uncle Rex and Mrs. Everybody. 9:35: Magic Island. 9:50: Music, relayed from 2UE. 9:15: Home on the Range. 10:0: News. 10:30: Close. 2KO NEWCASTLE, 1410 K'cycles [Commercial Station] Day Sessions as Friday, except: 2:15: Radio Pictorial of the Air. 3:0: Lights, Camera, Action! — An Exclusive Session on Hollywood, written and presented by Miss Elma Gibbs and Mr. Tom King. 5:45: Birthday Calls. OVERSEAS STATIONS WHAT'S ON THE AIR TO-DAY 5:20 a.m. (QJD 25.4, DXC 49.8, TJL 19.9): Contemporary Chamber Music. 9:0: "At the Carnival," Music, Song, and Golden Wine. 7:0: News. 7:20: Close. 6:0 a.m. (2HO 31.13): Varied Programme from Italian Stations. 6:0 a.m. (GSD 31.5, GSD 29.3, GSI 19.6, with GSC 31.3 at 7:0): Fred Hartley's Novelties Sextet, with Brian Lawrence Australian Band. 6:15: Musical Feasts. 6:30: Chopin Music. 6:45: Interval. 7:0: News. 7:20: The B.B.C. Theatre Orchestra. 6:0 p.m. (DIN 19.7, DIN 31.4, DJA 31.5): Women's Hour. 9:15: Concert of Request Music. 7:30: News in German. 7:45: A Press Review. 8:0: New German Law. 8:15: Close. 9:0 p.m. (GSI 31.3, GSC 31.5): The Big Band Special Broadcast from Cornwall. 6:20: The B.B.C. Empire Orchestra. 6:55: Talk on World Affairs, by H. V. Hudson. 7:10: A Folklore Programme from Italy. 7:20: Close. 6:30 p.m. (JRN 31.5): Records, Markets and Sporting Commentary. Music. 8:0: Relay of National News Bulletin from Daventry. 8:30 p.m. (WINK 31.3): Early Morning Music. 8:40: Market Reports. 10:0: News of the World from Boston. 10:15: Music. 8:30 p.m. (YDR 31.1, VPD 24.5, PLP 27.2, PMN 29.2, DYC 19.8): Evening Musical Programmes. 10:0: VPD2 closes. 9:0 p.m. (TPAZ 31.5): Concert from Radio Paris. 10:0: News in English. 10:15: Music. 9:0 p.m. (DUR 19.7, DJN 31.4, DJA 31.3, DJQ 15.6, DJQ 19.6, with GSC 31.56 at 9:1): Concert of Request Music. 9:0: News in English from Berlin. 10:15: More Light Music. 11:0: News in German. 11:15: Contemporary Chamber Music. 12:0: Midnight Music in English. 12:0: DNG, DJE, and DNM. 9:0 p.m. (GSD 31.5, GSC 31.56, GSI 19.6, with GSC 31.5 at 10:0): Dance Music. 9:30: Talk on World Affairs by H. V. Hudson. 9:45: News of the World from Chicago. 10:5: St. Columb Hurling Game. 10:20: Scenes from Julian. 9:43 p.m. (2RO 25.4): Variety Programme from Italian Stations. 11:20: News in Italian. Music. Midnight: English Session. News of the World from Boston. 11:30 p.m. (PCJ 19.7): Special English Session. Calls to World Listeners. Request Records. 2.0 a.m.: Close. 2MO GUNNEDAH, 1360 K'cycles [Commercial Station] Day Sessions as Friday. 6:0: Children's Music. 6:2: Serial Story. 6:15: Birthday Calls and Cheerios. 6:30: Entertainment, News, and Midnight Music. 7:0: Overture. 7:15: Light Music. 7:30: Stock Sales. 7:35: Light Music. 8:0: Heroes of Civilisation. 8:15: Moon Over Africa. 9:0: The Happy Hour. 9:15: Golf Links. 9:30: Weather Report. 9:30: Hollywood Review. 10:0: Close. 2TM TAMWORTH, 1300 K'cycles [Commercial Station] Morning and Midday Sessions as Friday, except: 12:30: Luncheon Music. 12:45: Horoscope Reading. 12:50: Music. 12:55: Canterbury's Session. 1:15: Music. 1:30: Close. 5:30: Children's Session. 5:45: Canterbury's Session. Adventures of Ben Sam, Gorkey and Tom. 6:30: Children's Birthday. 6:45: Canterbury's Session. 6:50: Canberra. 6:55: Dinner Music. 7:0: Vocal Hall Hour. 7:30: Sababonds of the Prairie. 7:45: Famous Dance Bands. 8:0: Famous Orchestras. 8:30: Favorites. Old and New. 8:45: Industrial Psychology—Talk by Mr. Keane. 9:0: Instrumental Quarter Hour. 9:15: Dad and Dave. 9:45: Relaxation Music. 10:0: Announcements and Close. 2GN GOULBURN, 1390 K'cycles [Commercial Station] Day Sessions as Friday. 5:30: Entertainment for the Tiny Tot. 5:45: Serial Story for Children. 6:0: Dinner Music. 7:0: Miniature Programme. 7:30: Popular Releases. 7:45: Famous Walzies. 8:0: Radio Snapshots. 8:15: Musical Miniatures. 8:30: Variety Programme. 9:0: Celebrity Artists. 9:30: Popular Releases. 10:15: Slumber Music. 10:30: Close. 2CA CANBERRA, 1050 K'cycles [Commercial Station] 12:30: Luncheon Music. 12:45: Horoscope Reading. 12:50: Music. 12:55: Canterbury's Session. 1:15: Music. 1:30: Close. 5:30: Children's Session. 5:45: Canterbury's Session. Adventures of Ben Sam, Gorkey and Tom. 6:30: Children's Birthday. 6:45: Canterbury's Session. 6:50: Canberra. 6:55: Dinner Music. 7:0: Vocal Hall Hour. 7:30: Sababonds of the Prairie. 7:45: Famous Dance Bands. 8:0: Famous Orchestras. 8:30: Favorites. Old and New. 8:45: Industrial Psychology—Talk by Mr. Keane. 9:0: Instrumental Quarter Hour. 9:15: Dad and Dave. 9:45: Relaxation Music. 10:0: Announcements and Close. 2LM LISMORE, 900 K'cycles [Commercial Station] 7:0: The Early Bird. Jim Sharp. 7:5: Musical Moments. 7:30: LM's Market Place. 7:45: Miniature Programme Selections. 8:0: News. 8:30: What's On To-night? 8:45: Music. 9:0: Close. 10:0: Jim Sharp Entertains You. 10:15: Tilt-Blitz of News. 2:0: The Music Box. 2:30: Let's Get Together. 2:30: Close. 5:30: Let's All Be Children, conducted by Aunt Mollie. 2RG GRIFFITH, 1470 K'cycles [Commercial Station] 5:0: Children's Session. 5:30: Music Lovers' Petpeturi. 7:0: Songs at the Piano, by Cyril James. 7:15: Snapshots of Canberra. 7:30: Popular Releases. Planoforte Recital. 9:30: One Hour Dance Programme. 10:30: Goodnight Song. 2AY ALBURY, 1480 K'cycles [Commercial Station] Mid-day Sessions as Friday. 5:30: Children's Session. 6:0: Dinner Music. 7:0: Station Features. 7:30: Sporting Session. 8:0: Gem of the Evening. 8:30: Bright Interlude. 8:45: The House of WEDNESDAY, FEBRUARY 10 CONTINUED Peter MacGregor. 9:15: Country Man's News. 9:30: Celebrity Recital. 10:30: Close. 3LO MELBOURNE, 770 K'cycles [Aust. B'casting Commission] Day Sessions as Friday, except: 9:35: The Radio Serial—"Cattle-Camp," by J. J. Hardie. Read by "Serena." 10:10: Current Happenings in Sport, by A.B.C. Commentator. 10:25: "After Hours—Really Real-Life," by Marie Leesman. 10:45: "Some Book to Read"—Mrs. K. McQuillan. 12:30: "The World We Live In"—Professor G. S. Browne. 12:45: From Sydney—"At Home and Abroad," by "The Spectator." 12:50: From the Rotary Club—Luncheon Speeches. 2:00: Vienna Philharmonic Orchestra, conducted by P. Weingartner—Symphony No. 9 in A Major (Beethoven). Eileen Joyce, Piano, with Orchestra—Rondo in A Major K. 386 (Mozart). 3:45: Copenhagen Orchestra, conducted by Albert Wolf: Spanish Rhapsody (Ravel), Dino Borghini (Tenor), Ivo Nelson (Piano); Gondoliera Veneziana (Sadero). La Barcarolle della Marzucca, Farr, Sadero. Milan Neapolis Symphony Orchestra, conducted by E. Ormandy: Die Fledermaus—Overture (Strauss). 4:15: Close. 5:30: Young People's Session—Overture. We say farewell to "The Wanderer" (composed by a Carter)—A Discussion. Melody from Bob Molyneux. Finale. 6:00: Dinner. Music. 7:40: "On the Beach at Ball Ball" (See 2FC). 8:00: National Military Band. (See 2FC.) 9:30: "These Were Hits," presented by Jim Davidson's A.B.C. Dance Band. (See 2FC. Interlude.) 9:35: "La Lune Bleu." (See 2FC.) 10:00: Koslov Trio. (See 2FC.) 10:45: Cricket—Detailed Scores of To-day's Play. The Mid-week Review of the News. 11:00: Walter Gieseking, Piano—Forte Sonata—Bergmansque (Debussy). Royal Opera House Orchestra, Stockholm: King Christian Suite (Rubbra); Toccata, Henshaw; and Yehudi Menuhin—Sonata in D Minor, Op. 10 (Schumann). Vienna Philharmonic Orchestra—Siegfried Idyll (Wagner). Fritz Kreisler, Violin Solo—May Breeze (Mendelssohn). M. Henry Merckel, Violin—Berceuse (Fauré). 11:30: Close. Three Modern Arts: "The Art of Listening," by Miss Peggy Walter Murdoch. 7:15: Interlude. 7:20: National News Bulletin, 7:30: Victorian News Bulletin. 7:35: Melbourne's Session. 8:00: Cricket—Detailed Scores of To-day's Play. 8:15: A Musical Programme by "The Eldon Trio": Trio No. 9 in E flat (for Piano, Violin and Cello) (Beethoven). Allegro moderato. Scherzo. Rondo—Allegretto. (Cello Solo, Elegie Faure). Violin: On Wings of Sound (Mendelssohn Act II). Scherzo-tarantelle (Wieniawski). Trio: Four Short French Tunes (Brahms). 8:30: "On the Witch's Broom"—In a series of vivid travel talks we will take you to the countries you have never visited. 8:45: Kathleen Goodall in a programme of Songs at the Piano—A Boy's Ball (Singing Wood); Struña (Bamford); The Wicked Chinaman (Dudley Gibbs); The Bend in the River (Edwards). 9:00: "The Mystery of the Marble Room," Adapted by Ruth M. George from Edgar Allan Poe's famous "Tales of Mystery and Imagination." The Characters are—Monsieur Auguste Dupin, a gentleman of eccentric habits, with a keen analytical mind, often consulted by the Police for reference to difficult cases; Edwin Morrell, an English barrister, friend and confidant; Marie Roger, a French gipsiete; Madame la Duchesse de Praline; Count Alfred de Montfaucon; Countess de la waiting to her Majesty Queen Marie H.M.; King Louis Philip King of the French; Count Bresson, French Ambassador to the Spanish Court. Scene: France in the year 1849. Production: Mayne Lynton. 10:00: Interlude. 10:15: News. Cricket—Detailed scores of To-day's Play. 10:30: Close. Broadcast Service, conducted by Rev. Percival Watson. 11:15: The Weekly London Letter. 11:30: Music. Miss Peggy Campbell—A Glimpse of Ireland. Talk by Miss Gwen Timbury, L.A.M., A.T.C.L.—Music and the Child. 12:00: From City Hall, Brisbane—Community Song Concert, with Daphne and Brian Lynch, Jeanette Gerrard with her Piano-Accordion, Clive Cook and Elsie Maude ("Entertainers") (Carl Jay (Vibra), John Victor, the Greer (Ventriloquist). Supported by Billio Smith and his Band. 2:00: During the afternoon, Descriptions will be given of the Ipswich Amateur Turf Club's Bundamba Races. 5:30: Children's Corner, conducted Cap and Bebe. 6:00: Dinner. Music. 6:45: Spelling Results: Music. 7:00: From 6WF—National Talk by Walter Murdoch, Professor of English, in the University of Western Australia—Three Modern Arts: The Art of Listening. 7:15: Interlude. 7:20: From 2BL—National News Bulletin. 7:30: News Memoranda of Air Mail Planes. Weather Data: Late Market Reports; General Commercial News; Light-houses; Railways; Programme Announcements. 8:00: From 2FC—The National Military Band. (See 2FC.) 8:30: These Were Hits, presented by Jim Davidson's A.B.C. Dance Band. 9:00: Legend of the Moonlight — A Play for the Orpheum. Produced by Dion Wheeler. 9:45: The Orpheans (Male Quartet), directed by Will Donald, Junior, in association with Helene Baird (Piano). Quartet: Max Lindy Lou (Schubert); Trio: The Agonist (Rich); Piano: Improvisation in B Flat. Op. 142, No. 3 (Schubert); Meditation (Chaminade). Trio: Quartet: White Dance (Leur); Give Me Back My Dear Old Sunny Spain (O'Hara); Romance: Value Paintings (Cundell); Romantic Box (Levenoch); Ballade (Granville-Bantock). Quartet: A Sea Tale (Rachmaninoff); Gentlemen, Good Night (Longstaffe). 10:30: Weather; Late Sporting Summary. 10:45: London Philharmonic Orchestra: Rossiniana — Selection (Rossini, arr. Respighi); Van Phillips, Quartet: Salut d'Amour (Elgar). Rudolf Hindemith (Cello) and Alice Ehlers (Cembalo): Sonata for Violoncello and Allegro (Marcelli). Edith Lorand and the Viennaer Orchestra: Spring Story (Weinberger). Eric Trege and Rosenberg Orchestra: The Jester of the Wedding (Coates). March: Valse. Van Phillips, Quartet: Niccolini (Van Phillips). Len Fills, and his Novelty Orchestra: Lullaby Land. 11:30: Close. 4QN Nth. Regional 600 K'cycles 7:0: Relayed from 4QG. 9:0: Close. 10:30: News. Music. 11:0: Relayed from 4QG. 4:45: Close. 5:30: Relayed from 4QG. 6:0: News. 6:15: Relayed from 4QG. 11:30: Close. 5CL ADELAIDE, 730 K'cycles [Aust. B'casting Commission] Main Features: 12:00: National Stock Exchange Information. During the afternoon, racing services from Strathalbyn and Geelong will be included, with music from the Studios. 5:30: The Children's Session, conducted by Marie Louise, with "Jestette" and "Old Pal." 6:00: Dinner Music. 6:15: The 5CL Bluebird Club, conducted by "The Bird Lady." 6:30: From Perth—National Talk "Three Modern Arts—The Art of Listening," by Professor Walter Murdoch. 6:45: From Adelaide—Sporting Results. 6:50: From Sydney—National News. 7:0: From Adelaide—Evening News. 7:15: Official Stock Exchange. 7:22: Cattle Market Reports. 7:30: From 2BC—The Morning Programme—The National Military Band, conducted by Stephen York. 8:00: "These Were Hits," a feature presentation, by Jim Davidson's A.B.C. Dance Band. 8:30: From Adelaide—Regional Programme—Recital by Paul Vinogradoff (distinguished Russian pianist), playing Appassionata in F Minor, Opus 57 (Beethoven). 9:00: Interlude. 9:0: From Torch Theatre—The A.B.C. (Adelaide) Theatre Orchestra, conducted by William R. Cade, with Thel Phillips, Operatic Soprano. 10:0: From 2BC—"Songs of the Sea," Recorded Interlude. 10:15: News Service and Weather 10:23: Dance Music. 11:28: Signature Melody. 11:30: Close. IF YOU ARE DEAF ENJOY RADIO PROGRAMMES WITH REIN'S "RADIEAR." This latest British instrument can be easily attached to ANY radio set. The whole radio programme is reproduced with absolute fidelity without batteries, microphones or cumbersome headphones. Consult R. A. NOEDL, LIB.S.T. (Call, write, or Phone MA3774 for full particulars.) Suite 805 T. & G Building, Park Street, Sydney. THE GREATEST RADIO DRAMA OF 1937 An absorbing drama of love, hate and intrigue in old Paris—more real than the book—dramatic as the picture—don't miss one episode of this thrilling play. "The Hunchback of Notre Dame" TUNE IN 2UW 8 P.M. EVERY NIGHT except Saturday and Sunday. THURSDAY 2FC 7.40 A.B.C. (Sydney) STRING ORCHESTRA Conducted by KENNETH MURISON BOURN Suite in C Major ........................................ Purcell, arr. Hurstone Prelude, Saraband, Minuet. Marche .................................................. Grieg Two Norwegian Melodies .................................. Grieg Air and Dance ........................................... Delius 2FC 8.0 THE A.B.C. (Sydney) STRING ORCHESTRA Conducted by KENNETH MURISON BOURN With RUTH PEARCE-JONES, Lyric Soprano ORCHESTRA— Concertina Ossia for Quartet and Orchestra Dennington (First Performance in Australia) Allegro, Passacaglia, Fugue. SOPRANO— L’Amaro (with Violin Obligato and Pianoforte Accompaniment) Mozart Alleluia (with String Accompaniment) Mozart ORCHESTRA— Serenade .................................................. Dvorak Moderato, Tempo di Valse, Scherzo, Larghetto, Allegro Vivace. 2FC 9.0 PIANOFORTE RECITAL By MADAM MAY HEINEKEN Sonata in A Major ........................................ Mozart Rhapsodie Hongroise, No. 8 ............................. Liszt 2FC 9.25 CURRENT TOPICS This space is reserved for a Topical Talk by a prominent speaker. 2FC 9.40 SALON RECITAL By FRED WILLIAMSON, Tenor VINA BARNDEN, Pianist And CONSTANCE PETHER, Flautist PIANO— Scherzo in C Sharp Minor ................................ Chopin TENOR— O Lady Rose ............................................ Quilter O Mistress Mine ......................................... Quilter Two Daisies ............................................... Quilter FLUTE AND PIANO— Concerto for Flute and Piano .......................... Chaminade FLUTE— Pastorelle .................................................. German 2FC 10.10 MELODIES AND MEMORIES A Special Programme by COLIN CRANE Conducted by PERCY CODE, In Association with KATHRYN DEMPSEY, Pianist. CHORUS— Suite—Scenes Poétique (Godard) Valse—Wine, Women, and Song (J. Strauss) PIANIST— To the Spring .......................................... (Grieg) Flower Garden ........................................... (Besly) Capriccio .................................................. (Bridge) CHORUS— Air de Ballet ............................................. (Chaminade) Serenade .................................................. (Borodine) Nocturne, from Lyric Suite ......................... (Grieg) PIANIST— Schumann .............................................. (Mendelssohn) Valse in A Major ....................................... (Leutitzki) Spanish Dance ........................................... (de Falla) CHORUS— Selection from The Chocolate Soldier ............ (O. Strauss) Marriage at Knightbridge (Cates) 4.00 KATIE ZAMMIT, Soprano— April Moon .............................................. (Batten) Dance Away ............................................. (Oliver) Easter Day .............................................. (Sanderson) 4.30: VERSE GRAVE AND GAY (C.) The Courtship of Haguain— The Sixth of a Series of TALES OF A WAYSIDE INN, Presented by ARTHUR GREEN-AWAY. 2FC 10.45: Interlude (r.). 3.30: CHAT OVER THE TEACUPS, by WILLIE BRENNAN (Our London Letter). 4.40: Trade Demonstration Music (r.) 5.00 Stock Exchange, Third Call. 5.5: Trade Demonstration Music (r.) 5.23: Weather Information. 5.30: Close. EARLY EVENING SESSION 6.0: Model Aeroplanes Session, conducted by WALTER. 6.12: MEN AND WOMEN—Macquarie’s First Wife. Continued from the Series of Talks by C. S. YARWOOD. 6.35: Stock Exchange Report. 6.35: SPORTING SESSION—Results and Comments. 6.45: Women’s Sporting Results, by Graham Johnson. 6.55: SPORTING SESSION (Continued). 7.00: MARRIAGE—ITS SUCCESS OR FAILURE. The Third Talk of the Series— The Other Side—By A WOMAN. 7.15: Interlude (r.). 7.30: NATIONAL NEWS BULLETIN. 7.45: Local News. 7.55: Weather Report, General Market, Produce and Fruit Prices, and late Sussex Street Prices, supplied by the State Marketing Bureau; Wool Sales, Railway Report, and Truck Bookings. 7.58: What’s on the air to-night? EVENING SESSION 8.0: THE DREAMING MAN—(See Panel) 9.0: Interlude (r.) 9.15: B.B.C. RECORDING—(See Panel) 10.0: SALON MUSIC (r.)—(See Panel) 10.29: Weather Information. 10.30: Close. 2NC NEWCASTLE, 1230 K’cycles [Aust. B’casting Commission] 7.45 to 9.30: Relayed from 2BL. 9.30 to 10.30: Relayed from 2FC. 10.30 to 10.40: News Service by courtesy of the “Newcastle Morning Herald.” 10.40 to 11.30: Relayed from 2FC. 12.0 to 1.30: Relayed from 2FC. 2.30 to 3.40: Relayed from 2PC. 3.40 to 3.45: Chat over the Teacups—by Colin Crane. 3.45 to 4.15: Relayed from 2FC. 5.30 to 6.0: Relayed from 2FC. 6.0 to 6.30: Relayed from 2BL. 6.30 to 10.30: Relayed from 2BL. 10.30 to 11.30: Relayed from 2FC. THURSDAY, FEBRUARY 11 . CONTINUED 2NR GRAFTON, 700 K'cycles [Regional Station] 7.0 to 7.25: Relayed from 2PC. 7.25 to 8.30: Relayed from 2BL. 8.30 to 9.30: Relayed from 2GC. 12.0 to 1.30: Relayed from 2PC. 1.20 to 1.30: Local News from "The Grafton Express." 1.30 to 2.00: Relayed from 2PC. 3.0 to 4.15: Relayed from 2PC. 5.30 to 6.30: Relayed from 2PC. 6.30 to 6.35: 2NR Studio—Militia Notes. 6.35 to 8.0: Relayed from 2BL. 8.0 to 11.30: Relayed from 2PC. 2UE SYDNEY, 950 K'cycles [Commercial Station] Day Sessions as Friday, except: 1.30: Acceptances for Canterbury. 2.30: Between Ourselves—An Incident Session, conducted by Frank Stuart, 8.00. Relaxation Music. 2.45: The Radio Adviser on Life's Problems. 3.15: Lifting Melodies. 4.00: Stories for Youth by Frank Hartly. 3.45: Dance Rhythms. 4.30: Celebrate Recital. 5.0: With the Lamp of Christmas. 5.30: Dance Rhythms of To-Day. 5.45: Films of Yesterday—Mississippi. 6.0: Dinner Music. 6.15: Acceptances for Canterbury. 6.30: Dinner Music. 6.35: SING AND BUDDY ENTERTAIN. 6.45: Bright Music. 7.0: The Sign of the Purple Spider — A.A.B.S.A. Production. 7.15: A Sign of Humor. 7.30: Film Rhythm. 7.50: Home, Sweet Home. 7.55: The Singing Troubadour. 8.0: At the London Palladium. 8.15: Build a Little Home—Music In Quiet Moments. 8.45: Closely Knitbridge and George Formby Entertain. 9.0: Musical Comedy Favorites. 9.15: BROADWAY PIE, FEATURING JACK LUMSDALE AND JAN DAILY. 9.45: Dancing Dance Night—Roy Fox and his Band. 10.0: Victor Sylvester and his Ballroom Orchestra. 10.15: Kenneth Powles Reads a Story. 10.30: Ray Noble and His Orchestra. 10.45: Robert Russell Dance Orchestra. 11.0: Eddie Duchin and his Orchestra. 11.15: Benny Goodman and his Orchestra. 11.30: Close. 2CH SYDNEY, 1190 K'cycles [Commercial Station] Day Sessions as Friday, except: 11.30: Our Legally Qualified Medical Practitioners Will deliver a Talk on Their Services. 1.0: From Pitt-street Congregational Church—Community Praise Service. 2.30: A DICKENS' READING, BY A. S. COCHRANE. 5.55: Children's Session, from the Fairy Corner. 5.30: The Hello Man's Children's Session. 6.15: Carson Robinson and Co.—The Show. 6.30: Knight Barnett at the Burwood Palatial Cinema. 6.45: Dinner Music. 6.50: Weather Report. 7.0: CONTRIBUTED COMEDY, COMPERED BY ROB STROTHER. 7.15: Paul Robeson, Basso—Ah Still Suits Me (Kern). 7.20: The Goodenoughs—The Little Piano Accordion Band—Mexico (Edgar). 7.30: Further Air Adventures of Jimmy Allen. 7.45: Bits and Pieces, presented by Oscar Lawson. 8.0: Fred and Maggie Everybody, featuring Edward Howell and Teresa Deans. 8.15: Raeburn and Landner—Live in Britain (arr. Raeburn and Landner). 8.15: Stars of the Network. 8.30: Point of View. Presenting— 2BL 8.0 THE DREAMING MAN A Fantastic Comedy for the Microphone by LEONARD CRABTREE We have all dreamt our dreams. This is a story of George, a simple soul, who fell asleep one summer afternoon in his garden, and, as he slept, he dreamt of success. CHARACTERS: GEORGE HARLEY. MR HARRIS TEDDY CARTER. MR. BURLINGTON. MRS. HARLEY. ELSIE MR. BLATCHETT. MRS. BROWN A BUS CONDUCTOR. A TELEGRAPH BOY. GERTIE HARLEY. PRODUCTION: JOHN GOULD. 2BL 9.15 B.B.C. RECORDING THE ROYAL NATIONAL EISTEDDFOD OF WALES, 1936 A Programme by Some of the Successful Competitors 2BL 10.10 SALON MUSIC (r.) VICTOR OLAF SEXTETTE Grieg Selection. Merchant's Venice—Incidental Music .... Rosse Prelude No. 1. Intermezzo—Portia. Prelude No. 2—Doge's March. 2SM SYDNEY, 1270 K'cycles [Commercial Station] Day Sessions as Friday, except: 1.5: Track Work. 1.45: Acceptances for Saturday's Session at Canterbury. 2.30: "The Pictorial of the Air"—Reading by John Tuttle. 4.30: "The Red Lacquer Table"—Radio Drama. 5.0: Uncle Tom and his Gang. 6.0: Angels. 6.1: Dinner Session—Dominic Harneitt. 6.15: To-day's Poultry Market and Report—M. Hennessy. 6.45: Music. 7.30: "This Age of Plenty"—Talk. 7.50: Walter Kingsley, Concert Bandman. 8.0: INSPECTOR ZONE in "ON THE SPOT"—Comedy with a Moral. 8.22: "The Grocer and Madame." 8.45: "The Romance of Ireland." 9.0: "IN TOWN TO-NIGHT," with John Dunne interviewing the City's Personalities. 11.0: "THEY MADE THESE FAMOUS," Compered by Jack Meaney. 10.25: When You Come to the End of the Day—Paul Oliver. 10.30: Close. 2GB SYDNEY, 870 K'cycles [Commercial Station] Day Sessions as Friday, except: 1.0: A Talk by Mr. Richard Want, M.A. 1.10: Mr. T. L. Laing, "Planning for Recreation"—A Talk under the auspices of Parks and Playgrounds Movement of New South Wales. 4.45: Auntie Goodie: Songs to the Tiny Tots. 5.0: Children's Session, conducted by Ursula George. 5.15: Union Jiggs. 5.30: Krays' College. 5.45: Growing Up. 6.0: Dinner Music. 6.15: Brushwood Sports—Mr. Oscar Lawson. 6.45: The Voice of England—A.B.S.A. Production. 6.55: A Little Humour. 7.0: Music. 7.15: Waltz Memories. 7.30: Bill, Mack, and Jimmie. 7.45: The Musicians—Featuring White Range Orchestras and Vocals. 8.0: Musical Highlights. 8.12: Modes and Moods of the Moment. 8.30: McColl Model Agency. 8.35: Build a Little Home. 8.45: Strange As It Seems. 9.0: Charm of the Orient. 6.15: Dinner Music. 6.30: The Talk of the Town. 6.37: Tea Time Topics. 6.45: Dinner Music. 7.0: Orchestral Selections. 7.30: Mrs. Arris and Mrs. Iggs. 7.35: Late Dinner Music. 7.45: The Singing Salesman. 7.50: Highlights of Harmony. 8.0: THE HUNCHBACK OF NOTRE DAME—GEORGE EDWARDS PRODUCTION. 8.15: Cole Porter Sucesses. 8.30: George Edwards and Neil Shirley in "Notre Dame Hunchback." 8.45: GARDENING TALK—S. H. HUNT. 8.55: Kennedy Arch. 9.0: Continental Dances. 9.15: Where to Fish—Mr. Oscar Lawson. 9.30: British Dance Bands. 9.45: TIER GAS. 10.0: New Versions of Old Favorites. 10.15: American Dance Bands. 10.45: Comedy Cameos. 11.0: Musical Miscellany. 12.0 midnight: Vitality in Rhythm—Henry Gregory Entertains. 12.30: Keeping the Party Going. 1.0: Cheerio Calls. 1.30: Winding Up the Clock. 2.0: News Flashes. 2.30: Hospital Nurses' Supper Session. 3.0: Relay—3rd Hospital Children's Bath Tub Session. 2KY SYDNEY, 1020 K'cycles [Commercial Station] Day Sessions as Friday, except: 1.0: Special "Argus" Broadcast. 4.25: Melody Session—John Harper. 5.15: Rion, and Eame—Children's Session. 5.30: Dinner Divertissements. 6.0: Stable Spy. 6.30: Dinner Entertainment—Goodo with Uncle Ben. 6.35: Spot of Honor. 6.45: Radio Show Shots. 7.0: Miss Melody Girl at the Piano. 7.10: Later News Commentary. 7.20: Musical Quiz—Unemployed Question. 7.50: Music. 8.0: Overture. 8.5: Music. 8.30: Musical Highlights. 8.45: "THE STORY BEHIND THE STAMP"—RADIO PLAY WITH 200 PLAYERS. 9.0: Music. 9.30: Melbourne Turf Topics—Mr. Francis. 9.35: Music. 9.45: Holiday Bureau—Secondo Cavalli. 10.0: Piano-Accordian. 10.15: Radio Rhythm. 10.30: Happy Hour. 12.0: Close. THURSDAY, FEBRUARY 11 2GZ Central N.S.W., 990 K'cycles [Commercial Station] Day Sessions as Friday, except: 6.45: Producers’ Service Session—Dancing and Pipes. 11.45: Our Round Babies. 3.30: Music. 3.40: Original Story. 3.58: Closing Out Section. 4.10: Royal Pavilion. 4.25: The Pictorial of the Air. 4.35: Fancy Free—Music. 5.30: Tiny Tots’ Session. 5.50: Children’s Session. 5.15: Backward and Forward. 6.0: Dinner Music. 7.0: Racing Final—Truk. 7.15: Wool Report. 7.24: Produce. 7.30: Weather. 7.30: Homebush and Newcastle Stock Sales. 7.50: Music. 8.15: Homebush and New South Wales. 8.30: The Smoke Social of the Air. ★ 8.15: HEROES OF MEDICAL SCIENCE. 9.0: Follow the Drum. 9.15: “Mr. and Mrs. Everbody.” 9.30: Swing Music. 10.0: Gallery of Fame. 10.15: Rhythm of Romance. 10.30: Close. 2HD NEWCASTLE, 1140 K'cycles [Commercial Station] Day Sessions as Friday, except: 11.30: Afterbreakfast Session. 12.5: Luncheon Music. 2.0: Waltz Memories. 3.15: Children’s Session. 6.0: Dinner Music Overture. 6.15: Soul. 6.20: Joyster Notes and News Paper. 6.30: Joe and Cynthia. 7.45: Sporting Review. 7.5: Ed and Zeb. 9.15: Torchbearers in History. 10.0: News. 10.30: Close. 2KO NEWCASTLE, 1410 K'cycles [Commercial Station] Day Sessions as Friday, except: 2.0: Radio Pictorial of the Air. 3.0: Light Entertainment Actuators!—An Exclusive Session on Hollywood, written and presented by Miss Elma Gibbs and Mr. Tom King. 5.45: Birthday Calls. 6.0: Dinner Music. 6.15: The Croonaders. 7.0: The Hunchback of Notre Dame. 7.30: Man About Town. 8.0: Inspector Bone. 8.15: The Law. 9.0: A Spot of Humor. 9.0: The Flams of Service. 10.0: The Love Corner, presented by Uncle Peter. 10.25: Good-night Melody. 10.30: Close. 2MO GUNNEDAH, 1360 K'cycles [Commercial Station] Day Sessions as Friday. 6.0: Children’s Music. 6.3: Serial Story. 6.15: Birthday Calls and Cheerios. 6.30: Entertainment News, and Bright Music. 7.0: Overture. 7.10: Light Music. 7.30: Stock Sales. 7.35: Light Music. 8.0: Shamrocks. 8.15: Music. 9.0: The Happy Hour. 9.30: Weather Report. 10.0: Close. 2CA CANBERRA, 1050 K'cycles [Commercial Station] 12.30: Luncheon Music. 12.45: Horoscope Reading. 12.55: Music. 1.10: News. 5.30: Children’s Session. 5.45: Children’s Store. 6.0: Adventures of Ben, Sam, George, and Tom. 6.30: Children’s Birthday Cheerios. 6.45: What’s On in Canberra? OVERSEAS STATIONS WHAT’S ON THE AIR TO-DAY 5.45 a.m. (DND 25.4 DJQ 49.8): A Radio Play. 6.49: Ballad, Intermezzo, and Romance (Brahms). 7.0: News. 7.20: Close. 5.50 a.m. (2DR 25.4): Variety Programme from Italian Stations. 9.0: News from Rome. 6.15 a.m. (GSB 31.5 GSD 25.8, GSI 19.6, with GSC at 7.0: The B.B.C. Symphony Concert, from Queen’s Hall, London. 6.45: International. 7.0: News. 7.20: Second Part of the B.B.C. Symphony Concert. 5.30 p.m. (DJB 19.7 DJA 31.3 DJN 31.4): An Evening in Old Berlin. Songs, Dances, and Berlin Humor. 6.30: The Colorful World of Opera. 7.30: News in German. 7.45: Hitler Youth. 8.0: Close. 6.0 p.m. (GSB 31.5, GSO 19.76): Big Ben. “Cue for Adventure,” a Play with Music. 6.31: The British Industries Fair for Young Instructors Talk by General Smuts. 6.46: A Studio Concert. 7.0: “Food for Thought.” Three Short Talks on Topical Matters. 7.40: News. 8.0: Close. 6.30 p.m. (3LR 31.3): Records, Commercial and Sporting Commentary. Music. 7.0: Relay of National Events Programme. 10.30: News Bulletin. 10.45: Music. 11.30: Close. 6.50 p.m. (JUN 28.14): Music. 7.0: Talks. 7.55: News of the World in English from Tokio. 8.0: Oriental Programme. 7.0 p.m. (3ME 31.5): Recorded Music. 9.0: Weather and Stock Exchange News. 9.10: Music. 10.0: Close. 7.0 p.m. (2DR 25.4): News Bulletin. 9.0: Programme from Homebush Studios. 10.0: Weather and Programme Announcements. 10.15: Music. 11.0: Relay of National News Bulletin from Daventry. 11.15: Music. 8.30 p.m. (WIXX 31.83): News. 9.0: Early Morning Music. 8.40: Market Reports. 9.0: Music. 10.0: News of the World from Boston. 10.15: Music. 8.30 p.m. (VPD2 34.6 YDB 31.1, PDP 27.3, PMN 29.2, YDC 19.81): Evening Musical Programme. 10.0: VPD2 closes. 9.0 p.m. (YDC 19.81): Concerts from Paris. 10.0: News of the World in English. 10.15: Music. 9.0 p.m. (DJB 19.7 DJN 31.4 DJA 31.3, DJQ 19.6, DJE 16.8, with DJR 16.56 at 10.0: Concert of Italian Music. 10.0: News in Italian. 10.15: Light Music. 11.0: News in German. 11.15: Women’s Session. 11.30: ‘Broad Meadows in Evening Dust.’ Lily Neitz. Sixteen Songs by Strauss. Midnight Song in English from DJQ. DJR 16.56 closes. 9.0 p.m. (GSB 31.5, GSG 19.8, GSH 19.39): Big Ben. Songs from the Early Talking Films. 9.30: The British Industries Fair—Talk by General Smuts. 9.45: Harold Coombes at the Piano. 10.0: News from Australia. 10.15: The Student Prince. Cannon Off the Cush. In the Mystic Land of Egypt. The Memoirs of a Tiny Shoe. 10.35: Cruel Dog Show. 10.45: Old Folks at Home. Memory of Popular Melodies of Yesterday. 10.55: News. 11.15: Dance Music. 11.23: Three Short Topical Talks. 11.45: Close. 9.43 p.m. (2RO 25.4): Variety Programme from Italian Stations. Midnight English Session. News of the World from Rome. 12.15: Music. 2GF GRAFTON, 1210 K'cycles [Commercial Station] 7.30 to 8.30: Merry Morning Session. 12.0: Weather and Markets. 12.30: Music. 1.15: Musical Box. 2.0: Close. 3.30: Aunt Judy, Uncle Col, and Ivan will Entertain the Children. 6.0: Dinner Music. 6.30: Music. 8.0: Short Story of the Air. 8.30: Music. 10.30: Close. 2LM LISMORE, 900 K'cycles [Commercial Station] 7.0: The Early Bird, Jim Sharp. 7.5: Wake-up Session. 7.45: The Homebush Flats. 8.0: News. 8.30: Music. 9.0: Close. 10.0: Afternoon Session, conducted by Miss Mollie. Beautiful Tunes. 2.0: The Music Box. 2.30: Tales and Tarraiddies. 2.45: Irish Music. 3.0: Yourself. 3.0: Close. 5.30: Children’s Corner, conducted by Aunt Mollie. 6.0: Dinner Music, with Keith Spencer. 6.30: Build a Little House. 7.0: Heart Songs. 7.15: Sporting Sessions. 20.0: Markets Reports. 7.30: Preface. 8.0: Harmony Isle. 9.0: Let’s Take the Music and Dance. 9.45: News. 10.0: Birthday Greetings. 10.0: Close. 2RG GRIFFITH, 1470 K'cycles [Commercial Station] 6.0: Children’s Session. 6.30: Music Lovers’ Concert. 7.0: Songs at the Piano, by Cyril James. 7.15: Snapshots of Comedy. 7.30: Market Reports. 7.45: Local News. 9.30: One Hour Dance Programme. 10.0: Golden-night Song. 2AY ALBURY, 1480 K'cycles [Commercial Station] Mid-day Sessions as Friday. 5.30: Children’s Session. 6.0: Dinner Music. 7.0: Station Notices. 7.30: News Bulletin. 8.0: Gems of the Evening. 8.15: Theatre of the Air. 8.45: Harmony Isle. 9.15: Country Man’s News. 9.30: Waltz Time at SAX. 10.30: Close. 3LO MELBOURNE, 770 K'cycles [Aust. B’casting Commission] Day Sessions as Friday, except: 9.35: The Radio Sketch “Charlie Campion” by J. J. Wolfe, Read by “Scribe.” 10.0: Current Happenings in Sport, by H. C. Cumberlidge. 10.25: “An American Luncheon in Melbourne”—Miss Marjorie McNeale. 11.3: Physical Training—Mr. P. W. Pearce. 12.0: “The Home Sympathy at Home and Abroad,” by “The Spectator.” 12.30: “Speech”—Mr. W. H. Frederick and M. T. H. Rose. 3.15: Philharmonic Orchestra, conducted by Dr. Weingartner. Prometheus. Overture. (Beethoven.) Lucrezia Bori, Soprano, and Lawrence Tibbet, Baritone. Call at the Night Club, No. 1 (Goetze), British Symphony OrWIRELESS WEEKLY, FEBRUARY 5, 1937 CHESTRA, conducted by Sir Henry Wood. Brandenburg Concerto, No. 3 in G (Bach). 2:45: Kirsten Flagstad, Soprano: Die Walkure—Brunnhilde's Battle Cry (Wagner); Der Zerlaeser, Op. 18, No. 6 (Strauss); Yehudi Menuhin and The London Symphony Orchestra, conducted by Sir Edward Elgar: Cello Concerto No. 1 in G Minor (Max Bruch): Allegro Moderato. Adagio Allegro. Finale. 1:45. Close. 6:30: Overture, A Voyage with "Sparks." The Show Must Go On—The Man Next Door," by Claire Mellion. Finale. 6:30: Dinner Music. 7:46: A.B.C. (Sydney) String Orchestra. (See 2FC.) 8:0: A.B.C. (Sydney) String Orchestra. (See 2FC.) 8:5: Interlude. 9:5: Irene Bennett, Soprano. (See 2FC.) 9:5: From Adelaide—"Current Topics"—This space is reserved for interviews with prominent persons, special talks on current events, and the like. 9:40: From Adelaide—A Salon Recital. (See 2FC.) 10:10: "Maidens and Memories." 10:30: Cricket—Detailed Scores A Summary of the Day's News. 10:40: Dance Music. 11:30: Close. MELBOURNE, 580 K'cycles [Aust. B'casting Commission] Day sessions as Friday except: 10:30: From the King's Theatre—Community Singing. 3:0: A Violin Recital by Tossy Spivakovsky. Raymond Lambert, Accompanist. Preludio and Allegro (Pugnani-Kreisler). Introduction et Ronde Chinoise (Pugnani-Kreisler). Note: During the day progress scores will be given of the Cricket Match England v. Country Team played at Canberra. Results of the Pakenham Races will be given as they come to hand. 4:45: From St. Paul's Cathedral—Evensong. 6:0 Music. 6:15: Evening Session—Dr. Anita Rosenberg. 6:31: An attempt will be made to re-establish contact with the Eastern Station—The British Industries Fair, 1937. An introductory talk by General Smuts, from South Africa. 6:48: Sportine Session, conducted by A.B.C. Commentator. 7:0: I Look at Life—Mr. Alan Moorehead. 7:15: Music. 7:20: National News Bulletin. 7:30: Victorian News Bulletin. 7:35: Countryman's Session. 8:0: See 3LO. Detailed Scores of Today's Play. 8:2: Unit One—Frédéric Earle at Electric Organ; while at Battling at the Piano: Serenade (White), Le Papillon (Lavallee-Cornicelli (Lee), Russian Romance (Prima), Russian Dance (Prima). 8:15: Interlude. 8:20: "The Cosmopolitan Four" in their Variety Show. Spinning Song from The Flying Dutchman (Wagner): On Wings of Song (Mendelssohn). In the Woods (Traditional). Spinning Song (Coningsby Clarke): Greeting (Schumann): Look, Sister, What Say You? (Smetana): Ave Maria (Schubert): Midnight Bells (Fritz Kreisler): Auf Wiedersehen (Hoffmann): Lithuanian Song (Chopin): Margaret at the Spinning Wheel (Schubert). 8:35: Interlude. 9:0: Alfred and Vera Field-Flache in a Comedy Sketch entitled "Two Sharps and One Flat," by J. F. Pritchard. 8:15 Interlude. 9:20: "Listeners Inviting Rhythmic Refreshment by the National Dance Orchestra, conducted by Al. Hamnett. 10:0: Interlude. 10:15: News. 10:30: Close. BRISBANE, 800 K'cycles [Aust. B'casting Commission] Day Sessions as Friday, except: 11:15: A Beauty Culture Talk. 11:40: Talk by Mr. H. M. Smith (arr. Australian and Society of Queensland) — Preparation for Sweet Peas. 12:00: Broadcast to Schools by H. J. McKenna, of the Education Department — Tour of the Atherton Tableland. 12:30: From Cunningham Club — Mid-day Lecture. 3:35: Broadcast to Schools by W. H. Haynes, B.A.—Senior Modern History: (2) Some European Descendants of the Mother Parliament. 5:00: Children's Corner, conducted by Capy and Bebe. 5:45: The Music Museum—Dinner Music. 6:45: Sporting Notes, Music. 7:0: Talk by J. Picot, B.A. — This Modern Poetry: What Is It? 7:15: Interlude. 7:20: From 2BL — National News Bulletin. 7:30: News. 7:30: Movie News. 7:40: Air Mail Times; Weather Data. Late Market Reports; General Commercial News; Programmes and Announcements. 8:00: From 2FC—The A.B.C. (Sydney) String Orchestra. (See 2FC.) 8:55: Recital by Bina Addy (Indian Soprano-Contraalto) and Peggy Woods (Violin). Soprano: Sundara Radhe (Tagore); Weep Not, My Country (Tagore); Violin: Minuet (Boccherini); Valse de Concert (Musini); Rondino Kreisler. Soprano: Boro Saha (Tagore); Tagore Gramchara (Tagore); Violin: Songs My Mother Taught Me (Dvorak); Rondino Kreisler. 9:25: From 5CL — National Talk: Current Topics. 9:40: From 5CL Recital by Fred Sullivan (Tenor), Spruhan Kennedy (Piano), and Constance Petrie (Flute). 10:0: A Recital of English Folk Songs, by Cleve Carey. (B.B.C. Recording). 10:15: Weather: Late Sporting. 10:40: From 2FC—Dance Music by Jim Davidson's A.B.C. Dance Band. 11:30: Close. SALE, 830 K'cycles [Regional Station] 7:0: See 3AR. 7:37: See 3LO. 8:0: See 3AR. 9:30: See 3LO. 11:30: Close. 12:0: See 3LO. 12:15: See 3AR. 12:30: See 3LO. 1:10: See 3AR. 2:0: Close. 3:0: See 3LO. 4:15: Close. 5:30: See 3AR. 6:0: See 3AR. 6:15: Local News Service including Gippsland Market Reports. 6:30: See 3AR. 8:0: See 3LO. 11:30: Close. COROWA, 670 K'cycles [Aust. B'casting Commission] 7:0: See 2FC. 7:25: See 3LO. 7:37: See 3AR. 9:30: See 3LO. 11:30: Close. 12:0: See 3LO. 12:35: Local News. 12:45: See 3AR. 1:0: See 3LO. 2:0: Close. "From the Hill-Billy Cabin" With the vogue for Hill-Billy music so strong, it has been arranged to present a session "From Hill-Billy Cabin," each Saturday evening, from 2FC, at 9, on February 6, 13, 20. It will be an atmospheric presentation by a group of vocalists and instrumentalists. Listeners may look forward to entertainment along the lines of one of the most successful radio sessions abroad, "By the Rocky Mountaineers." Broadcasting "From the Hill-Billy Cabin" will be male vocalists, an accordion player, a guitarist, a mouth-organ player and a country yodeller. Not only will they broadcast all the newest Hill-Billy numbers but many of the old favorites. The session will last for fifteen minutes and will be followed by the "Father Brown" series. REDUCE YOUR WAIST! Fat-infiltrated muscles cause that waistline bulge leading to weakened, abdominal sagging and causing morbid and dangerous changes in the position of the vital organs. The Governa Health Belt will reduce your waistline and give you a slim, athletic figure. Approved by medical men. Supports the delicate organs, and, by its gentle, changing pressure, waistline bulge is eliminated with every move you make. TRY THE GOVERNA BELT AT OUR EXPENSE In 7 days, your waistline will be 1 inch smaller, free of pain and strain—OR NO COST. Write for full details of FREE TRIAL OFFER and illustrated folder describing the Governa Belt. Mention this paper. Enclose 3d. stamp for postage. GOVERNA BELT COMPANY, 252-6 Castlereagh St., SYDNEY. P.M.G. TYPE MORSE CODE KEY 19/6 Adjustable Heavy Brass Fittings. 12/6 Like-a-Flash Professional Morse Keys, 1/6. P.M.G. Type, 19/6. Both adjustable and reset type made. Buzzers, 1/6, 2/6, 3/6, 5/6. High-pitched tone adjustable Morse Code Buzzer, professional type, 7/6. Emicel English Volt M/A Meter, Bench type, 18/6. Reads 0 to 5 v., 0 to 150 v., 0 to 50 ma., 0 to 2000 ohms. Emicel English Pocket Meters, 1/6, 2/6, 5/6. v., 0 to 15 v., 0 to 180 v., 0 to 30 ma., 12/6. Radio Saws, 3 Detachable Blades, cut metal, bakelite, or wood, 1/3. Batteryless Hand Microphone, just plug into pick-up terminals of set, 45/- value. Now 35/-. Music, Song, or Speech. Others 19/6. MORSE CODE PRACTICE SET FOR BEGINNERS, 4/- complete, as illustrated Headphones Caps, rubber pads for comfort, 1/3 pairs. Every Man's Wireless Book, giant size, just landed, 7/6. Telsen English Meters: Pocket type, bakelite case, 2 readings, volts, 7/6; 3 readings, volts and m.a., 10/6. Universal A.C./D.C. bench type, reads volts and m.a., 25/-. Rebuilt Battery Operated Metal Chassis, 2 valve March 4 1937 Radios, Floor Model Cabinet. Heavy Duty equipment, everything complete, £13/10/- value. Now 2/20/-. Plus packing and freight. Rebuilt Radios have all tested components—85 per cent, new material, accessories, and components. All Batteries and Valves Brand New. Mighty Mite Midget Crystal, 4 v., 15/-. Just a handful. Now 10/-. Complete with Phones, Aerial, Earth, 27/6. Miniature De Luxe Audio Transformers, replaces any other, 8/6. Pocket Volt Meters for A and B battery reading, 2/6 and 3/6. Giant Scell Focussing Torches VALUE NOW 6/6 Complete PHILIPS MID-POT SPEAKER ASSEMBLY, 7/6 Packing, 1/6 extra. All parts excepting input transformer and field coil. Limited supplies now remaining; order early. HEADPHONES Lightweight Continental Headphones, 10/6. 22/6 line, Now 18/-. 12 months' guarantee. Collapsible Headphones fold into small space, suitable 1000 ohm, 10/6. Triad, 1000 ohm, Professional Model, 4000 ohms, 28/6. Like-a-Flash Electric A.C. Gramophone Motors complete, humless, £3/5/- value. Now 35/-. Packing 1/- extra. Super De Luxe Model, 75/-. Valves, NEW AND TESTED. Electric Types, 245, 255, 224, 8/6; 246, 247, 2415 Type, 5/9; 250, 8/6; 246, 10/6. Mullard Battery Valves, 2 volts, PM2, PM12, PM22, 8/6. Mullard 4 volt Det. R.F. or Output Battery Valves, 8/6. WRITE FOR PUNCH BOARD AND NOVELTY GAME LEAFLETS. Fix a "Pix" in a jiff, and say goodbye to your wasted transmitting inter-station crosswires "Pix." The simple English wave trap with spring terminals and mounting clips, complete, 4/6. "Gold Seal" U.S.A. Manufactured, 280 Rectifying Valves, 13/6 Value. 8/6. Levenson's Radio Handbook Levenson's Radio Bargain Bulletin, 120 pages, 8d. post free. Also Levenson's Radio Buyer's Guide and Assembly Chart Handbook, 3/-. post free. New, Tested, Charged. 12 Months Guarantee Add 1/6 postage, plus freight. 2 Volt 160 Amp. 15/- 3 Volt 100 Amp. 15/- 4 Volt 60 Amp. 18/9 5 Volt 85 Amp. 27/9 6 Volt 40 Amp. 22/6 7 Volt 60 Amp. 23/9 Like-a-Flash B Batteries. Please add freight. 45 v. Heavy Duty 7/3 45 v. Heavy Duty 11/9 45 v. Triple Duty 14/6 C Batteries, 4½ and 9 v. 2/- Large Size Hydrometers, for testing accumulators, 3/6 value. Now 2/6. Levenson's Radio 226 Pitt Street, Sydney AND RADIO CHEAPSIDE, 240A Pitt Street. WRITE FOR Pin-Game, Totem, and Race Wheel LEAFLETS. P.M.G. TYPE SOUNDER 35/- Heavy Brass Fittings, heavy oak base, 63/- value. Now 35/-. Cosser New and Tested Valves, 610 xp. 2/6; 625P, 2/6; 411MRC, 2/6; 4 XP, 2/6; 4181, Rectifier, 2/6; 610 XL, 2/6; Volt. Det. and R.F. Valves, 5/6; DU2 and DU10, UX Base Rectifiers, 7/6; 1 volt. Bait. Valve Only, 1/6; Transistor, 2/6; 2 and 3 Volt English Base Valves, for experimenters, 2/6. "ADMIRALTY" RUBBER CLAD AERIAL 30 Strands 50ft. 2/6 52ft. 5/2 "COSMOCORD" 25 Amplion (English), 28/6. Gramophone Shipsets, B.T.H. Bakelite Model, 32/6. "Cosmocord" 25/-. "Cosmocord" De Luxe, 32/6. "Cosmocord" Super de Luxe, 45/-. "Pallard," U.S.A. make, with separate volume control, 28/6. Send for Leaflets of Radio Publications. "CLARION" PICK UP AND TONE ARM COMPLETE WITH VOLUME CONTROL BAKELITE MODEL BRITISH MADE 98% PURE TONE 22/6 Electric Klaxon Horn, with all fittings and push button, 7/6. Fitted in a jiffy. Splendid for autos, shops, offices and stores. Electric Light Generating Sets with Crystal Head and Tail Light, complete, with extra globe, 9/11. Super Grade de Luxe Model, 15/-. Packing, 9d extra, plus postage. A MODERN TELEVISION CIRCUIT The recently introduced television service in England has now been in operation long enough to give a fair idea of public reaction. So far television receiver sales have been very poor, only a handful of sets being purchased for household use. A number of people seem to feel that the price of television receivers will fall as soon as they are produced in large quantities, but a study of this circuit will show that it is not likely that television receiver prices will be lowered, for most of the component parts are already produced in mass quantities for ordinary radio receivers and costs have already been reduced to a minimum. There is, therefore, not the same scope for lowered prices with mass production, which was one of the features of the early history of radio broadcasting. It is said that the actual number of television receivers sold in England since the opening of the recent experimental service has not yet reached one hundred. With radio receiver sales amounting to thousands upon thousands every week, the figures for television receivers seem disappointing. The cathode ray tube unit, which requires a high tension voltage of 4500 volts, and a cathode ray tube costing many pounds. The time base unit, quite complicated and containing many adjustments which need to be operated in order to get the television image correctly screened. Circuit of the main portion of the television receiver. Five stages of intermediate amplification are necessary. A special form of vertical aerial, erected out in the open is required, and has to be accurately cut to length. The whole public acceptance of television has been disappointing, according to a radio man who returned from England last week, even the public demonstrations in the departmental stores failing to draw crowds. Perhaps one of the reasons why the public are hesitant about buying television receivers is the comparatively high prices at which these receivers are listed, mostly around £100 in English money, equivalent to more than £125 in Australian money. The development of radio receivers being comparatively recent, it is well within the memory of the public that radio receivers started with list prices around £100, but soon dropped to their present level of £15 to £30. Although the public don't appreciate the fact, those who have followed television feel confident that similar reductions in prices with television receivers will not be possible. To explain the reason we may point out that modern radio receivers employ just as many condensers and resistors as the early models; in fact, in most cases a great many more components are used. But the individual cost of each of these components has been reduced step by step from, in the case of fixed condensers, about 3/- to 4d each. But television receivers seem to employ about twenty times as many resistors and condensers as a radio broadcast receiver, and, since the unit cost of each of these units has already been brought to bedrock, there is not the same scope for a reduction in prices, unless a method is found for simplifying the design of the television receiver. At present there seems little chance of this. For those who may be interested we show the circuit of a television receiver, as designed by Mr. F. H. Haynes, one of England's leading designers, for the "Wireless World." This television receiver gives a fair idea of the necessary complexity, having over one hundred resistors, over one hundred condensers, and such bulky and expensive equipment as a power supply, including power transformer, rectifier, filter condensers, and other parts capable of working at a high tension voltage of 4500 volts. COMPLEXITY OF CONTROL Some idea of the complicated control mechanism is also given by the circuits, which show the main tuning control for the sound, three variable controls for the cathode ray tube unit, eight controls on the time base unit, and about three more on the television receiver proper. It should be remembered that the circuits only show the additional equipment necessary to add to a good broadcast receiver! Suitable filter condensers are not listed in Australia, but a 4 mfd. filter condenser to work on 3000 volts lists at £15/6/8. Large cathode ray tubes are not listed in Australia either, but there is a 9in. cathode ray tube, suitable for reproducing an image about 6in. x 8in., which lists at a little over £50. How would you like to drop one? SUMMARY In a nutshell, it looks as though our previous forecasts about television are not the slightest affected by the recent developments, and a cheap, satisfactory, and attractive television service is unlikely to be found in Australia for many years to come. The production of attractive programmes is one problem, and the cost of manufacturing television receivers is another. EXPERT RADIO REPAIR SERVICE. We specialise in repairing Philips, Alzone, Strons, and Carlton, Radios, Telecs or any Wireless Weekly Set. We have been established since 1926, and are competent to handle any class of radio repair or service in COUNTRIES ABROAD! We will send you a test report the same day as we receive your set. ELECTRICITY HOUSE. Robert T. A. Waters Manager (Incorporating The Retro - Electrical Trading Co.), 11 Surf Road, CRONULLA. Telephone, Cronulla 296. We pay freight in the Metropolitan area. CITY RECEIVING DEPOT, 340 Kent St., Sydney. MA2344. RADIO BARGAINS Philips Eliminators, 10/-; A.W.A. 15/-; Philips Audio Transformers, 10/-; A.W.A. 6/-; others from 3/-. Lightning Arrester and Switch, 6d; Variable Condensers from 2/-. Reliance Coils, 1/6. Valves: 12AT7, 12AU7, 6AU6, 6AU5, 221, 222, 3/6, 4/6, 32 6/6, EAT 7/6, 6BT 7/6, 2AS 6/-. Cord Grip Holders, 3d. All tested and guaranteed. Add Postage. RADIO SUPPLY STORE 7 ROYAL ARCADE, SYDNEY. Announcing the new CARBON TYPE POTENTIOMETER INCORPORATING FERRANTI ELEMENT Ferranti’s years of experience in the manufacture of highest quality radio components are behind the essential parts of this new Potentiometer. Silky in Action . . . Constant in Value . . . Robust in Construction. Stocked by all Leading Radio Stores NOYES BROS. (Sydney) LTD. 115 CLARENCE STREET, SYDNEY. 11 WATT STREET, NEWCASTLE. Phone: B7581 (10 Lines) 197 ELIZABETH STREET, BRISBANE. HAMMOND INDUSTRIAL and OFFICE ELECTRIC CLOCKS No Winding—No Ticking No Loss—No Gain—No Regulating ONLY OBSERVATORY TIME ALL THE TIME Priced from . . . £4/10/- Domestic Models also Available from 25/- BICHRONOUS TYPES. CLARION. A handsome Office Clock, combining appearance and utility with lift up face. A Bichronous (non-stop) clock. Colour: Walnut. Dimensions: Height 18in., Width 18in., Depth 4½in. Weight, 13½lb. Dial, 14in. Cord: Brown, 9ft. 6in. long. PRICE, £7/10/- SENTINEL Veneered walnut case with lift up front. Perfect visibility at 150 feet. Ideal for office or factory. A Bichronous (non-stop) clock. Colour: Walnut. Dimensions: Height 18in., Width 18in., Depth 4½in. Weight, 13½lb. Dial, 14in. Cord: Brown, 9ft. 6in. long. PRICE, £7/10/- CENTURION A new Hammond Industrial Clock of a synchronous type. It has a 11¾ inch face of German Silver with black Arabic numerals. The finest, cheapest industrial clock on the market. Colour: Marine Bronze. Dimensions: Overall Diam. 16¾in. Depth, 3¾in. Weight 5½lb. Dial, 11¼in. Cord: Brown, 9ft. 6in. long. PRICE, £4/10/- PUBLICATION 54B GIVING DETAILS OF BOTH INDUSTRIAL AND DOMESTIC MODELS GLADLY MAILED UPON REQUEST. DISTRIBUTORS QUEENSLAND: Edgar V. Hudson Pty., Ltd., 284 Edward Street, Brisbane, Qld. SOUTH AUSTRALIA: Newton McLaren, Ltd., Leigh Street, Adelaide, S.A. Gerard & Goodman Pty., Ltd., Synagogue Place, Adelaide, S.A. VICTORIA: Henry G. Small & Co. Pty., Ltd., 11 Hardware St, Melbourne, Vic. TASMANIA: W. & O. Genders Pty., Ltd., 53 Cameron St, Launceston, Tas. W. & O. Genders Pty., Ltd., 69 Liverpool St, Hobart, Tas. MAKERS OF THE FAMOUS AMPLION LOUD SPEAKERS. Advertisement of AMPLION (A'sia.) LTD., 66 Clarence St., Sydney Australia’s Pioneer Radio Training Institution ESTABLISHED 1913 Complete Courses of Instruction in all branches of Radio RADIO ENGINEER: Highly specialised training in every branch of Radio Engineering, including practical training at the A.W.A. Radio Centres at Pennant Hills and La Perouse, in addition to workshop instruction at the A.W.A. Radio-Electric Works and Laboratory. RADIO TECHNICIAN: Training includes instruction for the P.M.G.’s Broadcast Operator’s Certificate of Proficiency. Practical Instruction at the School and A.W.A. Radio Centres. The School is equipped with modern C.W. and I.C.W. and broadcast transmitters. Call, Write or Phone BW2211 For free 40-page illustrated prospectus. MARINE OPERATOR: The only School in Australia equipped with complete marine stations and auto alarm equipment to enable students to qualify for the P.M.G. Certificates. 95% of operators in the Australian Mercantile Marine are Marconi School graduates. TALKING PICTURE OPERATOR: Theoretical and practical training on standard theatre equipment. RADIO MECHANIC: Advanced theoretical and practical courses in broadcast receivers and servicing. RADIO SERVICEMAN: Correspondence Course in servicing broadcast receivers. Engineering and Technician sections are under the direct control of Dr. W. G. Baker, B.Sc., B.E., D.Sc.E. MARCONI SCHOOL OF WIRELESS 97 Clarence Street, Sydney Conducted under the auspices of Amalgamated Wireless (A'sia.) Ltd. A Revival of Direct-Coupled Circuits? Prominent Factory To Use Direct-Coupling For 1937 Models BACK in the year 1929 a "new" circuit was developed, one which was cheap to build, yet gave exceptionally fine tonal quality. It boomed and boomed for about a year, and then flopped. It looked so simple, and only a few cheap resistors were used, but how those amplifiers could play up if anything went wrong. So long as the valves had normal characteristics and the resistance values were true to label, the amplifiers were really good, but if they started to motor-boat, distort, or play up in any other way, it was almost impossible to say what the trouble might be. The two valves were so closely tied together that a fault in one would make it appear as though the other was the cause of the trouble, and, resistors and valves being what they were in those days, and with the lack of sound technical knowledge, which was also a feature of that period in radio's progress, WATCH YOUR WATCH A reader has suggested that we should issue a special warning about the way in which watches can be ruined if they are brought within reach of the magnetic influence of the speaker-field. Both wrist and pocket watches are inclined to become magnetised when worn by radio engineers or those working close to loud speakers. Even the modern permagnetics have such strong field magnets that a watch can be affected at a range of several inches. it was no wonder that direct-coupling circuits flopped out of existence. Memories of the tonal quality of the old "1930 Three" and "1930 Four" receivers brings to us a steady, if small, demand from readers for a modern version of a direct-coupled circuit. In last week's issue we let fall a mention that we had been toying with direct-coupled circuits again, and so we have had an increased demand from readers for further details of a modern direct-coupled amplifier. As it happens, the circuit we were playing with was taken from one of the 1937 model commercial receivers, being marketed by a firm which did quite a bit with the original direct-coupled circuits in the old days. Dealers who sold the old sets never forget the quality of reproduction, and so for the new season a direct-coupled superhet, is listed. On this page we show the circuit used for this receiver. A 75 type (2A6 could be substituted) valve is used as diode detector and supplies automatic volume control voltages and is directly coupled into the grids of a pair of 45 type triodes used in a parallel output circuit. A special field coil of 2200 ohms is used for the main biasing of the 45 type valves, giving them a filament potential of 150 volts in respect to earth, the plate current of the two valves being about 65 milliamps. Applied voltage to the plates of the 45's, in respect to earth, is 445 volts, giving an actual plate voltage (plate to filament) of 220 volts. Correct bias for the output valves is obtained by having the grids of the 45 type valves about 50 volts negative in respect to filaments, the latter being 150 volts above earth, meaning that the grids have to be about 100 volts above earth, this being also the effective plate voltage of the triode portion of the 75, and is obtained by the plate current flowing through the 1 megohm resistor, which serves as combined grid leak and plate feed resistor. CHOKE REQUIRED The maximum high tension voltage is 445 volts, a little higher than usual, but low enough to be used with good quality electrolytic condensers and not nearly as high as the voltages used in the original direct-coupled circuits with such tragic effects on filter condensers and other components. The speaker field cannot be used as a choke, however, meaning that a filter choke must be used in the h.t. supply. ISOLATION FOR THE R.F. The whole of the r.f. end of the receiver is entirely isolated from the audio amplifier in the receiver under discussion, the main high tension supply for the tuner portion being taken from the 445 volts of the amplifier, but through a 5000 ohm dropping resistor, suitably by-passed. CRITICAL FEATURES With an amplifier of this type it becomes most necessary for all valves to have their normal characteristics, and under no circumstances can the amplifier be operated without all valves in place and a field coil of the right resistance (although we found the latter sufficiently flexible to allow the use of a 2500 ohm field without seriously upsetting performance). A separate filament winding for the two output valves in the amplifier is essential, but the tuner end of the receiver can also operate from the same filament winding as the first valve. SOME CONSIDERATIONS Working on this amplifier we have found it possesses several advantages over the old systems, the first being the fact that a triode is used for the first amplifier, whereas the old systems always used a screen grid type of valve. The screen voltage was a critical feature, and variations in the plate current of the output valve affected the voltages on the screen, which in turn affected the plate current, thereby the grid bias of the output stage and its plate current in turn, being something like a dog chasing its own tail and causing an awfully messy tie-up to sort out when any troubles were being chased. On the other hand, it is possible that a screen-grid valve with a screen whose voltage can be controlled might be a definite asset, as the plate current of the valve, and consequently the bias of the output stage, can be controlled by this screen voltage. It is along these lines that we are working at present. Uniformly high Quality KEN-RAD Valves are recognised as world leaders today because of their superior construction, and the uniformity of quality that characterizes each single valve. In general replacement, and amateur set-construction, always use "The Fine Valves of Radio" for better performance and longer life. Factory Representatives: EASTERN TRADING CO., LTD. SYDNEY AND MELBOURNE Short-Wave Notes by R. N. Shaw From Czechoslovakia Some few months ago we reported the first tests carried out by the new broadcasting station at Poderusy in Czechoslovakia, when signals were heard equal to anything heard on short waves in Australia. It was then understood that after the beginning of this year the station would settle down to a regular schedule. It has been heard during the past month with variations in carrying strength, but at no time have its signals reached the strength observed on the opening date. This week we have received from the Czechoslovakian Broadcasting Foundation reports regarding the station. The broadcasting authority is known as the Czechoslovak Firm Radio Journal Praha, which announces itself as "Short Wave Station Prague." It sends out for the time being experimentally the following programmes:—Regular daily experimental broadcast on group programme from 3 to 6 p.m. on wavelengths of 25.24 metres and similar programmes daily on 49.05 metres between 8.25 and 10.30 p.m. This programme also contains programmes in German, French and English at 7 p.m., in German at 8.55, in French at 10.0, and in English at 10.15 a.m. Apart from these schedules, two hours' Thursday evening special broadcast is sent over the American zone between 1 and 3 a.m. This contains news in Czech at 1.25 a.m. and music at 2.40 a.m. The wavelength then used is 25.24 metres. We should be pleased to receive from listeners in various parts of Australia reports of reception during the various sessions. At our receiving station the Czechoslovak has been fairly well on the 25 metres band, but is very weak on 49 metres. Up in Western Queensland, Dr. Gaden (Camoeowall) has reported hearing the station in the early morning at very good strength. Last Sunday afternoon we heard the Czech on an announced wavelength of 25.24 metres at excellent strength, beginning at 7.30 p.m., but it was much weaker towards 9 o'clock. All the English announcements were made by a lady, speaking quite good English. Towards 8 p.m. the Czech was heard on maximum carrier strength, and was then transmitting every-possible choral numbers. In the half-hours the station was listened to announce as "Our Short-Wave Station Praha," but at 7 p.m. quite a lot of information was given, and reports from distant listeners asked for. Listening to other countries, one thinks in very nice terms for the assistance they had rendered the station engineers by forwarding valuable reports. The lady quoted in the programme concluded by expressing the hope that everyone "would have pleasant listening." SPECIAL COLOMBIAN PROGRAMME We have previously remarked upon the large number of new American stations which have this season been heard on various bands. Most of these are located in Colombia or Cuba, and we are also inclined to think that there is less money than a little friendly rivalry amongst the rival stations, who appear to be vying for popularity in overseas countries. These stations have been heard at all times of the day, afternoon, and night. And now the well-known Colombian, HJIABP, of Cartagena, and which operates on 22.2 metres, announces a special broadcast for Australian listeners. As a matter of fact these stations are being heard better in New Zealand than in Australia due to the nearer time difference. A letter to hand from a Press correspondent Mr. J. D. Watson, of the Manawatu "Evening Star," who has recently had lengthy communication from HJIABP regarding the special broadcast. This will take place at 10 p.m. on February 28. E.S.T., representing 1 p.m. on Sunday, February 21, in Western Australia. The station owner states he has received such fine reports from Australian listeners that he is anxious that this broadcast should be made widely known. He also very anxious to make contact with a Sydney or Melbourne radio engineer over the air, and mentions 8 p.m. Sydney time as preferable. We consider such contacts and broadcasts will tend to cement mutual understanding between Australia and South America. He desires to obtain detailed reports on his transmissions, and issues a rather fine DX card. On 31.2 metres a power of 1000 watts is used. We should be pleased to receive reports from any listener logging the special broadcast, and will forward such reports and the names of the authors to the stations. NEWS OF THE STATIONS Rather freakish conditions have prevailed during the late afternoons this week. As listeners know, most stations have disappeared in daylight until about 6 p.m., but on Sunday we had no difficulty in logging DJN and DJA, of Berlin, on the 31 band, and DJB on 19.7 metres, shortly after 4 o'clock. The Czech station on 25.24 metres was also coming through excellently during the afternoon. However, a search at 3 p.m. on Monday proved a waste, as there was not the slightest trace of any station, except our own 3LR, which was coming through at maximum strength. Stations audible in the early morning quite early, under the worst conditions, were as early as 6 a.m. But on Monday morning quite a number of stations were coming through excellently before 7 a.m. was reached. The three Empire stations GSE (31.5), GSD (29.8), and GSI (19.6), were all good strength, and a number of others on the 31 band could be followed excellently. Rome on 31.12 TPAA on 25.24 metres, WIXK on 19.71, WXXAD on 19.58, and Cubans on 31 and 30 metres, were quite audible. On the 31 band, stations were WXXAF and LJK on the 31 band, these stations badly heterodyning one another at 7 a.m. The German DJR on 19.8, was also heard, with a weak signal, and was heard on 49 metres which was probably DJC. News in English, and quite a good deal of it, is now heard from TPAA (25.24) at 6 a.m. Stations audible in the early morning now readily fade away, and at which hour GSD (31.8) opens strongly, but in half an hour has also disappeared. At 8 a.m. the only audible station is the American, WEKK, on 19.71 metres. The Japanese, JVN, on 28.14 metres, has almost suddenly come back into prominence this week, and can be heard daily from about 6.45 p.m. and sometimes even earlier. It reaches excellent strength at this hour, and gives an English news session at 7.55 p.m. Another Japanese is heard some evenings on 31.4 metres, but this station still does not dominate the position on that band, there being at least three stations operating on 31.4 metres during late evenings. Evening stations are again in abundance from Paris (TPAS) operate strongly at 7 p.m. until after midnight. On 31 metres the following were heard; GSB, DJN, DJA, ZBW, YDB, WIXK, JHJABP, CQCO, 3LR and 3ME, whilst on 34.3 metres VPDA of Suva, Fiji, provides a strong signal until 11 p.m. Another strange station, with good signal, has been heard on about 35 metres, but no announcement was heard, although many musical numbers apparently received were heard. On 25 metres Paris (TPAS) operates strongly at 7 p.m. with an excellent musical programme, with English news about 7.40, and closing at 8 p.m. There is then little to be heard on this band until 10.30 p.m. Among the stations heard until midnight included a Japanese, Radio Saigon, PHI and 2RO, whilst on Wednesday and Thursday the number is increased by Moscow (IRNE), between 9 and 10 p.m. in an English session. The 19 metres band still maintains its general superiority every evening, commencing with GSD and DJN at 6 p.m. Then at 7 p.m. and on until early morning, others heard strongly are DJQ, DJR, TPAA, PCJ, and YDC, whilst on 27 and 29 metres the two Dutch East Indian stations are excellent. The remaining stations are a mystery, but on Friday night 9ML, the transmitter on the steamer Kamisiba, was heard strongly around 10 p.m. Another French station has been heard in the evening on 29 metres, about 3 metres, or practically where our erstwhile friend, Radio Rabat, was heard in days gone by. Dr. K. B. Gaden (Camoeowall) reports the Japanese very strong on 28.14 metres, whilst JVN (28.4) was also heard strongly, and at which hour WXXAF (31.4) is also strong, sometimes better than GSC, which is fairly good till 8. Again DJN (31.4), which we cannot hear, is reported by this listener as being good, but the opposite prevails in early evening, when he reports GSO (19.76) as often inaudible, GSB (19.6) very weak, DJN (19.7) splendid. But by 7 p.m. QSO is very good, with GSB towards 7.30. Stations all that could be detected from 7 p.m. with DJN the best on, and ZBH first at 21. It is difficult GSO (19.8) is better than GSB on opening with ZRO (25.4) also very good, whilst Berlin on 19.56 is very fine. CLASSIC COILS • IF'S • SPEAKER TRANSFORMER WINDINGS "Classic" components revolutionise set-construction with increased efficiency, greater stability, and longer service. "Classic" Aerial, R.F., and Oscillator coils, and intermediate transformers, are designed for high fidelity results. All intermediates obtainable ordinary or Litz wound on genuine Hammarlund Isolantite base. "Classic" speaker transformer windings are manufactured in impedances to suit every speaker. JOHN MARTIN LTD RADIO & ELECTRICAL SUPPLIES 116-118 CLARENCE STREET, SYDNEY Telephone: BW3109 (2 lines) Telegrams: "Jonmar," Sydney Used by Electrical Manufacturers Everywhere— LEWCOS ENAMEL and PAPER COVERED WINDING WIRE Higher Insulation... Better Space Factor... Non-ageing and Flameproof BRITISH and BETTER 233 CLARENCE ST. SYDNEY 586 BOURKE ST. MELBOURNE THE LIVERPOOL ELECTRIC CABLE CO. LTD. AGENTS: Adelaide—Robert C. Forbes & Co., 30A Currie St. Brisbane—(Industrial)—Intercolonial Boring Co. Ltd., Ann St. Hobart—Wm. L. Buckland Pty. Ltd., 57 Liverpool St. Launceston—Wm. L. Buckland Pty. Ltd., Charles St., at Cameron St. Newcastle—Martin de Launay Ltd., Cnr. King and Bolton Sts. Perth—Carlyle & Co., 915 Hay St. K.P.M. (Watson's Bay) sends a circuit for checking. A.—Yes, the circuit appears O.K., except for the value of the bias resistor for the 2AS. We think that this should be changed to 400 ohms instead of 100 ohms. That is, no resistor between cathode of the 2AS and earth. The value of the resistor of which you are in doubt can be anything from 100,000 to 500,000 ohms, but being 100,000 will do. We regard 250,000 ohms as a happy medium. The circuit will not be suitable for headphone use unless there is a choke or transformer coupled for the coupling of the headphones, as otherwise the heavy plate current of the 2AS would burn out the windings. We doubt if it is likely that we shall see enough articles covering the whole of the knowledge needed to pass the A.O.P.C. exam, but you can get it from the handbooks which are available at city bookshops for 7/6 each. E.A. (Mortlake, Vic.) is a little worried about the picture diagram of the "World-Wide" converters... A.—The coupling units are supplied with the wires already arranged for coupling direct to the mounting lug, so that an extra wire earth on a terminal on the unit is unnecessary; in fact, only two terminals are fitted to the units in the set, as are supplied on the market. That explains why the earth wire was not shown on the picture diagrams. We do not follow the matter of the condenser gangs being out of step, and we feel sure you have been misinformed in this direction. Not that it matters, as a step or out the step sections should be perfectly O.K. Doubt if there is much chance of running an article on a suitable beat frequency oscillator and a detector, but we will try to give a suitable description in Melbourne, except in one of the handbooks. D.V.W. (Christchurch, N.Z.) knows a good circuit when he sees one, and is going to build up the 13-37. A.—The original coils were part of a "World Standard" Lekmek "Chasakit," supplied to us specially but doubtless you could get a similar ready-wired tuner to your special order. We are unable to supply a more detailed diagram than the one given, being so complicated it would take days to draw out all the wires and connections, and it is also against our policy to encourage the building of such a complicated tuner by those amateurs who do not feel confident to follow circuit diagrams. You might find the world standard diagrams in the Handbook, as the whole of the tuning and is almost ideal. Speaker would be O.K. on a baffle mounted away from the chassis speaker leads and the 13-37 would drive the 13-37 audio transformer you mention should be O.K., but not being acquainted with its quality we cannot say how good it will be in the matter of frequency response. J.M. (Northwood) submits a circuit of a two-tube beat frequency oscillator which has tuned in 31 stations and asks if we can suggest any improvement. A.—So far as the circuit is concerned we doubt if you can improve on it, although a very slight gain might be obtained by using more sensitive valves, such as the PMH1L for the detector socket and PMDDX to drive the headphones. The job of fitting the extra terminals for the double coil winding on the broadcast receiver should be quite a cheap one and we fancy that any dealer would undertake the job for about four shillings. The only complication of taking the earth wire off the primary of the aerial coil and running it to an insulated terminal is the same in principle as the bias. The cost of the terminal would be less than 1/-, but time is worth from 3/- to 10/- per hour and the job might take up to half an hour. J.F. (King's Cross) has seen a circuit in an Australian magazine for a receiver which tunes from 2½ to 200 metres and asks us to publish the coil winding data. A.—We regret that it would be quite useless to attempt to give the coil data. Down around 2½ metres coils aren't used at all, but the wiring is cut to length to have sufficient distributed capacity to give tuning range for the ordinary bands on which you would hope to get the set to work, and on which there is a chance of hearing signals, as given in the Radio Handbook. The range of any one tuner from, say, 10 to 20 metres or 10 to 40 metres depends upon the amount of capacity used. If you build the all-wave two-tube receiver in the Handbook you will get a very practical idea of the meaning of coil windings, the effects of distributed capacity, etc., and it will also reveal the difficulty of going outside the usual bands. W.P. (Wangaratta) has built up a set consisting of a superhet tuner feeding a 5J detector operated as in the Standard circuits to drive a pair of 2AS valves, and these turn drive a pair of 2AS through an AOPC. The difficulty is distortion at low volumes. A.—We don't think there is anything seriously wrong with the other circuits mentioned, but there is no doubt that you are likely to have low-level distortion when using such a low-input valve as the 2AS. When the 2AS valves are running lightly loaded with signal the diodes will be handling very small inputs. A simple way would be to use a 2AS as a trim potentiometer, as diode load resistor and auxiliary volume control, so that the diodes can be kept loaded up when the desired output is desired. If you are sure that the 2AS valves get their correct bias and plate voltage you need not worry about them. TO COUNTRY OWNERS OF BATTERY-POWERED RADIO RECEIVERS. The Carlson (Air) Cell Battery is the ideal "A" or "Filament" Battery. No charging facilities are required, as it is a chemical self-generating battery manufactured in and marketed in Australia for over 10 years. ILLUSTRATED FOLDER ON REQUEST. AMPLION (A.S.I.) LTD. 66 Clarence Street, Sydney. Enclose Stamped Addressed Envelope. FORWARD THIS COUPON To the "Wireless Weekly" Radio Information Service, Box 3386PP, G.P.O., Sydney. PLEASE ANSWER THE ACCOMPANYING QUERY: (a) Free of charge through the columns of "Wireless Weekly." (b) In a special letter. I enclose one shilling in payment. RULES: (1) Readers are requested to be as concise as possible; (2) Information cannot be given over the telephone; (3) No special correspondence and designs for transmitters, receivers, or other such apparatus can be given; (4) All questions are answered impartially and efficiently, the best of the ability or the Technical Editor, and to the extent of the channels of information available to him. No liability attaches to "Wireless Weekly" for any misinterpretation or misdirection of the information thus supplied. NAME (in block letters).................................................................. ADDRESS........................................................................................ STATE.......................................................................................... YOU CAN'T MISS IF YOU USE R.C.S. COILS FOR THE ALL QUALITY 4/5 RECEIVERS! Recognising leaders in all coil design, R.C.S. Coil kits used for Wireless Weekly circuits will make your job a certainty. Precision made and "Q" meter tested. Ask for R.C.S. Coil Kit Type K41D Price Retail 37.6 R.C.S. Superhet Coil Kit Type K41F Price Retail 37.6 See your local dealer or write to R.C.S. Radio Laboratories, 21 Ivy St., DARLINGTON Phone MA9011-2. Printed and published for the Proprietors by Sun Newspapers Limited at the registered office of the Company, 60-66 Elizabeth Street, Sydney. O.J.B. (Bundaberg) is building up the "High-fidelity" amplifier and is a little worried about two minor points. A.—The power transformer has a filament winding for the 250 type valves with a centre-tapping then you won't need to use a centre-tapped resistor, all that is required will be connected between the centre-tapping of the winding and earth. The centre-tapped resistor is shown on the circuit in case the filament winding is not tapped, and schematically illustrates the centre-tapping when the transformer winding is arranged that way. The resistance of the field coil can be either 100 ohms or 1000 ohms, the latter change being too small to notice in practice. A.G. (Mount Lawley) encloses a letter from a prominent Melbourne dealer who refuses to quote for an 8-valve short-wave superhet, unless supplied without any guarantee of performance. A.G. (Mount Lawley) encloses a letter from a prominent Melbourne dealer who refuses to quote for an 8-valve short-wave superhet, unless supplied without any guarantee of performance. --- 1937 "PALEC" HIGH-GRADE MULTI-TESTERS Featuring the new 5" METER (MODEL 475) To the Radio and Sound Engineer In Modern radio practice trouble tracing and development work necessitate an instrument that is Accurate, Reliable, of sound Design and Construction, and also of wide Application. The "Palec" Multi-tester has all these Qualifications and more. ACCURACY. The meter guarantee is exceptionally high (see Meter Catalogue). For range accuracy, only wire for shunts and voltage multipliers is tolerated (exception, 1000V. range). RELIABILITY. The best component's possible, ensure many years of profitable service. DESIGN. An automatic circuit of fool-proof operation. CONSTRUCTION. An instrument job inside and out. APPLICATION. Protection every test position has been catered for. \| Volts, D.C., 5-10-50-100-500-1000. \| Impedance, 0-1 megohm \| \| Volts, A.C., 5-10-50-100-500-1000. \| Electric Cond. A Good-Bad leakage test for electrics, also capacity measurements of same. \| \| M.A.s, 1-10-50-100-500. \| Ohms, 0-10-50-100-500-200,000. \| \| Megohms, 0-10. \| Megohms, 0-10. \| \| Capacity, .001-10 mfd. \| Inductance, 0-10,000 Henries \| Price complete in black leatherette lidded case; £13 10 0 Model SM DC Volts, m.a., and ohms only £10 10 0 Model JM DC Volts, m.a., and ohms only £8 8 0 NOTE.—On Model SM and JM, the ohms range is extended to 1 megohm. Analyser-Selector Unit in same case or separate, extra £3 0 0 All prices are Trade and subject to tax. --- PATON ELECTRICAL INSTRUMENT CO. SEND FOR PARTICULARS AND ILLUSTRATED CATALOGUE TO 90 VICTORIA STREET, ASHFIELD, N.S.W. Manufacturers of the well-known PALEC Meter and Cathode Ray Oscillograph (as supplied to the Commonwealth Government). Obtainable from: Sydney: Fox and MacGillycuddy, Bloch & Gerber, Lawrence & Hanson. Melbourne: A. H. Gibson (Electrical) Co. Pty., Ltd., Lawrence & Hanson. Brisbane: Lawrence & Hanson. South Australia: Newton McLaren Ltd. New Zealand: The Electric Lamp House, Ltd., Wellington. --- JOHN HEINE & SON LTD. METAL WORKING MACHINES for RADIO MANUFACTURERS We manufacture PRESSES and DIES, GUILLOTINE SHEARS, FOLDERS and BENDERS, HYDRAULIC PLASTIC MOULDING PRESSES (for Bakelite and similar powders). Terms Arranged OFFICES, WORKS AND SHOWROOMS: ALLEN STREET, LEICHARDT, SYDNEY N.S.W. PHONES: PETERSHAM 1697-1698. Representatives: McPherson's Pty. Ltd., Melbourne, Adelaide, Perth; E.C.S.A., Brisbane; John Chambers and Son, New Zealand. Inspect also our "USED-RE-BUILT" Machine Dept. --- A.—It is quite normal to find the more conservative firms refusing to take the responsibility of guaranteeing performance of a receiver, but the amateur and the average type of buyer is likely to be very discriminating about performance, being far more critical than the ordinary broadcast listener. Our experience has also shown that often when one uses a receiver of this type the best reports on results are from those who build them themselves. Not that you will have any difficulty in getting one properly built to order, and the Sydney firm that you mention should be able to do the job. We have already had reports from satisfied readers who have had this set built by the Sydney firm. Amateur activity in New South Wales is very keen at present. DX-Bug (Arncliffe) has built the four-valve short-wave superhet from the Radio Handbook, but finds that when he adds a 2A5 valve to the end of it the set suffers from motor-heating. A.—This is more or less to be expected when you use such a light gain audio system. The correct way to drive the pentode stage is to couple the output of the first section of the tube detector valve, thereby cutting out the gain in the second section. If you want to retain the gain, you will have to fit de-coupling resistors in both plate feeds to the 2A5s with suitable by-pass condensers, preferably a mild electrolytic. H.W. (Guildford) recently suggested fitting a Radiolite chassis into a big console cabinet and in reply to his inquiry we recommended him to use a chassis with the ideal. Another reader, from Wagga, has now written us to say that the job is not nearly as easy as it looks. The binding posts on the Radiolite having a shaft which is too short to take a standard dial. Frankly, we can't see the force of this argument. Attention paid for results can be obtained from any good radio store for a few pence each and they could be made to order for a shilling or two. With such a small chassis you could have speakers with extensions of three to six inches long without the chassis overhanging the back of the cabinet. J.C.C. (Edgecliff) wants to know when we published the circuit of the "17 Station" crystal set mentioned in the queries columns in our January 29 issue. A.—This circuit appeared in our December 25, 1936, issue. There is nothing particularly wonderful about the circuit itself, and at Edgecliff it might be possible to get a good signal from 2GB. But in certain country locations and over a period of some months, it is quite possible that twenty different stations could be heard by a keen listener, especially if it was used late at night after certain stations have closed down, thereby giving other stations a chance to be heard. F.H.A. (Mt. Roskill, N.Z.) writes an interesting letter pointing out the requirements for a receiver for obtaining quality from local stations. A.—We don't quite gather what article you are referring to, but your recent campaign has been along the lines of more quality at low cost, not good quality irrespective of cost. It is simple enough to design a receiver for either low cost or a higher price and quality, but we know that listeners like long-range if they can get it and also a kit of parts at a cheap price, and so we are trying to do the best we can with a five-valve receiver without sacrificing anything or increasing the cost at all, and yet getting a definite improvement in long-range. We feel that we have been quite successful in this direction, but hope to carry it even further. The power output of a single 2A3 when driving an efficient speaker, should be quite considerable; in fact, by the time the distortion content is up to the level of a single 6L6 the power output should be nearly the same. H.L. (Bendigo) is a keen enthusiast, having built an elaborate version of the "Prizewinner" with every imaginable refinement. A.—We don't know of any firm which will supply individuals with long playing recordings, or song hits, or gramophone or even magnetic wire recordings. These appear to be available only to broadcasting stations. Your pre-set circuit could be readily modified to take the 6L6 valve, but the quality would not be improved; in fact, there may be considerably more harmonic distortion, and frequency discrimination against the speaker input transformer is taken into consideration. We doubt if you can get 20 watts from any of the valves mentioned, even the 6L6. If you are going to retain the high-gain grid modulation. One possible solution is the use of four 2A3 type valves in a push-pull arrangement. Circuits suit the P.E. cell used are usually supplied with the cell or from the distributors of the cell in question. Different cells have different outputs, and consequently require a different amount of pre-amplification. 2CH Presents Knight Barnett at the Organ Your Favorite Melodies Mon. to Fri. 6:30 to 6:45 p.m. Oscar Lawson presents Bits and Pieces and Thursdays at 7:45 p.m. The Thrilling Half Hour Radio Play Realm of Drama Wed. 8:30 p.m. Dramas of Intrigue Spies, adventurers, stolen papers, keep the plots moving at a pitch of excitement. 2CH Entertains the Whole Family Amalgamated Wireless (A SIA) Ltd. Advent Radio Church 2UE Every Sunday 5.15—6.0 p.m. Pastor L. C. Naden. BINA ADDY, the Indian mezzo-contralto of European renown, who recently arrived in Sydney after completing a broadcast tour of New Zealand, has been signed up by the Broadcasting Commission to make a recital tour of its national stations. She will open in Sydney on February 4th from 2FC, at 9 o'clock. On this tour her programmes will consist exclusively of Bengali songs, which have not been sung previously in Australia. She is the only Indian woman who has sung professionally in Rome, Paris, Berlin and London. She has also broadcast from Berlin and has appeared in London before Royalty. Bina Addy sought Western culture and musical training in Europe. For that purpose she travelled alone from India to Rome, Leipzig, and London, studying the language of each country and choosing eminent artists as her teachers. Her voice first attracted public notice when she sang with the Wesley Church choir in Calcutta. She left India in 1928, and for the next four years studied under well-known masters in England and Europe, principally Elena Gerhardt (Leipzig) and Mario Cotogni (Rome). Miss Addy has a broad international outlook and while in London took an interest in all movements which touched on the subjects of relationship between East and West and particularly in those dealing with conditions in India. "It is my sincere desire to create a better understanding between my country and other nations," she says, "and if I could feel that I had in any way provided a link between the East and the West, I should be content." Although she received a European education at a school in Calcutta and was, therefore, familiar with foreign customs before she left India, she has never adopted European dress, and in her bright saris and fine gold bracelets Bina Addy makes a most picturesque figure. This is ye old Drum Major, Ray Sherwood, making his bow, I'll be marching my first big Kraft Music Parade into your home each Tuesday night!!! A full hour with the biggest and newest stars in radio, you'll be with us—won't you? Kraft Music Parade 2.U.W. FULL HOUR But not the two bears of the bedtime story. Above is Bruin, the polar variety, climbing up to give the photographer a hug. On the right is Teddy, the gum tree fellow, waiting for the photographer to give him a hug. Both photographs were entered in our weekly photographic competition by Miss M. Green, 4 Flat, 21 Edgware Road, Enmore, who may collect 10/6 for each by calling at our office (12th Floor, Sun Building), or, if that is not possible, by writing to us. ALL OVER SYDNEY ENTERTAINING DRAMATIC EDUCATIONAL 2SM 9p.m. EVERY WEDNESDAY Page Missing Page Missing Page Missing Page Missing Page Missing Page Missing Page Missing Page Missing TO AUTUMN the latest fashion news One of the Fashion Reporter's activities is the broadcasting of descriptions of the frocking at the big Randwick meetings from 2FC. PRINCESS LINE FOR AUTUMN DINNER GOWN The high shirred bodice line, the high braided collar to correspond with the hemline are two important features in this dinner-gown of crepe in the new vintage shade grapefane. —David Jones "Choose a Radio Career my Son! "You must think of the future and what it will bring. Radio offers you a variety of interesting jobs, good money and an assured future. Get into Radio and you’ll get ahead my boy." Splendid Opportunities for YOU The sky’s the limit in Radio for ambitious fellows. Think of the fine jobs you can train for. Broadcast engineer, service engineer, talkie expert, television engineer, Radio executive, factory engineer, operator, etc. TELEVISION This wonderful industry of the future is on the way. Soon it will be here, bringing thousands of openings for trained men. Would YOU like a good job in Television? It’s in YOUR hands now to prepare for one. JOBS FOUND The A.R.C. Free Employment Service will help you find a job. Do you know, so great has been the demand for A.R.C. students, we have at times been forced to refuse positions. EARN WHILE YOU LEARN A.R.C. methods will show you how to make good money in your spare time—to start a part-time radio business which can later grow to a full-time business. Think of this—for less than the average fellow spends on tobacco each week, you can train for a splendid career in Radio-Television. Post Passport NOW! FREE BOOK It’s worth your while investigating. Send the passport now for a free copy of "Careers in Radio and Television." Read all about your chances for success in this great industry. POST THAT PASSPORT NOW!
TRW's SUPERCHIP PASSES FIRST MAJOR MILESTONE PAGE 49 Take Control! Chart your own course in CMOS system power and performance. Harris gives you control. Our static CMOS circuits allow you to match system power and performance to your exact requirements, from DC standby to full-speed operation... - 80C86 and 80C88 static CMOS CPUs for flexible system design - Wide selection of CMOS peripheral and data comm circuits - Static CMOS memory for low-power operation and data retention - Low-power CMOS programmable logic (HPL™) - Advanced standard cell capability featuring 82CXX LSI peripheral cells Full range operation. Completely static design ensures stable, reliable operation in the toughest applications. And our new 82C85 Static Clock Controller gives you control of four low-power modes: fast for full-speed operation...slow for continuous operation...stop-clock for low power...stop-oscillator for the lowest possible power. You get low-power operation without sacrificing performance. Take control! Harris static CMOS technology delivers top performance in CPUs, peripherals, memories, programmable logic and standard cell design. And for board-level low-power solutions, specify our 80C86-based Multibus™ CPU board, the HB-986C05. Why wait? Take control of your designs now. For full details, call 1-800-4-HARRIS, Ext. 1355 (in Canada, 1-800-344-2444, Ext. 1355). Or write: Harris Semiconductor Products Division, P.O. Box 883, MS 53-035 Melbourne, Florida 32902-0883. Harris Semiconductor: Analog · CMOS Digital Gallium Arsenide · Semicustom · Custom · Hybrids For your information, our name is Harris. MEGABYTES NOT MEGABUCKS That's what you get from Infotek's new cost and space saving 2 MByte HP 9000-compatible memory. Now, you can have a full two megabytes of instantly accessible memory for your HP Series 300 and Series 200 computers for a price you'd expect to pay for 1 MByte. That's like getting half your memory free! And even better, you get those 2 MBytes on a single board, so all that extra capability occupies only one I/O slot. You just plug in the new board in seconds with no special tools required. The new Infotek AM 220 2 MByte memory board is ideal for all scientific and engineering applications where large memory requirements remain constant. A 1 MByte product is also available at a very reasonable cost; and, for maximum capacity, Infotek offers a 4 MByte single board memory. If your data storage needs are growing, you can have Infotek's unbeatable price/performance/size benefits in an expandable single board memory—1 MByte of memory upgradeable to 2 or 4 MBytes, or a 2 MByte board which expands to four when memory requirements increase. You can even get a floating point processor and 1 MByte of memory on a single board. Of course, all Infotek memory boards have the high quality you demand for your HP system. Their exceptional reliability is backed by a two-year warranty. So increase productivity, save time, and enhance the capability of your HP 9000 computer. Call Infotek toll free today. The only way to get MegaBytes without spending MegaBucks. Infotek Systems, 1400 North Baxter Street, Anaheim, California 92806-1201, (714) 956-9300, (800) 227-0218, in California, (800) 523-1682, TELEX 182283. Your Second Right Decision. Bypass Your Competition With T1/CEPT Access Components To ISDN By Mitel Demand for T1/CEPT communications facilities is opening unprecedented opportunities for your MUX, LAN, Data Switch, PBX or other communications product. T1/CEPT Interfaces belonging to the Mitel ST-BUS™ family of ISDN-compatible components can provide the competitive edge you need with: - The smallest size and lowest cost per function in the industry. - Redundant links at the same price as our competitors' single link interfaces. - DMI and CPI compatibility. - Software-controlled diagnostic and maintenance functions. - MH89740/50 – T1 Interface Module with Clock extraction and line equalization. - MH89756 – T1 Interface Module with line coupling transformers, microprocessor control port, time slot switching, clock sources and PLL synchronization. - MH89780/785 – CEPT family of 2.048 Mbit/s interface components that functionally are compatible with our T1 Interfaces. The following devices offer a variety of functional and packaging choices that will minimize the cost, time and complexity of your design-in: - MT8974/75 – T1 Interface ICs with extensive interface support functions. Don't let the opportunity bypass you - pick up the phone now and talk with us about worldwide Primary Rate Access. With this issue, Electronics returns to a biweekly publishing schedule, and it seems to be an appropriate time to step back and take a look at you, the reader, and just what you like and expect from your magazine. We have always maintained that the 300,000 people who make up our readership are unique in our industry. To find out just how special you are, we decided to ask you what you think of the job we have been doing, the information that we bring you in each issue, and the way we present that information. The method was to conduct a detailed telephone survey, calling several thousand readers over a period of several months. Now, with the results in, it turns out that we have been right. For example, you proved that we are right when we boast of having the most loyal readers in the business. When we asked you if you read your copy of Electronics cover to cover, 92% of you replied that you did, while just 5% said no. And what you want to read about most of all in each issue is technology: 81% placed technology news at the top of the list, 79% voted for in-depth technology developments and analysis, 43% said they look for business and industry news, and 22% went for company and marketing strategy. You are more interested in the type of information we bring you than in the way we present it. In fact, 67% of you want to see both in-depth reporting and brief, quick-paced presentation, while 17% prefer the shorter format and 16% prefer to read information presented in depth. And you are absolutely sure about what kind of information you need. A resounding 73% of you say what you need is related to technology trends, perspectives, and design implications. And some 19% believe that detailed and application- or specification-oriented articles are required. It was particularly gratifying for us to find that you are so pleased with what we produce. Your satisfaction level is high: 83% of you say that you are mostly or completely satisfied with the magazine and 11% pronounce yourselves somewhat satisfied. Only 5% are not satisfied—and we're determined to win you over. But you still have definite ideas about what Electronics should do in the future, with most of your preferences having to do with news coverage and analysis of technology developments. Some 71% of you want to see more technology news and developments; 78% prefer more in-depth technology analysis and perspectives; 21% want more detailed design and application information; and 28% seek more industry and business news. Finally, 82% of you welcome our new biweekly publication schedule. Some 31% prefer to see us every other week, and 51% like us no matter how often we come out. Eighteen percent prefer a weekly. So we think you're pretty special—even though there are 300,000 of you. We promise to return your loyalty by making Electronics more indispensable as the magazine you need and want. --- EDITORIAL PRODUCTION & COPY DESK Benjamin A. Mason (Director) Production Charles D. Ciatto (Mgr.), Kathy O'Neill Copy Editors Susan Levi Wallach (Chief), Jay J. Iorio, Jim Taibi ART Sachiko Soskin (Associate Director) --- NEWS BUREAUS Boston: Craig D. Rose (Mgr.), Debra Michals Chicago: Wesley R. Iversen (Mgr.) Dallas: J. Robert Lineback (Mgr.) Los Angeles: Larry Waller (Mgr.), Ellie Aguilar New York: Tobias Naegeli (Mgr.) San Mateo: Clifford Barney (Mgr.) Washington: George Leopold Frankfurt: John Gosch (Mgr.) London: Sarah Miller Paris: Robert T. Gallagher (Mgr.) Tokyo: Charles L. Cohen (Mgr.), Mari Matsushita, Michael Berger (World News Chief) --- EDITORIAL ADMINISTRATION Denise Giaimo (Administrative Assistant), Ina Gruber, Josephine Ortiz --- PUBLISHER Laurence Allman DIRECTOR OF OPERATIONS Thomas E. Vachon CIRCULATION MANAGER Leon Irgang PRODUCTION DIRECTOR Thomas Egan SALES ADMINISTRATOR Evelyn Schmidt --- EDITORIAL DEPARTMENT PHONES: Administration (212) 512-2645, News and New Products (212) 512-2685, Technology (212) 512-2666, Bureaus: Boston (617) 262-1160, Chicago (312) 751-3811, Dallas (214) 458-2400, Los Angeles (213) 480-5234, New York (212) 512-3322, San Francisco/San Mateo (415) 349-4100, Washington (202) 463-1650, Frankfurt 72-5566, London 493-1451, Paris 48-89-03-80, Tokyo 581-9816. BUSINESS DEPARTMENTS: (212) 512-6435 (Business departments follow the Advertisers' Index). ## NEWS Newsletters - Electronics, 19 - New approach to digital signal processing is coming from Zoran - U.S. and Japan set deadline for final terms on chip dumping - International, 42 - France’s CGE and ITT join to form telecommunications giant - Siemens and Philips plan cooperation in telecom chips - Aerospace, 27 - Control Data launches CMOS/SOS space computer - Tests on VHSCIC space computers begin this fall - R & D, 28 - Pulling space research and development down to earth - Networking, 29 - Low-speed LANs check in at $100 per connection - Semiconductors ICs, 30 - RCA builds a better place-and-route tool - Telephony, 32 - DSP chips start to poach in analog conferencing country - Optoelectronics, 36 - Second-harmonic generator promises denser optical disks - Consumer, 38 - The French spark a home-net standardization push - Military, 38 - USAF project weeds out bids too good to be true - IC testing, 40 - A faster way to open up plastic packages for fault analysis ## INSIDE TECHNOLOGY COVER: TRW’s superchip passes first major milestone, 49 The development of a family of superchips, each integrating up to 34.7 million devices, is off to a flying start. TRW Inc. has fabricated process-test versions of the submicron family and is now moving on to chips that test self-diagnosis and software-controlled reconfigurability. Breaking the speed barrier on the VMEbus, 58 Interphase Corp. has developed an interface that triples the VMEbus’s throughput to 30 megabytes/s. This puts the VMEbus neck and neck with rival Multibus II, currently rated at 32 megabytes/s. How Motorola moved BiMOS up to VLSI levels, 67 Motorola Inc.’s first commercial BiMOS VLSI chip is a 6,000-gate array that is the forerunner of a number of mixed-process products. To reach VLSI density, the company had to figure out how to trade off the conflicting requirements of bipolar and CMOS processes. Intelligence comes to laser soldering, 75 By automatically adjusting the heat to individual joints, Vanzetti Systems’s laser soldering systems produce perfect joints every time. Special report: Software productivity moves upstream, 80 Many tools aid programmers with the coding portion of software development, but only now are tools being developed for front-end tasks such as analyzing requirements and developing specifications. Special report: The military rushes into surface mounting, 93 The coming generation of military electronic systems more and more will be turning to denser packaging techniques, notably surface mounting using different types of chip carriers. ## PROBING THE NEWS U.S. Midyear Market Report, 105 The electronics industry hasn’t bounced back as strongly as it had hoped from the down year that it had in 1985. There have been some mildly encouraging results, but the most striking characteristic of the first six months of 1986 is the way market watchers have been downsizing their growth predictions. Most executives once again hope that, led by a recovery in computers, things will get better. Multibus II heads for the fast track, 116 Intel has produced the crucial message-passing coprocessor interface, and its synchronous bus standard for 32-bit microprocessors is starting to win the backing it needs to gain ground against the VMEbus. Europeans are counting on Unix to fight IBM, 121 European computer makers have tried—and failed—to form a united front against IBM. Now, a nine-member group is trying to adopt Unix as a minicomputer and work-station standard. ## NEW PRODUCTS Newsletter, 23 - Rockwell to offer CMOS/SOS gate arrays - Bolt Beranek & Newman is enhancing its parallel computer Computers & peripherals, 129 - Intergraph’s Unix work stations cost only $5,000 per mips - A CAD/CAE accelerator from Cadnetix roars along at 10 mips - Trilobyte’s MS-DOS package writes user interfaces - Package from Parameter Driven Software sets up Unix data bases Components & subassemblies, 136 - Crystal oscillator from MF Electronics suits surface mounting - Avantek’s surface-mount amplifiers have 9.0-dB gain - Dc-to-dc converter from Rifa has I/O isolation Semiconductors, 138 - Digital timer from AMD cuts the analog connection - Analog Devices’ multiplying DAC settles in 8 ns - Gould and ICT to sell EEPROM-based logic - Dallas Semiconductor’s FIFO has parity bit for networks Communications, 142 - Filter chip from Crystal Semiconductor attenuates jitter in T1 lines - Banyan’s latest server options link diverse networks Industrial control, 146 - Standard controllers cut automation costs - Motion controller is easy to assemble --- ## DEPARTMENTS Publisher's letter, 3 What our readers expect from us FYI, 8 Adding value is still the prime directive at Electronics; we don’t want to contribute to the growing stream of raw data that is being directed at our readers Letters, 12 People, 16 - Olof Soderblom rings up token-ring royalties - Edward Schneider’s letter gets a positive reply Meetings, 152 Electronics week - Motorola sets transition plans... - ...while two RCA executives are named GE vice presidents - Matsushita to make TV sets in Mexico WEEK 38 If you're racing along technology's edge, meet the most powerful 32-bit ALU around. The Am29332. It's the newest member of our Am29300 Family. Besides doing everything a less powerful ALU will do it also has a mask generator, funnel shifter and field logical operations. And it does it all in a single 80ns cycle. And every last nanosecond helps you beat the competition. Am29332 It helps to know a powerful family. The single-chip Am29332 uses everything it's got to achieve such incredible speed. To keep data racing through, it's designed with a built-in carry look-ahead and barrel shifter. A priority encoder insures that the most important data gets through first. Whatever design you're working with, don't worry. The Am29332 is designed to let you emulate almost any system. The Am29332 sets you free of the bonds of your present architecture. Because of the built-in high level instruction set and the three-bus flow-through architecture you get maximum performance and flexibility. And since correct breeding is important to powerful families, data integrity is insured through chip-level fault detection. The Am29332 is just one member of a complete family solution to high performance 32-bit processing. A family that's nice to know when you need more power than anyone else. Design and application seminars are available for this product. Write or call for information. WEEK 39 AMD announces the Am99C641 64K CMOS SRAM. With access times down to 25ns, no one can beat it. Am99C641 The pedal hits the metal. But don't forget, this is a CMOS SRAM. It uses 110mW (max.) on stand-by (TTL-level) and all of 8mW on data retention. Lots of speed and no power dissipation problems. The Am99C641 has a 64K x 1 organization in a stylish 22-pin 0.3" wide DIP. Best of all, it's made by AMD, the folks who design what you need to stay as far ahead of the competition as they are. So when you need 64K of blinding speed married to CMOS cool, check out the Am99C641 64K CMOS SRAM. Because there's no such thing as too fast. WEEK 40 At last there's a microprocessor that's an arch conservative when it comes to power issues and a flaming liberal when it comes to high performance: AMD's Am29C117 16-bit CMOS Microprocessor. It's fast, but it only uses a frugal 25% of the power of the bipolar Am29117. Am29C117 Cross party lines. Get ready to save a lot of time if you're designing a pipeline system. The Am29C117 has two separate I/O ports for simplified timing, so you can achieve up to a 15% increase in throughput on a system basis over an Am29C116. A 16-bit barrel shifter allows you to shift or rotate a word up to 15 positions in a single instruction cycle. That means you don't have to worry about extra parts or using valuable board space. It's plug compatible with the Am29117 and uses the same instruction set. It has a 125ns microcycle time. Or an 8MHz data rate. It's really a very simple concept. When parties cooperate, everybody wins. Advanced Micro Devices has broken the game wide open. After building a comfortable lead with a new product a week, every week—on the shelf, in volume—we called in the heavy hitters: ISDN. CMOS. 32-bit microprocessor chip set. High-speed RAMs. Modems. No one who follows the game closely is the least bit surprised. This team puts more dollars into R&D, as a percent of sales, than anyone else in the business. If you like the sound of extra bases, call Advanced Micro Devices. The International Standard of Quality guarantees a 0.05% AQL on all electrical parameters, AC and DC, over the entire operating range. The 5380 with INT•STD•500. Only from AMD. Two new non-volatile CMOS RAMS. Quick, quantity deliveries... ...of 256 and 1024 Bit NVRAMS are now available through selected distributors and representatives of Catalyst Semiconductor, your fine new source of CMOS nonvolatile memory devices. Both devices are pin-to-pin compatible with Xicor parts X2210 and X2212. The CAT22C10 NVRAM is a 256-bit device organized 64 x 4. Full static CMOS circuitry uses 40 mA in the active mode and standby current of 30 uA. Model CAT22C12, a 1024-bit device organized 256 x 4 uses 50 mA active current and 30 uA standby. Both Feature: - Write protect circuits. - Automatic recall on power-up. - RAM access times of 200, 250, and 300 ns. - TTL/CMOS compatibility. - JEDEC standard 18-pin, 300 mil package. Single power supply (5V ± 10%). 3-state output. 200ns/min short store and 300ns/min recall pulses. False store protection below 3.5V. 10,000 nonvolatile Store cycles per bit. Write or Call for our no-kidding data pack. Please address Bob Simon, Director of U.S. Sales, Catalyst Semiconductor, 4051 Burton Drive, Santa Clara, CA 95054. Phone (408) 980-9144. FAX 408-980-8209. TWX 510-601-7631. In Europe, contact Tekelec Airtronic, Cité des Bruyères, Rue Carle Vernet, 92310 Sèvres, France. Phone (I) 45-34-75-35. Adding value is still the prime directive at Electronics; we don’t want to contribute to the growing stream of raw data that is being directed at our readers. We had but one major goal in mind 14 months ago when I returned to Electronics: restore this magazine to its former greatness. That meant bringing back and updating the interpretive technical coverage it had been famous for over the past half century. Returning to our traditional biweekly publishing schedule is the final step in achieving that goal. Don’t look for major changes: we’re delighted with the progress we’ve made to date. So are the readers, based on the letters we’ve been getting. Back when the vacuum tube was king, this magazine earned its spurs by breaking important technology news and putting it in focus for all readers. Our editorial niche is still that: analyzing new developments and putting them in perspective. As one longtime reader puts it: “I read Electronics cover to cover for the big picture in technology.” We’ve brought all that back during the past year—and more. We’re tailoring coverage to fit the changing information needs of technically trained managers as they move into the complex world of the late 1980s and 1990s. These executives need to know more about more technologies than ever. Yet they have less time than ever to keep up. No publication is better equipped to help them than we are. Competing publications, which have changed little in recent years, fall into two extremes: the technical magazines with their long-winded, narrowly focused “nuts-and-bolts” articles and the tabloids and general news publications that run undigested press releases and nontechnology coverage. The nitty-gritty tech article is understandable only to a narrow band of engineers, while the news story based on a nontechnical announcement or press release doesn’t interpret what’s happening and often fails to give both sides. The biweekly Electronics will carry more of the technology news we’re known for—coverage the competition can’t provide. More staff-written Technology To Watch articles will announce new technology and put it into perspective. The New Products section will explode in size, and sharp-eyed readers will spot an old friend: the International Newsletter, something only this magazine can turn out, because no competitor has more than a token foreign staff. Adding value is still the prime directive at Electronics; we don’t want to contribute to the growing stream of raw data that is being directed at our readers. ROBERT W. HENKEL THE MASK ROM MCU ALTERNATIVE DEVELOPMENT TO PRODUCTION IN ONE DAY Hitachi’s 8-bit microcontroller gives you Zero Turn-Around Time. Now you can have fully-coded 8-bit ROM-on-chip microcontrollers in just one day. Hitachi’s ZTAT™ (Zero Turn-Around Time) products make it possible. ZTAT is a series of 6301/6305 compatible all-CMOS microcontrollers with one-time-programmable ROMs—in plastic packages. Since you can program them yourself, there’s no turn-around time, no mask charges, and no minimum order quantities. And, the plastic packaging makes ZTAT cost-effective for production runs as small as one, or as large as 20,000. Hitachi has the right CMOS 6301/6305 series device for each phase of your product’s life cycle: windowed ceramic package EPROM-on-chip devices for development, ZTAT for pilot production, and plastic package mask ROM devices for large-scale production. All are pin-for-pin compatible. Call your local Hitachi Sales Representative or Distributor Sales Office today, and reduce your turnaround time to zero with the 63701/63705 ZTAT devices. Fast Action: To obtain product literature immediately, CALL TOLL FREE, 1-800-842-9000, Ext. 6809. Ask for literature number R15. HITACHI We make things possible Hitachi America, Ltd. Semiconductor and IC Division 2210 O'Toole Avenue San Jose, CA 95131 1-408/435-8300 Stocking Distributors Almo ■ CAM/RPC ■ Cronin ■ Diplomat ■ IEC ■ Image Electronics ■ Industrial Components ■ ITT Multicomponents ■ Marsh ■ Marshall ■ Milgray ■ RC Components ■ Reptron ■ Resco ■ Ryno ■ Space Electronics ■ Sterling ■ Time ■ Vanrage Components ■ Western Micro Technology ■ Hitachi Regional Sales Offices Boston 617/720-2150 ■ Chicago 312/773-4864 ■ Dallas 214/991-4510 ■ Ft. Lauderdale 305/491-6154 ■ Los Angeles 818/704-6500 ■ San Jose 408/277-0712 ■ Circle 9 on reader service card At last, the ordinary microprocessor can take its rightful place in history. It had to happen—the conventional microprocessor has had its day. Relegated to the ranks of yesterday’s devices by the new transputer family from INMOS. It’s history in the making. The IMS T414 transputer is a fast, easy-to-use VLSI component, integrating a 32-bit processor, four inter-transputer communication links, 2K bytes Static RAM, 32-bit memory interface and DRAM controller. All on a single CMOS chip—offering execution rates up to 10 MIPS. While transputers excel in single-processor systems, their real power can be unleashed by connecting any number of transputers together via the high-speed serial links. Multi-transputer systems can deliver the performance you need today, and can be easily expanded in the future as your processing requirements increase. And there’s more. Programming multiprocessor systems has never been easier. The Transputer Development System (TDS) supports C, Fortran, Pascal and OCCAM, providing a complete software development environment, and is available for a number of popular hosts. Software developed on the TDS can be executed on one or more transputers, enabling cost-performance tradeoffs to be made. INMOS transputers are available now and have already found their way into companies who are evaluating, prototyping and manufacturing transputer-based systems. Applications include supercomputers, DSP, graphics, robotics, AI, distributed control systems and many others. Write or phone for more information on the transputer family and start making history yourself. \| TRANSPUTER PRODUCTS \| \|---------------------\| \| IMS T414 \| 32 bit Transputer—2K Ram —4 links \| \| IMS T212 \| 16 bit Transputer—2K Ram —4 links \| \| IMS M212 \| 16bit Disc Processor—1K Ram—2 links \| \| DEVELOPMENT TOOLS \| \|--------------------\| \| IMS D701.2 \| IBM PC—Transputer Development System. \| \| IMS D600 \| VAX/VMS—Transputer Development System. \| \| EVALUATION BOARDS \| \|-------------------\| \| IMS B002-2 \| Double Eurocard + IMS T414 + 2Mbyte DRAM + 2 x RS232. \| \| IMS B003-1 \| Double Eurocard + 4 x IMS T414 + 4 x 256Kbyte DRAM. \| \| IMS B004-2 \| IBM PC Format + IMS T414 + 2Mbyte DRAM. \| \| IMS B006-2 \| Double Eurocard + 9 x IMS T212 + 128K SRAM. \| \| IMS B007-1 \| Double Eurocard + IMS T414 + 0.5Mbyte DRAM + 0.5Mbyte Video Ram. \| High Performance VLSI Memories 16K + 64K CMOS SRAM, 25—70 ns 256K CMOS DRAM, 60—120 ns THE TRANSPUTER PARALLEL PROCESSING. UNPARALLELED POTENTIAL. INMOS, Colorado Springs, Colorado, Tel. (303) 630-4000; Bristol, England, Tel. 0272-290861; Paris, France, Tel. (14) 687-2201; Munich, Germany, Tel. (089) 319-1028. Circle 11 on reader service card CIA means quality and high reliability in MIL-SPEC. and custom hermetic connectors Connector Industries of America offers you the highest quality and reliability in military, QPL’d, and custom metal shell hermetically sealed receptacles. Product offerings include: - MIL-C-38999 ........... Series I, II, III - MIL-C-26482 .............. Series I, II - MIL-C-83723 ............. Series I, III - MIL-C-5015, MIL-C-26500, MIL-C-81703 For applications requiring high-density, high-reliability, and high performance, call or write for a free catalogue. CONNECTOR INDUSTRIES OF AMERICA 639 N. WAYNE AVENUE P.O. BOX 15278 CINCINNATI, OH 45215 (513) 733-3680 Circle 12 on reader service card --- LETTERS The HP 1000 marches on To the editor: Hewlett-Packard is not ceasing development of HP 1000 A-series machines used for industrial control [Electronics, June 9, 1986, p. 58]. The HP 1000 line will continue to serve factory-floor computer applications for years to come. There will be future products in the HP 1000 product line as well as continued enhancement of the network services and operating system. The HP 9000 model 840, reviewed in the same article, represents a convergence and an extension of the HP 1000 and HP 9000. In no way does its introduction imply any reduction of emphasis on the existing HP 1000s and HP 9000s. On the contrary, the HP 9000 model 840 has added appreciably to the acceptance of both by presenting tangible evidence of a smooth upgrade path for those that will require it. Brian Moore General Manager Manufacturing Systems Group Hewlett-Packard Co. Cupertino, Calif. Micrion 808 defended To the editor: The story on Seiko Instruments USA Inc.’s SIR-1000 mask-repair system [Electronics, March 10, 1986, p. 71] contained some incorrect technical points about the KLA/Micrion 808. The article states that somehow the 808 damages the substrate in an uncontrolled way, resulting in poorer-quality repairs than can be achieved with the SIR-1000. The 808 produces very high-quality repairs consistently and is the only production focused-ion-beam mask-repair system operating in the field today. Micrion perfected end-point detection of the mask-repair process early in 1984 and can offer this feature to our production customers at any time, unlike manufacturers that only talk about these topics. Users of the 808 are repairing mask plates on a number of printing systems with no problems. Images printed with repairs from the 808 are indistinguishable from unrepaired chrome. John A. Doherty Vice President, Marketing Micrion Corp. Beverly, Mass. We want smart power Correction: In “New RCA Process Halves Thickness of Gate-Oxide Layer in FETs” [Electronics, May 26, 1986, p. 9], we reported that “demand for smart power devices is still small” because of a perception that device reliability is a problem. Demand for smart power is high, RCA Corp. says, but demand for logic-level FETs (L²FETs) has suffered. From the United States you can call just about anywhere in the world through the vast, reliable AT&T Long Distance Network. So, whether you're doing business near the corn fields of Kansas or the rocky coast of Maine, the AT&T Long Distance Network is virtually as close as the nearest phone. And, you can always count on the nationwide AT&T Long Distance Network's quality service. Providing you with full service AT&T long distance operators ready to help place calls or give credit for misdialed long distance numbers. And, whether you're traveling around the country or to many foreign locations, it's reassuring to know you have the AT&T Card. Use the AT&T Card to plug into the AT&T Long Distance Network. To order your AT&T Card today, simply dial 1 800 CALL ATT, Ext. 229. AT&T The right choice. Digital has it now. Now there's a UNIX™ workstation that lets each graphics window tap the power of the network. Digital's VAXstation II/GPX™, with its innovative X Window System®, offers multiple, transparent windowing to any ULTRIX™ or UNIX (4.2 Berkeley) computer, anywhere on your network. A different application can run in each window, or within an application, individual windows can be linked to a larger computer. You can also access VAX/VMS™ systems via DECnet-ULTRIX™ networking software, or other vendors' systems via the SNA gateway, X.25 or TCP/IP protocols. And with larger systems handling compute-intensive tasks, the built-in MicroVAX II™ CPU and GPX graphics coprocessor can concentrate on what they do best: delivering exceptional graphics at tremendous speeds. VAXstation II/GPX lets you utilize your existing UNIX applications through ULTRIX, Digital's UNIX operating system. ULTRIX is a true Berkeley 4.2 BSD implementation and compatible with AT&T's System V™. VAXstation II/GPX. Everything an engineer wants in a workstation, backed by all the service and support he needs. For a brochure, write to: Digital Equipment Corporation, 200 Baker Ave., West Concord, MA 01742. Or call your local sales office. SODERBLOM RINGS UP TOKEN-RING ROYALTIES NEW YORK It was 3 a.m. and Olof Soderblom scurried out of bed to jot down an outline for a network that would come to be known as the token-passing ring. During the 18 years since, Soderblom has scraped and scratched to convince the world that the token ring was his idea. His story is one of persistence, and for Soderblom, 46, the battle is only now beginning to pay off. With 10 licensees signed up and six more expected to join the fold in the next month or so, Soderblom, whose highest educational degree is a high school diploma, has achieved at least a little vindication. He says the potential for licensees is limited only by the ability of Willemijn Holding BV to track down the companies that are using and selling token-ring technology worldwide. He shares the patent rights with the Dutch holding company, where he is vice chairman of the board. He estimates that at least 200 companies now use or plan to use the concept. Last year, Willemijn earned about $500,000 in royalties from its various international token-ring patents; this year's royalty income topped the $1 million mark in May. CLAIMING CREDIT. That money was a long time coming. Svenska Handelsbanken, the Swedish bank where Soderblom was working as a statistical analyst when he developed the token-ring concept, was the first to lay claim to the invention. Soderblom developed statistical models to help determine the bank's networking requirements. He was part of a group that worked with IBM Corp. and NV Philips on developing an addressed-polling scheme. The network allowed multiple terminals to share a single transmission line, but the computer power required to administer the polling protocols was enormous for the time. Moreover, the network lost much of its transmission time to routine protocol exchanges, and that meant system delays. Thinking the polling solution inadequate, he was suddenly struck with the idea of having terminals pass a permit—a token—among themselves to use the transmission line to which they were connected in series. That turned out to be the easy part. Unlike U.S. patent law, an employee in Sweden often has rights to an invention made while working for someone else if it is outside his area of specialization, says George Vande Sande, Soderblom's U.S. patent attorney. Nevertheless, it took almost four years to pry the patent rights from the bank. ONLY THE START. Once Soderblom had them, however, he found he still hadn't gotten very far. "As a private individual, you can't really support a 26-patent portfolio," he says. While struggling to get approval from the U.S. Patent Office—a drawn-out process that took 13 years to complete—he worked for IBM's Complex Systems Group in Stockholm and set up a pair of air-service companies to help generate extra capital. By the mid-1970s, IBM was using the token ring in its System 8100, and the company was loath to pay royalties. In 1976, the Patent Office began an interference inquiry to determine whether Soderblom's 1967 patent application had precedence over a patent issued to AT&T Bell Laboratories in 1971. Frustrated, he went back to freelancing and eventually sold his air services business—which held the licensing rights to the token ring—to Willemijn, a specialist in exploiting patent rights. In 1980, Soderblom got his first license: IBM signed a lump-sum agreement when it appeared that U.S. patent approval was imminent. In 1981, the U.S. granted the patent for which Soderblom first applied in 1968. Soderblom, who now lives in Surrey, England, with his wife and 12-year-old daughter, travels about 30% of the time, mostly to Europe, the U.S., and Japan. A native of Sweden, he moved to the U.S. in 1958 when his mother, who was in the Swedish foreign service, was sent to Washington. Count on OKI for world-class CMOS CSIC! On the leading edge and still expanding: your opportunities in OKI CSIC* product and performance—supplied by the VLSI source with major experience in single-chip logic integration. OKI's been applying the world's most advanced CMOS technology to Gate Array development since 1977; to Standard Cells and Customs since 1979. Today you can count on OKI for full-range CSIC capabilities. For quick turnarounds: very broad predesigned Gate Array options. To save silicon: easy conversion to Standard Cells, through OKI's fully-transparent data bases—exclusive OKI software you can design-interface with any major CAD workstation tool. And OKI Design Services are available throughout your development cycle. OKI: world leaders in packaging. OKI offers the most extensive packaging options in the industry today. You can specify your CSIC chips in any package type that will support your die. And you can count on OKI for technical assistance to help wrap up a precise fit with your specific gate-size and pin-count requirements. OKI: world leaders in plant automation. In one of the world's most highly automated facilities, fully robotized CSIC lines are assigned to fabrication, assembly and 100% testing—compatible line-automation dedicated to Gate Array, Standard Cell and Custom runs. OKI's invested over $500-million in plantwide automation to lower production time and costs, and to assure strict quality control. TURN THE PAGE FOR FAST ACTION. Customer-Specific Integrated Circuit Count on OKI to expand your options in CMOS Gate Arrays. \| 1.5-micron & 2-micron \| 3-micron \| \|-----------------------\|---------\| \| 70V000 Series* \| 70000 Series \| \| VERY HIGH SPEED \| HIGH SPEED \| \| # Gates \| # I/Os \| # Gates \| # I/Os \| # Gates \| # I/Os \| \| 700 \| 66 \| 3289 \| 86 \| 387 \| 44 \| \| 1000 \| 66 \| 4290 \| 104 \| 720 \| 62 \| \| 1500 \| 66 \| 6000 \| 130 \| 1445 \| 88 \| \| 2000 \| 80 \| 8024 \| 148 \| 2000 \| 106 \| \| 2500 \| 80 \| 10008 \| 172 \| 4205 \| 120 \| Expanded product families and packaging. For more cost-effective, more flexible single-chip logic integration, OKI supplies more pre-designed Gate Array options, in the most advanced package types: plastic or ceramic through-hole DIPs or pin grid arrays; or plastic surface-mount PLCC or flatpack. Expanded services, plus world-leading production automation. OKI CSIC* design expertise is accessible at any development stage. And OKI’s CSIC manufacturing/testing facilities—among the world’s most highly automated—assure high quality, high volume fabrication. RFQ OKI CMOS Gate Arrays For prompt response to your Request for Quote (returned with complete technical data), please fill out the brief specs needed below: Name/Title ________________________________________________________ Company ___________________________________________________________ Address _____________________________________________________________ City __________________________ State _______ Zip _________________ Tel: (______) _______________________________________________________ My application requires: 1. Gate count: ________ 2. Pin count: ________ 3. CMOS process: ( ) 1.5 micron ( ) 2 micron ( ) 3 micron 4. Packaging preferred: Surface-mount ( ) PLCC ( ) Flatpack Through-hole ( ) Pin Grid Array ( ) DIP 5. Anticipated volume: __________ pieces/month □ Please call me for immediate consultation. Request for Data Please rush complete technical input on: ( ) CMOS Gate Arrays ( ) CMOS Standard Cells ( ) CMOS Custom VLSI Logic To return your RFQ to OKI: Remove entire page and fold into thirds. Staple or tape together and mail to: OKI Semiconductor, 650 N. Mary Ave., Sunnyvale, CA 94086. Tel: (408) 720-1900. Count on OKI to deliver world-class CSIC! OKI SEMICONDUCTOR CSIC Marketing 650 N. Mary Avenue Sunnyvale, CA 94086 Customer-Specific Integrated Circuit Soderblom went to high school in Washington, decided he did not want to go to college, and upon graduating headed to Boston in search of a high-technology job. He landed with fledgling Wang Laboratories Inc. in 1958. He remains something of a free spirit, setting aside about 50 days a year for work as an independent consultant, and he says he misses the nitty-gritty work of designing solutions to technical problems. But Soderblom remains excited by the prospect that token-ring technology could find its way into other applications, such as a possible replacement for the popular RS-232-C interface. "Token-ring would fit the bill," he says, "because its a very simple protocol. If that happens, the licensing opportunities would be limitless." —Tobias Naegle SCHNEIDER'S LETTER GETS A POSITIVE REPLY POMPANO BEACH, FLA. Eighteen years ago, Edward J. Schneider wrote a letter to newly formed Computer Products Inc. suggesting that he was the man it needed to get its product ideas into production. Now he is the company's president and chief executive officer. Schneider, 56, who has a mechanical engineering diploma from the General Motors Institute as well as a BS in mechanical engineering and an MS in production engineering from the University of Michigan, recalls the letter well. "In March 1968, I saw an article about two fellows who had started a new company. One, David C. Yoder, was an engineer. The other, Daniel Wiper, was a marketer. So I wrote them a letter saying, 'Sooner or later you're going to have to make something, and that's my bag.'" Cofounder Yoder, who preceded Schneider as president and chief executive and is now chairman, says, "He was our first employee in operations, and much of the company's 19-year success can be attributed to his expertise and leadership." When Schneider wrote his letter, he was working as manufacturing director for the Instruments Division of Electronic Associates Inc., a West Long Branch, N.J., manufacturer of analog computers. Schneider had been there 12 years, starting as a sales engineer. He was promoted to chief of mechanical engineering and director of corporate planning before taking on the division's manufacturing post. Before that, he had served for three years in the Air Force as a research and development contracting officer. He started at Computer Products in 1968 as director of manufacturing. In 1979, he became director of the Power Conversion Group, which has six manufacturing units in the U.S. plus one in the Caribbean. The group, which makes converters, power supplies, and uninterruptible power supplies, accounts for three quarters of the company's revenue. The Measurement and Control Division, which accounts for the rest, has five manufacturing units that make process control and industrial automation products linking computers to process equipment. Computer Products today has 2,200 employees and had a net profit of $1.45 million last year on a gross of $88.5 million. But it was a difficult year for the company, says Schneider, because of the discovery of accounting irregularities in the Measurement and Control Division that resulted in a $4 million reduction in the division's revenue from the original results. There is a criminal action pending against the division's former comptroller, says Schneider. "But that's behind us now and in the hands of the courts," he says, "and we should top the $100 million mark this year. In fact, our Power Conversion Group ranks No. 2 in the market in the country. My goal is to move it to the top spot." —Howard Wolff SCHNEIDER: Computer Products' first operations employee is now president of the 2,200-employee company. When Firestone needed performance, Zenith had the track record. On its way to the races, Firestone made a stop for Zenith's Z-200 Advanced PC. And right away, it gave them the power for faster starts, along with IBM-AT compatibility. But Firestone picked up a lot more than speed. Like the ability to run almost all IBM-PC software. Up to 40MB of storage. And the built-in expandability to accommodate as many as eight simultaneous users. With Zenith, they also got performance and reliability proven among some of the largest and most demanding PC buyers. Which is why the Z-200 is now a PC standard throughout the Firestone corporation. But without a test drive, you'll never know what you're missing. For more information and the name of your nearest Zenith Data Systems dealer, call 1-800-842-9000, Ext. 1. The quality goes in before the name goes on. A new approach to digital signal processing, which promises to perform functions previously attainable only with systems consisting of numerous boards based on general-purpose DSP chips, is said to be scheduled to go into production next week. The manufacturer is Zoran Corp., a Santa Clara, Calif., startup. Industry insiders say the single-chip vector signal processor, or VSP, can perform complex operations such as fast Fourier transforms with only a few instructions and can process a 256-by-256-pixel image at video rates. What's more, the word is that it uses a relatively low integration level: a 2-$\mu$m CMOS process for densities of no more than 70,000 transistors. U.S. AND JAPAN SET DEADLINE FOR FINAL TERMS ON CHIP DUMPING Japanese trade authorities have a tense two weeks ahead of them. They must hammer out the details of an agreement by July 26 that will suspend two cases involving the dumping of Japanese erasable programmable read-only memories and 256-K dynamic random-access memories in the U.S. semiconductor market. The Commerce Department and the U.S. Trade Representative's office say they are close to agreement on many issues with the Japanese, but that they are still far apart on U.S. access to the Japanese semiconductor market and regulations to prevent future dumping. ESDI TO BE CONSIDERED FOR INDUSTRY DRIVE STANDARD After more than two years of independent development, ESDI, the Enhanced Small Disk Interface for 5$\frac{1}{4}$-in. Winchester drives, has achieved recognition as a possible industry standard. Next week's meeting of the ESDI steering committee in Irvine, Calif., will serve as an official working group for American National Standards Institute Committee X3T9.3 on device interfaces. Control Data, Electronic Processors, Fujitsu America, Maxtor, Priam, and Western Digital are the first members of the ESDI committee; others will be recruited in Irvine. One issue to be considered: upgrading the interface from 10 to 20 Mb/s. MACHINE-VISION STANDARDS COULD APPEAR NEXT YEAR The Automated Vision Association, Ann Arbor, Mich., expects to start reviewing first drafts of machine-vision standards this summer—which is good news for users frustrated by the myriad nonstandard and expensive custom approaches offered by some 200 vendors. Five subcommittees are working on the drafts. Final standards could be ready as soon as mid-1987, says James C. Solinsky, the Rockwell International official who heads the committee overseeing the work. THE FIRST CONDUCTIVE, TRANSPARENT THIN FILM IS DEMONSTRATED An organic thin film said to be the first transparent polymer capable of conducting electricity has been demonstrated at Honeywell Inc.'s Physical Sciences Center in Bloomington, Minn. One possible use is as a replacement for inorganic transparent conductors used in liquid-crystal displays and photovoltaic cells. Made from the polymer polydiiodocarbazole doped with bromine, the film can easily be formed into layers 1 to 30 $\mu$m thick with standard techniques such as spin coating, Honeywell says. The material's conductivity levels, ranging from 0.1 to 1.0 $\Omega^{-1} \text{cm}^{-1}$, can't match the 10,000-$\Omega^{-1} \text{cm}^{-1}$ levels of conventional, inorganic transparent films made from combinations of tin, indium, cadmium, and zinc. But Honeywell says its material is cheaper, and it expects conductivity to improve significantly. TANDON ABOUT TO INTRODUCE ITS OWN PERSONAL COMPUTER IN THE U.S. This week should see the long-predicted U.S. introduction of Tandon Corp.'s namebrand personal computer, more than a year after the company hired away some of the manufacturing and marketing talent responsible for IBM Corp.'s Personal Computer. In the interim, Tandon has become a supplier to Xerox Data Systems, Tandem, and Tandy (Radio Shack). The computer is virtually the same machine that Tandon introduced in Europe last autumn but with some slightly different features that the company will bill as improvements. OLD RIVALS TI AND HONEYWELL UNITE TO BID FOR GaAs LINE The Pentagon's latest plans to set up a third pilot production line for fast gallium arsenide chips is resulting in a partnership between two military semiconductor rivals. Texas Instruments Inc. and Honeywell Inc. have submitted one of a handful of bids that are expected. The Defense Advanced Research Projects Agency wants the production line for fast memories and logic necessary to support the 200-MHz 32-bit microprocessors being developed for the Strategic Defense Initiative. "Everything is leading to the GaAs being used on the advanced on-board signal processor [module] of SDI," says Sven A. Rossild, a deputy director in Darpa's Defense Sciences Office. Bidding will end next week, and Rossild hopes for an award by the fall. FAIRCHILD SIGNS HITACHI TO MAKE FACT CMOS LOGIC FAMILY Gaining access and credibility in the Japanese market, Fairchild Semiconductor Corp. has signed a five-year agreement with Hitachi Ltd. under which Hitachi will become the first alternate-source supplier of the Fairchild Advanced CMOS Technology (FACT) family of digital logic circuits. Hitachi will produce and market FACT devices and use the trademark, and both companies will have access to new devices developed for the family. FACT devices are made in a silicon-gate p-well process with a 1.3-μm channel length and double-level metal. Fairchild is producing the family at its South Portland, Maine, facility, and will start manufacturing at its new plant in Nagasaki, Japan, in the first half of 1987. Hitachi expects to have first samples of its FACT products by the end of this year, and it plans to go into production in the first quarter of 1987. RAYTHEON GETS RIGHTS TO BIT1 BIPOLAR PROCESS Raytheon Semiconductor Corp. has locked up rights to use the high-speed low-power bipolar process from Bipolar Integrated Technology Inc., a Beaverton, Ore., startup, in gate arrays and standard cells. Raytheon, of Mountain View, Calif., whose parent corporation has invested in the company, will use the BIT1 process [Electronics, April 7, 1986, p. 35] in a series of 3,500- to 5,000-gate products to be announced late this year. The emitter-coupled-logic process is specified at 300 ps for unloaded gate delays and toggle frequencies of 600 MHz, at 300 mW power dissipation. This yields a speed/power ratio of 300 pJ. NORTHWEST SIGNS TO MAKE SOFTWARE TOOLS FOR INTEL Intel Corp. will integrate a family of software tools and debuggers from Northwest Instruments Systems Inc. into its line of microcomputer development tools. Under a two-year $15-million agreement, Northwest, Hillsboro, Ore., will engineer and manufacture Intel-specified instruments that will be marketed by Intel's development system operation, also in Hillsboro. If your system is only using this many colors, isn't it time to grow up? The INMOS IMSG170 Color Look-up Table offers a grown-up solution to video display color enhancements. It lets you and your RGB analog display advance to a palette of more than a quarter million colors. This programmable DAC conforms to RS170A standards with pixel rates up to 50 MHz, in a 28 pin package. And to simplify things, the table integrates the functions of a 256 word x 18 bit color mapping table, three 6-bit DACs, 75 ohm drivers and microprocessor interface into a monolithic CMOS device. The result? Reduced board space and power consumption at a lower cost. The Color Look-up Table from INMOS. Designing with thousands of colors just became child's play. INMOS Corporation, Colorado Springs, Colorado, Tel. (303) 630-4000 Bristol, England, Tel. 0272-290-861 Paris, France, Tel. (1) 687-2201 Munich, Germany, Tel. (089) 319-1028 Compare your computer design colors to these crayons. If the crayons win, call INMOS. Crayola is a registered trademark of Binney & Smith, Inc. Used with permission. INMOS, IMS and IMS are trademarks of the INMOS Group of Companies. "16 Data Acquisition boards. Not bad." Fred Molinari, President Data Translation stretches itself to new limits for users of the IBM PC. No one extends himself for a customer like Fred Molinari. Data Translation offers you the broadest line of Data Acquisition boards for the IBM PC in the business — and we've just added four more. Now you can get low-cost single function boards, medium-performance multi-function boards, and high-performance boards with A/D throughput all the way to 130 KHz. All supported by subroutine libraries and application software. Not to mention advanced capabilities like continuous "no-gap" data collection, random channel/gain sequencing, five-level bus interrupt, and a programmable pacer clock. See the chart for details and remember: Fred Molinari will go to any lengths to provide the boards you need. Call (617) 481-3700 \| Data Acquisition Products for the IBM PC \| Model \| Price (U.S. $) \| A/D Channels \| A/D Resolution (Bits) \| Speed (kHz) \| D/A Channels \| D/A Resolution (Bits) \| DIO (Mines) \| Clock \| Terminal Panel \| \|------------------------------------------\|---------\|----------------\|--------------\|-----------------------\|-------------\|---------------\|-----------------------\|-------------\|-------\|----------------\| \| General Purpose \| DT2801 \| 995-2070 \| 1-16 \| 12 - 6 \| 13-27 \| 2 \| 12 \| 16 \| Yes \| DT707 \| \| \| DT2806 \| 295 \| t> 90 \| 12 \| 20 \| 10 to 24 \| 12 \| 10 to 72 \| Yes \| DT757 \| \| \| DT2808 \| 495 \| 16 \| 10 \| 3.3 \| 2 \| 8 \| 16 \| Yes \| DT707 \| \| Low Cost \| DT2814 \| 395 \| 16 \| 12 \| 25 \| \| \| \| \| DT757 \| \| \| DT2815 \| 499 \| \| \| \| 8 \| 12 \| \| \| DT757 \| \| \| DT2817 \| 199 \| \| \| \| \| \| 32 \| \| DT758 \| \| High Speed \| DT2821 \| 1345-2495 \| 4-16 \| 12-16 \| 5C-130 \| 2 \| 12 \| 16 \| Yes \| DT707 \| PCLAB and ATLAB driver software and DT Notebook, DT/ILS PC, Asyst, Snapshot and Labtech Chrom application software. See our new 646 pg. catalog/handbook or see us in Gold Book 1986. Or call for your personal copy today. DATA TRANSLATION World Headquarters: Data Translation Inc., 100 Locke Dr., Marlboro, MA 01752 (617) 481-3700 Tlx 951 646 European Headquarters: Data Translation Ltd., 13 The Business Centre, Wokingham Road, Wokingham, Berks, RG112QZ, England Tlx 851849862 (2 0) International Sales Offices: Australia (61) 2-6352829; Belgium (2) 2-352 X35; Canada (416) 625-1907; Chile (2) 2-255689; China (400) 777-8222 (86) 8/214017; Denmark (02) 187188; England (44) 0734-793838; Finland (358) 0-372-144; France (33) 146306839; Greece (30) 031 527039 (30) 13-614300, (30) 95-14944; Hong Kong (852) 3 324563; India (91) 2-231040; Israel (972) 3 324298; Italy (39) 2349751; Japan (81) 3 502-5550; (81) 3 348-8301; (81) 3 335-1111; Korea (82) 753-3101; Malaysia (60) 3 362-99; Morocco (211) 9-30 6949; Netherlands (31) 70996360; New Zealand (61) 2-663-5289; Norway (47) 10275950; Peru (51) (1) 31 8100; Spain (34) 818 0120; Sweden (46) 87617828; Switzerland (41) 17231410; (41) 2-2360630; Taiwan (886) 2-721 7844; (85) 2-531-2134; West Germany (49) 88905020; 12469221; Spain (34) 14558112; Sweden (46) 87617828; Switzerland (41) 17231410. (41) 2-2360630. Taejon (886) 2-721 7844. (85) 2-531-2134; West Germany (49) 88905020. IBM PC is a registered trademark of IBM. Data Translation is a registered trademark of Data Translation, Inc. ROCKWELL TO OFFER CMOS/SILICON-ON-SAPPHIRE GATE ARRAYS Gate arrays to be fabricated in CMOS/silicon-on-sapphire technology will soon emerge from Rockwell International Corp.'s Electronics Operation. The new arrays, which will be built in the Anaheim, Calif., division's 1.25-$\mu$m process, feature built-in self-test technology from Control Data Corp., Minneapolis. The first array will offer about 6,500 gates. Design support is available from Control Data's Technology Application Center in Plymouth, Minn. Prototypes can be delivered in one month; pricing has not been set. BBN IS MARKETING AN ENHANCED VERSION OF ITS PARALLEL COMPUTER Bolt Beranek & Newman Inc. has enhanced its Butterfly parallel-processing computer and has set up a separate subsidiary to market it. The system can now be fitted with up to 256 nodes that are built around the 68020 microprocessor and 68881 floating-point coprocessor. Each node can now have 4 megabytes of memory and operates at 1 million instructions/s. A four-processor starter system is available from BBN Advanced Computers Inc. for less than $40,000; additional 68000-based nodes are $6,500 each and the new 68020-based nodes are $9,500 each. The Cambridge, Mass., company says a typical 32-node system costs $375,000. A new Fortran 77 compiler is $9,000. CHIP LETS SLOWER PERIPHERALS WORK WITH THE NEW, FASTER Z80 A CMOS chip from Texas Arrays Inc. will allow existing 4-MHz peripheral circuits to be used with new 8-MHz versions of Zilog Inc.'s Z80 microprocessor. The 20-pin TA84HC04 emulator chip operates during input/output cycles, issuing wait states to the processor. Slowing down the microprocessor only during the I/O operations lets the Z80 run at full speed 95% to 98% of the time. In lots of 100, the TA84HC04 costs $6.15 each. The circuit is also available in the Carrollton, Texas, company's 3-$\mu$m CMOS standard-cell and gate-array libraries. DATA BASE CAN HANDLE BOTH PICTURES AND TEXT Computer Corp. of America is unveiling two products that let its model 204 mainframe data-base-management system handle both pictures and text along with traditional data. The Picture/204 makes it possible to store digitized pictures in the Cambridge, Mass., company's model 204 and to retrieve them from the system, using a personal computer attached to the mainframe. CCA's new Text/204 allows documents to be integrated with the model 204. Text/204 will be available in December for $47,000. Picture/204, which will ship in October, sells for $30,000; the color version of the imaging software is $950. COMPUTERTHOUGHT RELEASES NEW VERSION OF ITS EXPERT SYSTEM ComputerThought Inc. is releasing a new version of its OPS5+ expert-system development software, following a legal battle with former employees over ownership of the product. The OPS5+ package, which builds an expert system with up to 1,500 rules, was previously marketed by employees who left the Plano, Texas, company and formed Artelligence Inc. OPS5+ is now the exclusive property of ComputerThought, the company says, and the debugged Version 3.0 is priced at $1,850. OPS5+ is available for the IBM Corp. Personal Computer and the Apple Computer Inc. Macintosh. Versions for Apollo Computer Inc. and Sun Microsystems Inc. work stations will come out in two months. ANALOGIC OFFERS AN INTELLIGENT PROCESS MONITOR Analogic Corp. is applying its expertise in analog-to-digital converters to a line of intelligent process monitors and controllers for industrial and factory applications. The Danvers, Mass., company's new microprocessor-based Control Master can accept two different types of input signals—current and thermocouple—and provides four programmable digital outputs for controls or alarms. Control Master, the first product in the new line, also includes a 17-bit ADC that runs at 20 conversions/s, 10 times faster than competing products. Control Master will sell for $375 when available next month. NEWBURY DATA BUILDS 51-MEGABYTE 3½-IN. WINCHESTER DRIVE A 3½-in. Winchester disk drive with a 51-megabyte capacity may soon appear in equipment. Newbury Data Inc.'s drive is under evaluation by several major U.S. and European equipment manufacturers. The four-platter Penny drive, announced some two years ago by the Hermosa Beach, Calif., company, was delayed by development problems. It now boasts an average access time of 40 ms, transfers data at 5 Mb/s, weighs 1.65 lb, and consumes 12 W. Newbury Data's parent, Newbury Data Recording Ltd. of the UK, produces the drive. The price in large quantities is less than $1,000 per unit. WESTERN DIGITAL'S HARD-DISK CONTROLLER STORES MORE BITS Western Digital Corp.'s first hard-disk controller using the run-length-limited (RLL) encoding format can record up to 60% more bits than the widely used modified frequency modulation. Moreover, the WD1002-27X transfers data 50% faster than previously possible, says the Irvine, Calif., company. The controller incorporates two proprietary chips: Western Digital's standard WD1010 controller and the new hard-disk RLL device, which integrates a 56-bit error-correction code and support for 7.5-MHz data transfers. Samples of the WD1002-27X will be out next month. Specific pricing varies, but the controller initially will cost more than products now on the market. ROCKWELL SHIPS SINGLE-CHIP FACSIMILE MODEM... Rockwell International Corp.'s Semiconductor Products Division is shipping samples of a single-chip facsimile modem that transmits data over the public switched telephone network at 2,400 b/s. The R24MFX will cut the size and cost of facsimile equipment and could give a shot in the arm to the emerging market for personal facsimile machines. Optimized for use in compact Group 3 facsimile machines, the chip has two interfaces—a parallel microprocessor bus interface and an RS-232-C serial port. In 1,000-piece lots, the R24MFX sells for $40 each. The Newport Beach, Calif., division will begin volume shipments in August. ...WHILE AN NEC MEMORY WILL LEAD TO COLOR FACSIMILE NEC Corp. has boosted the size of its byte-wide dual-port memories fivefold, to 5,048 bytes, which should lead shortly to commercial color facsimile systems. Applications for the μPD42505C first-in-first-out buffer also include graphics, data compression and expansion, and buffering for the transmission of data between two systems operating at different speeds. The buffer is suitable for use in Group 3 and 4 facsimile systems, multifunction copiers using digital imaging, image scanners, and other digital systems. The sample price for devices with 50-ns write/read cycle times is $42.68 in Japan. Thinking about building a multi-user dial-up application, but concerned about the high cost of a superminicomputer or mainframe? Consider the GALACTICOMM BREAKTHROUGH, a 16-channel modem card for the IBM PC/XT, PC/AT, and compatibles. Now, using this card together with the SOFTWARE BREAKTHROUGH object library, you can easily put together high performance multi-user applications at minimum cost. The software is the key - Special interrupt-driven I/O method eliminates 92% of the interrupt overhead of conventional designs - DOS-bypass subroutines to access all modem functions (e.g. accept a call, place a call, send/receive ASCII data, send/receive DTMF data, check modem status, set baud rate and parity, and much more) - Assembly language automatic pending-input channel-scanning and circular/linear buffer management - Supports up to 64 channels (4 BREAKTHROUGH circuit cards) simultaneously at 1200 baud on the PC/AT - Fully PC-DOS compatible, bindings exist for most popular languages, not copy-protected, no royalties The hardware is incredible - The XECOM 1201 is the modem module described in Steve Ciarcia's November 1985 Circuit Cellar column - Each modem module is complete, USART to DAA, 212A compatible, FCC registered and ready to run - Adapters available for direct connection via 16 RJ-11's, or a single RJ-21X mass telephone connector - Port stacking technique allows direct addressing of each modem while consuming only 2 PC ports per card - DIP-switch selectable port-group addressing on any even boundary in the range from 200H to 3FEH The applications are limitless - Multi-user bulletin boards and EMAIL systems - Multi-user real-time teleconferencing - Multi-user customer order entry networks - Multi-user online expert systems - Multi-user classified advertising services - Multi-user database search & retrieval systems - Multi-user interactive adventure games - Multi-user sales information inquiry systems - Multi-user online educational services - Multi-user "simulated CB-radio" networks For some free sample time on a PC/XT running the GALACTICOMM BREAKTHROUGH, dial (305) 922-3901 during business hours. Have your modem ready to go online in originate mode at 300 or 1200 baud: 7 data bits, even parity, 1 stop bit. The Acquisition. With sweep speeds from days to nanoseconds and resolution up to 15 bits, the 4094 digital 'scope can capture the most elusive signals. Every plug-in has 16K of memory, viewable trigger set-up and independent pre- or post-trigger delay on each channel. Signal averaging is standard and our latest 10 MHz/12-bit plug-in even offers real time manipulation of the incoming signals. With two plug-ins the 4094 can record four channels simultaneously. Or even monitor two slow signals and capture high speed glitches at the same time. All under computer control or via manual operation: whatever your application demands. The Analysis. Expand and examine any waveform feature in detail. Use the dual cursors and numerics to measure the time or voltage of any point. Compare live or stored waveforms with each other or with pre-recorded references. Store signals on disk manually or automatically. Use pushbutton programs to manipulate the data or send it to your computer via GPIB or RS232 interface. Complete your report with a hardcopy plot using the XY/YT recorder or digital plotter outputs. First Time, Everytime. Don't miss important data because of set-up errors. From the World's first in 1973 to the latest models, Nicolet 'scopes are easy to use. Find out how they can be the quickest solution to your signal problems. For more information call 608/273-5008, or write Nicolet Test Instruments Division, P.O. Box 4288, 5225 Verona Road, Madison, WI 53711-0288. CONTROL DATA LAUNCHES CMOS/SOS SPACE COMPUTER 1750A MACHINE BEATS VHSIC-PROGRAM PROCESSORS TO MARKET MINNEAPOLIS By rolling out a radiation-hardened, spaceworthy computer based on a 2-$\mu$m CMOS/silicon-on-sapphire process, Control Data Corp. is hoping to steal a march on contractors involved in the Defense Department's Very High Speed Integrated Circuits program. The Minneapolis company this week will unveil its Spacecraft Control Processor, which implements MIL-STD-1750A, the Air Force's instruction-set architecture for 16-bit computers. Set to be ready in next year's third quarter, the SCP could offer a substantial improvement over today's state-of-the-art space machines. At the same time, it could beat to market by two years the first 1750A space computers built in rad-hard 1.25-$\mu$m Phase 1 VHSIC technology. There will be a market for the SCP, predicts Darrell J. Putnam, a member of the professional staff in digital design at the Missile Systems Division for Martin Marietta Orlando Aerospace, Orlando, Fla. "The VHSIC stuff will be a lot better when it does come out, but Control Data is about two years ahead, as far as I can see," Putnam says. SDI PROOF. In particular, Putnam says, the Control Data machine could play a major role in experimental systems that are aimed at early proof of concepts to be used in the DOD's Strategic Defense Initiative. Such systems might be launched as early as 1988, Putnam says. By contrast, computers built under a year-old DOD program to produce rad-hard 1.25-$\mu$m VHSIC-class 1750A space computers are not due to emerge until mid to late 1989 (see "VHSIC computer tests begin this fall," p. 28). The SCP is actually the first of a new family of planned space-computer modules from Control Data. It will run on a new interconnection platform that Control Data designed to meet the needs of future space programs, such as SDI, which are expected to increasingly require multiple processors on board. Future Control Data machines will be based on VHSIC-class 1.25-$\mu$m technology, the company says. But by establishing a presence and a reputation based on 2-$\mu$m technology, the company hopes to get a jump on what could be a $4 billion to $5 billion market for high-performance space computers by the mid-1990s, says Terry Petrzelka, vice president of Program Planning and Development for Control Data's Government Systems Group. Control Data also plans to adapt the SCP technology next year for use in military avionics. Using an 8-MHz clock, the SCP's two-chip central processing unit will execute about 1.75 million instructions per second using a standard floating-point Digital Avionics Instruction Set mix, says Andrew Wardrop, Control Data's product chief engineer for space engineering. Even after it has received its full rated radiation dosage of 200,000 rads, it will still hit 1.75 mips. The addition of single-bit-error correction and double-bit-error detection for memory will bring SCP performance down to 1.25 mips in a space-configured system, Wardrop says. That's well below the performance of the 3 to 4-mips single-chip processors expected from VHSIC-based space machines. But by offering the SCP in an optional dual-processor version, Control Data figures to boost SCP space-system throughput to 2.5 mips, or more than twice the speed of today's standard space computers, executives say. RAD HARD. Control Data Corp.'s Spacecraft Control Processor, with its 2-$\mu$m CMOS/SOS technology, will be ready two years before machines based on 1.25-$\mu$m VHSIC chips. Control Data has no plans to fabricate the SCP chip set itself but will instead rely initially on CMOS/SOS chips supplied by Rockwell International Corp., says Curtis L. (Pete) Nelson, the Government Systems Group's manager of new business development. Control Data is also working with RCA Corp. as a second potential supplier, he says. FIRST PASS. Working versions of the custom two-chip CPU set, each integrating about 10,000 gates, were obtained on the first processing pass, Nelson says. Three simpler circuits—a memory-management chip, a memory interface device, and an input/output chip—are under design using a semicustom gate-array approach. The firm expects to have working versions of all five chips by next year's first quarter. Engineering units of the SCP should be available in the third quarter of 1987, says Nelson. A brassboard version is scheduled to undergo Air Force 1750A certification tests at Wright-Patterson Air Force Base next month. The SCP will feature two independent memory-access buses for improved reliability. To be housed on boards measuring 3$\frac{1}{2}$ by 3$\frac{3}{4}$ in., a minimum three-board configuration—one board each for CPU, 64-K 16-bit words of error-protectVHSIC COMPUTER TESTS BEGIN THIS FALL While Control Data focuses on the near-term market for space computers built with devices featuring minimum 2-μm geometries, at least four contractors are already hard at work on radiation-hardened space computers made with chips using more advanced 1.25-μm technology produced under Phase 1 of the Very High Speed Integrated Circuits program. The work is being sponsored by the U.S. Department of Defense under a project known as the Generic VHSIC Spaceborne Computer (GVSC) program. Initial contracts awarded last July went to Harris, Honeywell, IBM, and RCA. Under the first phase of the effort—to be completed with testing in September—each of the four is to develop a space processor design based on the Air Force MIL-STD-1750A architecture. The contractors also will deliver several lots of test chips and wafers that will undergo government radiation testing, says Capt. Russell R. Herndon, GVSC program manager at Kirtland Air Force Base in New Mexico. Harris, Honeywell, and IBM are working in bulk CMOS processes, and RCA plans to implement its design in its CMOS/silicon-sapphire technology. The radiation data accumulated on each contractor's process under the initial phase will be used in selecting one or two of the four contractors for the second phase of work, Herndon says. Phase 2 aims to produce working space-qualifiable hardware using 1.25-μm technology 30 months after the contracts are let. "We're hoping to start the second phase during the first part of 1987," Herndon says. GVSC goals include single-processor performance in the range of 3 million to 4 million instructions/s. Herndon declines to reveal GVSC targets for radiation hardness, but the program is believed to be aimed at systems capable of withstanding a total dose of up to 1 Mrad. VHSIC-class 1750A chips produced outside the GVSC program might also turn out to be suitable for use in space. Texas Instruments Inc., for one, has already produced working 1750A processor chips built in 1.25-μm Schottky-transistor logic. The Dallas company's single-chip CPU is aimed at avionics applications and has not been tested for radiation tolerance, says Dennis Best, VHSIC program manager for TI's Defense Systems and Electronics Group. But Best notes that STL is inherently rad-hard, and tests run on other devices built with the technology have shown total dose tolerance levels beyond 1 Mrad. ed memory, and I/O—will cost about $350,000 for an engineering unit. A maximum fully space qualified 12-board system using two CPUs and eight cards of random-access memory totaling a 512-K words will go for about $1.25 million. Control Data says a maximally configured system will weigh about 7 lb, require 10 W of regulated power, and occupy about 120 in³. For applications that can tolerate 10% to 20% lower performances, an SCP could also be equipped with additional boards, bringing maximum addressable memory up to 1 million words, Wardrop says. Both logic-design techniques and the inherent radiation hardness of CMOS/SOS are responsible for SCP's 200,000-rad rating, CDC says. Currently available memory devices can't provide equivalent radiation hardness at the 64-K densities and 45-ns access times needed by the SCP. But such chips will be available by the time the system is ready to fly, CDC officials expect. BEYOND CONTROL. In the near term, Control Data says the SCP will serve as a replacement for its current-generation 10-μm p-MOS-based 469R² space computer used for spacecraft navigation and control. For applications beyond control, a second SCP aboard a spacecraft could also be tied in using the Control Data interconnection scheme for handling payload data-processing tasks, and also could work with other specialized processors. "Our goal is to build a family of products, using the 1750A [SCP] as both the control processor and as a near-term data processor, and then replace it with a larger 32-bit data processor in the future," says Nelson. Beyond that, officials are mum on plans for the company's new space-computer architecture, except to say that future announcements will include a new specialized signal-processing module for crunching satellite-sensor data. Space-industry officials give Control Data high marks on its game plan. "It looks like they've really put a lot of thought and homework into their architectural design to address a lot of the issues unique to flying computers in space," observes Lt. Col. Ralph Gajewski, SDI manager of signal-processing technology. The SCP is "a start in the right direction," Gajewski says. "You're always ahead by having working hardware." But to be successful in the long term, the SDI official warns that CDC must keep moving. "Space-computer users are pretty fickle, and they always want the latest and the best," Gajewski notes. "Control Data will have to follow through on their product-improvement programs and be able to adapt to higher-performance circuitry as it becomes available." —Wesley R. Iversen RESEARCH & DEVELOPMENT PULLING SPACE R&D DOWN TO EARTH LOS ANGELES The wealth of technology generated at the nation's heavyweight research centers offers a tempting prospect for industry, particularly smaller companies with limited budgets. But moving that technology from the lab into a commercial-product environment has always proved a difficult task. Now a nonprofit Los Angeles group called Rimtech (short for Research Institute for the Management of Technology) is launching a new drive to forge a link between small local companies and scientists from the National Aeronautics and Space Administration's Jet Propulsion Laboratories in Pasadena, Calif. Four companies have already signed up in the Rimtech-JPL/NASA Product Development Program since its start earlier this year, and the group plan a stepped-up effort to sell the program to others at a public briefing scheduled for July 7 in which government and industry luminaries extolled the project's virtues. Prior attempts to get government researchers and private industry executives to join hands have had little success because they've been too limited in scope. In 1980, the federal Technology Innovation Act established the Center for the Utilization of Federal Technology, but since then there have been only a few successful examples of technology transfer, next to none of which has been electronics-related. Other programs went under for various reasons. "Many of these failed in the past because they were some researcher's pet project in search of a market," says Rimtech president Steven M. Panzer. He and other officials make it clear that the developers of valuable research have historically not found success in their attempts to distribute their findings into the hands of commercial users. Instead, they say, the impetus for successful transfer must come from individual companies, especially those with strong entrepreneurial spirit and a product and market in mind. What these companies lack is specific technology assistance to get the job done, and that's where Rimtech is seeking to make its presence known—as a technology broker of sorts. Rimtech officials will help companies identify useful technology generated within JPL, and then help arrange to channel that technical know-how into the company's hands. This can involve large or small group presentations, one-on-one engineering laboratory demonstrations, a transfer of documents, or a careful orchestration of any of these methods. The contract is then sealed with a $25,000 annual fee (regarded as a bargain by all hands, considering time and administration costs alone) and an agreement on a schedule of specific items to be delivered. Rimtech closely monitors the accompanying timetable. TYPICAL CUSTOMER. Because it carries little risk, Rimtech officials expect the plan to draw widespread interest. Emulex Corp.'s reaction was typical of the four companies to get involved so far. The Costa Mesa, Calif., maker of data-storage subsystems and peripherals quickly opted to take part after an initial presentation on data processing and a follow-up talk. Specifically, Emulex is looking for advanced work on data compression and encryption, a field in which JPL is an acknowledged leader because of its accomplishments in transmitting data to Earth from unmanned probes in deep space. "This could have real downstream impact for us," predicts Emulex chairman Fred B. Cox. Optical-memory research is another hot topic, he adds, but one that Emulex, with only about $100 million in sales, cannot pursue on its own. "We can't afford to look that far out, so we need to take advantage of [JPL's] forward thinking." Cox maintains that without a player such as Rimtech organizing the transfer, JPL's technology would be lost to Emulex and others like it. But industry observers familiar with the fate of earlier efforts to merge the labs with industry are not yet convinced this one will succeed. "It's no easy task to take sophisticated technology and convert it to the real electronics world," says Matt Crugnale, a Mountain View, Calif., marketing consultant. For one thing, he says, such an exchange must emphasize verbal communications, "and engineers are terrible at communications." Moreover, getting the right information often requires more time than either side can devote to digging out facts from records and developing them for a specific purpose. And prickly professional pride can raise a barrier that may go unnoticed until it's too late, Crugnale says. "Lack of mutual respect at the technical level is one of the elements that makes [transfers] most difficult." Rimtech officials admit success will not come easily. The group notes in its prospectus that "management practices and guidelines for technology transfer—ownership, decision-making, and publication rights—have not been studied, developed, or disseminated in depth." No systematic approach has emerged and techniques vary widely, it continues. But the officials point out that initial responses are encouraging. JPL is pushing this sharing program and investing some $200,000, in part because getting more deeply involved with industry presents a way to keep its engineering and scientific teams intact in the face of NASA budget uncertainties. Even though no blame was laid on JPL for the Challenger disaster—the Pasadena lab has been very successful with its unmanned missions—a planned schedule of projects is on hold because there are no means to launch. —Larry Waller NETWORKING LOW-SPEED LANs CHECK IN AT $100 PER CONNECTION MOUNTAIN VIEW, CALIF. Just as low-cost local-area-network solutions are beginning to emerge from the haze of protocol wars and competing access technologies, a new class of software-based LANs is appearing to make the job even harder for mainstream networking companies. The new LANs are slow (115 kb/s and under), they accommodate only a few users, and, because they run under the MS-DOS operating system, they are limited to use with IBM Corp. Personal Computers, PC ATs, and compatibles. But they support the kind of file-transfer and resource-sharing functions that make up the bulk of networking requirements for small installations. Most important, these new LANs are priced at the magic figure of around $100 per connection. The cheapest mainstream LANs still cost $750 or more per connection, and even the Starlan, backed by AT&T Co. and billed as a low-cost approach, is priced at just less than $600 [Electronics, June 9, 1986, p. 20]. "Why spend $1,000 so that a PC can share a $500 drive?" Server Technology, like others in the field, substitutes twisted-pair telephone wiring for coax and links its EasyLAN to a computer through an RS-232-C connector and the serial port, thus eliminating the need for a separate network card. Network control is performed by memory-resident software. EasyLAN is priced at $180 for the first two connections and $109 for each additional node. But because it is a series of point-to-point connections, it requires the server to have a separate serial port for each node, and these ports add to the overall cost of the network. UNDER $100? The first true $99 LAN may be a new offering from Applied Knowledge Group Inc. of Mountain View, whose Knowledge Network runs in any configuration—bus, star, tree, or ring—without a server. Stations plug into a simple RJ-11 bus box. Knowledge Network incorporates a simple form of the Ethernet contention scheme, carrier-sense multiple-access with collision detection. Remote disk drives and peripherals are identified as virtual devices, and users access them with standard DOS commands. Simple error correction is performed with cyclic redundancy checks, and users may protect or hide files. Other low-cost LANs that operate as a collection of point-to-point connections from a central server include offerings from Avatar Technologies Inc. of Hopkinton, Mass.; The Software Link Inc. of Atlanta; IDEAssociates of Billerica, Mass.; and Digital Products Inc. of Watertown, Mass. Avatar charges $895 for an eight-user system. Digital Products sells its peripherals-sharing subLAN for about $150 per connection and a version that adds file sharing for about $230. IDEAssociates' IDEAshare is priced at $595 for two users, and LANLink from Software Link at $695 for four users. Cables, in all cases, are extra. Because they operate with a central controller, these LANs avoid the problem of network contention. But they must accept a star configuration with a multiport server that runs at all times. The servers range in complexity from Easynet's, which is a nondedicated PC with multiple serial ports, to Avatar's Alliance, a separate box that employs an 8-bit Hitachi Ltd. 64180 processor plus a gate array wired as a 1-Mb/s switch. COLLISIONS. Knowledge Network achieves CSMA/CD operation through a simple switch in the RS-232-C connector that echoes all transmissions back to the sender. If two stations attempt to send at once, the voltage on the circuit will not support the required 7-V peak-to-peak level, and the echo will be squelched. Both stations will then wait for a random time before trying to retransmit. The speed of the networks ranges from 19.2 kb/s for subLAN and Easy-LAN implemented on an IBM PC, to 37 kb/s for Knowledge Network and Easy-LAN on a PC AT, to 115 kb/s for Alliance and the Software Link's LANLink. Alliance accommodates the most users, 20, though a 31-user Knowledge Network is promised for the fall. The LANs are all extendable. Knowledge Network, for instance, can use a modem as one of its nodes and through it dial up other network clusters. LANlink, which has eight satellite stations per cluster, allows the satellites themselves to become servers for additional clusters; Software Link says it operates a 42-station system of its own. Avatar's Alliance can also be daisy-chained. Easynet can also make its node connections through private-branch-exchange switches, increasing the potential number of nodes. Server Technology expects to announce an agreement with a PBX maker within 60 days. Because the low-cost nets were developed without regard for the transmission protocols being developed by the International Organization for Standardization, they do not interface easily with Ethernet or IBM's Token-Ring Network and PC-Net. Alliance and the Knowledge Network, however, could be linked to such nets by attaching a network controller. And Avatar plans a 3270 link for Alliance. —Clifford Barney SEMICUSTOM ICs SPEEDING UP PLACEMENT AND ROUTING MOORESTOWN, N. J. RCA Corp. is continuing its drive to become a dominant force in application-specific integrated circuits with a new design tool that can place and route large numbers of standard cells and larger macrocells while automatically designing the chip's power bus. Dubbed Happi, for Hierarchical Automatic Partition Placement Interconnect, the software produces layouts for RCA's double-metal single-polysilicon process and can be adjusted for different geometries. Happi is embedded into Vital, the system RCA's Solid State Division will use in its upcoming ASIC Silicon Circuit Board products. Communicating with Vital by means of an ASCII data base, Happi uses that program as an interactive front end and is transparent to the user. But the results are far from invisible. Running on a Digital Equipment Corp. VAX 8600, the new program produced one chip design incorporating three macrocells and 1,500 standard cells in 97 minutes—about half the time it took to produce the same design using RCA's older and more established MP2D placement and routing program. "We are faster than MP2D, there's no doubt about it," says Rathin Putatunda, a technical staff member with the ASIC, CAD, and VLSI Design, Development, and Support Lab at RCA's Advanced Technology Laboratories in Moorestown. "We're about 50% better in speed." But Putatunda adds it's not just speed that makes Happi special. "Our previous programs, like MP2D, could only handle up to four macrocells—and even then, with restrictions on the locations where those cells could be positioned on the chip. This program has more flexible power distribution, more flexible chip topology, and it's much better in terms of placement and routing." Happi does not represent the death When you need certified AC Power Sources, we deliver hardware—not promises. DEPARTMENT OF THE AIR FORCE HEADQUARTERS AERONAUTICAL SYSTEMS DIVISION (AF/SC) WRIGHT-PATTERSON AIR FORCE BASE, OHIO 45433-6501 REPLY TO: ATTN: AEGB SUBJECT: MATE Certification TO: California Instruments 5150 Convoy Street San Diego, CA 92111 1. Reference: California Instruments' AC Power Supply with Sperry Corp Test Module Adaptor (TMA) (docket #ACPSI/qsi/1) 2. The MATE Program Office (ASD/AEGB), Wright-Patterson AFB, Ohio please to inform you that your AC Power Supply with Sperry Corp Test Module Adaptor, model 847M, (Serial Number 90501) has successfully completed MATE Verification Testing. 3. The testing was completed on 08 January 1985 at the Module V Test Station within the MATE Support Center at Sperry Electric Great Neck, New York. The testing was witnessed by Air Force personnel. 4. We have enclosed a MATE Certificate for your AC Power Supply with this test. In addition, you are entitled to purchase MATE Test Placards for each of your manufactured 847M AC Power Supplies. 5. You are free to advertise your model 847M AC Power Supply and MATE module as it has completed all phases of MATE testing and A-10 AFCS Application Testing. By virtue of the successful product data will be entered into the MATE Data System inventory of the module. 6. We look forward to evaluating more of your products! LOUIS M. SALERNO, JR. Deputy, MATE Program Office Subsystems/Support Equipment SPO Attachments: MATE Certificate, 30 May 85 MATE Placards When your testing requirement demands proven capabilities to meet tough specifications, call California Instruments. Our complete line of AC power sources are the confirmed and certified choice in military applications. California Instruments also offers the broadest line of standard AC power sources including the versatile T Series Invertrons, the TPM models for high instantaneous power and the new XP series with packaging efficiencies and control capabilities that lead the industry. For custom applications our depth of experience (over 25 years as a supplier) is unmatched. Our technical staff has the expertise to assist you in the following areas: • Avionics Testing • Frequency Conversion • Instrument Calibration • DC Power Supply Testing • Ground Power Simulation • Position Sensor (LVDT/RVDT, Synchros, etc.) Testing • Circuit Breaker and Fuse Testing • Military Electronics Testing • Gyro Testing—Microsyn and Wheel Supplies • Watt Hour Meter Testing Take command of the testing power in your application. Call or write today for assistance and a complete catalog. An affiliate of California Instruments A Division of Amstar Technical Products Company, Inc 5150 Convoy Street San Diego, CA 92111 (619) 279-8620 Telex 695047 Call Toll Free 800-356-2244 In California 800-821-1634 Electronics / July 10, 1986 knell for MP2D, however, says Putatunda. MP2D has a large installed base, both in and out of RCA. And the Happi system is designed for double-metal processes, while MP2D is a single-metal program. RCA has no plans yet to market the program, but that is an option it will study—especially because Happi will be able to interface to Silicon Compilers Inc.'s macrocell library. Happi's high degree of automation comes from its highly structured topology, Putatunda says. Following a binary-tree pattern, the chip is laid out section by section such that certain domains within the chip's architecture must be in place before the program can proceed to other domains. With this scheme, Putatunda and his research team were able to develop a flexible system that simplified the automatic-routing task. "We chose structured chip placement to facilitate automatic routing and automatic power bus distribution," Putatunda says. "The binary tree structure also helps to minimize area and the interconnect distance between domains." "It makes routing much easier—it simplifies it," adds Stephen McNeary, who along with Putatunda and David Smith took the project from start to finish over the last 18 months. Happi's placement component is responsible for generating and placing the domains and determining the cell-row spans of the standard-cell domains. Macrocells are always domains unto themselves, but the placement program can choose the cell's orientation—whether it is to be flipped or turned into any one of eight positions. CLUMPS. Standard cells, on the other hand, are arranged in clumps to make the best use of real estate while keeping trace lengths to a minimum. "We like to see the standard cells as a big soup," says McNeary, explaining that standard cells are "malleable" compared to the macrocells, which are fixed in shape and dimensions. "With the standard cells, it's putty logic," he says. "You can squeeze them around the macros." Happi's routing component generates the power-bus network and performs the detailed routing of signal and power lines automatically. It is here that the responsibility for low channel density falls. Long multiple-connection nets are broken down into segments and assigned to routing channels. Unlike conventional approaches, which use additive power routing, Happi uses a subtractive process, starting out with an overconcentrated power bus and gradually taking out connections when it becomes apparent that they are unnecessary. "The philosophy behind the power-bus generation method is to maximize the sources feeding power to every domain while attempting to minimize the impact of power bus routing on chip area," Putatunda says. "The channels that affect the X and Y dimensions of the chip are the critical channels. The main aim is to delete as many of the power-bus segments in these channels as possible." —Tobias Naegele --- TELEPHONY DSP POACHES ON ANALOG CONFERENCING DALLAS The telephony world is getting a bit dizzy as the digital revolution whirls along. Most digital private branch exchanges implement conference calls by converting voice signals back into analog form for tried-and-true, low-cost analog summing. And now a small Dallas-area company is going the other way—introducing a teleconferencing product that connects to older analog exchanges but makes its line connections digitally. The digital-analog anomaly results from familiar tradeoffs between the use of new technology and higher costs. DSP Technology Corp.'s digital Confer teleconferencing bridge is further evidence of inroads made by programmable digital signal-processing chips, which were once considered too costly and exotic to be used for end-user telecommunications equipment. FALLING PRICES. "We aim to take advantage of the falling prices of DSP chips, which until the past year have been too costly to consider for most telecommunications applications," says Paul Pandian, president of DSP Technology, Carrollton, Texas. Texas Instruments Inc.'s TMS32010 DSP chip has dropped from $120 in 1983 to under $10 today in large quantities, according to TI. The growing ranks of DSP suppliers are expected to spark price competition in the coming years. DSP Technology is applying the TMS-32010 in its six-party teleconferencing subsystem, which will be available next month for $1,495. Pandian says the Confer conference bridge, which installs on the central-office trunk-line side of an analog key-service unit, targets offices with 30 to 50 workers. The 19-by-10-in. wall unit is controlled by the telephone's tone-key pad. With it, workers can set up conference calls using phones either inside or outside the office. The ideal installation would have a 10-line key-service unit, says Pandian, who notes that about 200,000 analog exchanges are still shipped each year. The reverse of DSP Technology's analog-to-digital approach is found in digital PBXs. For example, Shared Resource Exchange Inc., Dallas, has been shipping a small bus-based digital exchange that connects calls in an analog manner [ElectronicsWeek, March 18, 1985, p. 17]. Shared Resource uses standard pulse-code-modulation codec chips to change digital voice signals into analog form. The analog signals are then combined using a conventional resistive summing network, says Charles B. Johnson, cofounder and executive vice president. The company's office system has a standard module supporting three-party conferences but more lines can be connected, he adds. "It is a cost issue," says Johnson, referring to the use of DSPs in conferencing hardware. "It is a nice..." Digital Signal Processing Single-Chip DSP Processors and Parallel Processors September 22-26, 1986/Washington D.C. October 27-31, 1986/Dallas, Texas November 10-14, 1986/Anaheim, California Texas.Inst. TMS32010, TMS32020, TMS320C25 Analog Dev ADSP2100 NEC upd7720, upd77230, upd7281 National LM32900 Fujitsu MB8764 AT&TWE DSP32 Philips PCB5010, PCB5011 OKI MSM6992 Kurzweil KSC2408 Zoran ZR33481, ZR33881, ZR34161 NCR NCR45CG72 INMOS A100, T414 This seminar introduces you to the theory of digital signal processing, and available DSP processors. Each device mentioned above will be discussed and applications will be presented. For more information, please contact: DSP Associates 18 Peregrine Rd. Newton, MA 02159 (617) 964-3817 Yes, I will attend the seminar to be held at: ☐ Washington D.C. ☐ Dallas ☐ Anaheim ☐ Yes, I will require hotel accommodations Name: ______________________________________ Company: ____________________________________ City ________________ State ____ Zip _______ Telephone: ____________________________ Registration Fee $995, a 10% discount for early reg. and to companies sending 3 or more employees For additional information circle 33 on the reader service card ONE DAY GATE ARRAYS. GUARANTEED. Prototypes in 24 hours? Only from Laserpath, the gate array prototype specialists. Lost time is the biggest risk in gate arrays. Now, we guarantee you'll know in one day—in silicon—if your design works. Because prototypes are the only way to know if your whole system works. No worry about circuit changes, either. Revised prototypes are discounted and get the same one day turnaround. You'll design with industry standard 2µ and 3µ HCMOS families. Use extensive macrocell libraries supporting the leading CAD systems. And get volume shipments from our production partners, who are major IC manufacturers. For details, call 408-773-8484 and ask for Earl Watts. Or write him at: Laserpath, 160 Sobrante Way, Sunnyvale, CA 94086. Give us a day, and we'll save you weeks. LASERPATH ONE DAY GATE ARRAYS™ Circle 158 on reader service card THE ELECTRONICS INDUSTRY’S MOST DRAMATIC INTEGRATED TESTING STORY BEGINS WITH A LINE LIKE THIS: Design Automation Products HILO® Universal Logic Simulation System HICHIP™ Universal Hardware Modeling System HIPOST™-ST HILO to GenRad VLSI Tester Interface VLSI Test Products GR125 VLSI Test System GR160 VLSI Test System GR180 Complex VLSI/VHSIC Test System Incoming Inspection Products 1657 & 1658 Digibridge® RLC Testers 1687-B 1MHz LC Digibridge 1689M Precision RLC Digibridge 1731M Linear IC Test System 1732M Digital IC Test System 1734M Memory Test System SCAN™ Semiconductor Component Analysis Network 1752 Elevated Temperature Handler 2295 Programming Station Board Test Products 2272 High Pin Count Incircuit/Functional Tester 2276 XP High Performance Incircuit/Functional Tester 2276E Expandable Incircuit/Functional Tester 2276 ICA Incircuit/Functional Analyzer 2275 Low Cost Incircuit/Functional Tester 2293 & 2294 Programming Stations BusBust®, Scratchprobing and AC Quadrature Bridge Diagnostic Tools Automatic Test Generation (ATG) SpeedPlus™, J11™ Processor, Productivity Plus™ and Analog Instrument Options Bantam™ Lightweight Fixture TRACS® Test and Repair Analysis/Control System DataTRACE™-32 TRACS Software Enhancement Package Structural Test Products 2511 & 2514 Vibration Control Systems 2515 Computer-Aided Test System Service Test Products 2225 Portable Service Processor 2235 Logic Test System 2255 Incircuit Test Adaptor 2610 & 2620 Test Systems Networking Products 229X Central Host Computers GRnet® Baseband Local Area Network It takes experience with a full line of test products to have a full understanding of integrated testing. That’s why our integrated testers of today are designed to fit into the completely integrated testing environment of tomorrow. And that’s no story. The state of the art of integrated testing. Electronics / July 10, 1986 sophisticated approach, but a lot of times it ends up costing more than the standard analog solution." His company uses three $\mu$ codecs to perform digital-to-analog conversion for conferencing and then analog-to-digital to return the combined conversation to the digital PBX system. But DSP Technology executives believe the cost barrier is starting to crumble. Inside Confer, the 32010 processor is a multitasking workhorse, not only combining up to six phone conversations but also playing digitized music for those on hold. The chip also handles the complex overhead of converting logarithmically companded voice bits into a mathematical form better suited to its real-time switching algorithms. The 32010 uses a proprietary algorithm to quickly switch speakers in and out of the conversation. The loudest speaker is switched onto the line at any given time. The chip also automatically adjusts each speaker's volume, compensating for losses in long-distance lines. The 32010 estimates speech energy levels using the recursive filtering algorithm. It samples speech from each line every 4 ms to decide which party is speaking the loudest. The software is held in 2-K 16-bit words of off-chip programmable read-only memory. But connecting conference callers is the easy part, says T. Raj Natarajan, vice president of engineering. The toughest part of software development came in the overhead needed to use inexpensive codecs, he says. DSP Technology opted to use single-chip codecs over other ADCs because of hardware costs. The penalty was additional development time needed to write the code for converting the companded digital output of $\mu$-law pcm codecs into the linear values used by the DSP chip. "That conversion is a big task, but we perform it routinely as part of the processing of all six channels," says Natarajan. Nonconference calls pass through Confer without processing. Calls being connected first pass through an analog interface and then into a bank of six codec chips, which digitize speech. After voices are digitized by the codecs, a serial-to-parallel shift register sends 8-bit blocks of voice data 8,000 times a second for each of up to six channels to the 32010. The DSP, which has a 200-ns instruction cycle time, converts the $\mu$-law voice data to a linear format, switches in loudest speakers in real time without clipping off the first syllable of words, and then places speech back into a companded form so that the codecs can perform digital-to-analog conversion. The combined conference-call signal is then placed back on the analog network. —J. Robert Lineback OPTOELECTRONICS DEVICE PROMISES DENSER OPTICAL DISKS OSAKA, JAPAN If anyone knew how to make them, short-wavelength semiconductor laser sources would be just the ticket for such applications as higher-density optical-disk recording systems and more efficient laser printers for facsimile and typesetting systems. Another approach promises to become practical in the short term: halving the wavelength of the light from near-infrared laser diodes from 0.84 to 0.42 $\mu$m with second-harmonic generators. Matsushita Electric Industrial Co. will offer samples of its second-harmonic generators later this year [Electronics, June 23, 1986, p. 13], heralding the first low-cost, high-efficiency products after more than two decades of development in Japan and elsewhere. Volume production of the chips should bring prices down low enough for use even in such consumer products as Compact Disc players, says Takao Kajiwara, director of the Matsushita Semiconductor Research Center's Optoelectronics Laboratory, Osaka. Early production devices would provide 1-to-3-mW output levels for a 30-mW input, an adequate efficiency level for playback-only disk systems or optical printers, say the company's researchers. Over the longer term, efficiency will be improved, providing output levels of up to 10 mW for the same input power. The current experimental devices were fabricated by a group headed by Matsushita researcher Tetsuo Taniuchi in cooperation with Yoichi Fujii, a professor at Tokyo University. The lithium niobate devices have optical waveguides 2 $\mu$m wide by 0.55 $\mu$m deep that run the chip's full 6-mm length. When an IR beam flows into the waveguide, part of its energy is converted to the second harmonic by the LiNbO$_3$'s nonlinear optical characteristics and radiated at a 12.5° angle from the original beam path. Because attenuation losses should lead to higher efficiency because the input beam's diameter currently exceeds that of the optical waveguide. Efficiency also increases at higher input power; Matsushita has measured values as high as 25% for a 1-W input. Higher efficiency is the result of increased optical nonlinearity at higher input power. The optical-grade LiNbO$_3$ wafers used as a starting material are about 2 mm thick. Photolithography processes selectively etch an aluminum mask with a window slit where the waveguide is to be. The wafer is then immersed in heated benzoic acid for a proton-exchange process that partially replaces lithium atoms inside the waveguide with hydrogen atoms. This step increases the optical waveguide's index of refraction to 2.33 from the 2.2 value of the original LiNbO$_3$. The higher-index waveguide surrounded by the lower-index chip material and air is similar to the core of an optical fiber surrounded by its cladding layer. Devices for second-harmonic generation that are currently available commercially use bulk single crystals of potassium titanate phosphate or potassium dihydrogen phosphate. The price of such devices is high—about $6,000 each. In bulk devices, the power density of the fundamental beam is lower, reducing the magnitude of the material's nonlinear behavior because the beam is not confined within a waveguide. Furthermore, the fundamental and second-har"In ASIC, the customer has to be more than a job number." "What do application specific IC (ASIC) users need? We've been answering that question for 20 years and three basic beliefs hold fast. One, no single ASIC option is optimal for every circuit. So we developed them all — gate arrays, standard cells, and cell-based custom. Two, we can't expect customers to adapt to us. So we assembled the most flexible design interfaces in ASIC. Including gate array and cell libraries for workstations, PC-based systems and most others. Finally, to keep customers ahead, we must be a leader in design technology. Our cell-compilation work is the natural next step. Keeping pace with change is one of the hardest skills to learn. Gould AMI is leading the pack. Our customers have the advantage." To optimize your ASICs, call us at 408-554-2311. Gould Inc., Semiconductor Division, manufacturer of Gould AMI semiconductors. Gould AMI ASIC: Depend on it. High Performance Solutions in Factory Automation, Computers, Instrumentation, Defense and Semiconductors. Circle 159 on reader service card Robert Penn President Gould AMI" monic beams have different phase changes with temperature, necessitating expensive close-tolerance (on the order of 0.1°C) temperature control for phase matching. Even so, efficiency remains orders of magnitude lower than it is for the Matsushita device. Other researchers have used waveguides, but the mode of operation they have used is different and results have been inferior. Researchers at NEC Corp. have developed a device with a sputtered LiNbO₃ waveguide on top of a magnesium oxide crystal, but losses in the waveguide are high. Researchers at the Musashino Electrical Communications Laboratories of NTT Corp. developed a second-harmonic-generation device in which a waveguide is formed by diffusing titanium as a dopant into a LiNbO₃ crystal, but it is susceptible to damage from incident light, and the index of refraction varies with wavelength. —Charles L. Cohen CONSUMER FRENCH SPARK HOME-NET STANDARDIZATION EFFORT PARIS The French electronics industry is taking the initiative in establishing a European standard for tomorrow's domestic networks. The move aims to gain an edge on Japanese and U.S. competitors—Mitsubishi, Sharp, Apple Computer, GE, and Honeywell, among others [Electronics, July 1, 1985, p. 45]—that are expected to attack the European home-network market with products based on their respective Home Automation and Smart House projects. Simavelee, the consumer electronics group of France's Fédération des Industries Electriques et Electroniques last month presented a working prototype of a home network put together with the cooperation of more than 30 native electronics firms, including all principal companies. The French will present the prototype to standards organizations in an effort to gain Europe-wide acceptance for the approach. They will also try to put together a project under the aegis of the European Eureka high-technology research and development initiative to promote further home-networking developments. A great deal is at stake. Though executives hesitate to estimate the size of the market, they generally agree the home network would spur the consumer electronics market out of its doldrums within the next five years by giving consumers a reason to buy goods with more modern features. One beneficiary could be the home computer, which might serve as the network's programming station. France is in a strong position to push standardization. Some eight years ago, the government accepted the Peritel interconnection system as the national standard for linking TVs with external equipment. More than half of the TVs in France are now equipped with a 21-pin Peritel socket, which was made an obligatory accessory on all sets sold since 1980. The connector was made a European standard in 1984. REMOTE CONTROL. The Peritel connection makes it possible to receive and transmit signals in a number of formats: composite video, red-green-blue video, stereo audio, switching voltages, and digital data. To exploit the Peritel, the French have now developed a system called Sidav, from the acronym for dynamic audiovisual interconnection system, which can switch the different signals on the plug to and from external devices. Sidav is a programmable network node controlled remotely by a TV system with a more versatile keyboard than is usually found on a set. Peritel and Sidav form the center of the home-networking proposal. French equipment manufacturers are now in the process of defining links between this Peritel-Sidav audiovisual core and other necessary parts of the home network, such as systems for security and surveillance, domestic robotics, and external communications. Under development are a gateway to two-wire telephone networks and coaxial video networks, interfaces to electrical power wiring and alarm systems, and a filter for the home's electric meter to prevent commands traveling on the electric wiring from going beyond the home. Electronics executives feel they could be available as early as next year. The user will be able to configure his electronic equipment any way he wants. He could, for example, watch TV while switching the program's sound so that he hears it through his stereo system. He could watch one program and record another, or broadcast a recorded program to all the TV sets in the house. The connections with the home's electrical power wiring will create an interactive interface with Sidav so that, using the TV remote control, the user can control lights and appliances. The Simavelec (for Syndicat des Matériels Audio-Visuels Electroniques Grand "Why compromise in ASIC, when you don't have to?" "Don't let anyone tell you an ASIC is an ASIC is an ASIC. If that were true, you'd only need one choice of application specific ICs. But an ASIC solution is more than a gate array. Or a standard cell. Or a cell-based custom circuit. These are simply different approaches to your system solution. Time-to-market, development cost and performance needs make every application unique. So no single ASIC option can work best every time. There's no problem, though, unless your vendor offers just one basic ASIC option. You'll get an ASIC design all right, but odds are it won't be the optimal one. At Gould AMI, you don't have to compromise. We offer the complete continuum of ASIC options. I know some vendors make you do ASIC their way. It's just not our way." To optimize your ASICs, call us at 408-554-2311. Gould Inc., Semiconductor Division, manufacturer of Gould AMI semiconductors. Gould AMI ASIC: Depend on it. High Performance Solutions in Factory Automation, Computers, Instrumentation, Defense and Semiconductors. Bruce Bourbon VP, Marketing Gould AMI Circle 39 on reader service card Public) demonstrated the equivalent of an apartment whose living room and three bedrooms were wired with a cable network and telephone lines and equipped with a Minitel, France's low-cost videotex terminal. A country home was connected in the same way and the two residences were linked to each other and to a private security agency through a cable network. Shown in operation were interconnections among a wide range of audio and video equipment as well as sensors to signal intrusions and failures of equipment, such as the freezer. As an example of remote control, the Simavellec showed how the country home's garden could be watered in response to commands issued at the primary residence. —Robert T. Gallagher IC TESTING A FASTER WAY TO OPEN UP PLASTIC PACKAGES BERN, SWITZERLAND If it meets our stringent demands, it is sure to make it on the outside." So argues a small Swiss electronics producer that has developed, originally for its own use, a fast, cost-saving, and environmentally clean etching method for opening up plastic packages so failures can be analyzed on bare chips. The decapsulation method is the result of a one-year development effort at Gfeller AG, a $65 million communications equipment maker in Bern. It relies on a stream of hot sulfuric acid squirted against the plastic package to etch it away where the circuit is embedded and leave a well-defined hole extending down to the chip's surface. The jet-etch process takes 2 to 3 min. "We wanted a safe and reliable technique for in-house use to open up plastic-encapsulated circuits for failure analysis," says Hans Köppel, head of Gfeller's materials test laboratories. "The results our technique yields have convinced us that we can commercialize it for the open market." Köppel believes the method is superior to other etching schemes used for this job. The Swiss technique has been implemented in an experimental etching system called the Jetty, whose manufacture Gfeller will either handle on its own or license to another company. Gfeller says several semiconductor makers, among them West Germany's Siemens AG, are interested in the method. Quality-control engineers are increasingly turning to this type of fault analysis because electrical tests at the pins often cannot pinpoint the cause of a failure. In such cases, it is necessary to bare the chip so it can be visually examined, viewed through an electron microscope, or tested with microprobes. TOUGH TO OPEN. Unlike metal or ceramic housings, which can easily be removed or cracked apart in mechanical operations, plastic packages are hard to open because of the material's cohesive nature. Chemical etching must therefore be applied, and of the three techniques that are possible—plasma, ion, and jet etching—the latter is by far the fastest and least costly, Köppel says. Fully automatic to provide reproducible results, Jetty ensures that the hot sulfuric acid comes in contact with the chip for the shortest possible time to prevent damage to the bonds and passivation layers. After the etching process, the chip is automatically rinsed to remove traces of the acid. Of particular note, Köppel says, is that the technique is environmentally clean—an important requirement these days, given the sensitivity of governmental authorities, and one that some competing sulfuric-acid-based systems don't meet. The heart of the Jetty system is the etch head, upon which the package sits with its pins up. Protruding through the etch head is a 1-mm tube inside a slightly larger outer tube. All parts of the equipment that come in contact with the acid are made of glass. A thin jet of sulfuric acid, heated by a quartz heating system to 285°C, or just below its boiling point, shoots through the inner tube, hits the plastic material, and etches it. A suction pump draws off the waste acid through the outer tube and cools it to about 40°C. The etching process typically stops after 2 to 3 minutes, when the Jetty senses that the electrically conductive sulfuric acid has made contact with the package pins. GAS CONTROL. A microprocessor controls and monitors the etching, the heating of the acid, the operation of the system's electromagnetic valves, and the level of the acid in its containers. Acid temperature is monitored independently of the other functions to ensure that the acid does not escape as a gas. Before the package is opened, the chip's location in it must be determined. This is best done, Köppel explains, by means of X-ray equipment with fine focusing; ultrasound methods can also be used. Then plastic is milled off the package to reduce the amount that must be etched away. —John Gosch OPENER. To expose a chip and one or both ends of the bonded wires (above) to allow fault analysis with a microscope or microprobes, Gfeller's Jetty acid stream system (right) can etch away a plastic package. "My philosophy on ASIC design? Whatever works for you." "Some application specific IC (ASIC) houses are very rigid. You follow their rules—no questions asked. At Gould AMI, you make the rules. For example, we know learning an unfamiliar CAD system takes time. So we support the workstations and PC systems most designers use: Mentor Graphics, Daisy Systems, P-CAD, Viewlogic and FutureNet, to name a few. Plus our own terrific, low-cost software for PCs—Sceptre II. You can even get translator programs for proprietary systems. And as much design help as you want. Why reinvent the wheel? Use our proven gate array and standard cell macros, and you can quickly design an application specific IC to meet your exact performance, time to market and cost needs. Call me. I'm committed to making your design efforts easy and successful." To optimize your ASICs, call us at 408-554-2311. Gould Inc., Semiconductor Division, manufacturer of Gould AMI semiconductors. Gould AMI ASIC: Depend on it. High Performance Solutions in Factory Automation, Computers, Instrumentation, Defense and Semiconductors. Jerry DaBell Manager, ASIC Design Tools & Methods Gould AMI Circle 41 on reader service card Telecommunications companies the world over will have to rethink their strategies now that ITT Corp. and Compagnie Générale d'Electricité, the nationalized French switch maker, have joined forces to create the world's second-largest telecom-equipment firm. In addition to receiving a $1.8 billion cash payment, ITT will maintain a 30% share in the company. The remaining 70% will belong to a European holding company in which CGE will have a controlling interest. CGE will likely own a full 50% of the new firm; the rest will be held by a consortium that includes the banking group Société Générale de Belgique and Telefónica SA, the Spanish telephone company. With activities in 75 countries and annual sales approaching $9.6 billion, the new company will be second only to AT&T Co. in the world telecom-gear market. Philips of the Netherlands and West Germany's Siemens AG, which have already teamed up to develop megabit memories, are hammering out details of an agreement to cooperate in telecommunications chips. Sources close to the two companies expect a second-source deal to be announced later this month. For starters, the Dutch company will produce two of Siemens's chips for integrated services digital networks, the 2070 communications controller and the 2080 interface circuit [Electronics, Sept. 30, 1985, p. 46]. For its part, the Munich company is likely to second-source certain Philips chips for use in private branch exchanges. The deal is part of a strategy by the two companies to thwart U.S. and Japanese producers that are trying to gain a foothold on the Continent's telecom chip market. Victor Co. of Japan (JVC) has signed on a heavyweight ally for its campaign to snare market share in the compact video cassette camera/recorder market from Sony Corp., the pioneer with its 8-mm camcorder. The marketing minds of Minolta Camera Co., Osaka, have followed JVC's lead and become the second Japanese company to produce VHS-C camcorders, which JVC developed to compete head on with 8-mm camcorders. Minolta is buying the mechanical recording and playback deck assembly from JVC on an original-equipment-manufacturer basis and manufacturing the electronics and other parts itself. Major Japanese consumer electronics houses—among them Hitachi, Matsushita, Mitsubishi, Sharp, and Toshiba—plan to sell private-label VHS-C camcorders bought from JVC starting this summer. Both Hitachi and Matsushita intend to go into production themselves next spring, and the others presumably will follow suit if the market takes off. Meanwhile, Minolta plans to make 2,000 units a month, mostly for export to the U.S. Whatever the outcome of the camcorder campaigns, Hitachi Ltd. figures to end up with the winners. Already set to sell JVC's VHS-C gear under its own label, the Tokyo company will start manufacturing 8-mm camcorders in September, but only for private-label products that Japanese camera makers such as Minolta Camera Co. will export. Production will be 5,000 units per month. Hitachi's move follows that of Matsushita Electric Industrial Co., which supplies Eastman Kodak Co. and Olympus Optical Co. with 8-mm camcorders on a private-label basis for the American market. Although Hitachi and Matsushita so far are marketing only VHS-C camcorders under their own names, they are positioning themselves to jump into the 8-mm market, too, if Sony's future success forces them to do so. First, clear your mind of all thoughts of using a plotter. There's a faster, more colorful, more controllable way to get color hardcopy. The new Seiko Hardcopier. It gives you solids like you've never seen before. In shades you've only thought about. Because this hardcopier lets you work with 4,912 colors. And adjust the individual intensity of each one. Without relying on your host to do the work. Plus you can use the copier's own intelligence. Doing color assignments, dithering techniques, scaling, rotation, aspect ratio changes, printing density adjustments and special effects with easy front panel control entries. Then you'll see your A or B size copies (and A size transparencies) in as little as 45 seconds. Interfacing is easy, too. You can work with either video (no need to write software) or parallel interfaces. And get copies the same morning your hardcopier is delivered. One more thing. Your hardcopier can be a shared resource with Seiko's new multiplexor. Because it lets you connect up to four different signal sources at the same time. So make a note. Look into the Seiko Hardcopier today. You'll see the fastest, most colorful way ever to get your pen on paper. Call Martin Nelson at (408) 943-9100 today. THE CURSE OF A lot of people choose Pyramid computers because they're fast and inexpensive. There's just one little catch. When your needs grow, you may wind up buried in your Pyramid. THE PRACTICAL PARALLEL® GROWS WITH YOU, NOT AROUND YOU. Sequent's family of parallel computers will keep you from getting buried no matter how much your needs grow. It lets you go instantly from 4 all the way up to 30 processors in a single system. From 3 to 21 MIPS. The best any Pyramid system can do is 5.5 MIPS. And the Practical Parallel has a lot better price/performance than any Pyramid. Naturally, the Practical Parallel runs UNIX® 4.2 bsd and System V concurrently. And offers Ethernet™ with both TCP/IP and NFS™. So it runs a ton of software. In seven languages. And because it uses parallel technology it's ideal for relational databases like Informix®, Ingres™ and Unify™. BUILT FOR THE AGES. The Practical Parallel has a proven MTBF of over 11,000 hours. That's more than a year without a failure. And thanks to our parallel architecture and automatic reconfiguration, even if your system does fail, you should be able to bring it right back up again. Okay, we know all this sounds almost too good to believe. But AT&T Bell Labs and Tektronix and Purdue University and dozens of other major companies already have Sequent systems up and running at over 80 sites around the country. For your copy of our 1986 multi-user benchmark report, call 800/854-0428. Or write to us at 15450 S.W. Koll Pkwy., Beaverton, Oregon 97006-6063. And stop slaving away over your Pyramids. FROM THE FIRST FAMILY OF GATE ARRAYS... A PROUD, NEW ADDITION I can remember when it all began at Hughes in 1979. With a single metal HCMOS 1000-gate array. Since then the Hughes family of military gate arrays has grown. In 1985 the 2 micron ($l_{eff} = 1.2$ microns), double metal U-series of channelless gate arrays was born, descended from the broadest, high performance line of HCMOS arrays in the semiconductor industry. And now, our latest arrival, the alternate sourcing of LSI Logic Corporation's entire military CMOS line: L5000, L7000 and L10,000 Series Meet the members of our family. We have one for every military application. And that's just for starters. Hughes channelless gate arrays open the door to full-custom ASICs, with unique, monolithic CMOS combinations of EEPROM, analog and digital. H-Series Array Family Double metal HCMOS, $l_{eff} = 1.8$ microns, $t_{ox} = 500$ A, clock frequency to 35 MHz. Five array sizes ranging from 1000 to 8000 equivalent gates. V-Series Array Family Double metal HCMOS, $l_{eff} = 1.2$ microns, $t_{ox} = 300$ A, clock frequency to 50 MHz. Five array types ranging from 1000 to 8000 equivalent gates. U-Series Array Family Double metal HCMOS II, $l_{eff} = 1.2$ microns, $t_{ox} = 300$ A, clock frequency to 100 MHz. Ten channelless array types with 1000 to 41,000 gate complexities. Pin counts to 256. Each output drive capability of 7 TTL loads. *L5000-Series Array Family Five double metal HCMOS types. Complexity from 880 to 5902 equivalent 2 input gates. L7000-Series* Array Family Eight double metal 2.0 micron HCMOS types ranging from 880 to 10,013 equivalent 2 input gates. L10,000-Series* Array Family** A future channelless HCMOS III array family with 1.5 micron design rules and 129,000 equivalent gates. We're always growing at Hughes. The First Family of gate arrays gets bigger and better with each new generation. Call or write us today with your application needs. Hughes Semiconductor Division, 500 Superior Ave., Box H, Newport Beach, CA 92658-8903, (714) 759-2727. The L5000, L7000 and L10,000 array families are products licensed to the Hughes Aircraft Corporation by the LSI Logic Corporation for sale to military and government contractors. A SLICE OF THE FUTURE HUGHES SEMICONDUCTOR HUGHES AIRCRAFT COMPANY SEMICONDUCTOR DIVISION Industrial Electronics Group Joe Angleton, assistant laboratory manager for VLSI design at the Missiles System Group of Hughes Aircraft. Mr. Angleton holds several patents in the field of semiconductor circuit design and is a primary originator of the Hughes gate array development efforts. Circle 47 on reader service card Circuit-Board-Artwork Software: $895. And guaranteed. smARTWORK® lets the design engineer create and revise printed-circuit-board artwork on the IBM Personal Computer. You keep complete control over your circuit-board artwork — from start to finish. And smARTWORK® is reliable. When we couldn’t find a package that was convenient, fast, and affordable, we created smARTWORK® to help design our own microcomputer hardware. We’ve used it for over two years, so we know it does the job. That’s why we offer every design engineer a thirty-day money-back no-nonsense guarantee. smARTWORK® advantages: - Complete interactive control over placement and routing - Quick correction and revision - Production-quality 2X artwork from a pen-and-ink plotter Prototype-quality 2X artwork from a dot-matrix printer - Easy to learn and operate, yet capable of sophisticated layouts - Single-sided and double-sided printed circuit boards up to 10 x 16 inches - Multicolor or black-and-white display The Smart Buy At $895, smARTWORK® is proven, convenient, fast — and guaranteed. Call us today. And put smARTWORK® to work for yourself next week. Try it for 30 days at absolutely no risk. That’s smart work. System Requirements: - IBM Personal Computer, XT, or AT with 320K RAM, 2 disk drives, and DOS Version 2.0 or later - IBM Color/Graphics Adapter with RGB color or black-and-white monitor - IBM Graphics Printer or Epson FX/MX/RX series dot-matrix printer - Houston Instrument DMP-41 pen-and-ink plotter - Microsoft Mouse (optional) Wintek Corporation 1801 South Street Lafayette, IN 47904-2993 Telephone: (317) 742-8428 Telex: 70-9079 WINTEK CORP UD In Europe contact: RIVA Terminals Limited, Woking, Surrey GU24 5JV ENGLAND, Telephone: 04862-71001, Telex: 859502 "smARTWORK", "Wintek" and the Wintek logo are registered trademarks of Wintek Corporation. At last, the superchips are solidly on their way. And they bring with them vast improvements in computing power. TRW Inc. has already fabricated process-test versions of the ultrahigh-speed submicron devices. Work on the next step, chips to test the twin concepts of self-diagnosis and software also accrue—superchips dissipate only 0.25 W per square inch, against 50 W/in.$^2$ for VHSIC Phase 1 packages, according to Zimmerman. The resulting superchip packages are the same as those for Phase 1, only larger (Fig. 3). Still awaiting approval by the DOD are three other TRW superchips: a bipolar fast Fourier transform circuit, a 32-bit CMOS data processor, and a 1,024-word-wide associative processor, also fabricated using the company's CMOS process. The most impressive of them is the associative processor, which incorporates a parallel architecture. Processors of this type typically incorporate a parallel architecture that consists of a single-instruction, multiple-data-stream array of processing elements and a content-addressable associative memory. The TRW associative processor can execute a sequence of instructions on stored data in parallel at throughput rates of up to 10 billion operations/s and is aimed at artificial-intelligence applications involving complex pattern matching and analysis. An associative processor executes a sequence of pattern-matching operations on all data within its content-addressable-memory array. Whenever the matching succeeds, some manipulation is performed on those data words found to match. The von Neumann-type architecture used in most general-purpose microprocessor CPUs executes a sequence of arithmetic and logical operations, each containing the memory addresses where the operands are assumed to be stored. The fast Fourier transform chip is a systolic array that can perform a 1,024-point operation in 41 $\mu$s. At the system level, this translates into a throughput of about 1 billion complex operations/s. The chip also provides premultiply and presummation windowing with on-chip trigonometric recombination. The data processor, says Zimmerman, is similar to a reduced-instruction-set computer and has an architecture and instruction set optimized for implementing floating-point functions at data rates up to 20 million instructions/s. It also contains a 1-Mb SRAM with an access time of 20 ns. TECHNOLOGY TO WATCH* is a regular feature of Electronics that provides readers with exclusive, in-depth reports on important technical innovations from companies around the world. It covers significant technology, processes, and developments incorporated in major new products. HOW TWO AT TRW ARE TACKLING A ONCE-IN-A-LIFETIME PROJECT Both Fred L. Alexander and Thomas A. Zimmerman are keenly aware that the superchip VHSIC Phase 2 project they direct at TRW Inc. is that once-in-a-lifetime opportunity engineers dream about. First of all, the fault-tolerant redundancy feature of the advanced chips involves refinements designed to make concepts first advanced in wafer-scale integration work—a hitherto unaccomplished feat. In addition, the 1.4-by-1.4-in. chip size, larger than anything else in sight, goes against the technology grain of ever-smaller devices. Enthusiasm for superchips continues to build throughout the large technology team effort orchestrated by the two TRW veterans, they report. It began when military-electronics project leaders were called in to write the formal proposal for the Defense Department in 1984, recalls Zimmerman, who manages the company's Very High Speed Integrated Circuits program projects. A widespread feeling of taking part in a seminal development effort indeed exists among engineers and scientists involved with superchip, adds Alexander, who is vice president and general manager of the Electronics & Technology Division. "It will be a tremendous breakthrough, if we can pull it off," he notes. Though no serious pitfalls now loom, the concept will be proven absolutely only when the devices are built during 1987-88. Alexander and Zimmerman, both midwestern natives, have been immersed in VHSIC since it began more than five years ago. Alexander, who joined TRW in 1963, managed VHSIC programs before promotion to his present post. He holds BSEE and MSEE degrees from the University of Missouri. Zimmerman was a prime mover in formulating TRW's VHSIC strategy and previously managed the Charge-Coupled Device LSI Products Department. He holds a PhD in electrical engineering from Purdue University. He joined TRW in 1971. MORE BANG. LESS BUCK. Why weapons systems designers are using Rantec standard military power supplies. Rantec's standard military power supplies save weapons systems programs money. No costly non-recurring engineering expenses. No costly retooling. And, with over 600 different models to choose from, no costly design compromises on your part. Also, no costly time delays waiting for product to develop or for manufacture. Which is very important when you consider it takes from 13 to 18 months and between $125,000 and $300,000 to get a custom military power supply.* Giving up reliability by going standard? No way. First of all, our quality program complies with MIL-I-45208 and meets the intent of MIL-Q-9858. Secondly, all of our power supplies meet or exceed MIL-E-5400, MIL-E-16400, MIL-STD-704, MIL-STD-1399, MIL-I-6181, and MIL-STD-461. This extensive product line includes single and three phase AC input power supplies and DC-DC converters in single and multiple output power ratings to 1KW. To find out how you can save money like your contemporaries by using Rantec standard power supplies which are qualified for airborne, shipboard, ground-fixed and ground-mobile use, contact us today. U.S. Navy/Industry Power Supply Ad Hoc Committee Estimates for Time and Cost. Arium. What microprocessor development systems? Ask a design engineer to draw up a "wish list" for his dream microprocessor development system, and you'll probably get something like this: Give me a preview window, so I can see not just where my data train has been, but where it's going. Give me a system with a debugging turnaround time of 15 minutes (end of debug session until start of emulation). I like taking coffee breaks, but not for two hours. Give me a system that lets me work in high-level language or assembly language. I hate manuals. Get me a system with drop-in menus. In English. Give me something I don't have to share. A standalone transportable unit that lets me work, not wait. Get me a system with on-screen time stamping so I know how long each routine takes. And please, please... give me a system with power. Say 20MB disk memory, super-high speed compilers and powerful editors. Introducing Arium Echo. The dream development system at a dream price. Arium's new Echo development system answers yes to all of the above. Tek™ has nothing that can match it. Neither does Intel. Echo delivers all the critical features of the HP 64000. And it does it at a fraction of the cost. Echo arms you with more power than you've ever known. It gives you Fast Break™ triggering, so while you're still executing code you can break on a register value. Another Echo exclusive is Auto-Patching™. When a patch file aided edit is completed, hit one key and presto! Echo recompiles the whole program, relinks it, relocates and starts emulation. Tek is a registered trademark of Tektronix. Unix is a registered trademark of AT&T Bell Laboratories. HP is a registered trademark of Hewlett Packard. should have been from the beginning. Echo includes a true multitasking UNIX™-like operating system. C and Pascal compilers are built in. And Echo is universal. It will handle virtually every major 8- and 16-bit microprocessor. Now for the good news: price. Best of all, you can put Echo on-line for less than $9,800. That's one fourth the cost of an HP system. Echo is a one-engineer system. No sharing, no waiting, no lost man-hours. And Echo comes from America's fastest-growing digital instrument manufacturer. In just eight years Arium has established itself as a leader in logic analyzers (Number Three behind Tek and HP) and a formidable innovator. Phone the Arium Hot Line. 800-862-7486. As you can imagine, we can't fit all the good news about Echo into one small ad. We'll be glad to send our full information pack, answer questions by phone or set up an appointment to give you a demonstration. Give us a call and learn more about the new Arium Echo. It's what microprocessor development systems should have been from the beginning. ARIUM CORPORATION 1931 Wright Circle Anaheim, CA 92806 (714) 978-9531 Intel is a registered trademark of Intel. BREAKING THE SPEED BARRIER ON THE VMEBUS A new bus interface uses packets and a delay-line asynchronous state machine to send data at more than 30 megabytes/s—near the top speed of Multibus II. Interphase Corp. has good news for system builders using or considering the VMEbus. The Dallas company has developed the BUSpacket, a VMEbus interface that more than triples the bus's throughput to more than 30 megabytes/s. That puts the VMEbus neck and neck with rival Multibus II, which boasts a speed of 32 megabytes/s. The BUSpacket sends preformatted data packets across the VMEbus as fast as the bus can take them. The packet approach keeps the bus-transfer rate from dropping to the speed of the slowest device on a board interfaced to the VMEbus. Interphase, a maker of high-performance peripheral controllers, added the BUSpacket to its existing Storage Module Drive disk controller (Fig. 1) and hiked its throughput from 5 to more than 30 megabytes/s. That performance sets a new speed level for VME peripherals. The VMEbus's theoretical throughput under ideal conditions is about 40 megabytes/s. Most VMEbus interfaces run at 5 or 10 megabytes/s, however, because they transmit data in streams. If the random-access memory used to buffer data is much slower than the maximum bus-transfer rate, for example, it will lower the bus-transfer rate to its own speed. "It is criminal to use the bus at 10 megabytes/s when it is rated at 40," says Michael E. Cope, Interphase president and founder. System designers and users seem to agree. They are starting to demand performance as high as 30 megabytes/s on system buses to match new peripherals, such as high-performance disk drives, that are coming along. These peripherals cannot reach their full potential if bus throughput stops at 5 or 10 megabytes/s, as is the current state of the art in VMEbus-based systems. In such systems, the disk drives often use as much as 35% of the bus time. Therefore, for good system efficiency—or balance—the disk subsystem must make effective use of the bus when it has it. A subsystem that wastes bus time is a drag on a system. The BUSpacket allows slower devices to work off its first-in first-out bus-isolation buffer at their speeds while allowing the state machine and the VMEbus to work off the other side of the FIFO at their maximum speeds. Thus BUSpacket adds the message-passing features of Multibus... II to the best aspects of the asynchronous VMEbus. Message-passing enhances the performance of multidevice systems by decoupling the operations of the devices from each other, from on-board buses, and from the system bus. The BUSpacket's two keys to faster bus traffic are the FIFO buffer and an asynchronous state machine (Fig. 2). In the upgraded version of its controller board, Interphase also increased the size of the multiport RAM buffer from 16- to 128-K bytes to add flexibility and efficiency to data buffering between high-speed peripherals and the system bus. The 256-word FIFO buffer builds up the packets before they burst out on the bus at high speed. Based on a delay line, the asynchronous state machine controls the flow between the FIFO and the bus. It actually controls the active VMEbus signaling, quite unlike its synchronous counterparts in older VMEbus schemes. Synchronous state machines are triggered by a clock signal and must march in step with the bus signaling without exerting any control. The BUSpacket interface achieves its high rates by formatting data packets and storing them in the FIFO before acquiring the bus. Then it acquires the bus, the FIFO empties at more than 30 megabytes/s, and the bus is released for other uses. Therefore, the bus activity is entirely decoupled from the other functions on the board. Without decoupling, achieving the VME specification's maximum burst-transfer rate would take RAMs with access times in the tens of nanoseconds. But wringing such performance from the current generation of RAMs is too costly in terms of hardware and board real estate. So slower parts are being used, though throughput suffers because of the coupling between on-board functions and the VMEbus. The BUSpacket architecture decouples the controller board functions from the bus. This means that the operation of the components on the local buses of the controller card are not tied to the operation of the system bus. Without this independence or decoupling, the slowest function would set the maximum speed. Decoupling allows fast devices to operate at full speed; for example, the interface to the VMEbus can run at the fastest speed that data can be burst onto the bus, even if the buffer RAM is much slower. The BUSpacket's FIFO allows the use of slower RAMs and other components on the peripheral controller and interface board without affecting bus speeds. It also frees up the processor time that would be taken by transferring data directly to and from the VMEbus, thus netting more available computing power for the functions on the board such as the peripheral interface and virtual-buffering control. The decoupling FIFO is made from 40-ns word-wide registers. The words are 2 bytes wide, so the FIFO can put out 2 bytes every 40 ns. But because the bus interface logic introduces delays, the transfer speed drops to between 30 and 33 megabytes/s. A FAST STATE MACHINE The asynchronous state machine in the BUSpacket architecture transfers data to and from the VMEbus considerably faster than the synchronous state machine in most architectures, including other VMEbus implementations. Traditional synchronous state machines run off the on-board clock or, alternatively, the bus clock. Running off a clock requires that all devices on the bus conform to the rated clock speed before accessing the bus. Instead of running off a clock, the BUSpacket state machine runs off a tapped delay line and ![Diagram](image) 2. NEW PARTS. A first-in first-out buffer for assembling packets plus a delay-line state machine dramatically upgrade the performance of this SMD disk controller. bus signal transmissions. Metastable states common to asynchronous signaling are avoided entirely. At the same time, the delay-line state machine is not loaded with the wasted clock cycles that are tied to an arbitrary clock rate. The delay-line approach takes better advantage of some VMEbus characteristics. For example, VME is an asynchronous edge-driven bus, so the tapped delay line can sample and set all edges at any time. This gives the state machine control over acquiring the bus and executing burst transfers. Clock-driven synchronous state machines have finite times—the clock pulses—in which to sample the state. Then the machine has to wait for the next state to start transferring data. This procedure slows down the state machine, making it unable to transfer data as fast as the bus can accept and deliver it. The asynchronous state machine does not incur this penalty. BIGGER BUFFER RAM When an event such as the FIFO packet transmission occurs, the new state machine does not have to wait. It can synchronize with a clock so it can act immediately to sample and set the state of the bus and start the data moving. In other words, the delay-line state machine accommodates to the ever-changing bus timing without being constrained by fixed sample points. To further push performance of the disk controller and to make room for enhancing its functions, Interphase's engineers increased the size of the multiport RAM buffer eightfold—from 16-k bytes to 128-k bytes. This makes available a larger pool of buffers for the controller's virtual buffering scheme as well as creating room for more elaborate caching schemes that can further improve performance. Users of this product and others soon to follow will have no trouble seeing the current improvement in bus throughput, Cope says. "The FIFO and the new state machine make all the difference in the world. Users will stand up and notice because a 6:1 improvement in throughput is immediately noticeable on systems." Interphase's designers did not have to use faster RAM chips with this new architecture. In the old controller design, the speed of the RAM was the limiting factor in holding the bus transfer speed down to 5 megabytes/s. In fact, the RAM chips used in Interphase's enhanced disk controller, the V/SMD 4200 Cheetah, are slightly slower than those used previously but have no effect on performance. The first product to use the new interface is a speedier version of the company's Storage Module Drive disk controller. The V/SMD 3200, which runs at 5 megabytes/s, was the first high-performance 32-bit Storage Module Drive disk controller for the VMEbus and became the prominent one in the VMEbus marketplace, the company says. The high-speed V/SMD 4200 Cheetah continues to offer such proven features of the original V/SMD 3200 as the multitasking virtual buffer architecture, intelligent caching optimized for AT&T Co.'s Unix operating system, and zero latency reading and writing. All this and high speed, too, will be available for $2,350 in 100-piece quantities. --- HOW A SMALL COMPANY'S IDEA TURNED INTO A BREAKTHROUGH When Tom Thawley identified what could be a breakthrough in VMEbus technology, excitement raced through Interphase Corp. Thawley, who cofounded the Dallas company with Michael E. Cope in 1977, has the commonplace title of executive vice president but the rather unusual role of Interphase's floating creative engineering executive. Freed from everyday project demands, he focuses on the big picture in developing tomorrow's technologies. Determined to gather elements of a potential new design that would significantly increase VMEbus throughput, Thawley locked himself in a room this past January to be isolated from distractions. Just days later, he emerged with some ideas. Thawley, who holds a BSEE from the University of Maryland in College Park, showed his findings to Ed Gross, Interphase's peripherals engineering manager. Gross instantly recognized the potential of Thawley's suggested architecture. "It wasn't hard to see that this would create a new performance level for the VMEbus," says Gross, who has been with the company for two years and holds a BSEE from Carnegie-Mellon University. Cope, as president of Interphase, is excited about the overall potential of the new interface. Cope, who holds a BSEE from Vanderbilt University, says, "this really sets the state of the art where it needs to be to realize the long-range potential of the VMEbus. And it is vitally important to protecting VME markets from competing buses in the future." In fact, the BUSpacket interface came in part from Cope's urging that one of the primary concepts used in the Multibus II message-passing architecture be applied to the VMEbus. Thawley figured out just how to do it and the result is now ready to go to market. Design engineer Craig Cantrell is responsible for much of Interphase's bringing the product to market ahead of schedule. Cantrell, who received his BSEE from the University of Texas at Austin, has been with the company for more than two years. The Video Telephone: ways to create time and money. The new Photophone™ sends clear, still video pictures over normal dial-up telephone lines, letting your experts "visit" remote sites without the down time and expense of travel. 1. Manufacturing Let your best engineers see and repair machinery at multiple plants. Install equipment with visual assistance from the manufacturer. Monitor construction without sending an architect to the site. Visually check tooling and parts before ordering full-scale production. Fortune 500 companies are already proving Photophone's™ value. 2. Engineering Share pictures of drawings, schematics and models. Visually link dispersed groups working on subsections of a design. Check a subcontractor's quality and compliance with your drawings. Make design changes and quickly inform people. 3. Project Management Place video conference calls to design schedules, check progress, review budgets and set goals. Illustrate your point using objects, charts, computer spreadsheets and graphics. 4. Training Let your best instructors teach from their home base. Deliver training to multiple points at the same time, without traveling. Interactively define course content with customers and get immediate approval. Photophones are already creating time and saving money for companies around the world. Think of what they can do for your business, for only $8550 ... plus the cost of a phone call. The Photophone™ from Image Data Corporation. 7986 Mainland Drive • San Antonio, Texas 78250 • (512) 680-2727 Circle 61 on reader service card Two-piece connections for every itinerary. Get on board now with the world’s biggest selection. Name your application—our two-piece connector schedule covers every stop on the map. Need maximum reliability and flexibility in a medium-to-high pin-count application? AMP Box Contact Connectors and high-density AMP-HDI Connectors feature four-way contact on every pin. Very reliable. Very forgiving of pin angle during mating and unmating. And both also offer power and coax contacts—big design help in those crowded little corners. Headed for design-wide compatibility? AMPMODU two-piece connectors are part of a complete, cost-saving, modular system, featuring shortened signal paths for high-speed designs. Or, go Eurocard. The European standard for over 10 years, now used everywhere. And now available everywhere—from AMP. Whatever your destination, AMP has the two-piece connectors you need, engineered for quality, reliability, and—especially with our compliant-pin option—increased productivity. At left: AMP-HDI Connectors • To 684 positions on standard .100" grid • Selective gold plating • Press-fit ACTION PIN contacts available • 10A and 30A power contacts • Miniature coax contacts Box Contact Connectors • .100", .075", .050" center lines • MIL-C-55302 versions available • Power and miniature coax contacts Eurocard Connectors • Compatible with DIN 41612 connectors • Selectively gold-plated press-fit contacts • Standard and inverse mating AMPMODU Connectors • Standard .100" grid, 12-200 positions, horizontal or right-angle versions • Duplex or selective gold plating • Press-fit ACTION PIN contacts available Call (717) 780-4400 and ask for the AMP Two-Piece Connector Desk. AMP Incorporated, Harrisburg, PA 17105. AMP, AMPMODU, AMP HDI and ACTION PIN are trademarks of AMP Incorporated. How long would it take you and your staff to solve this problem? Problem: solve the differential equation: \[ \frac{dY}{dT} + Y^2 + (2T + 1)Y + T^2 + T + 1 = 0 \] subject to the condition that $Y(1) = 1$. The first complete AI program for applied mathematics. MACSYMA™ is the world’s first commercially available software capable of handling most applied mathematical problems. Originally developed at MIT through research in Artificial Intelligence, MACSYMA has been continuously developed and enhanced since 1969. Now you can put MACSYMA to work on any of a growing list of workstations and mainframes. MACSYMA saves you time and money. MACSYMA saves you time. More time than you can count. Which adds up to more dollars than you can imagine. Without previous programming experience you can interact with it in an almost conversational way while you use it to explore scientific problems. Problems that you couldn’t even begin to approach using just a pencil, chalk or ordinary numerical software. Hundreds and hundreds of real-world applications. MACSYMA is a total mathematical programming environment that gives you problem-solving power of virtually unlimited proportions. Here are just a few of the applications it can perform. For you. - Do algebra, trigonometry and calculus - Provide standard numerical techniques - Generate FORTRAN output from MACSYMA expressions - Simplify, factor or expand expressions - Solve equations analytically or numerically - Expand functions in Taylor or Laurent series - Compute Fourier and Laplace transforms - Manipulate matrices and tensors - Plot functions in 2 and 3 dimensions - And many, many more Throughout the world, thousands of scientists, engineers and mathematicians are using MACSYMA in such diversified applications as electronics, structural engineering, CAD, acoustics, plasma physics, antenna theory, VLSI circuit design, control theory, numerical analysis, fluid mechanics, genetic studies, ship hull design, maximum likelihood estimation, ballistic missile defense systems design, underwater shock wave analysis, helicopter blade motion studies, and atomic scattering cross section analysis. And the range of use is expanding every day. MACSYMA gives you the answers. In symbolic form. (C1) DEPENDS(Y,T)$ (D1) [Y(T)] (C2) DIFF(Y,T)+Y^2+(2T+1)Y+T^2+T+1; (D2) dy/dT + Y^2 + (2T + 1)Y + T^2 + T + 1 (C3) SOLN:ODE(%,Y,T): (D3) Y = - %C T%E - T - 1 (C4) SUBST(T = 1,Y = 1), %); (D4) 1 = - %E %C - 2 (C5) SOLVE(%,%C),NUMER; (D5) [%C = 0.5518192] (C6) SPECIFIC_SOLN:SUBST(%, SOLN); (D6) Y = - 0.5518192 T%E T - 1 0.5518192 %E T - 1 In FORTRAN code. (C7) FORTRAN(%)$ Y = - (0.5518192TEXP(T) - T - 1) 1 / (0.5518192EXP(T) - 1) In graphics. Put MACSYMA’s Artificial Intelligence to work for you. If it is still taking you weeks or months to handle rudimentary algebraic problems and extremely complex analyses or to generate FORTRAN code out of symbolic expressions, then make it your business to find out about MACSYMA mathematical software. It is available right now from Symbolics, Inc., the leader in symbolic processing. For a complete kit including a full capabilities brochure, just clip and mail the coupon or call toll-free. 1-800-MACSYMA In Mass., Alaska & Hawaii only call (617) 577-7500. MACSYMA™ The most comprehensive software approach to symbolic mathematical computations. symbolics™ Your next step in computing.™ MACSYMA and SYMBOLICS are trademarks of Symbolics, Inc. Copyright 1986 Symbolics, Inc. Symbols, Inc. Dept. M-EL2 11 Cambridge Center Cambridge, MA 02142 Name ____________________________________________ Title _____________________________________________ Company __________________________________________ Address ___________________________________________ City State Zip ______________________________________ Telephone Extension ________________________________ MACSYMA is available to colleges and universities at special rates. Circle 65 on reader service card THE BEST. ..... in automation LeCroy 9400 DUAL 125 MHz DIGITAL OSCILLOSCOPE ACQUISITION PARAMETERS \| VERTICAL \| Chan 1 \| Chan 2 \| \|----------\|--------\|--------\| \| Fixed V/div \| 8 mV x10 \| 1 mVx100 \| \| Total V/div \| 80 mV \| 11.8 V \| \| Offset \| -2.0 mV \| -342 mV \| \| Coupling \| AC-DC \| AC-DC \| TRIGGER \| Time div \| 5 μs \| \| Delay \| 90.00 Pres \| \| Level \| DC V \| \| Coupling \| DC \| \| Source \| EXT \| \| Slope \| + \| \| Mode \| Normal \| Trigger Level has absolute meaning with DC-Coupling only DISPLAY CONTROLS SELECT TEST SIGNAL POSITION INTERFERENCE CURSOR POSITIONS TIME/DIVISION VERTICAL CHANNEL 1 CHANNEL 2 TRIGGER POWER 125 MHz BW; 100 MS/s ADCs; 5 GS/s Interleaved Sampling; 128 k Waveform Memory; ± 1% Accuracy; Summation and Continuous Averaging; Arithmetic Processing; Fully Programmable. PROGRAMMABILITY. All settings of the 9400 Digital Oscilloscope are programmable; you can operate the instrument in unattended areas or as a computer-aided testing station. You can redefine the softkeys and send messages to the display, in order to step inexperienced operators through complex test procedures. And even without a computer you can store and quickly reproduce up to eight front-panel setups, for your most common measurements. SIGNAL CAPTURING. Long 32 k memories and versatile programmable triggering makes capturing any signal sure and easy. Pre-trigger sampling over 32,000 samples captures not only the triggering signal but also its cause. Post-trigger delay is programmable up to 10,000 divisions. INTERFACING. Unequalled computer interface capability is provided by one GPIB (IEEE-488) and two RS-232C ports. Simple English commands send any portion or all of data record to a computer at speeds as high as 400 kbyte/sec. ARCHIVING. On-board firmware makes archiving of measurement results by a digital plotter fast and easy. The entire display may be dumped in any size to a wide range of high-resolution multicolor plotters — ready for publication or immediate presentation. ★And there is much more to say about this versatile and cost effective ($9900 base) DSO. Call us now...for details and a demonstration!! USA price list only. Top: Acquisition Parameters listing enables the 9400 user to precisely set and check front panel settings, all of which can be remotely controlled. Middle: Window mode trigger set at ± 3.5 divs from center grid captures switching transient. 50% pre-trigger shows contact bounce prior to trigger moment. Below: Crosshair marker, acting as a precise timer and DVM, gives time from trigger (arrow) and absolute voltage. LeCroy 700 S. Main St., Spring Valley NY 10977, (914) 578-6038; Geneva, Switzerland, (022) 823355; Heidelberg, West Germany, (06221) 49162; Les Ulis, France, (1) 6907-3897; Rome, Italy, (06) 320-0646; Botley, Oxford, England, (0865) 727275. Representatives throughout the world. HOW MOTOROLA MOVED BIMOS UP TO VLSI LEVELS VLSI mixed bipolar and MOS technology is about to go commercial in the U.S. in a big way. Motorola Inc. is entering the market with a bipolar-CMOS very large-scale integrated circuit: the MCA6000ETL, a 6,000-gate array featuring an average gate-level power dissipation lower than that found in an equivalent bipolar or CMOS structure. For a two-input NAND gate with a load of 2 pF, for example, a BiMOS structure dissipates 50 μW/MHz versus 70 μW/MHz in a CMOS structure. What makes the VLSI MCA6000ETL possible is a careful working out of the conflicting process requirements of bipolar and CMOS technology. The MCA6000ETL is only the beginning, says Robert Jenkins, Motorola's vice president for technology development at the Semiconductor Products Sector, Phoenix, Ariz. The company plans to introduce a whole range of BiMOS macrocell gate-array products. It is also looking to apply the technology to standard cells, static random-access memories, and custom circuits in telecommunications, graphics, and tape and disk-drive circuits, as well as in consumer applications. While there has been increasing interest in fabricating LSI and VLSI digital logic circuits by applying mixed bipolar and MOS processes, most commercial applications have until recently been the exclusive domain of Japanese companies. But now Motorola is challenging the Japanese. Motorola's proprietary process (Fig. 1) merges a 2-μm n-well CMOS structure with a p-type epitaxial bipolar transistor featuring walled polysilicon emitters, says Patrick Hickman, bipolar semicustom design manager at the company's application-specific IC division in Mesa, Ariz. The MOS device incorporates silicided gate, source, and drain structures for reduced contact and drain-source resistance. The entire process, including two layers of metal interconnection, requires 14 mask steps, three of them for the bipolar devices. The chip architecture (Fig. 2) consists of an internal array of 3,072 basic macrocells containing both CMOS and bipolar devices, 1,400 dedicated wiring channels, and 202 input/output cells implemented in bipolar to pro- 1. BIPOLAR-CMOS. Motorola's 2-μm process with n-wells and silicide gates, drains, and sources merges CMOS and bipolar. provide TTL or emitter-coupled-logic compatibility or both, as well as a high off-chip drive capability. Although the introduction of the BiMOS gate array represents the company's first attempt to use this technology in the digital VLSI arena, "it is not our first effort at mixing bipolar and MOS on the same chip," says Thomas George, vice president of the IC Wafer Manufacturing Group, Mesa, Ariz. "The technique is really far more flexible and adaptable than that [just digital applications] and has been applied within the company to a wide range of products, including discrete power transistors and analog circuits" [Electronics, Dec. 23, 1985, p. 35]. In general, Jenkins notes, the company's strategy is to offer BiMOS as one of a number of process solutions. "BiMOS is not being considered as a replacement either for MOS or bipolar but as an alternative and complement. There are places where neither CMOS nor bipolar is suitable, and a combination of the two, taking advantage of different characteristics, is the best fit." Be that as it may, Kenneth Wolf, vice president and general manager of the division, argues that the application of CMOS technology to VLSI systems has been investigated intensively because of its inherent advantages: low power and high noise immunity. Yet CMOS circuits have only limited drive capability, which degrades performance in systems where signals frequently drive large parasitic capacitances. On the other hand, bipolar circuits can drive the capacitive loads with much less degradation of speed but can achieve only limited integration because of their higher power dissipation. By combining both technologies, designers can attain high output drive, reduced gate-interconnection delays, and lower power dissipation simultaneously. WALKING A PROCESS TIGHTROPE* Developing a mixed bipolar/MOS process appropriate for digital VLSI applications involved "walking a tightrope, making tradeoffs between conflicting processing requirements," notes Hickman. "The idea was to retain the individual benefits of CMOS and advanced bipolar technologies, without compromising either and without increasing process-flow complexity too much." In Motorola's approach, where the n-well or n-type epi layer serves as the foundation both for the npn bipolar and MOS transistors, most critical tradeoffs involved the n-well structure. In the p-channel devices, for example, the n-well concentration under the field oxide determines the parasitic field threshold, which in turn sets the minimum n-well impurity concentration deep below the surface. Punch-through voltage requirements also set a limit on the minimum impurity concentration. At the same time, the p-MOS FET's threshold voltage, set by a threshold-adjustment implant, limits the maximum n-well surface-impurity concentration. Slightly below the surface, a lower level of doping is needed to minimize body effect, junction capacitance, and breakdown voltage. This combination of constraints sets a limit on the maximum concentration of bulk impurities. As a result, the n-well impurity profile is tightly bracketed, with only the threshold dosage open to adjustment by the process engineer. Motorola designers also faced a number of constraints on the design of the npn transistors. To minimize high-level injection effects in the collector, Motorola uses a thin epi layer and a heavily doped n-well, limiting the doping level possible for a given collector current density. But such considerations conflict with the requirements for npn breakdown voltage and junction capacitance—requirements that demand a high level of bulk n-well doping. Another problem that must be overcome with BiMOS is the collector impurity gradient, which can create excessive hole storage at the interface of the n-well and the buried layer. The bottom line on the npn structures, says Hickman, is tight bracketing of the optimal n-well impurity profile, with even fewer degrees of freedom available to process engineers than they could get with p-channel devices. By contrast, n-channel MOS FET structures involve the fewest compromises in the BiMOS environment, he says. The one significant restriction is that the epi layer must be thick enough that autodoping and the n-well diffusion do not cause the bottom buried-layer-to-buried-layer boron channel stop to affect the n-channel device performance. Though the n-MOS FETs are affected by the same restrictions limiting the p-MOS FETs, the impact is much slighter, because these characteristics are coupled only weakly to the n-well. Motorola's BiMOS process attacks these problems on two fronts. Whenever possible, says Anthony Alvarez, staff scientist and developer of the process, it minimizes performance tradeoffs by merging process steps. The process also reconciles conflicting requirements by decoupling them during processing while adding as few process steps as possible. This balancing of performance tradeoffs and conflicting requirements is accomplished using a 15-mask, 2-μm twin-tub CMOS process that is coupled to an n+ buried layer with a p-type epi layer, and three mask steps (an n+ buried layer, an inactive base, and an active base) combined with a silicided CMOS process that has a Schottky diode option for the npn transistors. An n+ buried layer runs underneath the p-MOS devices to reduce latchup susceptibility. Though the n-MOS devices are built in the p-epi layer, an n-well diffused into the n+ buried layer serves as the foundation tub for both the npn and p-MOS transistors. Motorola arbitrated the conflicting requirements in epi-layer fabrication—thin, low-resistivity layers for good bipolar performance versus thicker, higher-resistivity layers for good CMOS performance—by optimizing the n-well drive capability depending on the thickness of the epi layer, says Alvarez. The active base is formed in a way that recalls the n-MOS buried contact step, with the subsequent poly deposition simultaneously forming the npn emitter contact and the MOS gate electrodes. The process is synergistic, a characteristic illustrated by the n-MOS source and drain implant, which also dopes the emitter, and the p-MOS source and drain implant, which also dopes the base contact. Another example is the platinum-silicon silicide layer, which reduces sheet resistance of the single-crystal and poly regions, while forming Schottky diodes in the lightly doped epi regions. Motorola uses two-layer metallization with a polyimide interlayer and metal pitches of 4 and 5 μm, respectively, on the metal layers. The designers improved bipolar performance with a thin epi npn structure consisting of a very shallow poly emitter—oxide-walled on all four sides—with silicided sources, drains, and gates used to enhance MOS FET performance. They eliminated latchup by combining the thin epi layer with a retrograde-p-well structure, effectively shunting the parasitic well and substrate resistances that contribute to the problem. The first array fabricated with this process, the MCA6000ETL, consists of 3,072 macrocells, organized in 24 rows of 128 cells each, connected by 580 first-metal and 896 second-metal routing channels. Usually, says Hickman, the fabrication of any logic circuit, including a macrocell array, imposes certain limitations on device sizing. These limitations create differences in the output impedance of various CMOS functions, and these differences in turn usually cut drive capability. In Motorola's BiMOS gate arrays, the basic transistors are configured as push-pull devices, which isolate the CMOS circuits from loading and therefore make the unit-load degradation both very low and essentially the same for all circuit functions. Each of the internal array's 3,072 macrocells, containing both gate-isolated CMOS and bipolar devices, can be configured into a two-input CMOS NAND gate or one input BiMOS NAND. The BiMOS NAND gate (Fig. 3) was chosen for high performance, low power, and ease of configuration. The circuit uses a pair of push-pull bipolar devices to provide the needed driving capability. However, the low power requirements of CMOS are preserved because the circuit draws no direct current. This allows any CMOS circuit to be modified into a BiMOS structure. Because the MOS devices drive only local macrocell interconnections, not extended lines or fanout, sizes can be significantly lower than they would be in a purely MOS implementation. Moreover, the performance of mixed-process BiMOS structures is a good deal better than that of CMOS devices of comparable size. For example, while a BiMOS NAND using 30-μm p and n devices... is about the same size as conventional CMOS NANDs made of 60-$\mu$m p- and n-channel devices, it will outperform them on all loads above 0.3 pF. Up to loads of 0.3 pF, the CMOS circuit is marginally faster than the BiMOS implementation. In medium to heavy loading conditions—from 0.3 to 4 pF—the BiMOS circuit is two to four times better than CMOS. The bipolar push-pull devices isolate the CMOS circuits from the loading, so the unit load degradation is the same for all circuit functions. In addition, dissipation is lower than in an equivalent bipolar circuit, and average dissipation is even less than in an equivalent CMOS gate. There are two reasons for the low power dissipation. First, diffusion capacitances are smaller than in CMOS circuits of comparable area. Second, the fast edge rates of BiMOS under loading conditions reduce the through current component, a significant element in CMOS circuits. To take full advantage of the BiMOS array, the bipolar I/O cells come in either TTL or ECL, as well as in a mixed TTL-ECL mode, and are configurable as inputs, outputs, or bidirectional pins. This combination produces a range of speed and power-dissipation options unmatched by either bipolar or CMOS implementations, says Hickman. For example, in an all-TTL mode (100 high-impedance inputs and 100 outputs), the MCA6000ETL features output delays of 5.5 ns and an active power dissipation of 1.7 W at 50 MHz. In an all-ECL mode with 100 inputs and 100 outputs, output delays are reduced to 1.5 ns, with a total dissipation of 4.7 W. In an all-ECL mode with 150 inputs and 50 outputs, dissipation is cut to no more than 3.3 W. Finally, in a mixed mode (50 ECL inputs and outputs and 50 TTL inputs and outputs) operating with 1.5-ns delays on the ECL outputs and 5.5-ns delays on the TTL outputs, dissipation remains reasonably low: 3.3 W. SUBMICRON BiMOS COMING Currently under development and scheduled for introduction later this year are a number of follow-ons using the same 2-$\mu$m BiMOS process, says Hickman. They include the MCA6000ALS—identical in configuration to the MCA6000ETL except that its die size is smaller and it is limited to TTL on inputs and outputs—and the MCA3000-ETL, with half the number of configurable ECL/TTL I/O lines. A 1.5-$\mu$m version of the BiMOS process will be used early next year in two additions to the macrocell family: the MCA1000ALS, with 10,000 internal and 220 TTL I/O cells, and the MCA6000-RAM, with 6,000 internal gates, 200 reconfigurable TTL/ECL cells, and 4-K of reconfigurable static RAM with single, dual, or triple port capability. Motorola is also working on a submicron version of the BiMOS process, which should almost double gate density and further reduce both gate and output delays. Also under investigation is the replacement of local-oxidation isolation techniques by some form of trenches isolation. --- IN THE MIXED-PROCESS WORLD, EVERYONE’S ON A FIRST-NAME BASIS Even among the select group of engineers skilled in developing mixed bipolar and MOS digital VLSI circuits, Anthony Alvarez, staff scientist at Motorola’s Bipolar Technology Center, Mesa, Ariz., is a rarity. He has worked on nothing but mixed processes since joining Motorola in 1979 after earning his BSEE and MSEE from the Georgia Institute of Technology. The process used in the MCA6000ETL is the third generation of digital BiMOS processes that he has worked on at Motorola. “When I started at Motorola, fewer than half a dozen people at the company were involved in mixed-process development,” says Alvarez. “And around the world, there could not have been more than five or six centers, academic or industrial, involved in similar efforts.” Though the number of engineers involved in such efforts has increased by several orders of magnitude, Alvarez says there are still relatively few: “Everyone knows everyone else on pretty much of a first-name basis.” In the MCA6000ETL project, Alvarez worked with some relative newcomers to BiMOS: Patrick Hickman, design engineering manager for gate arrays; Frank Ormerod, design manager for BiMOS; and Douglas Schucker, ASIC LSI design engineer. Hickman received his BSEE degree from Purdue University in 1977 and joined Motorola immediately thereafter. Ormerod received his BSEE from Strathclyde University, Glasgow, Scotland, in 1979, working at STL Research Laboratories in Harlow, England, for five years before joining Motorola. Schucker received his BSEE degree from Pennsylvania State University in 1982 and has worked at Motorola since graduation. Hickman, Ormerod, and Schucker all have extensive experience in the design of gate arrays. As geometries move below 1 $\mu$m and the advantages of mixed bipolar-CMOS processes become more obvious, Alvarez expects the tight BiMOS community to expand dramatically. “There is still a substantial difference between the way you build CMOS and bipolar structures. As a result, mixed-process circuits are still more expensive and implemented only where the performance merits it. However, as we move below 1 $\mu$m, these differences become negligible.” Alvarez believes that at the VLSI level, BiMOS will emerge as a mainstream process used to fabricate a wide array of products including standard-cell ASICs, structured custom circuits, static RAMs, and even microprocessors. GOOD MIX. Motorola’s BiMOS development team, clockwise from top left, includes Ormerod, Hickman, Schucker, and Alvarez. FEW REMEMBER WHO WAS SECOND TO SOLO THE ATLANTIC... OR SECOND IN THE MARKET... AT&T DOESN'T IMMORTALITY... WAY TO MAKE Today, the product that's first in the market is likely to win the biggest share of market. So when you're racing to get a great idea off the ground—ahead of your competitors—you need more than just a "supplier." You need a company with a broad line of high-performance components and electronic systems. A company with a networking point-of-view, an end-to-end capability, and the people and resources to assure all-out technical support. AT&T. Ready now to offer you the total commitment to quality and reliability that we've always insisted on in the systems and products we develop for ourselves. Ready now to deliver solutions. We'll keep you on the leading edge. Because that's where AT&T Bell Laboratories keeps us. Our new 32-Bit UNIX™ Microsystem, for example, delivers performance others only promise. It's a chip set that's 100% complete, 100% CMOS, and 100% TTL-compatible—fully able to reduce your design time by as much as 50 percent. In data communications devices, we're state-of-the-art every step of the way. (As you might expect from the company that developed the world's biggest, most reliable communications network.) Right now, for example, AT&T is the only company in volume production of a microprocessor-controllable, single-package modem that can handle up to 2400 bits per second. Our Digital Encryption Processor is the only software-programmable, encryption processor available. And our X.25 Protocol Controllers offer the widest range of applications, from PCs to satellites. In memory, not only are we currently delivering an American-made 256K DRAM, but we offer a range of PROMISE YOU JUST THE FASTEST AN IDEA FLY. leading-edge specialty memories. And our third-generation, single-chip Digital Signal Processor gives you the edge you need to design-in superior system performance at a competitive price. Count on AT&T's advanced technology to help you move your product out the door on schedule. Advanced custom design capability. Nobody knows better than you—moving to market first, and staying there, can require custom designed components. At AT&T, we'll get involved at any stage of your product design—from layout to prototype to production. We'll support you with the industry's most advanced CAD/CAM software, rapid prototype turnaround, and unsurpassed volume manufacturing capability. Everything you need to ensure the success of your product—on your manufacturing line and in the market. Your specialized interconnection needs will be taken care of—from flexible printed wiring to multilayer boards to hybrid ICs—all produced with our high standards of quality and reliability. We'll meet your application-specific IC needs with expert design and engineering personnel. And with powerful CAD software that helps make sure your devices work the first time. The commitment and resources to deliver solutions—that's what makes AT&T, AT&T. It all takes power. AT&T's board-mounted power products cut design time with unmatched flexibility. Our low profile power converters are modular in design and about one quarter the size of conventional DC/DC circuit board converters. So they can be mounted in more places. And in the tightest situations. Whatever your power needs, from board-mounted to off-line switchers, we can work with you to develop a system to meet your needs. Networking products that lock-in the future. For local or long-haul transmission, AT&T offers a complete family of fiber optic products and apparatus. In local area networks, our new ST™ Connector can actually double your network connections—or double the distance between connections—without affecting the fiber optic cable or electronics. And for high performance data transfer, AT&T's ODL® 50 and ODL® 200 Lightwave Data Links, with bit rates up to 200 Mb/s, incorporate the latest optical and integrated circuit technology. These products, as well as our ODL RS232-1 Fiber Optic Modem, a high performance data interface, readily mate with the ST Connector. Our newest addition to the AT&T lightwave family—the ASTROTEC™ ceramic laser module—is a reduced-size, long-wavelength laser that offers highly reliable, low cost performance. Advanced processing technology provides superior buried microvias in multilayer printed wiring boards. News in ac plasma. AT&T has just introduced a new lightweight, compact ac plasma display for alphanumeric and graphic images. It offers excellent visual characteristics in virtually unlimited applications. We'll work with you, all the way. Call us right from Day One, and we'll put our expertise at your disposal. Call us when you hit a snag, and we'll work out a solution together. After all, we've got the world's greatest problem-solvers on call—the men and women of AT&T Bell Laboratories. For more information, phone AT&T at 1800-372-2447. We'll help you put wings on your concept, and "first" on your product. (In Europe, phone AT&T Microelectronics in Munich, Germany, at 089/95970. Telex 5216884 attm d.) © 1986 AT&T Technologies, Inc. AT&T The right choice. Circle 73 on reader service card LEARN WHY AT&T IS THE ONLY WAY TO FLY. MAIL THIS COUPON FOR MORE ABOUT AT&T PRODUCTS THAT CAN HELP GET YOUR IDEA OFF THE GROUND. (For fast action, call us at 1800 372-2447.) ☐ UNIX® Microsystem (32-bit) ☐ MOS Integrated Circuits ☐ Communications ICs ☐ Digital Signal Processors ☐ Memory ☐ Application Specific ICs ☐ Fiber Optic Products —ODL Data Links, ODL Data Interfaces, ASTROTEC™ Ceramic Laser Modules, Active and Passive Components, ST™ Connector, Lightguide Fiber Products, Lightguide Apparatus ☐ Interconnection Products —Multilayer Printed Wiring Boards, Hybrid Integrated Circuits, Multi-Fiber Array Connectors ☐ Linear Integrated Circuits —Interface and Telecom, High Voltage Solid State Relays, Semi-Custom Arrays ☐ Power Products —Board Mounted Power Modules, Custom Offline Switchers, dc/dc Converters, dc Reserve System, Power Protection Systems for ac ☐ ac Plasma Displays ☐ Opto-Isolators ☐ Miniature Voltage-Controlled Crystal Oscillators ☐ Capacitors ☐ Miniature Mercury Wetted Relays ☐ Electronic Circuit Transformers and Inductors ☐ Cable & Wire —Electronic Wire, Cable, Mechanical and Electrical Protection Devices, Cross-Connection, Premise Distribution Components, Connectors, Network Interfaces ☐ Please send me a complete list of AT&T components and electronic systems. Mail To: AT&T, Dept. KB 555 Union Blvd., Allentown, PA 18103 NAME ____________________________________________ TITLE ____________________________________________ COMPANY _________________________________________ ADDRESS _________________________________________ CITY ___________________ STATE ______ ZIP _______ PHONE (______) ____________________________________ AT&T The right choice. INTELLIGENCE COMES TO LASER SOLDERING Surface mounting and very large-scale integration have hiked dramatically the density of printed-circuit boards. The ultradense boards now in production carry expensive parts that can be damaged by the soldering process, resulting in costly rejection rates. Vanzetti Systems Inc. claims it has solved this problem with a family of intelligent laser soldering systems that produce perfect joints every time by adjusting the heat they deliver to individual joints. The Stoughton, Mass., company calls its ILS 7000 series of soldering systems intelligent because they use an infrared detector and mini-computer system to monitor and control the laser soldering process. The IR detector monitors the formation of each joint as the laser heats the solder to its melting point and beyond. The detector senses when the optimal solder-joint temperature has been reached—where the solder just becomes molten—and triggers control circuitry that closes a laser shutter. Vanzetti Systems says its intelligent laser soldering provides several major benefits. For one, the prospect of product damage is virtually eliminated. Also, users can check a joint's formation as the solder process proceeds, eliminating the need for a separate inspection step. Finally, data gathered during the soldering process can be used for immediate process control to track down the cause of solder abnormalities and correct them systematically. These advantages may well turn into necessities with the advent of extremely dense surface-mounted pc boards. Some board assemblers have already ruled out conventional soldering methods such as IR and vapor-phase reflow for surface mounting, pointing to an unacceptably high level of joint rejects. Joints done with the vapor-phase or IR reflow method often exhibit undesirable large-grain recrystallization because of the reduced solder mobility at low temperatures and because of the length of time required to cool the joint through the solidification process. In addition, both techniques hold circuit components and boards just above the solder's melting point for several moments, compromising their long-term reliability. Laser soldering systems do not encounter these problems. Based on carbon dioxide or ne- 1. TIME BASED. The infrared signatures of solder joints with identical laser exposure times show overheating in one. The cause is insufficient solder. dymium yttrium-aluminum-garnet lasers, these systems afford higher heating rates but use optics to focus the beam only on the solder joints. Boards and components are spared from heat, improving long-term reliability. In addition, air quenching follows the heating process in laser systems, producing finer-grained, stronger solder joints than can be achieved by either the vapor-phase or IR techniques. The speed of joint formation with laser soldering virtually eliminates intermetallic layers, resulting in solder joints that are less brittle and more ductile. Nonetheless, appropriate control of the laser is required. Soldering of two nominally identical joints, laser-heated for the same period of time without intelligent feedback, can produce a normal joint in one case and board damage in another (Fig. 1). Conventional laser soldering systems apply the same amount of heat to each task regardless of the joint-to-joint variations that occur in the real world of board assembly. That means some joints with excess solder or slightly larger surface dimensions won't reach the necessary solder-melting temperature and become cold solder joints. On the other hand, joints with below-average solder coverage, slightly smaller dimensions, or both, are vulnerable to overheating, which can damage a lead, the component, or burn up very expensive boards. The ILS 7000 series enables a user to thermally characterize the work to be performed based on the usual conditions that will be encountered in a normal solder joint for a specific application. The characterization process takes into account the geometry of the solder joint under test; the mass of the solder, lead, and pads; the heat sinks that may be attached to the pads; and the surface condition of the target. The system is taught to understand the typical thermal-signature variation—the radiation versus time profile—that may be expected. Time limits and thermal thresholds are set up for each type of joint. By monitoring the amount of available solder, the Vanzetti systems deliver precisely the correct amount of laser energy to form an optimal joint. Comparing the thermal signature of each solder joint to preestablished norms thus makes it possible to flag abnormally formed solder joints. The X-Y coordinates of abnormal joints then are stored in the computer's memory for later reporting or retrieval. The key building blocks of the 7000 series are a helium-neon laser used for accurate positioning; a dual-shuttered Nd:Yag laser with a typical output power of 12.5 W and a minimum spot diameter of 24 mils for soldering; an IR detector; a servo-controlled positioning table; a computer based on a Digital Equipment Corp. PDP 11/23 minicomputer; and precision optics (Fig. 2). The systems are designed to work with components that are precluded from a mass-soldering process. The leads of these components must be prepasted to pads. For example, temperature-sensitive components that would not withstand either the heat levels of vapor-phase or IR reflow processes must be attached after these tests have been completed. Also, parts missing at the time of the initial soldering must be attached without reflowing all other joints on the pc board. IR 'EYE' RUNS JOB During soldering, a component-laden pc board is inserted into the system's chamber and attached to a fixture plate. Then the positioning table moves to a programmed location and stop. The control shutter then opens, allowing focused laser energy to irradiate the target. At the same time, the IR detector views the target, determines when the solder liquefies, and sends a control signal to close the shutter. The table then proceeds to the next target. The computer records critical measurements taken during the soldering operation, printing reports that detail abnormal joints and storing information in its data base for later retrieval. The finished board then passes to an external defect-marking system or rework system. Here the data base would be interrogated and the defective joints either marked for future rework or reworked immediately. The Vanzetti soldering systems can be used on pretinned parts and with solder forms. They can be applied to most chip packages, connector lead configurations, and passive surface-mounted components. To date, soldering has been done on 12-mil leads on 40-by-100-mil pads and 28-mil lead wire in 37-mil-diameter plated through holes, according to the company. Soldering speed ranges from 50 to 150 ms. Soldering a small-outline integrated circuit, for example, takes 100 to 150 ms per lead. In a connector-soldering application, the soldering rate was five joints per second. CHARACTERIZATION ON THE FLY In another case, a Vanzetti system demonstrated its ability to characterize abnormalities when it soldered seven pins on a pc-board connector. It monitored the formation of these solder joints, compared them with the limits established, and found a wide variation (Fig. 3). In joints 1 and 2, the proper thermal level was not reached in the allotted time, so the laser heating was terminated and an abnormal flag set. The cause of the trouble probably was excess solder. In joint 3, the shutoff thermal level was reached too quickly and laser heating terminated. Again, the abnormal flag was set. The cause of the trouble was determined to be the absence of solder. Joints 4 to 7 were normal. The thermal shutoff level was reached within acceptable time limits and laser heating terminated. In practice, Vanzetti's engineers say, soldering is a function of the solder paste type, component type, and the pc board's layout and construction. --- 3. THERMALLY MONITORED. The monitoring of seven identical joints revealed the probability of excess solder in the first and second joint and the absence of solder in the third. In a board they encountered having three identical pads, one pad showed a radically different thermal signature. The abnormal signature was caused by a missing via. The Vanzetti units are ready for delivery and start at slightly more than $200,000 for the fixed-head 7005. Rounding out the series are the 7010, which features a tilting head, and the 7012, which has a tilting head and a larger bed. Vanzetti says considerable interaction will be needed between the machine supplier, paste manufacturers, and the equipment users before optimal thermal signatures can be set up for each application. Also critical are the conductor pattern of the pc board, the type of IC package used, and whether through-the-board assembly or surface mounting is used. --- IR EXPERTISE LEADS TO IMPROVED LASER SOLDERING Riccardo Vanzetti came up with the idea for the ILS-7000 in 1982, when Vanzetti Systems Inc. introduced its Laser/Inspect. That system used lasers to illuminate solder joints on printed-circuit boards, thereby generating infrared signatures that were used to check the quality of joints on a production line. The company founder noted that with a higher laser power level, the same principle of operation could be extended to create optimized solder joints. He assigned chief operating officer Ashod S. Dustoomian to guide the IR-based intelligent laser solder program through development. Dustoomian, who helped found the Stoughton, Mass., company in 1968, is the originator of numerous inventions in both IR instrumentation and applications. His awarded or pending patents include an IR temperature probe for high-pressure use, means for detecting changes in skin temperature, a band-ratio radiometer, a method of controlling quality in spot-welding operations by means of phosphorus and IR thermal sensing, and a laser system for automatic solder-joint inspection. "Our biggest problem in the development of the laser solderer was to identify the proper thermal characteristics of a good solder joint," Dustoomian notes. He believes this unit will raise solder-joint yield considerably over existing methods, along with radically increasing quality and lowering inspection costs. Apply within. No experience necessary. That's another wonderful benefit of our exclusive EPLD technology. You can move right into user-programmable logic design. Right now. Because we have a complete family of high density products (from 300 to 2000 gates), plus the easy-to-use development software you need to turn your assignment into a working prototype. In a matter of days. And for a fraction of what you'd expect to invest. Best of all, you no longer have to enter the frightening realm of gate arrays to get a high level of integration. Just look into our newest family member: the EP1800. You'll work with 2000 gates. In a tiny J-lead package. Without worrying about speed (it has plenty), power (it desires very little) or noise (it's quite deaf). What's more, design development is a breeze. Even on the EP1800. With our A+PLUS software and your IBM PC, prototypes can be produced in days. Sometimes hours. Modifications can be made in minutes. And with the software automatically placing your design into the right chip, you don't waste time figuring out what goes where. One last step. Volume production. Also no problem. Because every one of our EPLDs is available as a One-Time-Programmable (OTP) plastic part. So you can get to market in less time. For a lot less money. So apply today. We've got immediate openings for your design. Call (408) 984-2800 Altera 3525 Monroe Street Santa Clara, CA 95051 A+PLUS is a registered trademark of Altera Corporation. IBM is a registered trademark of International Business Machines Corporation. CMOS is a patented process of Intel Corporation. Distributed in U.S. by Alliance, Pioneer, Quality Components, Schweber and Wyle. In Canada by Future. © 1986 Altera Corporation. Openings. ALTEA EP1800 JC ALTEA EP1210 DC ALTEA EP600 DC ALTEA EP900 JC Large-scale software projects are still very much hit-or-miss propositions. The reason: though many tools have been available to assist programmers with the downstream, or coding, portion of software development, little help has been around for the more abstract upstream tasks, which demand that software engineers analyze requirements, develop specifications, and come up with workable designs. "What we have been doing in the large-system area has been by the seat of our pants," declares Laszlo A. Belady, vice president and program director of software technology at the Microelectronics and Computer Technology Corp., Austin, Texas. "Even after 30 years of research, more of the efforts have been directed towards programming," which falls in the downstream area of the software life cycle. Today, however, many groups are beginning to tackle the upstream problems of software development. They're also taking a big picture approach, hoping to tie together a full range of ON THE WAY. Programmers will soon be getting help with abstract upstream software-development tasks, such as requirements analysis. tools to manage projects throughout the entire software life cycle—from its start through its maintenance. High-profile efforts to improve the software-development process—by such research groups as the MCC and the federally funded Software Engineering Institute, as well as by such major corporations as Lockheed, TRW, and IBM Japan—include a work-station-based software-development environment, a generic set of development tools called the software factory, a knowledge-based tool that automatically generates documentation from source code, and a cost model that helps manage software-development projects. Though some of the work is just gearing up, it is sure to have a major impact on software development in years to come. In late 1984, the MCC's software technology group embarked on the ambitious Leonardo (for Low-cost Exploration Offered by Network Approach to Requirements and Design Optimization) project. Like other MCC projects, it targets the upstream portion of software design. The goal was to develop a complete environment for a team of software engineers to use in designing large software packages of the highest complexity—programs for real-time, embedded, distributed software systems. The Leonardo environment will be a networked work-station-based system for use in the early stages of a system's design. Through the work stations, software engineers will have access to knowledge bases containing extensive expertise on software development in general as well as on the software project under development. That information will be continually updated as a project progresses to help manage the task; it will also monitor the work stations and automatically determine what activities it should initiate. "The whole architecture is centered on an information data base, and attached to it is what we call a progress promoter," says Belady. The progress promoter is a large module that manages the Leonardo's activities. For example, it will decide whether there is source code that has to be compiled or whether it must provide information in response to a question asked by one of the software engineers. The Leonardo system goes beyond the two-dimensional interface of standard work stations. It will support voice communications, provide an electronic blackboard to allow engineers to draw diagrams for one another while sitting at their own computers, and even accept video inputs. One of the best ways of communicating in software design is to use scenarios, or storyboard-like renditions of a complicated design procedure, says Belady, adding that video scenarios could well become part of the knowledge base at the center of a Leonardo-based system. Though conceived originally as an eight-year project, Leonardo is now open-ended. MCC's 10 sponsoring shareholders have already begun to receive formal transfers of spinoff technologies from the effort. And the MCC's software group is now looking at creating a Leonardo kernel project that could make a subset of the full system available, on a widespread basis, before completion of the project. "The Leonardo environment is designed to take on more machine-activated progress as time goes on," says Belady. "So even a system with little machine intelligence today could be helpful in the early stages of software design." All told, the MCC's software technology program consists of four subgroups: architecture, design information, design process, and design interface. The architecture subgroup is developing the modules of the Leonardo prototype design system. The design information subgroup is attempting to represent information and software engineering know-how in knowledge bases. They are also researching distributed systems. The MCC's design process subgroup, which is theoretically oriented, is trying to understand what the software design process really is. To this end, it is tracking existing design technologies under way in other research groups. And the design interface subgroup is working on graphics interfaces and exploring the use of animation and other ways of synchronizing designers on massive projects. A different type of software-design project is under way at the Software Engineering Institute in Pittsburgh, the nation's first federally funded research and development center for software technology [ElectronicsWeek, Feb. 4, 1985, p. 27]. Recently, it completed its software-factory project, one of its first important efforts to study software engineering tools. "We're trying to develop the concept of the software factory as a means to automate all the activities involved in the production of software," says associate director Mario Barbacci. That full range of activities includes the complete software lifecycle: requirements analysis, design, implementation, testing, integration, and maintenance as well as managerial activities such as scheduling and budgeting. Actually, Barbacci says, "software factory" is something of a misnomer. "There is no word in the English language that will express the concept," he says. Factories, like software developers, use standard components and methods to produce their products. (In software, those standards include languages and design methodologies.) But in a factory, raw materials go in at one end, and finished products come out the other. In the case of software, no raw materials are used to reproduce the product—a string of data is simply duplicated. "Instead, a lot of activity goes into things that look more like engineering design than actual manufacturing," says Barbacci. "We've kicked it around for a while. I don't think anybody has come up with a better idea. We're still calling it a factory." The task of pulling together a full complement of automated aids for the software factory is more difficult than it might first appear. Though the market is chock-full of software tools, many are aimed at the design and implementation phases of the software lifecycle. Lifecycle phases, which entail more abstract thinking—most notably, requirements analysis—are much more difficult to automate. In addition, many of the currently available tools serve a narrow audience. "We still don't know how to provide the right support for different kinds of users," says Barbacci. "For instance, a lot of work still has to be done on different types of user interfaces. If you're a manager, you want to deal with the computer differently than if you're a programmer." SHOWCASE ENVIRONMENT Using development tools already on the market, the Software Engineering Institute has built a prototype software factory, which they call a showcase environment. Using work stations linked by means of Ethernet, it incorporates facilities for coding, debugging, documentation, and communications. "The factory itself is a generic thing," says Barbacci. So the software factory will be different depending on where it is being used. When a project gears up, its particular needs will be assessed and the appropriate tools will be selected and added to the factory. "For example, if you are building prototypes, perhaps you give less emphasis to formal specifications up front. On the other hand, if you are going to build something that will become commercially available, then you might want to pay a lot more attention to budgets and schedules. All of those things will affect what kind of factory is ideal for you. And there isn't usually a single factory that will solve all problems." Today, the software factory exists more in theory than in practice. "At the institute, we're using the idea of a software factory as a guiding force to select projects, because building a software factory with today's technology would be very difficult," says Barbacci. "Instead, what we're doing is selecting smaller pieces of what could be in a software factory in the future.” Right now, the results of SEI’s year-long software-factory project are being used to determine where to pursue further investigations. An area under study is the automation of tools that generate software conforming to DOD-STD 2167, a Department of Defense software standard. Another project is evaluating programming environments for use with the Ada programming language. And several researchers are working on applying artificial-intelligence techniques to project-management tools. Researchers are also looking at AI to automate the requirements-analysis phase at the frontend of the software-development process, where errors are particularly costly. Errors here will ripple through the engineering chain and can crop up as bugs once the software is out in the field, making errors in requirements analysis the most expensive to fix. A form of AI called knowledge-based design is being used by IBM Japan Ltd. to develop a software productivity tool called Prompter. The still-experimental Prompter, developed by a group led by Koiehi Fukunaga at IBM Japan’s Software Engineering Group (see “The toolsmith’s tale,” p.84), is a support tool for code understanding. As such, it attempts to make programs easier to follow by producing annotations—the comments following the actual code on each line of a listing—that adhere to a uniform style (Fig. 1). Prompter’s knowledge base is an a priori collection of information about how programs work. The knowledge base is used by Prompter as a sort of reference book to help understand how the program it is analyzing works, analogous to the real-world information a programmer must possess to be able to comment a program properly. It contains three kinds of knowledge—hardware knowledge, such as storage and instruction formats; programming knowledge, such as coding techniques and control and data flow information; and knowledge of conventions, such as control-block definitions or module structures. Prompter’s knowledge base must be customized to enable it to analyze programs written in a particular language. Prompter simulates the execution of a program. Notably, this simulation is run not at the binary-code level but rather symbolically. “The simulation is at the same level at which a programmer thinks when writing a program,” says Fukunaga. So when the program being simulated adds the value 1 to an address, Prompter understands that the program wants to increment that address and will generate a comment accordingly. “The output comments of Prompter are more coherent than those which a human programmer could produce,” says Fukunaga. Typically, a programmer’s comments will vary in style from coding session to coding session, and they may become difficult for other programmers to follow. "The advantage of the computer output is that it is always the same, and even first-time users can soon understand them. The disadvantage of the computer is that it may be verbose." Uniform comments prove most important in the maintenance phase of the software life cycle, when programmers who often had nothing to do with the software's development must step in and fix bugs or incorporate modifications. Rather than first having to reconstruct a previous programmer's thought processes, these maintenance programmers could get right down to the business of changing code. Notably, Prompter works with assembly language programs—typically the toughest to comment well and the toughest to maintain. OBJECT ORIENTATION During symbolic execution, Prompter applies the information in its knowledge base to understand how the program under simulation works. Many knowledge-based expert systems store information using the rule-based approach. For Prompter, Fukunaga decided to use an object-oriented approach, which allows knowledge to be decomposed into collections of objects and classes. This was an ideal fit for software: for example, registers can be classified as objects that belong to the class "storage"; the instruction LDA (load register A) is an object that belongs to the class "load-instruction." To implement object-orientation in Prompter, Fukunaga developed his own programming language, Spool, an object-oriented extension to the AI programming language Prolog. "'Object-oriented' is a metaphor to describe a mode of operation similar to a conversation among people," says Fukunaga. "An object does tasks of which it is capable, but for tasks it cannot perform it sends a message to another object asking it to perform the task." For example, an assembler executes a load instruction by copying data from one location to another. In an object-oriented approach, "load" is represented as an object and both the source and memory locations are also objects. The load object first asks the source object what its content is. Then, after learning the content of the source, the load object asks the destination object to store that value. In addition, Spool incor- --- THE TOOLSMT'S TALE Koichi Fukunaga is firmly against rushing headlong into building large software tools that could turn out to be white elephants. "Tools are different from other programs," says the head of the group at IBM Japan Ltd. that developed the Prolog and Prompter programs. "You have to work with them before you know if they will be useful." That's why he thinks it's best to start off with a simple version and get colleagues to try it out. Then if it works, enhance it. The merit of that philosophy was recently borne out in Fukunaga's software group. "One member of our group had an idea for a tool to understand the operation of Prolog programs. The backtracking mechanism in Prolog can make the search for bugs very difficult because the same type of backtracking can be caused by searching for a rule and by a bug." A tracer is available in Prolog, says Fukunaga, but it spews out great volumes of paper as it prints out a complete trace of chronological program execution. But his colleague thought that a top-down search for the error would be more efficient. "He proposed that the program under test be completely executed but that only the execution of the top level should be checked," Fukunaga says. This would enable the programmer to find approximately where the error is. Then that portion of the program could be reexecuted with the next level checked, and the process repeated until the error is found. Initially, the software group did not take to the proposal. "I told the person who came up with the idea that the only way to check a tool is to try it, so he quickly made a prototype and showed it to the others. The others tried the prototype, and they all changed their minds," he says. The proof of the project's success is that the tool, called Proedit, is in widespread use in the group's lab at IBM Japan. A paper on the tool, entitled Visual Debugger for Prolog by Masayuki Numao and Tetsunotsuke Fujisaki was presented last December at the IEEE Second AI Applications Conference. Fukunaga took the team approach in developing his idea for Prompter, and the group labored for 18 months on the project. Prompter's knowledge base was developed with the assistance of a consultant, Yoshinobu Yamamura, who is now studying for a doctorate in computer science at Princeton University. Prompter's syntax analyzer was developed by Linore Cleveland, an American woman who worked on the project. Code for the generation of comments was written by Arthur Ozeki, a colleague at IBM Japan. Fukunaga, who holds a master's degree in computer science from Tokyo University, wrote the preprocessor software. The big problem in developing successful tools, according to Fukunaga, is the limits of imagination. He says that if an engineer can imagine a tool, it's probably already available. But unconventional tools may be just what industry needs. "A tool that changes one's viewpoint can be very useful. Perhaps what is really needed are tools that are still unimagined, rather than tools that are merely unavailable." porates the backtracking mechanism that is built into Prolog. As for Prompter's future, no comments are forthcoming from IBM Japan. For now, Fukunaga has been assigned to other projects as the new manager of the Knowledge-Base Applications Group. ACQUIRING TECHNOLOGY Stateside, a number of private corporations are working on products to help engineers develop software more efficiently. Among them, Lockheed Corp. is taking a leading role by aggressively acquiring productivity-improvement technology from a host of smaller firms. Lockheed's Software Technology Center, Austin, makes heavy use of the Rational R1000 Ada-language development system (Fig. 2), acquired from Rational, Mountain View, Calif. [Electronics, July 6, 1985, p. 36], and of the Refine knowledge-based software development technology from Reasoning Systems Inc., Palo Alto [Electronics, Nov. 4, 1985, p. 30]. Lockheed has also invested in the AI-tools company Inference Corp., Los Angeles, and will use Inference's ART (for automated reasoning tool) to develop expert systems. In addition, the company is considering investments in four or five other companies, which it declines to name. "What we're looking for are companies that have advanced technologies on how to improve the ability to produce software," says center director Winston Royce. Those productivity-enhancing technologies are beginning to proliferate, according to Royce. "What's generally happening is that detailed design and coding productivity is improving by leaps and bounds." The improvements have resulted from the increased use of aids such as structured design tools and of better programming languages, including Ada and Lisp. But a heightened awareness of the problems involved in designing high-quality software could account in equal measure for the gain. "The whole underlying basis of reasoning about computer programs has changed," says Royce. "In the past, all of us looked at computer programs as procedural—devices in which you thought through in a serial way the functions at each step of the computer program. A new philosophy has been developed—the new design approach is based on declarative analysis in which you treat the things being processed as objects and you describe what must be done to them, but not how it's to be done." The "how" refers to the implementation details, or coding, of a software project. Automating the code-generation portion of software design is something that's within the reach of such current technology as expert systems and Reasoning's Refine. Lockheed is using Lisp to build up a knowledge base in a computer so that it can implement low levels of design and code. The objective is to allow software engineers to work in what is best described as a midlevel design language (as opposed to a very detailed, procedural, Fortran coding approach). Programmers would develop software using this midlevel language. The language would tie in to the knowledge-based system from which the final code would be produced automatically. Using such a system could quintuple overall life-cycle productivity, according to Royce. Lockheed has already successfully used the Refine technology in Link11, a software-development project for the U.S. Navy. The distributed message- and communications-processing package contained 16,000 lines of code. Because Link11 involved real-time processing with many protocol handling considerations, Royce says it is a serious test of the system. Already, Lockheed is gearing up to start a second project. The development of an all-purpose system must proceed incrementally, incorporating additions to the knowledge base as software engineers learn more rules about code generation. "In the beginning, although the system is built, there are no rules, so you stick in some rules," says Royce. "As time goes on you put in more complex rules. Over a decade, the rule base will get very complex." At the beginning of a project, coding proceeds pretty much the way it did in the past, says Royce, because there are very few rules. "But as you build up the domain knowledge base, that allows you to shift up into higher semantically meaningful statements and get a better gain in productivity in comparison with the number of lines of code that are produced." Gains in productivity do more than lower the cost of developing software. They also help companies cope with the current scarcity of software designers. According to Barry W. Boehm, chief engineer of TRW Inc.'s System Development Division in Redondo Beach, Calif., "Managers still are facing a personnel shortage, so getting productivity out of the people you have is important." To help use the resources more efficiently, TRW uses cost modeling to help in line management of its software projects. Currently, for all large and midsize programs, the company requires its managers to use its Cocomo cost model and to define each program's major milestones: plans and requirements, product design, detailed design, code and unit testing, and integration and testing. They continually compare where they are on the project against where they expected to be in the model. This helps keep them on schedule and supplies data to update the model to match the actual project conditions. TRW plans to continue to develop Cocomo, which it started in 1981. "It's providing valuable insight on how to reduce costs," says Boehm. Using Cocomo, TRW has realized a 40% productivity improvement over previous programs. In the software design phases, TRW is also applying automated design tools. "We find that the formatted structure they impose encourages better and cleaner designs," says Boehm. "We're doing more planning in the design phase for cost reasons because later, when you have to fix something, it's more expensive." Despite the resolute efforts of software companies to turn the development process into a predictable, easily manageable exercise, large-scale software development is likely to remain problematic. "The reason is that program demands increase as fast as the improved design methodologies—it's a constantly moving target," says Boehm. He points out that programs written for mainframes could evolve into software for mobile computers and then migrate to distributed microcomputers. Each of these machines poses new sets of problems, with different functions and testing requirements. "The typical job is always bigger and more complex than its predecessor." In the past, software managers paid lip service to the need for testing as a way to catch errors before code went out into the field. But testing budgets usually weren't up to snuff, and managers depended on their programmer, not on their test engineers, to make sure that the software worked. That is beginning to change. Today, testing and quality assurance are becoming critical items on the software manager's agenda. A powerful reason behind the shift, according to Lockheed's Royce, has been the increased awareness generated by complicated upcoming software projects for the National Aeronautics and Space Administration's space station and the Strategic Defense Initiative. As a result, customers are demanding that higher-level, more thorough tests be run on software. "Furthermore, they're going to demand that logical testing be done to prove that under all possible inputs, certain classes of errors cannot take place—that's a new kind of testing, that's tough," Royce says. BUILDING IN MAINTENANCE Maintenance, or the continued monitoring and upgrade of software once it's out in the field, is also becoming more important. Says TRW's Boehm, "In the past, project managers charged with budget and scheduling primarily made it work first and skimped on making it easy to maintain. But recently, maintenance has been built into development plans." Building-in the ability to maintain code requires cleaner design and high-quality documentation—which are easier to do with the generation of automated design tools now becoming available. Nevertheless, even as the art of software productivity advances, it is the abilities of the people using the tools that count the most. And no matter how advanced the tools, one fundamental rule of software development still seems to hold true: the old saw that 20% of the people do 80% of the work. So though advances in software development can help plodding programmers, their biggest impact is in helping the smart work even smarter. "The difference in people capabilities is still the biggest variant," says Boehm. "It can amount to a 4:1 difference in productivity, even with design aids. As more advanced technology comes along, the variant gets bigger because the good people use the tools better." Additional reporting provided by Charles Cohen in Tokyo, J. Robert Lineback in Dallas, and Larry Waller in Los Angeles. polyFORTH GETS YOUR PROGRAM FROM CONCEPT TO REALITY 4 TO 10 TIMES FASTER THE ONLY INTEGRATED SOFTWARE DEVELOPMENT PACKAGE DESIGNED FOR REAL-TIME APPLICATIONS If you're a real-time software developer, polyFORTH can be your best ally in getting your program up and running on time. In fact, on the average, you will develop a program 4 to 10 times faster than with traditional programming languages. But you don't have to take our word for it—just ask our customers. We'll tell you who they are. They're some of the largest and most successful companies in high technology fields such as aerospace, robotics and machine vision, instrumentation, process control, and more. polyFORTH shortens development time by making the best use of your time. There are no long waits while you load editors, compilers, assemblers, and other tools, not to mention the long waits while they run—because everything you need is in a single, easy-to-use, 100% resident system. Using polyFORTH, you take a raw idea to fast, compiled code in seconds—and then test it interactively. polyFORTH's total programming environment has all the tools you need to develop real-time applications: fast multi-tasking, multi-user OS; FORTH compiler, interpreters, and assemblers; editor and utilities; and over 400 primitives and debugging aids. And because of its unique modular structure, polyFORTH even helps you test and debug custom hardware interactively. polyFORTH is available for most 8, 16, and 32-bit computers. FORTH, Inc. stands behind its customers with complete professional support services including custom application programming, plus polyFORTH programming courses and the FORTH, Inc. "Hotline" to help programmers learn to use polyFORTH. For more information and a free brochure on how you can cut development cycles 4 to 10 times, contact FORTH, Inc. today. FORTH, Inc., 111 N. Sepulveda Blvd., Manhattan Beach, CA 90266. Phone (213) 372-8493. One of the ‘big’ attractions at the Expo ’86 World Fair in Vancouver, Canada, is the world’s largest electronic billboard. It’s called Philips VIDIWALL. It incorporates no less than 108 computer-controlled television monitors, creating a single-picture display area 3.5 meters high by 10 meters long! The screen can be electronically divided into individual picture configurations of different size and shape. The result is a highly dynamic display of prerecorded or live video images - some moving, some still - that will astonish and entertain visitors to the World Fair. Philips VIDIWALL represents the perfect integration of advanced computer techniques with television, video and LaserVision technologies. Trust Philips to keep Another innovative development is the Philips thermal imaging system. The thermal energy radiated by all objects and living creatures is picked up by a heat-sensitive camera. It is then converted into pictures that can be shown on a TV monitor and recorded on a VCR. Using the most advanced techniques, the system provides excellent pictures, even over long distances, in very bad weather or in total darkness. Crime prevention, aerial surveillance, industrial troubleshooting and national defence are just a few of the many fields of application for Philips thermal imaging. So you see, whatever your point of view, you can trust Philips to keep you in the picture. Philips. The sure sign of expertise worldwide. How we tailor VMEbus solutions to your industrial applications VMEbus and Intelligent I/O Channel... With the VMEbus finding overwhelming acceptance worldwide, PEP is now enhancing VMEbus versatility with the Intelligent I/O Channel (IIOC). The IIOC provides you with lower cost interface modules typically used in industrial applications. It can also be furnished with its own CPU which can execute tasks locally, offloading the VMEbus master CPU. The IIOC can also be used stand alone. With or without CPU, the IIOC is accessed transparently (as a 64 Kbyte memory mapped space), eliminating the need for complex multiprocessor operating systems. Furthermore, there is no limitation in the number of IIOC’s on the VMEbus other than for physical space. ...all in Single Height format PEP delivers both VMEbus and VME Related Products on the IIOC modules in a consistent, single height format. This gives you the physical and functional configuration flexibility ideally suited for industrial applications. In addition, single height modules are more reliable in harsh environments because of greater resistance to vibration, torsion and shock. PEP Modular Computers, Inc., 600 North Bell Avenue, Pittsburgh, PA 15106 Phone: (412) 279-6661, (Tollfree): 1-800-228-1737 (outside PA), 1-800-255-1737 (inside PA), Telex 6711521, Telefax 412-279-6860 PEP Elektronik Systeme GmbH, Am Klosterwald 4, D-8950 Kaufbeuren Phone: (08341) 8974, Telex 541233, Telefax (08341) 4022 Modular Computers® Circle 91 on reader service card 2. LEADED CARRIER. Intel's 68-lead quad pack eliminates the troublesome problem of thermal mismatch between leadless carriers and their boards. Commercial and industrial applications. Because the packaging density of military equipment calls for many interconnections, it relies more and more heavily on multilayer boards and flexible circuitry of all types. There are several kinds of surface-mounted chip carriers. The original version was the flatpack—a square package with flattened leads coming out of two opposite sides. This package is no longer recommended for new designs: it is rarely made anymore, it is hard to handle, and it does not lend itself to the automated assembly process. Today's leader in the military chip-carrier market is the leadless ceramic chip carrier with input/output pads on 50-mil centers. The leadless carrier dominates in spite of certain significant problems. These include a mismatch of thermal coefficients of expansion with standard pc-board laminates, which causes solder-joint failures during thermal shock testing, and solder joints that are difficult to inspect. This unit has the smallest size and weight of all military carriers and can be designed for a fine pitch. It possesses excellent thermal conductance, is easy to heatsink, and imposes no handling problem. There are four ways to solve the thermal-mismatch problem caused by the leadless ceramic carrier: using a constraining metal core such as copper-Invar-copper or copper-molybdenum-copper; using a low-TCE dielectric such as Kevlar to approximate the TCE of alumina; adding an elastomeric layer to the top pc-board layer; and mounting leadless carriers on a leaded ceramic motherboard. The most successful and common solution is the copper-Invar-copper method, which has two variations. In the simpler one, two multilayer boards are laminated to a central metal core and communicate through a card edge connector. In the more complex variation, copper-Invar-copper layers are used as ground and power planes in a multilayer structure. Both constrained-core and Kevlar boards, extensively tested by many U.S. companies, have exhibited reliable operation with no solder-joint failures over a wide range of thermal shock (Table). One of the largest programs using leadless ceramic chip carriers on metal-core thermally compensated multilayer boards is the defensive electronics for the B-1B bomber, which includes 5,000 lb of avionics gear. It is in production at the Airborne Instruments Laboratories (AIL) Division of Eaton Corp., Deer Park, N.Y. In this program, two four-layer, polyimide-glass boards, each with 8-mil printed traces and spaces and 13-mil vias, are laminated to each side of a copper-molybdenum-copper core. A special surface-mounted connector links the two circuit boards to each other and acts as a connection to the backplane. Thermal vias under each chip carrier conduct heat down to the metal core of the composite board. AIL's designers chose a copper-molybdenum-copper core rather than the more heavily used copper-Invar-copper for two reasons. First, the moly-based material is stronger, with a higher Young's modulus. This results in a thinner core and, in turn, a thinner overall board structure. AIL deliberately kept its boards thin to control impedance and to hold down the aspect ratio of the board's vias. Second, moly has much better thermal conductivity than Invar. Interestingly, these advanced boards plug into wire-wrapped backplanes rather than into a multilayered pc-board backplane—the more modern approach. Finally, a large percentage of the B1-B's other boards are Multiwire types that carry DIPs rather than the leading-edge leadless chip carriers. THE LEADED ALTERNATIVE There are about 24 production programs based on leadless ceramic chip carriers and a similar number in preproduction. Many more military programs are in development. However, even the leadless ceramic chip carrier's strongest advocates admit that there are potential attachment problems for very high-power, high-pin-count VHSIC devices. This situation dictates the use of some form of the leaded carrier. Because of the thermal-mismatch problem with leadless chip carriers, leaded ceramic chip carriers that are often used in custom versions on supercomputers are now being considered as an alternative to leadless types, particularly when the number of I/O pins exceeds about 44. The leaded carrier should theoretically solve the TCE board-mismatch problem if the right degree of compliance is designed into the unit's leads. And with a J-lead configuration, a leaded carrier has the same footprint as the leadless type. But the leaded carrier is two to three times more expensive, 1.5 times heavier, and has a vertical profile two to four times higher than the leadless version. What's more, it is difficult to handle, has limited availability, and has yet to be standardized by the Joint Electron Device Engineering Council. Nonetheless, Intel Corp. is already producing standard military versions of its microprocessors in a 68-lead quad pack (Fig. 2), and it appears that the leaded ceramic chip carrier could yet find a place in pc-board applications requiring at least 44 leads. Usually, leaded carriers for surface mounting have their pins shaped into either a gull-wing or a J-lead configuration. The J-lead takes the least space but is not easily inspectable. The gull wing, by contrast, is easy to inspect but has a larger footprint than the J-lead. Both leads have a relatively high profile. Several companies, among them General Dynamics, Fort Worth, Texas, are exploring a solution to this problem: butt mounting. This technique involves mounting a component to board pads on extremely short vertical leads, resulting in a low-profile package with the footprint of a leadless carrier. One approach to butt mounting shows up in a developmental avionics board being assembled at Hamilton Standard Corp., Farmington, Conn. It combines standard surface mounting of leadless chip carriers that have up to 84 I/O pads with a mix of butt-mounted components such as semiconductors in TO-18 cases, radial- and axial-lead passive components, and a 132-lead ceramic quad flatpack on 25-mil centers. Butt-mounted components constitute about 15% of the total number of components. All components are vapor-phase reflow-soldered. The finished structure will consist of two polyimide-glass multilayer boards with copper-invar-copper cores laminated to the opposite sides of an aluminum stiffener that also acts as a heatsink. Already, the multilayer boards with a mix of leadless and butt-mounted surface-mounted parts have gone through design and development, notes Jim Long, an assistant design project engineer at Hamilton Standard. Now, they must prove out in thermal shock and vibration tests. Lab tests show promising results. Eventually, Long would like to go to 100% butt-mounting and eliminate the leadless carriers for the avionics board, but for now, component availability limits this possibility. Although avionics are well into the rush toward surface mounting, there is another class of airborne equipment—missiles—that is much slower to incorporate the technology. But, as missiles are redesigned for greater functionality, surface mounting often finds its way into the new designs. "There are two kinds of airborne equipment: avionics meant to last 10 years or more, and the one-shot missile," says Classon, manager of electronic packaging for Martin Marietta. "The latter is built, tested, stored, and fired. You can't apply the same principles of packaging as used in avionics. This would be a waste of money in expendable equipment. For example, in missile electronics we try to avoid putting in a lot of expensive connectors and only use the minimum number of this type of component." Martin Marietta is currently producing two tightly packed guidance packages—one for the Copperhead guided projectile and another for the Hellfire antitank guided-missile system. The Copperhead (Fig. 3) is launched from a 155-mm howitzer and must withstand 9,000-g forces for 13.5 ms. In the past, cannon-shell electronics were always packaged in small potted modules, notes Classon. In the Copperhead, there is a sizable amount of electronic circuitry and it is not potted. "The Copperhead was the high point of leaded packaging and it has excellent producibility," says Classon. A laser receiver in the nose of the shell uses surface-mounted passive components and thick-film hybrids. The guidance package contains interconnected annular pc boards with section interfaces that are connected by a rigid-flex motherboard, which is supported by aluminum rings that transfer the shock load into the housing. The assembly is compressed and bolted to prevent breakup at firing (Fig. 4). Holes in the center of the electronics assembly permit a shaped-charge warhead to pass through without interference. The Hellfire missile, by contrast, encounters a comparatively normal missile environment: 100 g plus aircraft vibration. Martin Marietta's solution is to repackage the Copperhead's circuitry using simpler circular rather than annular boards plugged into a simpler rigid motherboard. The circular boards could be used because the Hellfire has a fixed warhead behind the laser-seeker section. In the late 1970s and early 1980s, Martin Marietta started increasing its use of surface-mount technology. A case in point was the Adats (Air Defense Anti-Tank System) that the company developed with Oerlikon-Buhrle of Switzerland. Martin Marietta initially tried to package the guidance circuitry of the Adats missile using the Copperhead's technology. This resulted in a design requiring 14 Copperhead-like boards where there was room for only 10. A complete redesign using surface-mounted leadless chip carriers on both sides of polyimide multilayer boards with a central copper-Invar-copper core reduced the number of boards from 14 to 4. Martin Marietta has many new programs using surface-mount technology. The most significant is for avionics equipment for the F-16—Lantirn (Low Altitude Navigation and Targeting Infrared for Night System)—using leadless chip carriers on polyimide boards. The company has also done a great amount of development work on all phases of the leadless chip carrier's thermal-mismatch problem. Packaging large arrays of leadless ceramic chip carriers on compensated boards is still mainly the field of the very large companies, including AIL, Collins, Hughes, Martin Marietta, RCA, and Tracor. Many small to medium-size companies still are turning out cleverly packaged equipment based mainly on ICs in dual in-line packages. But these companies are aware of the advantages of surface mounting and are either working on developmental equipment using chip carriers or are even establishing separate facilities to carefully get the know-how to apply the new techniques needed. A significant part of this expertise involves developing complex interconnections for the increasingly complex chips. FLEXIBLE CIRCUITRY A case in point is Gull Inc., Smithtown, N.Y., a midsize company that builds airborne avionics for fuel gauging, fuel management, engine instrumentation, and displays. Equipment in production uses a mix of DIPs and TO-18 packages on pc boards with up to 12 layers. Gull excels in the design and use of flexible and rigid flexible circuitry. It has actually made flexible circuits with up to 26 layers. For example, the densely packaged IFMP (Interface Fuel Management Panel) for the KC-135 tanker is full of both flexible circuitry and rigid-flexible circuitry (Fig. 5). In this equipment, even the circuitry for transient suppression and electromagnetic-interference suppression uses rigid-flexible boards to save space and ease manufacturability (Fig. 6). Gull is now developing an Integrated Fuel and Engine Indicator for the F-18 fighter; this project constitutes the company's first contact with SMT. The IFEI has a complex liquid-crystal display composed of two separate backlit LCDs with both analog and digital presentations (Fig. 7). Each LCD has more than 100 input connections, forcing Gull to design both a special copper-Invar-copper board to house four leadless chip carriers for custom driver chips. Two of these boards are required to drive each LCD. In addition, the connectors from the two surface-mounting boards to the LCDs could not be standard connectors because conventional connectors were much too large for the space allotted. Gull's solution was to go to a Zebra-type elastomeric connector (a rectangular strip with alternating layers of conductive and nonconductive silicone rubber) that easily met the F-18's interconnection density, size, and volume specifications. The rest of this avionics package consisted of two standard non-surface-mounting boards plugged into a rigid motherboard. Gull is already planning to introduce more surface-mounted technology into its next generation of equipment. Similarly, Telephonics Corp., a subsidiary of Instruments Systems Corp., Huntington, N.Y., is testing the waters of surface mounting. The company is heavily engaged in avionics, aerospace, and Army electronics systems. It builds the sophisticated central integrated test system (CITS) for the B1-B [Electronics, June 25, 1985, p. 48D]. As is the case with practically all of this company's equipment, the CITS uses proven through-the-board DIP-based multilayer technology. Many small to medium-size companies, aware of the advantages of surface mounting, are starting to replace their DIPs with the new technology. But Telephonics has realized that surface-mounted technology would be needed for some of its coming programs, and six months ago it built a small but complete SMT lab-prototype facility with such equipment as a screener, vapor-phase oven, and pick-and-place machine. Soon, engineers at this facility will assemble the first batch of thermally compensated test boards for leadless chip carriers. After the surface-mounting process and boards are proven satisfactory, the new facility will be used to build boards for a new project. After an extensive survey of existing military SMD work, Telephonics will go to a method similar to that used by AIL on its SMT boards. They build up two separate chip-carrier-laden multilayer boards and then laminate the boards to opposite sides of a copper-Invar-copper core. "We prefer this method because we will be able to test both halves of the board before we laminate them to the core," says Al Comins, director of mechanical engineering at Telephonics. This would not be possible in the other thermally compensated board approach that uses the metal layers as ground and power planes. Also moving into surface mounting are makers of militarized microprocessor boards, though the move is slower than elsewhere. Nonetheless, the pace is quickening as companies move into VHSIC design. Bob Harris, director of computer programs at Titan Severe Environment Systems Corp. (Titan Sesco), Chatsworth, Calif., points out that as military systems incorporate increasingly sophisticated technology to improve system technology and as the current fiscal backdrop calls for the leveling of defense expenditures, many defense contractors are turning to militarized microprocessor boards to make the highest technology available for their designs. The use of a sophisticated militarized board allows systems designers to focus their efforts on the design and performance of the system and not on the design of the microprocessor boards. Titan Sesco is one of the companies that will supply militarized boards and full microcomputers based on ICs from the Intel military operation in Chandler, Ariz. What Titan Sesco does is to repackage Intel's larger microcomputer boards in Sesco's 6-by-9-in. format. These boards have stiffeners to resist vibration and a thermal frame to allow operation at elevated temperatures. On some boards, it has been necessary to replace small-scale-integration glue chips with either gate arrays or programmable logic arrays to maintain the 54-in.$^2$ card area. Now there are militarized boards based on the Intel iAPX286, iAPX186, iAPX86, and the 8080 and 8085 processors. In the future, a militarized version of the 32-bit iAPXX386 processor will become available. Ordinarily, the militarized boards are designed for DIP packages. The forthcoming 32-bit processor in a leadless chip carrier will be socketed onto a conventional through-hole board. VHSIC Phase 1 and 2 devices will radically affect pc-board design, processing, assembly, and even laminates. To prepare, the Aeronautical Systems Division of the U.S. Air Force at Wright-Patterson Air Force Base in Ohio has begun a program for sophisticated electronic assemblies. This program, managed by ASD's Materials Laboratory, is establishing manufacturing techniques, processes, and controls for economical, high-quality production of VHSIC assemblies. The work is being done through a cost-sharing contract with Martin Marietta. With a team of subcontractors, Martin Marietta is trying to improve production of VHSIC chip and pc-board assemblies. This involves establishing materials, processes, and controls for soldering VHSIC packages to boards and making necessary intrachip connections. The dense boards will use both leaded and unleaded chip carriers. Martin Marietta has over 40 subcontractors, including GE, National Semiconductor, and Westinghouse. All participants will share in the technology transfer and benefits from the program. Many experts believe that in about five years, packagers will have to go to the chip-on-board technique—wire-bonding bare chips directly to pc-board pads—to cram everything in the volumes called for. This in turn will introduce new problems such as thermally matching a large silicon chip to a pc laminate. Martin Marietta Aerospace and other companies are already looking at chip-on-board and uncased tape-automated chips as possible solutions to tomorrow's even more dense boards. Many aerospace firms are already dealing with VHSIC Phase 1 chips with more than 100 leads, which call for pc boards with increased interconnection density. The high-speed performance and high power dissipation of the VHSIC chips (now readily available) compound the immediate future's packaging problems. Boards for VHSIC chips will have to have finer lines and lower dielectric constants. CTS is the source for standard military hybrids Need special screening, environmental testing, data logging or specific electrical testing? Contact us. What a great alternate source for standard pin-for-pin Military Hybrid replacements! When you are faced with the "take-it-or-leave-it" attitude that some major suppliers have adopted on Military Hybrid standards—look to CTS Corporation, Microelectronics Division. CTS will provide Military Hybrids just as you want them. And, our prices are very competitive. CTS has been a leading producer of custom Military Hybrids for over 20 years. Our hybrids are designed and screened to Mil-Std-883 requirements. We have complete quality conformance evaluation capabilities as required in Mil-M-38510. Don't settle for standard hybrids that don't meet your individual needs. Ask about the CTS0002, 6, 8, 21, 32, 33, 34 and 41 as alternates for the LH and DH series. We'll make your standards to custom requirements. Write today for a product-for-product comparison chart, and technical data on customized "standards" that meet your specs. Contact: CTS Corporation, 1201 Cumberland Ave., West Lafayette, IN 47906. (317) 463-2565. Advanced Air Force Communications System uses CTS Military Hybrid The prime contractor for the Air Force SEEK TALK anti-jam voice communication system selected a CTS Hi-Rel unit as a primary microcircuit component. The need for high reliability in this system linking fighter aircraft with command stations was a major factor in the selection of CTS. CTS MEANS RELIABILITY CTS CORPORATION • ELKHART, INDIANA Voltage Controlled Crystal Oscillators Standard and hybrid designs. Phone: (815) 786-8411 CIRCLE NO. 160 Custom Hybrid Circuits for demanding medical electronics. Phone: (317) 463-2565 CIRCLE NO. 161 PC Boards Complex double-sided and multilayer. Phone: (415) 659-1770 CIRCLE NO. 162 Connectors Custom PC board and military styles Phone: (612) 941-9100 CIRCLE NO. 163 WHY MULTIWIRE DIVISION BOARDS ARE BETTER THAN MULTILAYER. POINT BY POINT. Computer-generated patterns of insulated wire form permanent interconnections. Wires cross over each other without shorting. X and Y axis routed on the same plane. Easily handles high-speed signals like ECL and FAST. Multiwire Division's electroplated circuit board construction provides surface features for surface-mount components as well as traditional through-hole-mount components. Our experience in new package types encompasses both leaded and leadless devices with pin pitches of 18 mils or more, pin grid arrays and discrete surface-mount components. We have successfully designed and manufactured boards to accommodate pin grid arrays and chip carriers with 130 pins and greater. Because our insulated wires can be crossed without shorting, board real estate can be used for components instead of via holes. This allows increased packaging densities and decreases the need to interconnect between cards, which can result in greater system reliability and lower costs. Over the last 16 years, Multiwire Division has designed and manufactured thousands of Multiwire boards. We have produced boards for through-hole and surface-mount components, military and VHSIC/VLSI applications and used our expertise and experience to solve system and circuit design problems for high-speed logic and thermal management applications. Our sales, design and manufacturing centers located across the country are ready to prove to you the advantages of Multiwire boards from Multiwire Division. For a free copy of our brochure, just fill out and return the coupon. Meets or exceeds ANSI/IPC-DW-425 (DOD) and the performance requirements of MIL-P-55110. Accommodates component densities of 0.4 square inch per IC and greater. Fast and economical design and design revisions. Electroplated surface features mechanically and electrically accommodate surface-mount components. Superior electrical characteristics. Crosstalk and impedance levels tightly controlled. Designed from as little input as a schematic or netlist. MAKE IT A POINT TO BUY MULTIWIRE® BOARDS FROM MULTIWIRE DIVISION. MULTIWIRE 250 Miller Place, Hicksville, NY 11801 ☐ Please send me a free copy of your brochure. ☐ Please have a salesman call. Name Title Company Address City State Zip Telephone MULTIWIRE/EAST, 250 Miller Place, Hicksville, NY 11801 (516) 933-8300 MULTIWIRE/NEW ENGLAND, 41 Simon Street, Nashua, NH 03060 (603) 889-0083 MULTIWIRE/WEST, 3901 East La Palma Avenue, Anaheim, CA 92807 (714) 632-7770 Multiwire® is a U.S. registered trademark of the Kollmorgen Corporation. Other ASIC companies brag about how many gates they can pack into an array. Or how much logic they can cram on a PLD. They think it's a big deal if they can shrink 3 boards down to 3 chips. Let's get serious. VLSI nails the competition to the wall. Our tools let you build a chip so far ahead of anybody else's they won't know what hit them. A single chip with logic, memory, processors, and peripherals. And you'll do it faster than you ever thought possible. Wang, Tandy, and over seventy other leading companies have already used our tools to build hundreds of chips. Everything from gate arrays to megacells to compiled silicon. One chip we developed with NCUBE packs over 160,000 transistors and requires no glue logic because the processor, communication logic, and all interface logic are integrated on each chip. What good is a tool set without nuts and bolts? Our cell libraries are compatible since they're produced in the same 2μ CMOS. So you can even combine standard cells, megacells, and compiled cells on a single chip. You can expect chips designed with our tools to work the first time in your system. You can use them with Apollo, DEC VAX™ and MicroVAX™, Elxsi, HP 9000 Series 300, Mentor, Ridge, Sun, you name it. And whether you buy them, lease them, or use them at our design centers, our ASIC team will support you all the way. If you want the right ASIC solution to your problem, talk to the only place that has them all: Tools. Programmable logic. Gate arrays. Standard cells. Megacells. Compiled silicon. And a complete fabrication facility. Call VLSI at 800-535-1860. (In California, call 800-862-1860, Dept. 701.) Or write to us at 1109 McKay Drive, San Jose, CA 95131. And start hammering away at your competition. VLSI Technology, Inc. Not just your basic ASIC. THE ONLY OTHER TOOL THAT COMBINES LOGIC, MEMORY, PROCESSORS, AND PERIPHERALS. MORE JUICE FOR A TIGHT SQUEEZE. ABBOTT'S NEW TRIPLE-OUTPUT 100 kHz 200-WATT SWITCHING DC-DC CONVERTER. When an application calls for a small yet powerful converter, Abbott delivers the juice. That's why Abbott has created the model BC200. With a 2.5 watt/cubic inch power density, it delivers a reliable 200 watts with low ripple/noise and EMI within the limits of MIL-STD-461B. Its mechanical design protects it from the most adverse environmental conditions and thermally allows for optimal heat dissipation to keep junction temperature below 110°C. A field-proven design and construction process provides dependable performance under severe shock and vibration. Its low profile makes it compatible for ATR sized applications. All this adds up to a triple-output converter that lives up to our reputation—"When reliability is imperative." Standard features include automatic short-circuit protection and current limiting. Plus, a wide array of options make model BC200 adaptable to just about any aerospace application. Soon, Abbott will offer a complete family of model BC converters including 60 and 100 watts with single and quadruple outputs. Call or write for complete specifications and our new Abbott Military Power Supply Catalog today. Abbott Transistor Laboratories, Inc., Power Supply Division, 5200 West Jefferson Blvd., Los Angeles, CA 90016 (213) 936-8185. Eastern Office: (201) 461-4411. Southwest Office: (214) 437-0697 London Office: 0737 82 3273 MODEL BC200 SPECIFICATIONS \| DIMENSIONS \| 1.875" x 5.75" x 8" \| \|---------------------\|---------------------\| \| Input voltage \| 22 to 32 V dc \| \| Outputs \| 5 V, ±12 V; or 5 V, ±15 V \| \| Switching frequency \| 100 kHz \| \| Efficiency \| 60% minimum \| \| Ripple/noise \| 100 mV peak-to-peak maximum \| \| Line regulation \| 10 mV (22 to 32 V dc) \| \| Load regulation \| 10 mV main (0.5% aux outputs) \| \| EMI \| Meets MIL-STD-461B \| \| Environment \| MIL-STD-810C, MIL-S-901C \| \| Input protection \| MIL-STD-704D \| \| Operating temperature range \| -55°C to +100°C \| \| Storage temperature range \| -65°C to +125°C \| \| MTBF * (Ground benign) \| Standard 140,000 hours \| \| \| ER option 1,100,000 hours \| \| MTBF * (Air inhabited) \| Standard 23,000 hours \| \| \| ER option 160,000 hours \| * MIL-STD-217D (50°C Baseplate) WHEN RELIABILITY IS IMPERATIVE. abbott MILITARY POWER SUPPLIES Circle 104 on reader service card The electronics industry hasn’t bounced back as strongly as it had hoped from the down year that it had in 1985. There have been some mildly encouraging results, but the most striking characteristic of the first six months is the way market watchers have been downsizing their growth predictions. What will the second half bring? Most executives once again hope that, led by a recovery in computers, things will get better. Test and measurement companies expected 1986 to wipe out memories of a 1985 that saw business grow an inconsequential 1%. They figured that U.S. orders for automatic test equipment would lead the way to a 12% overall increase in sales. Now they see a flat year, with only overseas sales keeping the picture from getting worse. A good indicator of the state of the U.S. market is rentals, which tend to be high in a growing market—and they are down, says Art Husami, vice president of marketing at Electro Rent Corp. in Burbank, Calif. "The decline has been going on since the last quarter of 1984, coincident with the general decline in the computer industry," he says. However, Husami adds, the market seems to have bottomed out: "There is no shrinkage, but there is no fast growth either." He bases his belief on the behavior of rental pricing. In the throes of the downturn, there was fierce price competition. "Per-month rental charges fell steadily as [rental] companies competed to cover cost of the equipment they had in inventory," he explains. Since then, rental costs have become more realistic. Also, supply is more in line with demand, hence closer to an equilibrium state. But Husami knows the real cure. "What will move the industry out of its doldrums is a recovery in the computer business." The U.S. test and measurement market is weaker this year than last, agrees Fletcher Chamberlin, corporate communications director at Tektronix Inc., Beaverton, Ore. He says the company's aggregate order rate is down 3% from the same quarter last year. Moreover, he adds, business in Japan is worse than that in the U.S. Comments analyst Daniel H. Rosenblatt, assistant vice president at Merrill Lynch Pierce Fenner & Smith, "Tek management describes a very flat order environment and expresses little hope for a strong near-term recovery." The only bright spot is the European market, Chamberlin says. "If the local European economy were not strong, the dollar decline would not be helpful. But the economies have been good, hence the business in Europe has been up." Chamberlin's assessment of general business conditions in the U.S., Japan, and Europe is right on the mark, says William Terry, vice president and general manager of Hewlett-Packard Co., Palo Alto. Terry, however, sees a potential new market in the People's Republic of China. "HP has been doing business for four or five years in China," he says. The Chinese market is no dumping ground for outdated equipment. "The Chinese have 75 instrument factories of their own turning out equipment," he says. "What they want from the U.S. is the newest, best, and most advanced equipment." The Chinese plan on moving into the 21st century on the leading edge of technology, he adds. A STRONG SECTOR. Terry says that instruments for engineers is still a good business, but that tools with greater utility will loosen up tight spending policies of companies with frugal capital-equipment budgets. "Bench instruments will always be in demand," he says. "With more engineers coming into the market, each will need the basic tools, so the replacement and outfitting of new employees will keep this market segment growing at a modest 6% to 8% per year." Chamberlin says digital oscilloscopes are one of Tektronix' stronger product lines among bench instruments. Sales of portable digital scopes are up slightly over last year. One exotic test-equipment area that is expected to show growth near term is tools for fiber-optic systems, he says. "There's somewhere between $60 million and $80 million worth of yearly sales and an annual growth rate of around 20% to be found in this business." One of the healthiest lines at Tektronix is communications test equipThe components business struggled through a forgettable 1985 of barely 2% growth, and manufacturers were hoping for an upswing by the second half of 1986 that would yield a 10% overall increase in business. Now they're beginning to see some evidence of an upswing, with inventories having been worked off, distributors coming back into the market, and end users increasing production. However, the upturn looks more modest than hoped for, in the neighborhood of 6% for the year. Inventory work-off has meant something of a turnaround for Bourns Inc., the Riverside, Calif., producer of trimming and precision potentiometers, panel controls, and resistor networks. Although there has been improvement, the change has not been spectacular, says Curtis Jones, corporate vice president for electronic component sales. Growth was 5% to 10% in the first half. "Overall, this year will be higher in terms of demand and need for components to supply the equipment manufacturers," he says. Also experiencing first-half improvement is Burr-Brown Research Corp. in Tucson, Ariz. Shipments will continue to increase in the second half, says Dennis Haynes, manager of applications engineering. However, Haynes refuses to characterize what he calls the slight increase as a turnaround. "In the past, as soon as people sensed that a turnaround was going to take place, they ordered in anticipation of the wild upswing. That in turn created a wilder upswing," he says. "I believe there's a lot less of that going on now." For connector maker ITT Cannon Inc., Fountain Valley, Calif., "military funding needs to be released. And, on the commercial side, there needs to be a stronger confidence so that the economy will continue to grow," says Charles Hofbauer, director of marketing and business development. The first half of 1986 looks as though it will turn out to be very slow, says Hofbauer, particularly among original-equipment manufacturers, which are spreading delivery time over 12 to 18 months. He says he hasn't seen any growth in the first half of this year, but the company expects a 5% to 10% increase in the second half. Business has been improving gradually since the first of the year for connector maker Molex Inc., Lisle, Ill. Each month's sales through April have exceeded those of the previous month, says board chairman John H. Krebbiel Sr. It has been gradual, and he expects that trend to continue. Krebbiel says Molex's 1986 revenues will be about 14% to 15% ahead of 1985's. For Allen-Bradley Co., the outlook is a tale of two product lines: resistor networks and surface-mount passive components, both of which it makes in Greensboro, N.C. The resistor network business bottomed out at the end of last year, says marketing director Jack R. Polakowski. The first half of this year has been "flat to maybe showing just a slight improvement." Polakowski attributes this change to some firming in the computer business as well as slight improvements in the automotive business. Telecommunications, the third major application where resistor networks are used, has been "mildly soft." Surface-mounted components are a different story, however, and have been showing continuing quarter-to-quarter improvements of 10% to 20%. The reason, says Polakowski, is that more companies are taking the plunge into surface-mounted technology. "When a large organization makes a commitment to do it, it comes along with a bang," says Polakowski. However, surface mounted components are still a very small part of Allen-Bradley's business. For the second half, Polakowski sees "a little bit of gradual growth. The prospects for the future are better because sooner or later, the computer business has got to get better," he observes. There are some signs of this, he says, in terms of quote activity. Because computer makers' inventories are low, a new product creates an immediate need for components. There are no signs of an upturn for Peter B. Cherry, president of Cherry Electrical Products Corp., Waukegan, Ill. "I'm really pleased that things have stopped getting worse. But I don't see any reasons for us to be jubilant." Nevertheless, there are a few bright spots compared with six months ago. The automotive business, to which Cherry supplies components, "continues to be good, but not great," he says. Cherry adds that he does see "some firming" in some of the company's traditional markets such as office equipment, home appliances, and in some segments of the data-processing-equipment business. Reporting was provided by Clifford Barney, Wesley R. Iverson, J. Robert Lineback, Jonah McLeod, Tobias Naegele, Craig Rose, and Larry Waller. Score A Winning Touchdown with UMC With backing from UNITED MICROELECTRONICS CORPORATION, one of our customers scored a winning touchdown with a voice control IC last year. He netted a profit of 2.5 million dollars on just one product. This is one of the many examples of how UMC helps its customers score financial goals. UMC scored its first touchdown by becoming profitable 6 months after it went into operation and has been making a profit and registering phenomenal sales growth annually since then. Last year, 4 quarters of penalties left most companies sitting on the bench and several others were ejected from the game. UMC, however, still romped to a sales growth rate of 24.4%, which was the fourth best in the world and outscored 92% of the IC industry. If you want to be on a winning team, go with a proven winner......UMC. CORPORATE HEADQUARTERS: NO. 3, INDUSTRIAL EAST THIRD ROAD SCIENCE BASED INDUSTRIAL PARK, HSINCHU CITY, TAIWAN, REPUBLIC OF CHINA SALES OFFICE: 9TH FLOOR, NO. 201-26, TUNHUA NORTH ROAD TAIPEI, TAIWAN, REPUBLIC OF CHINA TEL: (02)7152455 TLX: 28560 UMCTPE FAX: (02) 7186291 U.S.A. HEADQUARTERS: NMC CORPORATION 3054 SCOTT BLVD. SANTA CLARA, CA95054 U.S.A. TEL: 408-7279239 TLX: 172730 NMC SNTA FAX: 408-9700548 Circle 112 on reader service card JUST PUBLISHED! EASY, RELIABLE SOLUTIONS TO YOUR DESIGN PROBLEMS Focused strictly on design problems, CIRCUITS AND SOFTWARE FOR ELECTRONICS ENGINEERS delivers professional, innovative solutions for your most demanding projects. CIRCUITS AND SOFTWARE FOR ELECTRONICS ENGINEERS covers a vast array of design problems conveniently organized into 25 vital categories by function. CIRCUITS AND SOFTWARE FOR ELECTRONICS ENGINEERS contains hundreds of circuit schematics and computer programs that have been proven to work and meet the highest standards of performance. These creative, new ideas and approaches keep you on top of what's happening in the latest circuitry developments. So whether you are a design engineer, technician or supervisor—don't take the chance of being less than on top of the latest circuitry developments! Order your copy today! Before you tackle your next project, order this valuable resource today! Use the coupon or send in your company purchase order. Electronics Magazine Books P.O. Box 541 Hightstown, NJ 05820 (609) 426-5070 Send me _______ copy (copies) of CIRCUITS AND SOFTWARE FOR ELECTRONICS ENGINEERS for $19.95 each. U.S. residents please include local sales tax. ☐ Payment enclosed (postage & handling included) ☐ Bill me (postage, handling & tax will be added) NOTE: All orders under $25 will be sent a proforma invoice requiring prepayment. ☐ Bill my company. Purchase order is attached. Ship to: Name Company Street Address City __________ State ________ Zip ________ Country Please allow four to six weeks for delivery. McGraw-Hill knows you will be satisfied, but if for some reason you're not, there is a ten-day money-back guarantee that applies on all books. Coming in September 1986-'87 Electronics Buyers' Guide The industry's most often-used directory: • Lists more than 4000 products. • Lists over 5000 companies. • FREE current catalog retrieval service. Price: $40 USA & Canada $50 elsewhere (add $30 for airmail) Send order with payment to: Regina Hera Electronics Buyers' Guide 1221 Avenue of the Americas, N.Y., N.Y. 10020 UM65xx Series IC The one and only perpetual supplier of 65xx series IC \| P/N \| Description \| \|---------\|----------------------\| \| UM6502/7\| 8 Bit CPU \| \| UM6512 \| 8 Bit CPU \| \| UM6520 \| PIA \| \| UM6521 \| Enhanced UM6520 PIA \| \| UM6522 \| VIA \| \| UM6532 \| RAM I/O Timer Array \| \| UM6551 \| ACIA \| UNITED MICROELECTRONICS CORPORATION U.S.A. HEADQUARTERS: NMC CORPORATION 3054 SCOTT BLVD, SANTA CLARA, CA. 95054 U.S.A. TEL: 408-727-4339 TLX: 73150 NMC SNTA FAX: 408-9700548 Circle 113 on reader service card After a stuttering start, Multibus II has begun to build up a head of steam. Intel Corp., the prime mover behind the synchronous bus standard for 32-bit microprocessors, came through this spring with the crucial chip for Multibus II—a message-passing coprocessor interface that enables both the system and local buses to work at top speed. As a result, Intel's bus has started to find the backing it needs to gain ground against the entrenched VMEbus. "Our best guess is that there should be about 50 vendors out there waiting to announce Multibus II products," says Fredrick J. Mazanec, president of the Microcomputer Interface Group, San Diego, which publishes several buyers guides for Multibus I, VMEbus, and other board-level products. If Mazanec's numbers pan out, Intel and its fellow Multibus supporters will have achieved the critical mass they have been striving for ever since they first proposed the standard some three years ago. But there is much catching-up to do. Mazanec says the recent industry downturn slowed all board-level business, and Multibus II caught the brunt of it because there was little motivation to go to market with boards based on a standard that had gotten off to a slower start than the rival VMEbus [Electronics, Nov. 25, 1985, p. 48]. VMEbus, which represented a $125 million market last year, now has the backing of over 150 vendors with 1,100 products and is the acknowledged leader so far in the battle of the high-performance buses. Multibus II currently has 40 announced vendors offering some 100 products. Multibus II's fortunes should see a tremendous turnaround in the next few months, Mazanec predicts. He expects that at least 50 from among the 1,500 board manufacturers he is surveying will be jumping on the Multibus II bandwagon. "By the fall, we should see Multibus with critical momentum," he says. NEW PLAYERS. Some of the latest Multibus recruits are Siemens AG, Munich, and Interphase Corp., Dallas. The West German firm joined the camp in early April with a line of central processing unit boards, memory and input/output cards, and a line of communications and peripheral controllers. Interphase will be ready to announce a high-performance Multibus II controller early in 1987. Interphase is a VMEbus player as well. It is adding speed to the asynchronous standard with a 30-megabyte/s interface (see story p. 58); one of the VMEbus's weaknesses was its slow speed for many applications. Another prospect for the Multibus II camp is AT&T Information Systems, Morristown, N. J., which is sizing up both competing buses. The dramatic improvement in Multibus II's fortunes can be traced directly to the long-awaited message-passing coprocessor [Electronics, April 21, 1986, p. 17]. Intel is touting the MPC as the keystone in a multiprocessing architecture—a single-chip hardware solution to the problem of controlling the communications between independent processors (be they 8-, 16-, or 32-bit) on the backplane. "The MPC has changed the picture dramatically," says Rick Main, president of Zebu Corp., Sunnyvale, Calif., and independent analyst who follows the single-board-computer market. "It has changed the opinion of a lot of board makers. If I were starting a new company today, I'd probably do it with a Multibus II system." A coprocessor like the MPC became essential because systems designers were putting so many intelligent functional elements, such as disk controllers and local-area-network controllers, on the bus that handling communications among them encumbered the host processor. Before the MPC came along, message-passing communications on Multibus II's parallel system bus were limited. Intel was not the only outfit to realize that the host processor on a busy bus needs help with communications to prevent contention among the devices tied into it. Central Data Corp., a Champaign, Ill., board maker, was also working on a system bus interface but stopped development earlier this year when it became clear that Intel had its one-chip solution in hand. NCR Corp. has a three-chip set under development to handle message passing, but the set will not be ready until next year. One chip is already completed, and the other (which will be used twice in the interface) is still under development, according to Rodger Banta, director of external operations at NCR's engineering and manufacturing facility in Columbia, S. C. Banta says the NCR solution will have all the elements of the Intel implementation as well as a feature to optimize memory-to-memory data transfers. Previously, Intel was offering a two-chip solution of its own, a Bus Arbiter/Controller and a Message-Interrupt Controller. But the combination did not catch on. "I think a lot of vendors smelled a rat on the BAC/MIC," says Multibus Manufacturers Group executive director, Rob Hughes, in Aloha, Ore., "because the chips didn't go far enough in terms of their functionality. The MPC is what a lot of manufacturers have been waiting for." INCOMPLETE. The Bus Arbiter/Controller and the Message Interrupt Controller did not have full message-passing capability. (Message passing generates an unsolicited message that is used to set up a data transfer and then transfer the data without the intervention of the CPU.) The BAC controlled the bus in multiprocessor systems by managing control lines and checking for errors, while the MIC generated interrupt messages and handled received messages from processor modules. "Without message passing, Multibus II really would have little advantage over Multibus I," explains Jack Blevins, manager of new-product development at Central Data. "All Multibus II modules are treated as independent processors, albeit single-function processors. Message passing allows them to communicate without the slowest processor tying up the bus. You are going to get 100-ns transfers regardless of whether you have a 5-MHz, 10-MHz, 16-MHz, or 20-MHz processor. It makes no difference to the bus-transfer cycle." Before the one-chip solution emerged, message passing typically was treated as a software problem that was solved during system integration. The MPC not only eases the burden on software designers but also ensures that block transfers of data packets will be handled in the same way on boards from different vendors. To do so, the MPC takes over interprocessor communications from the CPUs on the modules tied into the system; it also decouples the local bus from the parallel system bus. The decoupling of the local and system buses frees each bus to operate at maximum speed. Without CPU intervention, the MPC controls both message and data transfers across the bus, handling all the formatting, arbitration, transmission, and parity generation needed to send the packets at the maximum bandwidth and speed of the bus. The MPC interfaces to the on-board CPUs, the parallel system bus, and the interconnect space—the address space reserved in Multibus II for board identification, configuration, and diagnostic functions. The MPC offers its full 32-bit data and burst transfer capabilities to any processor accessing the bus. Small data transfers, such as status or service requirements, are handled as part of an interrupt—Intel says this can save up to three central-processor cycles on an operation. Another lift in performance comes from the message space defined in Multibus II for interprocessor communications. By exploiting this feature, the MPC offloads the CPU and eliminates the bottlenecks associated with a dual-ported architecture. In addition to the extra cycles available from a processor freed from communication chores, the decoupling has other advantages. Resources contending for time on either the local bus or the system bus are not forced into wait states while arbitration is occurring. And because the MPC defines all the protocols for transfers, the system software does not have to coordinate shared-memory structures; this means that software applications can run independently on different processors. Central Data's Blevins says that Intel's delivery of the MPC came none too soon for the marketplace. He points out that his company was anxious to migrate its Multibus I customers to the newer standard; but when Intel proved slow to deliver the part, other plans were made. "In late November or early December, it wasn't clear that Intel would deliver. They were already late. We began to design our own chip as a fallback position, though we didn't have the buffering or the speed independence of the MPC. We completed our design down through schematic capture and timing, but we stopped in early February when we got assurance that we could get the chip." Interphase, too, had to bide its time until Intel came through. "We had to wait for the MPC to become available before we could get started" on the SMD controller, says Michael E. Cope, the firm's president and chief executive officer. With the MPC, the controller will be able to take full advantage of the message-passing capability of the Multibus II specification, Cope says. The chip—packaged in a 140-pin grid array—is produced for Intel by VLSI Technology Inc., Phoenix, Ariz., using 2-$\mu$m design rules in a double-metal CMOS process. Sampling is under way and operations are going according to plan, reports Tom Kinhan, general manager of Multibus II operations at Intel. "We are shipping the chip and a board stuffed with an MPC and a 386 [80386 microprocessor]." THINK FAST. TOSHIBA 70ns 55ns 35ns 25ns 45ns AREA SALES OFFICES: CENTRAL AREA, Toshiba America, Inc. (312) 945-1950; EASTERN AREA, Toshiba America, Inc. (617) 272-4352; NORTHWESTERN AREA, Toshiba America, Inc. (408) 244 4070; SOUTHWESTERN REGION, Toshiba America, Inc. (714) 752-0373; SOUTH CENTRAL REGION, Toshiba America, Inc. (214) 480-0470; SOUTHEASTERN REGION, Toshiba America, Inc. (404) 493-4401; MAJOR ACCOUNT OFFICE, Poughkeepsie, New York, Toshiba America, Inc. (914) 462-5710; MAJOR ACCOUNT OFFICE, BOCA RATON, Florida, Toshiba America, Inc. (305) 298-2004; REPRESENTATIVE OFFICES, ALASKA, Montgomery Marketing, (206) 830-0430; ARIZONA, Summit Sales Engineering Concepts, (602) 275-0000; ARIZONA, MIE REP Associates, (602) 275-0000; CALIFORNIA (SOUTHERN), Toshiba America, Inc. (310) 474-1100; CALIFORNIA, A. & E. County, Bay Area Electronics, Inc., (415) 771-7711; (714) 957-3367; San Diego County, Eagle Technical Sales, (619) 743-9550; COLORADO, Straube Associates Mountain States, Inc. (303) 426-0890; CONNECTICUT, Datcom, Inc. (203) 288-7005; DISTRICT OF COLUMBIA, Arctopics, (301) 825-0775; FLORIDA, Sales Engineering Concepts, (305) 834-7656 (305) 426-4601; (813) 963-1043; GEORGIA, Montgomery Marketing, Inc. (404) 447-6242; IDAHO, Components West, (509) 255-6224; ILLINOIS, Carlson Electronic Sales, (312) 956-8240; R.E.P.S. (215) 529-5711; INDIANA, Leslie M. DeVos Company, (317) 842-3245; IOWA, C.H. Horn, (319) 393-8703; KANSAS, D.L.E. Electronics, (316) 744-1229; KENTUCKY, Leslie M. DeVos Company, (317) 842-3245; LOUISIANA, MIE REP Associates, (713) 444-2557; MAINE, Datcom, Inc., (617) 891-4600; MARYLAND, AeroTek, (301) 825-0775; MASSACHUSETTS, Datcom, Inc., (617) 891-4600; MICHIGAN, Action HIGH SPEED STATIC RAMS. From bit-wide to byte-wide, from 25ns to 70ns, Toshiba is the power—and the speed—in static RAMs. When you think fast, think Toshiba. We have the organizations and the speeds to meet your requirements. In the popular 4K x4, we offer a 35ns version in the standard 20 pin type and in a 22 pin configuration with output enable function. The output enable helps eliminate bus contention, increasing speed. In addition, we will soon introduce a 25ns version in both configurations. We were first with 2K x8 devices which are ideally suited for cache memory applications. Now—to complement our 35ns versions—we are developing a 25ns device. For high speed bit-wide applications, we are in full production of two versions of our CMOS 64K x1 SRAMs. One with access times to 45ns, and one with the lowest standby power in the industry at only 100µA. Another new product family being introduced is the 16K x4 CMOS static RAM. Both the standard 22 pin device and the 24 pin device with the additional output enable function have speeds to 35ns. A 25ns version is in development. For additional byte-wide applications, we are introducing an 8K x8 device with access times to 35ns. Packaging is standard 28 pin 600 mil plastic dip. For applications where parity is required, we have an 8K x9 with speeds to 35ns in a 28 pin 300 mil ceramic dip. So when you’re thinking fast, think Toshiba, the power—and the speed—in static RAMs. Toshiba. The Power in High Speed RAMs. \| FAMILY TYPE: HIGH SPEED MOS STATIC RAMS \| \|----------------------------------------\| \| PART \| ORGN SPEED \| PACKAGE \| SAMPLES \| PROD'N \| \|------\|------------\|---------\|---------\|--------\| \| TMM2068D \| 4KX4 \| 55 \| 20P \| CDIP \| NOW \| NOW \| \| TMM2068D \| 4KX4 \| 45 \| 20P \| CDIP \| NOW \| NOW \| \| TMM2068D \| 4KX4 \| 35 \| 20P \| CDIP \| NOW \| NOW \| \| TMM2068AD \| 4KX4 \| 25 \| 20P \| CDIP \| JUNE \| 3Q'86 \| \| TMM2078D \| 4KX4 \| 55 \| 22P \| CDIP \| NOW \| NOW \| \| TMM2078D \| 4KX4 \| 45 \| 22P \| CDIP \| NOW \| NOW \| \| TMM2078D \| 4KX4 \| 35 \| 22P \| CDIP \| NOW \| NOW \| \| TMM2078AD \| 4KX4 \| 25 \| 22P \| CDIP \| JUNE \| 3Q'86 \| \| TMM2018D \| 2KX8 \| 55 \| 24P \| CDIP \| NOW \| NOW \| \| TMM2018D \| 2KX8 \| 45 \| 24P \| CDIP \| NOW \| NOW \| \| TMM2018D \| 2KX8 \| 35 \| 24P \| CDIP \| NOW \| NOW \| \| TMM2018AD \| 2KX8 \| 25 \| 24P \| CDIP \| 3Q'86 \| 4Q'86 \| \| TMM2089C \| 8KX9 \| 45 \| 28P \| SBDIP \| NOW \| 3Q'86 \| \| TMM2089C \| 8KX9 \| 35 \| 28P \| SBDIP \| NOW \| 3Q'86 \| \| TMM2088P \| 8KX8 \| 45 \| 28P \| DIP \| 3Q'86 \| 4Q'86 \| \| TMM2088P \| 8KX8 \| 35 \| 28P \| DIP \| 3Q'86 \| 4Q'86 \| \| TC55416P \| 16KX4 \| 45 \| 22P \| DIP \| JUNE \| 4Q'86 \| \| TC55416P \| 16KX4 \| 35 \| 22P \| DIP \| JUNE \| 4Q'86 \| \| TC55417P \| 16KX4 \| 25 \| 22P \| DIP \| UNDER DEVELOPMENT \| \| TC55417P \| 16KX4 \| 45 \| 24P \| DIP \| JUNE \| 4Q'86 \| \| TC55417P \| 16KX4 \| 35 \| 24P \| DIP \| JUNE \| 4Q'86 \| \| TC5561P \| 64KX1 \| 70 \| 22P \| DIP \| NOW \| NOW \| \| TC5561P \| 64KX1 \| 55 \| 22P \| DIP \| NOW \| NOW \| \| TC5562P \| 64KX1 \| 55 \| 22P \| DIP \| NOW \| NOW \| \| TC5562P \| 64KX1 \| 45 \| 22P \| DIP \| NOW \| NOW \| NOTE: DIP = PLASTIC CDIP = CERDIP SBDIP = SIDE BRAZED CERAMIC TOSHIBA. THE POWER IN HIGH SPEED SRAMS. © 1986 Toshiba America, Inc. VMIC™ “THE VMEbus DRIVER” Don’t Miss the Next VMIC Bus! Our full line of VMEbus Modules includes: • Host Computer Interfaces • VME-to-VME Links • Digital I/O • Analog I/O • Change-of-State • Counter/Pulse Rate • Synchro/Resolver • Intelligent I/O Controllers • Serial I/O • VMEbus Systems • Analog Backplanes • Synchro Backplanes One Year Warranty On All Modules VME MICROSYSTEMS INTERNATIONAL CORPORATION 12021-N SOUTH MEMORIAL PARKWAY HUNTSVILLE, ALABAMA 35803 (205) 880-0444 Circle 120 on reader service card INTERNATIONAL SALES DIAMOND POINT INTERNATIONAL, INC. 11351 PEARL ROAD CLEVELAND, OH 44136 (216) 238-0200 TELEX: 4332049 (DIAPT) EUROPEANS ARE COUNTING ON UNIX TO FIGHT IBM X/OPEN GROUP WORKS FOR A STANDARD OPERATING SYSTEM by Robert T. Gallagher PARIS European computer makers, with some help from U.S. companies, are once again trying to unite and head off IBM Corp. Although the field is littered with the remains of failed attempts at transnational cooperation, this time the companies have a plan that is simple and so far appears workable. The manufacturers—seven from Europe plus two from the U.S.—want to keep IBM from dominating the European minicomputer and high-level workstation market as it does the mainframe business (with 65% of the total) on the Continent. They plan to do this by establishing AT&T Bell Laboratories' Unix operating system as a Europe-wide standard. That would enable each company to retain its identity in the fiercely nationalistic European markets even as it produces systems that are compatible with other systems. To accomplish this, the nine firms have formed the X/Open Group. It is a logical extension of a European standardization movement that is far more advanced than anything that has so far developed stateside [Electronics, April 28, 1986, p. 48]. The first major step of that movement was the establishment of the Open Systems Interconnection protocols by the International Organization for Standardization and the International Telegraph & Telephone Consultative Committee. The protocols serve as standards for interfacing and networking data-processing systems, and they have been widely accepted worldwide for all levels of computers. With the success of OSI in mind, the X/Open group is set on extending the standardization effort to operating systems. The reasoning behind the move is simple enough: with both networking interfaces and operating systems standardized, purchasers of data-processing systems will be able to enjoy a significant degree of supplier independence. With X/Open, users could connect heterogeneous equipment through OSI links and then achieve applications portability with Unix. European equipment manufacturers reckon X/Open will effectively preempt any attempt by IBM to establish de facto minicomputer standards, as it has for mainframes and personal computers. Many industry observers feel that the move toward Unix is so strong that IBM itself will have no choice but to offer it for its own minicomputer line. The idea of forming the X/Open Group was a 1984 brainchild of Robb Wilmot, who was chief executive officer of England's International Computers Ltd. plc before cofounding Silicon Structures [Electronics, Sept. 8, 1985, p. 25]. It began with only five members. It now includes, in addition to ICL, Dutch multinational Philips, France's Bull, Italy's Olivetti, Sweden's Ericsson Information Systems, and West Germany's Nixdorf Computer and Siemens, as well as the European subsidiaries of America's Digital Equipment and Sperry. EXCLUSIVE CLUB. The X/Open member firms expect their group to expand over the years, and they already have applications from a number of companies, although they decline to name them. They will make a point of adding new members slowly, however, as they feel that the group's exclusivity has been one of the keys to its operational efficiency. The group's initial objective was twofold: to promote a standard version of Unix and to foster the acceptance in business applications that it has in the \| Maker \| Country \| Minicomputers and work stations \| Operating systems \| \|-------\|---------\|---------------------------------\|-------------------\| \| Bull \| France \| SPS5, SPS7 \| Unix \| \| \| \| SPS9 \| Ridge \| \| \| \| DPS-9 \| GCOS or Unix \| \| \| \| Questar 400 \| Unix \| \| DEC \| U.S. \| VAX \| VMS, Ultrix 32 \| \| \| \| Microvax \| MicroVMS, Ultrix-32M \| \| \| \| VAX stations \| MicroVMS, Ultrix-32M \| \| \| \| PDP-11 family \| RT-11, RSX-11, RSX-11M, RSX-11M+, RSTS/E, CTS-300, MicroPower Pascal, COS-310, Ultrix-11 \| \| Ericsson \| Sweden \| 2500 \| Eritron \| \| \| \| Work stations (licenses from Sun) \| Unix \| \| \| \| 286 \| MS-DOS, Xenix \| \| ICL \| England \| CLAN 3, CLAN 7 \| Unix \| \| \| \| DRS 300 \| Unix and C-DOS \| \| \| \| 3300 \| Unix \| \| Nixdorf \| West Germany \| 8810 \| MS-DOS \| \| Olivetti \| Italy \| M-30, M-40, M-60 \| MOS \| \| \| \| 3B2/300, /310, /400; 3B5, 3B15 (from AT&T) \| Unix \| \| Philips \| Holland \| P800 \| MAS \| \| \| \| P4000 \| DINOS \| \| \| \| PTS 6000 \| TOSS \| \| Siemens \| West Germany \| SICOMP M20, M30, M60-70, R10 \| AMBOSS-4 \| \| \| \| SICOMP R20, R30, R40 \| AMBOSS-3 \| \| Sperry \| U.S. \| 5000/20, /40, /50, /60, /80, /90 \| Unix \| \| \| \| 1100 \| SX1100 \| SOURCE: ELECTRONICS Unsurpassed CMOS experience gets you off to a good start. RCA offers the world's broadest range of speeds in CMOS logic, including high-speed CMOS and new Advanced CMOS Logic (ACL) that's as fast as FAST.* Our ASICs offer designer-friendly CMOS gate arrays and standard cells, with a growing library of state-of-the-art macrocells. We offer CMOS microprocessors, with a complete line of peripherals and memories. And we're getting you off to a good start in CMOS flash converters. FAST is a trademark of Fairchild Semiconductor Corp. New partnerships are rolling out the products for your future. It takes innovation and cooperation to stay ahead today. That's why we've teamed up in the RCA/Sharp Microelectronics joint venture. Our factory-of-the-future will be capable of producing submicron geometries in space-age clean conditions. But the benefits of our shared expertise are available now. We've just rolled out a new line of EPROMs and macrocells (including a 32-bit slice µP) resulting from an alliance with Wafer Scale Integration. And our new modem-on-a-chip comes from cooperation with Silicon Systems. Shouldn't you have a CMOS partner who knows how to make partnerships work? 50 ppm quality gives you the jump on the competition. We're breaking through to new levels of quality and reliability. In "quality partnerships," we're helping customers eliminate incoming inspections. And helping them squeeze more performance from their designs by tightening specs to the RCA quality zone. If you're interested in just-in-time delivery, you need RCA quality standards. Your winning designs deserve world-class CMOS support. Today’s designs must compete with the best in the world. Be sure yours are backed by a CMOS supplier whose products, research, manufacturing and quality live up to the world-class challenge. Call your RCA sales office or distributor to get a CMOS partnership started today. Or write: RCA Solid State, Box 2900, Somerville, NJ 08876. Together, we can take on the world. TIME EXPOSURES It took NASA years to get these shots. We helped keep track of the time. - Precision Timing Instrumentation - GPS Time/Frequency Monitors - Video Data Equipment - Tape Search Systems For information, call or write. Datum Inc Timing Division 1363 S. State College Boulevard Anaheim, California 92806 714/533-6333 Taiwan Electronics Show 1986 Oct. 7-13, 1986 Exhibition Hall, Taipei World Trade Center More than 500 exhibitors will be displaying their wares — consisting of electronics products of all kinds — at this outstanding international show. See for yourself why Taiwan is emerging as a major electronics production center for the world market. Organizer: china external trade development council 9-10th Fl., 201 Tun Hwa N. Rd., Taipei, Taiwan, R.O.C. Telex: 21676 CETRA. Tel: (02) 7151515. Sponsor: Taipei World Trade Center 5 Hsin Yi Rd., Sec. 5, Taipei, Taiwan, R.O.C. Telex: 28094 TPEWTC. Tel: (02) 7251111. Branch Offices: New York: Telex: 426299 CETDC NY. Tel: (212) 532-7055. Chicago: Telex: 253726 FAREAST TR CGO. Tel: (312) 321-9338. San Francisco: Telex: 171521 FETS SF. Tel: (415) 788-4304, 788-4305. UNIX WORK STATIONS COST ONLY $5,000 PER MIPS INTERGRAPH USES CLIPPER CHIP SET TO BOOST PRICE-PERFORMANCE RATIO Last spring, a team from Fairchild Semiconductor Corp. flew to Huntsville, Ala., to sell Intergraph Corp. on the idea of designing a new generation of graphics work stations around the chip maker's then-unannounced Clipper 32-bit microprocessor. Thus were born the work stations that Intergraph introduced last week at the Design Automation Conference in Las Vegas. Not only was the Fairchild trip successful, but the two companies then joined in a strategic partnership that Bruce E. Imsand, Intergraph's vice president for systems development, calls a "true symbiotic relationship." The teaming has already produced an amazingly fast product-development effort. In just one year, Fairchild has won its first Clipper design-in and Intergraph has introduced a low-cost, stand-alone Unix work station it claims will cost customers far less per million instructions/s than any competing product. In addition, Fairchild got an alpha-test customer for the Clipper. "The Clipper has the speed and horsepower needed for today's design-analysis-simulation problems," says Hal Barbour, Intergraph's executive manager of electronics marketing. A year ago, customers were demanding a lot more power for stand-alone applications; to accomplish this, Intergraph was looking at microprocessor designs from Intel Corp. and National Semiconductor Corp., as well as from Fairchild. But when Howard G. Sachs, general manager of Fairchild's advanced processor division, launched his pitch, it sounded very familiar to Imsand; indeed, he had already pitched the same approach to his bosses at Intergraph. "Both Bruce and I had the same idea of architecture," Sachs adds. "We were even planning to use the same operating system [AT&T's Unix System V] and compilers [from Green Hills Inc.]," says Imsand. The 32C family, which executes instructions at an average of 5 mips, is priced starting at $25,000, or only $5,000 per mips. This figure compares with a cost of $10,000 to $13,000 per mips for such competitive systems as Apollo's, Intergraph engineers claim. The system also performs double-precision floating-point calculations at 2 million whetstones/s, they maintain. Up until two weeks ago, though, Barbour didn't know if the Clipper-powered work station would be ready in time to announce at the design automation conference. But Intergraph decided to go for it, he says, when it was able "in the past month or so to get some of its applications up and going on Fairchild's first silicon." CLIPPERS ARE READY. Teaming up, Barbour says, helped both companies accelerate their product development, each saving from three to six months. Intergraph plans to begin production shipments by October, and Sachs says Fairchild will be in volume production by then. "We've already got several hundred Clipper CPUs built up right now." Intergraph's InterAct 32 and InterPro 32 work stations, introduced a year ago and designed around National's 32-bit 322032 microprocessor, can be upgraded by simply plugging a Clipper module into the main CPU board. Intergraph's Clipper module is packaged slightly differently than the standard module that Fairchild sells. Intergraph had to write a new code generator for the compiler to produce code for the Clipper instruction set and do a little tweaking of the operating system. The new work stations run on the basic Unix System V operating system, can emulate Digital Equipment Corp. VT-100 and -220 terminals, and come with screen-management software. Each has 6 megabytes of main memory, an 80-megabyte hard-disk drive, and a 1.2-megabyte floppy-disk drive. For full-time production environments, the InterAct 32C, starting at $40,000, has dual screens, ergonomic design, and a built-in digitizing tablet. For the designer, the InterPro 32C, beginning at $25,000, features a compact desktop design suitable for the office and is available with either 15- or 19-in. color monitors that can display 322 colors from a palette of 4,096. Both work stations have addressable resolutions of 1,184 by 884 pixels with 60-Hz noninterlaced refresh rates. A 10-MHz 80186 microprocessor runs all I/O operations and gives them the ability to run PC-DOS applications. Built-in local-area networking is provided through an IEEE 802.3 (Ethernet) link to access a data base in a DEC VAX host computer. The work station architecture consists of three independent subsystems—processor/memory, networking, and graphics management—with defined interfaces. Users can therefore implement new technology in one subsystem without affecting the rest of the system. Intergraph is just the first of several customers that will announce commitments this summer to use Clipper in their product families, according to Fairchild. Clipper consists of three CMOS chips, a CPU with on-chip floating-point execution unit, and two combination cache/memory-management ICs. The two cache chips are linked to the CPU over a dual-bus architecture; one 32-bit bus is dedicated to instructions, the other is dedicated to data. —Robert W. Henkel Intergraph Corp., 1 Madison Industrial Park, Huntsville, Ala. 35807. Phone (205) 772-2000 [Circle reader service number 340] CAD/CAE ACCELERATOR ROARS ALONG AT 10 MIPS Users of Cadnetix' soup-to-nuts line of computer-aided-engineering and design equipment can add a general-purpose accelerator that performs 10 million instructions/s. The CDX-760 general-purpose engine, a card with up to 8 megabytes of on-board memory, is the first system to make use of the high-speed, reduced-instruction-set-computer technology chip set from Mips Computer Systems Inc. An option permits the CPU card to be used in parallel-processing applications with other Cadnetix design tools. The accelerator, one of two major Cadnetix introductions at the Design Automation Conference last week in Las Vegas, is designed to be fully compatible with the company's application-specific engines, such as the CDX-77000 Simulation Engine, the CDX-7900 Physical Modeling Engine, the CDX-79000 Analysis Engine, and the CDX-7100 and 7200 Database Servers [Electronics, Feb. 17, 1986, p. 54]. The company also announced a reconfigured version of its analysis engine that allows users to select a configuration exactly suited to their needs. The analysis engine, which can include the CDX-760 GP Engine, comes in setups for simulation, physical modeling, compilation, or any combination of the three. The CDX-760 GP Engine is designed for applications where the overhead required for application-specific engines does not justify their use. That's why the company is initially offering the general-purpose accelerator with Berkeley Spice and data-base compilation—applications where the use of a general-purpose accelerator is most cost-effective. PARALLEL POTENTIAL. The GP Engine has an interprocessor-communications port built into it, making parallel processing possible. In addition, Cadnetix modified another product, a microcoded bit-slice application-specific engine, to include an interprocessor communications facility. In the future, it will be possible to configure any combination of up to five engines in a single chassis. This would give the engineer local access to physical modeling and data-base sharing, and local acceleration of compilation, logic simulation, and Spice execution. System performance would be boosted because the user would not encounter network traffic or competition between nodes for remote resources. The parallel-computing capability is something Cadnetix is now putting into its full product line in anticipation of applications to be released in the future. The CDX-760 gets its speed from the Mips Computer architecture, which uses RISC technology [Electronics, May 5, 1986, p. 56]. The Mips chip set offers high speed at a low price, and Cadnetix says its new system will operate 3 to 10 times as fast as a Digital Equipment Corp. VAX 11/780. Implementing the Mips RISC architecture in a system compatible with Cadnetix' other products without sacrificing speed was simple, the company says, thanks to Mips Computer's optimizing compiler system, which translates high-level-language instructions into machine code suited to the chip set's RISC design. The compiler consolidates redundancies and passes data to on-chip registers, where it is accessed more easily, to keep processing speed up. Cadnetix also plans to offer a C compiler and debugging utilities for the GP Engine later this year. The GP Engine is priced from $29,900. The data-base compilation application costs an additional $5,000, and pricing for Spice has not yet been set. A Unix operating system with a C compiler and debugging utilities will cost $120,000. Pricing for the Analysis Engine ranges from $84,800 to $127,900, depending on the configuration chosen. All the new offerings will ship in the fourth quarter. —Tobias S. Naegele Cadnetix Corp., 5757 Central Ave., Boulder, Colo. 80301. Phone (303) 444-8075 [Circle 343] MS-DOS PACKAGE WRITES USER INTERFACES Part of writing any application program is setting up the screens and control routines that form the user interface. This universal programming job is greatly simplified by Trilobyte Software Systems' software-development package called Access, which is now available for MS-DOS-based computers. Access not only automates the processes of designing screens and determining security levels, it also serves later to perform remote maintenance over asynchronous telecommunications links. Access was originally written to run on IBM Corp. mainframes under the Customer Information Control System. Its author, Stephen J. O’Kane, has now produced a functionally identical version in 8088 assembler language, giving programmers the same capabilities on an IBM Personal Computer or compatible. It’s intended for corporate management information system departments, value-added resellers, and original-equipment manufacturers. Access includes separate modules for security, screen control, and maintenance. The security system, which protects application files from unauthorized access, is an assembler routine that uses hidden directories and files to conceal passwords and data files. Only a sophisticated programmer who knows how to disassemble files can access the data, says Trilobyte president Jerome Draper. WINDOW CONTROL.* The screen-control module lets the programmer set up fields and windows, manage the cursor, define function keys, set system parameters, and do other jobs necessary to let a user run an application. Like the security module, the screen-control segment is menu-driven. The third module sets asynchronous communications parameters and makes it possible for Trilobyte, or a value-added or OEM reseller, to perform remote support over dial-up lines. Rather than walk a naive user through an unfamiliar program on the telephone, the seller can take control of the remote system and fix bugs on-line. Access interfaces with compiled programs written in Basic, Pascal, Cobol, and C. With a compiler plus software for manipulating files, the system can be used to write tailor-made applications on the IBM PC, Draper says. Remote access requires that the remote station have a run-time version of Access, which is provided free by Trilobyte. Access itself, which is not copy-protected, sells for $245 and is available now. The program includes source code for 75 assembler subroutines that set up screens and perform other functions. A 300-page documentation book accompanies the software. Trilobyte Software Systems, 295 Los Angeles Blvd., San Anselmo, Calif. 94960. Phone (415) 457-3431 [Circle 341] MENUS SET UP UNIX DATA BASES An application-program generator and data-base manager cuts by 80% the time it takes to develop customized Unix applications. Parameter Driven Software’s PDS-Adept Unix product complements the company’s earlier Adept packages for MS-DOS and PC-DOS-based machines such as the IBM Corp. Personal Computer, PC/XT, and PC AT, as well as for systems using hardware from Convergent Technologies that rely on the CTOS or BTOS (Burroughs version) operating system [Electronics, Nov. 4, 1985, p. 34]. The addition of PDS-Adept Unix to the Adept family will be especially useful to companies that have a variety of small computers based on different operating systems, says PDS president Patrick K. Comeaux. That’s because the company offers an optional conversion filter with which programs created with PDS-Adept Unix can be converted automatically for use on systems that run under MS-DOS, PC-DOS, CTOS, or BTOS. Similar conversion filters sold for the company’s earlier MS-DOS/PC-DOS- and CTOS/BTOS-based Adept packages also allow automatic conversion from either of those operating systems to either of the others. Like its predecessors, PDS-Adept Unix enables computer users who are not programmers to create customized, ready-to-run programs simply by filling in the blanks on display-screen menus. For example, the user might be asked, “Should customers’ names appear in alphabetical order?” Or the user might be directed to “Fill in the name and model number on this product.” The parameters established by the user’s answers then drive the Adept package’s processing functions for manipulating the data base. A program that might take two to three months to write using conventional programming languages such as Cobol, Pascal, or C could be created in only two weeks using PDS-Adept Unix, says Comeaux. Some simple programs can be created in minutes, he says. PDS-Adept Unix will run on computers operating under AT&T Bell Laboratories’ Unix System V, including the AT&T Co. 7300 and 3B series, the Burroughs Corp. XE 550 and Convergent Technologies Inc. Miniframe, Megaframe and Mightyframe series. PDS will also offer a suite of 10 Unix System V-based general-purpose business-application programs created using the Adept package. Business users can use PDS-Adept Unix to easily modify these programs to suit their needs and can also use the new Unix-based program generator to interface Adept-created packages with existing in-house software. Package prices depend on the target machine. For Unix-based personal computers such as the 7300 or 3B/1, the package is priced at $995. For larger multiuser systems such as the 3B/5, the XE 550 or Megaframe-based systems, the package sells for $4,000. The conversion filter for PDS-Adept Unix sells for $750. PDS-Adept Unix is available now from 180 U.S. dealers, and internationally through a network of distributors and original-equipment manufacturers. –Wesley R. Iversen Parameter Driven Software Inc., 30800 Telegraph Rd., Birmingham, Mich. 48010. Phone (313) 540-4460 [Circle 342] REVOLUTION BY ANALOGY. REVOLUTION BY ANALOG. Our revolutionary new 12-bit, 5μs CMOS A/D converter eliminates specification compromises. It gives you the speed and resolution you need without the power, size and cost penalties of hybrids. Made possible by our advanced monolithic BiMOS technology, the AD7572 features: low 135 mW power dissipation; 0.3" 24-pin DIP or 28-pin LCC packages; on-chip buried zener reference; and 90 ns bus access time. The most revolutionary specification of all is the price: 5μs versions from $39.10 (1,000's) and 12.5 μs versions from $29.75 (1,000's). The AD7572 is one of our hundreds of innovative data acquisition components, all aimed at making it easier to design signal processing subsystems. For more information, call Applications Engineering at (617) 329-4700 ext. #3596 or #3597. Or write Analog Devices, Inc., P.O. Box 280, Norwood, MA 02062. TOUCH TOUCHNOROLOGY 1: the blending of innovative concepts and advanced engineering design, to create the ultimate in touchscreen technology 2: offering seven (7) standard sizes (from 5" to 13"), as well as customs of virtually any size 3: continuous resolution with excellent linearity over the full scale 4: the flexibility of providing custom designs quickly, at little more than the cost of standards 5: available as screens alone, screens with controllers or the Prototouch 1 Kit PRO•TO•TOUCH 1 KIT 1: the beginning of your next generation of products 2: contains two (2) touch screens, one (1) controller, connecting hardware and applications software 3: get started today ...Call (619) 268-8555 for more information KENNEDY TOUCH SCREENS A product of KENNEDY TECHNOLOGY, INC. 4830 Viewridge Avenue • San Diego, CA 92123 • (619) 268-8555 • TWX: 4994941 KTECH Circle 135 on reader service card CRYSTAL OSCILLATOR SUITS SURFACE MOUNTING MF ELECTRONICS PACKS QUARTZ CRYSTAL AND ALL CIRCUITRY IN PLASTIC PACKAGE THAT IS SAFE FOR PICK-AND-PLACE GEAR A tiny thick-film hybrid oscillator for computer clocking applications resolves a problem in surface-mount technology. MF Electronics Corp.'s series M1000 crystal-stabilized oscillator is housed in a J-leaded plastic package that is handled without damage by SMT pick-and-place machines. Other companies have supplied quartz crystals housed by themselves in surface-mount packages. And some complete crystal oscillators are available in 24-pad ceramic leadless chip carriers. But ceramic, unlike plastic, is abrasive, and wears down the pick-and-place system's "fingers." In addition, fragile ceramic packages easily crack during shipping, handling, and assembly on pc boards. With only three circuit elements—a quartz crystal, a bypass capacitor, and MF Electronics' custom Clockchip—the oscillator is inherently reliable since it minimizes the number of wire-bond connections. ALL THE CIRCUITRY. The Clockchip contains all the oscillator's analog and digital circuitry. This chip uses an internal reference source to define the crystal-excitation voltage. It also has a gain-controlled amplifier functioning within an error-detecting feedback loop to sustain crystal excitation through temperature and supply-voltage variations and component changes due to aging. The Clockchip also permits either a fundamental or third-harmonic mode of oscillation to be selected by wire-bond connections to the desired filter. The chip has a Schmitt trigger and Schottky gate to produce fast-rise-time square waves for circuit timing. The Clockchip, along with the other two elements, are attached to a thick-film circuit pattern on an alumina hybrid substrate. After wire bonding and testing, the finished hybrid is hermetically sealed in a metal can, and the entire assembly is then molded into the plastic surface-mount package. A high-melting-point solder is used in the oscillator's construction, permitting the device to undergo prolonged exposure to the high temperatures of SMT soldering systems. Overall, the M1000 oscillator, including its J leads, measures only 0.45 by 0.49 in., which amounts to an area of less than $ \frac{1}{4} $ in.$^2$. Its seated height is less than 0.2 in. Because the J leads are resilient, they can flex during pc-board insertion into system slots and provide protection against vibration during shipping and field use. The M1000 parts run at any customer-defined frequency from 4 to 48 MHz and have guaranteed frequency stabilities to within 0.01%, 0.05%, and 0.1%. Power consumption from the 5-V supply is typically 35 mA; the oscillator delivers a 4-V square-wave output that can drive up to 10 TTL loads. The surface-mount oscillators sell for less than $2 in production quantities. Evaluation units at standard frequencies are available from stock and production orders can be shipped in four to eight weeks. The oscillators are supplied in plastic antistatic tubes, or in 24-mm tape and reel form. Jerry Lyman MF Electronics Corp., 10 Commerce Dr., New Rochelle, N.Y. 10801. Phone (914) 576-6570 [Circle 400] SURFACE-MOUNT AMPS HAVE 9.0-dB GAIN Two surface-mountable amplifiers have a minimum gain of 9.0 dB. Typically, the PlanarPak devices have gains of 10.0 dB over their full 500-MHz-to-2-GHz frequency range. The models PPA-2012 and PPA-2013 have maximum noise figures of 4.0 and 5.5 dB and typical power output of +13.0 and +21.0 dB, respectively. Both amplifiers have maximum voltage standing wave ratios of 2.0:1. The PlanarPak package is only $ \frac{3}{8} $-in.$^2$ and weighs 0.5 gram. Its surface-plane leads may be trimmed off for reflow soldering. In lots of one to nine pieces, the PPA-2012 is priced at $290 each and the PPA-2013 at $330 each. Both are available now. Avantek Inc., 3175 Bowers Ave., Santa Clara, Calif. 95054. Phone (408) 970-2583 [Circle 403] DC-TO-DC CONVERTER HAS I/O ISOLATION By using a new design for the feedback loop, Rifa is able to add input/output isolation to its PKA line of 25-W miniature dc-to-dc converters without cutting back on the targeted reliability level of 200 years (at +45°C) mean time before failure. The isolated versions are available in 24-V and 48-V input ratings. Three models of each type are available: one with +5 V at 5-A or +12 V at 1-A outputs; one with only 5 V at 5-A outputs; and one with +12 V at 2.5-A outputs. The converters measure 2.9 by 2.9 by 0.7 in. and are available on printed-circuit boards or as a chassis-mountable version housed in an aluminum case for convection cooling. The isolated PKA power supplies sell for $122 each in small quantities. Samples are available now. Rifa Inc., Greenwich Office Park #3, P.O. Box 3110, Greenwich, Conn. 06836. Phone (203) 625-7300 [Circle 404] PANEL-MOUNT PRINTER USES THERMAL PAPER The Mini-Printer-24 digital panel printer connects to any Centronics parallel port and produces dot-matrix printouts on thermally sensitive paper. The 24-column printer has a character set that includes 250 letters and symbols. A built-in memory and character generator simplifies the interface to the computer or data-logging device. The printer allows upright and inverted printing and it can also print double-wide characters. Characters are printed in a 5-by-7-dot matrix. Available from stock, the unit sells for $199 each. MetaByte Corp., 440 Myles Standish Blvd., Taunton, Mass. 02780. Phone (617) 880-3000 [Circle 405] $186 CRT DISPLAYS 1,024 BY 1,024 PIXELS The 300MDX monochrome CRT, which has a 1,024-by-1,024-pixel resolution, is priced as low as $186 each in large quantities. The CRT comes in 9-, 12-, or 14-in. sizes and can be provided in kit, chassis, or cabinet configurations. The 64-kHz CRT's deflection system is compatible with all popular 90° and 110° CRTs. It accepts either Schottky TTL or emitter-coupled-logic video signal inputs with full gray-scale display and requires a 24-V input. The 300MDX is available now. Digitran/Computron, 3100 New York Dr., Pasadena, Calif. 91107. Phone (818) 791-5600 [Circle 406] THERMAL PLOTTER DOES A PAGE IN 15 S Superplot-80L, an 80-column thermal printer and plotter, can produce a typical page of graphics in less than 15 seconds, thanks to built-in vector graphics routines. The plotter, which measures 4.2 by 10.7 by 10.5 in., can be mounted in a front panel. The plotter uses a linear-array thermal printhead that does in excess of 71 dot-lines/s and near-letter-quality alphanumerics at over 500 characters/s. The company says the printhead has a resolution of 100 dots/in. and a life of 100 million dot-lines or 17 miles of paper. The Superplot-80L has an 8-in.-wide printable field on 8½-in.-wide thermal roll paper. A top-of-form optical sensor indexes page tops for proper alignment. The unit comes with three standard interfaces: Centronics, 8-bit parallel, and page-level vector graphics. Available 90 days after ordering, the Superplot-80L is priced at $2,095 in single quantities. Gulton Industries Inc., Graphic Systems Division, Gulton Industrial Park, East Greenwich, R.I. 02818. Phone (401) 884-6800 [Circle 408] CMOS REGULATORS ARE 85% EFFICIENT Two low-power CMOS dc-to-dc inverting converter/regulators are targeted for operation in the range from 5 to 500 mW. A single inductor, a diode, and two set resistors are the only external components required to convert an unregulated positive voltage to a regulated negative voltage. The MAX634 and MAX4391 achieve efficiencies up to 85% and can operate over a voltage range of +2 V to +16.5 V and +4 V to +16.5 V, respectively. The output voltage may be up to −20 V, with the restriction that the total input-output voltage differential must never exceed 24 V. The chips, which can replace bipolar parts, have an operating current of only 70 μA. Housed in 8-pin Cerdips, the devices are available in the commercial, industrial, and military temperature ranges. The MAX4391 is priced starting at $2.71 and the MAX634 starts at $2.90, both in lots of 100 pieces. Maxim Integrated Products, 510 N. Pastoria Ave., Sunnyvale, Calif. 94086. Phone (408) 737-7600 [Circle 411] SMALL CONVERTERS PUT OUT 150 W The ERD series of dc-to-dc converters is available in models that range in power from 30 to 150 W and feature 300-kHz switching frequencies. The 30-W models accept inputs of either 24 V or 40 V and are priced at about $135 each. The 60-W supplies go for $160 each and the 150-W models at $225. The high frequency results in significant size and weight reductions compared with earlier designs. The converters are hybrid microcircuits containing surface-mounted devices on a ceramic substrate. They boast a mean time before failure of 250,000 h. The converters are available from stock. Kepco Inc., 131-38 Sanford Ave., Flushing, N.Y. 11352. Phone (718) 461-7000 [Circle 409] TRANSDUCER WARBLES BETWEEN TWO TONES The KBT-33SB-2T piezoelectric transducer has a dual resonant frequency near 1 kHz and a resonant impedance of 3.5 kΩ maximum. The device's low, easy-to-hear tones are produced when sound is warbled between frequencies. The solid-state device is designed for computer printers, telephones, modems, and security systems. It features four pins for mounting stability and a 12.5-mm stance for easy insertion in 30-mm lead spacings. The transducer, which operates without inductive coils, has a sound pressure level of 70 dB minimum and an input voltage of 20 V peak-to-peak. In lots of 10,000 pieces, the transducer sells for 58¢ each. Delivery takes eight to ten weeks after ordering. Kyocera International Inc., Electronic Components Group, 11425 Sorrento Valley Rd., San Diego, Calif. 92121. Phone (619) 454-1800 [Circle 407] DIGITAL TIMER CUTS THE ANALOG CONNECTION AMD's PROGRAMMABLE EVENT GENERATOR PUTS OUT 12 WAVEFORMS TO TACKLE TOUGH TIMING PROBLEMS BIG SAVINGS. AMD's programmable event generator replaces $30 worth of components. Even in the most advanced digital systems, analog circuitry plays a major role in timing. But now a monolithic digital timer from Advanced Micro Devices lets engineers rid their designs of hard-to-use analog delay lines, cumbersome counters, and RC networks. In addition, the AM2971 programmable event generator can serve as a general-purpose user-programmable timing and waveform generator. The Am2971 achieves a resolution as precise as 10 ns between system timing events, enabling designers to tailor system timing for peak performance. It offers 12 programmable, registered output waveforms, giving users the flexibility to define 12 independent timing signals. "The chip is the first in a family of monolithic chips for solving timing problems previously handled by hybrids," says Larry Wittenbaugh, the chip's marketing manager. START/STOP CONTROL. The chip also features programmable control of the timing sequence for start and stop functions. Users can pick one of eight start addresses to begin the timing sequence. The event generator's programmability and its many outputs mean it can replace several delay lines at one time. The chip suits a wide variety of timing-control applications in digital systems. It can replace analog delay lines and combinatorial logic required to accurately generate the complex clock signals required by address-multiplexed dynamic RAMs. For example, it can generate timing for the Am2968 Dynamic Memory Controller and Am2969 Memory Timing Controller. Typically, the timing controller generates the row-address strobe input signal whenever a refresh or a memory cycle is requested. This signal provides a leading edge to the Am2971's trigger output, thus initiating the user-defined timing sequence. The chip is ideal for designing page-mode-addressed systems, according to Bruce Threewitt, manager of product planning. "In that situation, the Am2971 would replace two delay lines and some combinatorial logic. When given the signal from a microprocessor to start, the 2971 bursts out a canned waveform until the processor says stop." In contrast, a delay line only advances one clock edge. Threewitt says that although the part was originally designed for DRAM timing applications, AMD is finding that it has a number of other applications. These include bus timing, peripheral control, and timing control for bitmapped graphics. An on-board 70-to-100-MHz phase-locked-loop crystal oscillator can clock the event generator to an output 1/5 or 1/10 the PLL frequency. Clocking can also come from an existing system. Programming the Am2971 is done in the same way as a PROM. The chip's fuse links are made using AMD's platinum-silicide fuse technology. The 2971 is available now in 24-pin ceramic DIPs for $17.75 each in lots of 100 pieces. The handful of components it replaces averages between $25 and $30. In addition to cost savings and increased system reliability through the reduced parts count, there are a couple of second-order benefits, Threewitt says. For one, the part improves system performance. "If you use external logic and delay lines, you are still stuck with skew problems between the individual circuits," he says. "Also, because the Am2971 is purchased blank, parts inventory is reduced since it replaces all of your timing components." —Steve Zollo Advanced Micro Devices Inc., 901 Thompson Pl., P. O. Box 3453, Sunnyvale, Calif. 94088. Phone (408) 732-2400 [Circle 360] FASTEST MULTIPLYING DAC ACCEPTS TWO INPUTS With the introduction of the HDM-1210, Analog Devices Inc. boasts the industry's fastest 12-bit multiplying digital-to-analog converter. Typical analog settling time to within 1% is only 85 ns. Up to now, comparable DACs have been in the 500-ns range. The current-output hybrid DAC has a maximum digital settling time of 110 ns, a maximum analog settling time of 120 ns, and a 10-MHz analog bandwidth. Because of its small size—just 1.2 by 0.7 by 0.1 in.—the 1210 is an ideal replacement for the converter modules currently used in satellites. The modular converters, commonly called hockey pucks, measure about 2 by 2 by 1 in. Unlike its fixed-reference counterparts, a multiplying DAC operates with varying (or ac) reference signals. Such DACs put out a signal proportional to the product of the reference (the analog input) voltage and the fractional equivalent of the digital input number. The 1210 is designed for waveform generation and other applications in which one or two analog inputs must be multiplied by scale factors established with digital input words. The analog input signal can be a sine, triangle, or sawtooth wave. The output is a scaled version of the input, with digital input used as the scale factor. In a second operational mode, the anaVOLTAGE TOO. Analog Devices' 85-ns multiplying DAC has a current output but can be set to put out a voltage. log input voltage serves as the scale factor for the digital input code. The analog input, digital input, analog output, and external amplifier can be combined by using the hybrid's offsetting capabilities. These features let the 1210 work with unipolar or bipolar inputs. By itself, the DAC can perform one-quadrant multiplications. In tandem with a second HDM-1210 in a circuit using standard binary coding, two-quadrant multiplication is possible. Users can adjust analog output ranges by changing feedback resistors in the output driver. Analog Devices' Computer Labs Division in Greensboro, N.C., developed the converter using a thin-film substrate and laser-trimmed resistors. The biggest obstacle in designing the part, says design engineer Larry Barnes, was developing the operational amplifier. "We needed wide bandwidth and fast settling time," says Barnes. He designed the op amp using discrete transistors. The HDM 1210 comes in 24-pin ceramic or metal DIPs to provide a choice of temperature ranges. The BD grade is said to operate from $-25^\circ C$ to $+85^\circ C$. Analog says its SD/SDB-grade 1210 will operate from $-55^\circ C$ to $+100^\circ C$. It is housed in a hermetically sealed ceramic package at a military-standard facility. The unit puts out 10.24 mA. Designers requiring voltage outputs can connect the signal-out pin to the load-resistor pin. This would yield 1.024 V out, with no degradation in settling time. Running the HDM-1210 output into an amplifier would also yield voltage. Single-quadrant multiplication can range up to $-5$ V. The HDM 1210 has a guaranteed monotonicity over temperature and maximum gain $\pm 0.5\%$ of full scale. Guaranteed differential and integral linearity over temperature is $\pm 1$ least-significant bit at a maximum. In addition to waveform generation, applications include CRT display, vector generation, and megahertz-rate analog and digital attenuation. The converter is available for immediate delivery at $188 each. -Craig D. Rose Analog Devices Inc., 2 Technology Way, Norwood, Mass. 02062. Phone (617) 329-4700 [Circle 361] GOULD, ICT TO SELL EEPROM-BASED LOGIC Two more semiconductor companies have begun offering programmable logic devices based on EEPROM technology instead of fuse arrays. Fledgling IC house International CMOS Technology Inc., San Jose, Calif., has announced a family of products it calls PEELS, for programmable electrically erasable logic. The PEELS will also be available from Gould Inc.'s Semiconductor Division, where they are being fabricated in 2-$\mu$m CMOS. First on the scene was Lattice Semiconductor Corp., Portland, Ore., with its generic-array-logic chips [Electronics-Week, June 3, 1985, p. 56]. VLSI Technology Inc., also of San Jose, recently added GALs to its portfolio as a second source. PEELS and GALs both seek to replace PLDs based on fuse arrays, the lion's share of which are the programmable-array-logic parts sold by Monolithic Memories Inc. Unlike PALs, EEPROM-based PEEL or GAL chips can be reprogrammed quickly and repeatedly during the design-and-debug process; they are also useful in low-power designs where bipolar PALs are impractical. PEELS are said to consume only one third the power of their bipolar competition, even at the maximum clock rate. EEPROM-based PLDs can also be 100% tested at the factory, whereas fuse-based PLDs can only be fully tested after the designer has blown a pattern of fuses on the chip. This eliminates a hidden cost that can crop up with fuse-programmed parts—they may not work properly after programming. "I've heard horror stories where there were 25% programming rejects" with fuse-array parts, says Drew Osterman, president and cofounder of ICT. Furthermore, one PEEL can emulate some 20 different PALs, thanks to the use of I/O cells that can be configured for bidirectional I/O, registered or combinatorial feedback, and active-high or active-low outputs that are either combinatorial or registered types. This drastically reduces the number of part types that must be kept in inventory. PEEL chips can also implement over 100 logic configurations not possible with earlier-generation PLDs, says ICT. PEELS and GALs are in many ways similar; current versions of each, for example, are specified as having 25-ns maximum propagation delays. But ICT says it has achieved a PEEL die size... some 50% smaller than Lattice's equivalent GALs, and that it has been able to streamline the production process, reducing the number of masks required for fabrication by about 30%. The first PEEL to hit the market will be the 18CV8, which has 18 inputs and 8 outputs and can replace 20-pin PALs. Samples will be available in July, and parts in plastic DIPs will cost less than $5 each in lots of 100. Following later this year will be the 22CV10, which fills in for 24-pin PALs, and two PEELS that replace 20- and 24-pin field-programmable logic arrays from Signetics Corp. Third-party vendors of development tools and PLD programmers will support PEELS, says ICT. In addition, ICT and Gould will offer an evaluation system for about $800 that includes a programmer board for IBM Corp. Personal Computers and compatibles, an external programming module, and software. The software allows designers to develop their own logic patterns, or to translate conventional PLD design files into the PEEL format. --Jeremy Young International CMOS Technology Inc., 2031 Concourse Dr., San Jose, Calif. 95131. Phone (408) 434-0678 [Circle 363] Gould Inc., Semiconductor Division, 3800 Homestead Rd., Santa Clara, Calif. 95051. Phone (408) 246-0330 [Circle 364] DATA-PATH IC HAS THREE PORTS A 32-bit data-path chip from Weitek Corp., the WTL 3332 offers three ports for I/O-intensive signal-processing applications that also involve the processing of large quantities of data. Such applications include image processing, sonar, radar, motion control, and high-end telecommunications systems. The chip's architecture prevents I/O bottlenecks and allows most algorithms to run at maximum speed, the manufacturer says. It is fabricated in CMOS technology and comes housed in a 168-pin grid array. The device dissipates less than 1 W. For less computation- and I/O-intensive purposes, Weitek provides the model WTL 3132, a one-port version. Samples of both will be out this month. In quantities of 100, the WTL 3332 costs $425; the WTL 3132, $350. Weitek Corp., 1060 E. Arques Ave., Sunnyvale, Calif. 94086. Phone (408) 738-8400 [Circle 365] DENSEST FIFO YET HAS PARITY BIT FOR NETWORKS Dallas Semiconductor Corp. is jumping into the first-in first-out chip business with the highest-density FIFO on the market. Built in CMOS, the DS2001 has 2-K by 9 bits of FIFO storage and a 120-ns cycle time. Like standard FIFOs, the 28-pin 2001 has two ports and can accept data while emitting bits at a different speed, making the chip suitable as a buffer between asynchronous communications devices. The Dallas-based company designed the 2001 with transmission and networking applications in mind. The FIFO's 2-K-by-9-bit organization allows an extra error-checking-and-correcting parity bit to be added to conventional 8-bit byte-wide systems. Previously, the highest-density FIFOs available were 1-K-by-9-bit parts. MOSTEK SUPPORT. Thomson Components Mostek Corp., Carrollton, Texas, will second-source the 28-pin FIFO. In return for the support, Dallas Semiconductor is receiving laser-based chip-production equipment. "Swapping a chip design for production equipment is a bit unusual," admits marketing vice president Michael Bolan, who left a position at Mostek in 1984 to help found Dallas Semiconductor. "System engineers are faced today with the challenge of connecting different types of systems together. FIFOs are a response to that, supplying memory that can be viewed as a connector device or adapter," notes Bolan, referring to the demand by system designers for higher-density FIFOs. The higher densities yield greater tolerance for different input and output data rates in FIFOs, he adds. Dallas Semiconductor is now designing a 4-K-by-9-bit FIFO that will be plug-compatible with the 2001. The 2-K-by-9-bit FIFO will sell for $42.80 each in 100-piece quantities when it enters volume production in September. Dallas Semiconductor will make samples available this month. The new FIFO is made from a 2-μm CMOS process that produces a twin-tub, single-metal-layer, double-polysilicon structure. The die measures 44,800 mils². It has a 400-mW active power dissipation and 2.5-mW in standby mode. The chip works in the commercial range of 0°C to 70°C, but the company intends to make industrial-grade chips for operation at -140°C to +85°C. --J. Robert Lineback Dallas Semiconductor Corp., 4350 Beltwood Parkway, Dallas, Texas 75234. Phone (214) 450-0431 [Circle 362] PARITY BIT. Dallas Semiconductor's dense FIFO is organized as 2-K by 9 bits for parity checking. 16-BIT DAC IS SELF-CORRECTING The AD1147 16-bit digital-to-analog converter eliminates manual trimming of initial offset and gain errors by performing internal digital correction—an industry first, the company says. Two 8-bit latched-input DACs perform this correction through microprocessor control. Both differential and integral nonlinearity are ±0.00076% maximum (±½ least-significant bit). Settling time is 20 μs. Data may be loaded into the AD1147's input latches from 8- and 16-bit buses. The analog output range can be set for either voltage or current. The AD1147, housed in a 32-pin triple-width DIP, costs $152 each in hundreds. A less accurate—15 bits linear—model, the AD1148, features a separate data bus for the correction DACs; this is convenient in automatic-test-equipment applications with different cycles for correction and test functions. The price of the AD1148 in hundreds is $138, and both versions are available from stock. Analog Devices Inc., Literature Center, 70 Shawmut Rd., Canton, Mass. 02021 [Circle 368] Electronics Buyers' Guide Reservation Card If you Specify, Purchase, or Approve the purchase of Electronics Products, you are eligible to receive a free copy of the new 1986 Electronics Buyers' Guide. 1. Please answer all questions on this card. 2. Give your plant address, where your EBG will be delivered. 3. Mail this postage-paid card today. 1. Name ____________________________________________ Title _______________________________________________ Div. or Dept. _________________________________________ Company ____________________________________________ Company Address _____________________________________ City _________________________________________________ State Zip ____________________________________________ Signature Date _______________________________________ 1a. Do you wish to receive continue to receive the Electronics Buyers' Guide? □ Yes □ No. 2. Do you specify, purchase or approve the purchase of electronics products? □ Yes □ No. 3. Please check which best describes your company's business at your location: □ Manufacturing □ Distribution □ Retailing □ Other Specify _____________________________________ 4. Indicate your principle job function (place applicable number in box). 1. General and Corporate management 2. Design and development engineering 3. Engineering service (evaluation, quality control, reliability, standards, test) 4. Basic research 5. Manufacturing and production 6. Engineering support (lab assistant, technician) 7. Purchasing and procurement 8. Marketing and sales 9. Other Specify _____________________________________ 5. Indicate the primary product or service performed at your plant. (place applicable letter in box). Manufacturing A. Large computers B. Mini-computers C. Computer peripheral equipment D. Data Processing systems (systems integration) E. Office and business machines F. Test and measuring equipment G. Communications systems and equipment H. Navigation and guidance or control systems I. Consumer entertainment electronic equipment J. Other consumer electronic equipment (appliances, autos, hand tools) K. Industrial Controls, systems and equipment L. Components and Sub assemblies M. Materials and Hardware N. Aircraft, Missiles, space and ground support equipment O. Oceanography, Geology, Astronomy and support equipment P. Medical electronics Q. Industrial companies in the OEM incorporating electronic equipment in their end product(s) not elsewhere classified R. Research and development lab at an educational institution S. Public Administration and military T. Independent R&D laboratory or consultant U. Broadcasting, sound and motion picture and recording studios V. Commercial users of electronic equipment (railroads, pipelines, police, airlines, hospitals, banks) W. Other Specify _____________________________________ Non-Manufacturing X. Utilities Y. Broadcasting, sound and motion picture and recording studios Z. Other Specify _____________________________________ 6. How often do you use EBG? A. □ More than once a week B. □ Once per week C. □ Twice or more per month D. □ Once per month 7. How do you use EBG? (check as many as apply) A. □ Select suppliers. B. □ Locate sales office and reps C. □ Locate phone numbers D. □ Send for catalogs E. □ Find the name of an individual to contact F. □ Other Specify _____________________________________ 8. Does anyone else use your copy of EBG? A. □ Yes □ No B. □ If yes, how many others? ____________________________ Electronics Buyers' Guide P.O. Box 511 Princeton Road Hightstown, NJ 08520-9938 Postage will be paid by addressee BUSINESS REPLY MAIL FIRST CLASS PERMIT NO. 42 HIGHTSTOWN,NJ No Postage Necessary If Mailed in the United States For A Limited Time—This $40 Directory is Yours If you specify, purchase, or approve the purchase of electronic products, you'll be eligible to receive the Electronics Buyers' Guide ABSOLUTELY FREE if you act now. If you wait, you'll have to pay $40.00. Copies of the Guide are limited and are distributed to qualifying applicants on a first-come, first-served basis. What's In The New '86 EBG? The new EBG is an easy-to-use, single-volume information source designed to help you make the dollar-wise buying decisions needed to stay on top in 1986. Actually three directories in one. EBG contains: - Directory of Products. Over 4,000 products, materials and services listed alphabetically with the names and locations of the companies manufacturing them. Also, end-use terms, synonyms and newly devised terms are cross referenced to the generic names of the products. - Directory of Manufacturers. An alphabetical listing of manufacturers with sales offices/reps and distributors, local addresses, telephone numbers and the name and title of the individual to be contacted for sales information. - Directory of Catalogs presents a comprehensive listing of companies offering current catalogs. So you won't be swamped with catalogs you don't want, you will receive only those catalogs you request by using our unique "no waste" Catalog Reply Card Order Form. And, as a recipient of the EBG, you will also receive our exclusive semi-annual catalog update. Both the Directory of Catalogs and the semi-annual update contain six postage-paid catalog response cards good for up to one year! Who's Eligible To Receive The EBG Free? If you specify, purchase, or approve the purchase of electronic products, you are eligible to receive a free copy of the EBG. However, you must also fulfill the following requirements: - Please answer ALL questions on the attached qualification Reservation Card. - The EBG will be delivered to plant or business addresses only. - Drop the postage-paid order form in the mail. - But remember, quantities are limited and are distributed to qualified applicants on a first-come, first-served basis. FILTER CHIP ATTENUATES JITTER IN T1 LINES CRYSTAL SEMICONDUCTORS' BUILDING BLOCK USES DIGITALLY ENHANCED ANALOG TECHNOLOGY A single-chip filter from Crystal Semiconductor Corp. automatically attenuates jitter and provides steady clock and data signals for 1.544-Mb/s T1 phone lines. The CSC61600 eliminates jitter with an internal phase-locked loop and an elastic 16-by-1-bit first-in first-out shift register that can get data in at one speed and out at another. The CMOS filter is the latest addition to the series of digitally controlled analog telecommunications and signal-processing chips that the company is introducing this year. Crystal calls its digitally enhanced analog technology Smart Analog [Electronics, Jan. 20, 1986, p.21]. The new chip, intended to be a building block for T1 applications, is a subset of Crystal's recently introduced CSC61544 monolithic T1 interface. Earlier last month, Crystal began making available its T1 line interface circuit, which has not only the jitter attenuator on chip but other blocks needed to attach to the phone network. The chip has a digitally programmable pulse-shaping line driver that transmits signals meeting T1 standards. The 28-pin 61544 sells for $23 each in 1,000-piece quantities. In orders of 1,000, the 61600 will sell for $4 each. Intelligence inside the 61600 attenuator comes in the form of a digital logic block, which detects T1 line jitter and automatically adjusts the frequency of the filter's voltage-controlled oscillator. The chip uses switched capacitors to increase or decrease the capacitance load from the system's crystal, slightly varying the speed of the outgoing clock signal and smoothing out jitter. ELASTIC. The 16-bit FIFO shift register, which acts as an elastic storage band, also helps smooth the clock frequency on T1 lines. The FIFO has two pointers that keep track of how many bits are held in the shift register. The number of bits in the elastic FIFO depends upon the frequency on the line. The 61600 accepts 1.544-Mb/s data and clock signals as input and removes up to 14 unit intervals of jitter, peak to peak, before putting out the data and clock. "If bits are coming into the 16-by-1-bit FIFO faster than they are going out, the logic will slightly adjust the outgoing clock," explains Robert F. Bridge, Crystal's telecommunications product marketing manager. "If the FIFO empties out, that means bits are spilling out faster than they are coming in, and the chip very slightly slows the output clock signal." The result is a filter chip that will eliminate jitter with frequencies of 15 Hz. The chip can be used in systems to clean up T1 transmission lines at the subscriber termination site as well as multiplexing applications, such as converting 1.544 to 45 MHz for T3 systems. In such multiplexers, the 61600 would help compensate for missing bits on the T1 side of the transmissions. The extra bits are part of the multiplexed T3 transmission scheme. When demultiplexed, missing bits can cause jitter at the receiving T1 termination. The chip, the smallest yet from Crystal at 130 mils on a side, is fabricated from 3-μm p-well CMOS using double-polysilicon, single-metal technology. The chip dissipates 25 mW when active and operates from a single 5-V power supply. The initial versions operate in the commercial temperature range of 0°C to 70°C. Crystal also plans to offer chips that work from -140°C to +85°C. Samples of the 61600 are now available in plastic 14-pin DIPs, and the company plans to begin volume deliveries in early September. Crystal will soon offer samples of a similar jitter attenuator for token-ring local-area networks, designated the CSC80600. Instead of a 16-bit FIFO, the part will have an 8-bit FIFO and will run at 8 MHz instead of the 1.544-MHz T1 rate. --J. Robert Lineback Crystal Semiconductor Corp., P.O. Box 17847, Austin, Texas 78760. Phone (512) 445-7222 [Circle 440] TWO SERVER OPTIONS LINK DIVERSE NETWORKS Banyan Systems is expanding its line of high-performance wide-area networking products with a software-based network server that runs on the IBM Corp. PC AT and a desktop hardware server. The company targets its networking products at large corporations that demand the integration of numerous local-area networks of various types and want to link those networks to a mainframe computer. As such, Banyan's networking products have features not typically found on low-priced LANs, including built-in microcomputer-to-mainframe links, server-to-server communications, and a node-naming and addressing scheme called Streettalk that allows users to locate and access any resource without having to know the network topology. The Banyan/DTS is a high-performance, 32-bit, desktop network server, and the Vines/286 is a software package that converts a PC AT into a multifunction network server. Both products are based on the company's own Virtual Networking Software (Vines). Vines supports multiple LANs, so users can choose networks based on performance and application needs. In addition, Vines lets users migrate to new LAN technologies as they emerge. Vines currently supports the IBM PC Network, Interlan Ethernet, the Standard Microsystems Corp. version of Datapoint Corp.'s Arcnet, Corvus Omninet, Proteon ProNET, 3Com Ethernet, Ungermann-Bass Net/One, and Allen-Brad... 27 million Americans can't read. And guess who pays the price. While American business is trying to stay competitive with foreign companies, it's paying an added penalty. The penalty of double-digit illiteracy. Believe it or not, 27 million American adults can't read and write. Another 47 million are literate on only the most minimal level. That adds up to almost one third of our entire population...and probably a disturbing number of your employees. What does illiteracy cost you? Get out your calculator. Illiterate adults make up 50%-75% of our unemployed. Every year they cost us an estimated $237 billion in lost earnings. They swell our welfare costs by $6 billion annually and diminish our tax revenues by $8 billion. Illiteracy costs you through your community, too. It robs the place where you work and live of its resources. It undermines the potential of the people who make your products and the people who buy them. No dollar figure can be assigned to this. But over the years, this may be the costliest loss of all. What can your company do about this? It can join in local efforts to fight illiteracy. It can volunteer company dollars and facilities for better school and tutorial programs. It can invest in a more literate community. The first step is to call the Coalition for Literacy at 1-800-228-8813 or fill out the coupon below. Do it today. You may find it's the greatest cost-saving measure your company has ever taken. ☐ I want my company to join the fight against illiteracy. Please send brochure with additional information. ☐ We want to discuss funding the Coalition for Literacy. Please have a representative contact me. Name ____________________________________________ Title ______________________________________________ Company ___________________________________________ Address ____________________________________________ City ______________________ State ______ Zip ______ Phone ____________________________________________ Please return to: Coalition for Literacy Business Division PO Box 81826 Lincoln, NE 68501-1826 A literate America is a good investment. Banyan Systems Inc., 135 Flanders Rd., Westboro, Mass. 01581. Phone (617) 898-2404 [Circle 441] UNIT RUNS ETHERNETS ON MAP NETWORKS The Ethermodem Remodulator allows users to operate broadband Ethernets on the same broadband cable as Manufacturing Automation Protocol factory networks. The manufacturer targets the remodulator at broadband network users who also want to add Ethernet connections and for those who want to extend the distance between Ethernet nodes over broadband cabling. The remodulator, which lets Ethernets coexist with all three MAP frequency channels, eliminates the need for guard bands between communication services. The Ethernet Remodulator lets users select either of two channels for their Ethernets while reserving the three MAP-specified channels. The unit is transparent to high-level network software so networks can run DECnet, XNS, Transmission Control Protocol/Internet Protocol, and Technical and Office Protocol software on broadband nets without modification. The remodulator costs $5,900 and will be available in September. Chipcom Corp., 195 Bear Hill Rd., Waltham, Mass. 02154. Phone (617) 890-6844 [Circle 445] HUB CONTROLLER SUPPORTS 8 NODES The 82C551 Starlan hub controller supports up to eight nodes downstream and one upstream. Up to five controllers can be cascaded to support a maximum of 40 nodes. The IC receives data, retimes it, and transmits it to the other nodes as well as to the next-level hub; the controller performs the reverse function. The 82C551 features collision detection. Its built-in jabber function detects transmissions over 65 ms, then ignores that line until the transmission ceases, which helps to isolate faulty nodes from the rest of the network. It has a diagnostic loop-back for fault detection and isolation. The CMOS chip comes in a 40-pin plastic DIP for $56.70 in quantities of 100. Samples are available now. Chips & Technologies Inc., 521 Cottonwood Dr., Milpitas, Calif. 95035. Phone (408) 434-0600 [Circle 442] T1 PROCESSOR LINKS WITH AT&T'S ACCUNET The AccuPac-1.5, a T1-format processor, creates a direct-access path between high-speed data-terminal equipment and AT&T's Accunet. Typical applications include video teleconferencing, high-speed host-to-host transmission, and terrestrial extension of satellite or microwave circuits. Accupac-1.5 sells for $3,995 in single-port configuration. An internal-linedriver option allows it to be used as a dual-port, high-speed local data-distribution unit. Delivery is in 30 to 60 days. Avanti Communications Corp., Aquidneck Industrial Park, Newport, R.I. 02840. Phone (401) 849-4660 [Circle 446] CHIP BUILDS T1 REPEATER A pulse-code-modulated repeater IC, the MP5262, contains all the active functions required to build one side of a T1 or 2-Mb/s PCM repeater. For stable operation and low temperature drift, the chip uses an advanced thin-film-resistor technology of silicon chrome and bipolar processing. The MP5262 operates with a single 6.8-V power supply and with a typical low-current drain of 13 mA. The bipolar output drivers are designed to go automatically to their off-state when there is no input signal present. In a 16-pin Cerdip, the 100-piece price of the MP5262 is $6. If the customer wishes it in die form, the company will quote prices. Production quantities take four weeks to deliver, and samples are available immediately. Micro Power Systems Inc., 3100 Alfred St., Santa Clara, Calif. 95054. Phone (408) 727-5350 [Circle 444] ONCE, YOU NEEDED A COMPUTER COMPANY, A GRAPHICS COMPANY AND A WORKSTATION COMPANY. NOW ALL YOU NEED IS APOLLO. On the next three pages, Apollo presents its newest products. Systems with capabilities and price performance that are making our customers shout and our competitors shudder. In the tradition of every Apollo product that’s come before, these offerings let technical people work together as teams. More united than ever. Supported by unprecedented power. Able to work in multi-vendor environments. More importantly, they establish Apollo as much more than a workstation manufacturer. Now we’re a graphics company. A computer company. A total solutions company. With a broader product line and better technology than anyone else in scientific and technical computing. For more information, call (617) 256-6600 x 4889. Or write Apollo, 330 Billerica Rd., Chelmsford, MA 01824, MS 40. Allen-Bradley is lowering the price of building automated factories with its line of standard work-cell controllers. The Base Cell line is a low-end version of the company's 16-month-old Vista 2000 cell-control computer, which until now has been used to set up more costly custom implementations. The Base Cell Program features standard hardware and software packages that the customer can configure himself, eliminating the manpower costs associated with customized solutions. Allen-Bradley is also introducing higher-capacity RAM boards and disk drives that further improve the price-performance ratio of the Vista 2000, an industry-hardened version of the 32-bit Masscomp model 500 computer from Massachusetts Computer Corp., Westford, Mass. The machine relies on multiple 68000-family microprocessors. In the Allen-Bradley scheme, the Vista 2000 oversees a number of programmable controllers, computerized numerical controllers, and industrial personal computers that interface with sensors and machinery in factory work cells. The cell controller coordinates the activities of its lower-level counterparts, making production adjustments and issuing reports, while providing the link between the factory floor and computers at higher levels in the hierarchy. The combination of reduced Allen-Bradley support requirements and the use of denser RAM and hard-disk technology makes the Base Cell line about 25% cheaper than comparable custom Vista 2000 implementations, says Ralph Waite, vice president and general manager for the company's Industrial Computer Group in Highland Heights, Ohio. Until now, cell-level control has been a custom proposition, he says. So far, the company has sold about 75 Vista 2000s, but all factory installations required the presence of Allen-Bradley personnel to aid in customizing systems, Waite says. The Base Cell Program represents the experience Allen-Bradley gained in those installations. MENU DRIVEN. The company developed standard hardware and software packages that can be configured by the customer for basic cell-control tasks without the aid of Allen-Bradley personnel, Waite explains. The Base Cell software features a menu-driven format and graphic aids that walk the user through the setup task. "All the user has to do is interact with the system," Waite says. Initial Base Cell products come in two configurations. Vista Base Cell 1 includes a Vista 2000 computer based on AT&T's Unix operating system, software, power supply, hard-disk memory, and a 19-in. color control monitor with 832-by-600-pixel resolution. The package sells for less than $50,500, including a 4-megabyte RAM board and a 71-megabyte hard-disk system. That compares with 1-megabyte RAM boards and a 42-megabyte hard disk in previous Vista 2000 systems, Waite says. Vista Base Cell 2 adds an Access Machine, a 32-bit computer designed to extend the system's capability. The Access Machine starts at less than $20,000 and acts as a database manager while enabling the system to operate on two Data Highways. (The Data Highway is Allen-Bradley's factory local-area network.) The Base Cell 1 can handle one Data Highway and provides monitor and control capability for up to 1,250 functions. With Base Cell 2, the number of functions extends to 5,000. Though the Base Cell line enables users to configure basic cell-control jobs, more sophisticated applications will still require Allen-Bradley support personnel at the customer site. But as Allen-Bradley gains experience, it plans to likewise convert the more advanced capabilities into standard products in the Base Cell line, Waite says. Future Base Cell products will interface with both broad- and carrier-band networks based on the Manufacturing Automation Protocol. They will also link to robots and computer-aided design systems from various vendors, allowing users to create true computer-integrated manufacturing systems, Allen-Bradley says. —Wesley R. Iversen Allen-Bradley Co., 747 Alpha Dr., Highland Heights, Ohio 44143. Phone (216) 449-6700 [Circle 460] MOTION CONTROLLER IS EASY TO ASSEMBLE A modular motion-control system that comes with keyed interconnection hardware and a detailed connection diagram enables a designer with no previous experience in this area to quickly assemble a working unit. The BBS/Series components include the motion controller, motor, feedback sensor, and amplifier. Motors come in torque ranges from 5 to 500 oz-in., with encoder resolutions up to 1,000 pulses per revolution. The user can select controllers for the IBM Corp. Personal Computer and standard bus systems; pulse-input format is also available for applications requiring a direct interface to an indexer or pulse generator. Typical position accuracy at "stop with no load" is 0.18". The amplifier puts out up to 10 A at 70 V; higher current and voltage limits are optional. Features such as coordinated motion, nonvolatile memory, uncommitted I/O channels, and sequencing are also available. Systems start at $995, with delivery in three weeks. Galil Motion Control Inc., 1928-A Old Middlefield Way, Mountain View, Calif. 94043. Phone (415) 964-6494 [Circle 462] SOFTWARE ENHANCES IMAGE PROCESSING An image-processing operating environment called PC Semper contains over 100 image-processing and -analysis functions. The package runs on IBM Corp. Personal Computers and compatibles. The software uses IEEE standard 32-bit floating-point arithmetic for most appliThe personal power shortage is over. Introducing the Apollo DOMAIN® Series 3000. A family of personal workstations that are priced like PCs, packaged like PCs, easy to use like PCs. Yet they perform with the authority of machines twice the size and three times the price. Consider their raw power. Despite the fact that DN3000s fit comfortably on the corner of your desk, they have room inside for a 32-bit, virtual memory engine that includes a Motorola 68020 and 68881 floating point chip. Sufficient force to deliver processing speeds that overwhelm the IBM PC. Their processing proficiency is equaled only by their graphics. A monochrome DN3000 exhibits 1280 x 1024 resolution on a 19-inch monitor. While a color version, with its four-bit planes, displays 16 colors simultaneously on a high-resolution, 15-inch flicker-free screen. While such qualities alone are cause enough for celebration, the DN3000s were notably designed with an understanding that technical professionals work as much with each other as they do with their workstations. Hence, like all Apollo workstations, DN3000s inherently let you transparently share information and resources across a high-speed local area network. An IBM PC/AT™- compatible bus assures the ability to integrate options into the system. And an open architecture supports such industry standards as UNIX™ and ETHERNET®, providing easy access to computer systems made by IBM® and DEC®. The DOMAIN Series 3000. In monochrome from under $10,000. In color from under $15,000. Proof that it doesn’t take a lot of money to attain a position of power. We let the cat out of the bag. Millions of children might never have met Dr. Seuss and his famous Cat had it not been for The New York Public Library. That's because our renowned children's librarian, Anne Carroll Moore, gave his first book an outstanding review. He became famous overnight, and Dr. Seuss himself credits her with his success. Others too, credit our librarians. "The people who work here," says author David McCullough, "provide a service to the whole of society that's never sung enough." Indeed, the librarians are the very heart of the Library. They are the ten librarians who answer five million telephone inquiries a year. They are the librarians who introduce children to the magic of literature and help adults find employment through our Job Information Center. They are also the librarians who select and preserve books for users of tomorrow. The Library can help just about anyone with anything. Even a Doctor with his Cat. The New York Public Library WHERE THE FUTURE IS AN OPEN BOOK A line of five pulse-width-modulated current-amplifier/motor drives amplifies three-phase torque command signals from Gould's Flexible Automation Controller to drive its brushless motors. The Cyberline 1000 drives use a proprietary direct-numerical-processing technology, and they produce outputs of 1 to 20 kW. The drives all feature diagnostic circuitry to protect against short circuits, over- and undervoltage, and overheating. They support a maximum bus voltage of 325 V dc and are said to have 30% fewer components than drives in conventional brushless motor-control systems. There are no potentiometers, for example, since the drives amplify torque and command signals rather than generate them, the company says. Prices range from $1,000 to $4,000. Delivery takes two weeks. Gould Inc., Industrial Automation Systems, 105 Delta Dr., Pittsburgh, Pa. 15238. Phone (412) 963-1444 [Circle 464] DATA COLLECTOR SCANS 100 CHANNELS/S Increased speed and performance over its predecessor are the calling cards of the Netpac 2 remote front-end module for industrial data measurement and collection. When connected to a host computer or data logger with an RS-422 communication link, Netpac 2 can scan over 100 channels/s. The module features complete on-board signal processing and can measure a variety of inputs with up to 0.03% accuracy without any preconditioning. Standard I/O options consist of millivolt, milliamper, contact I/O, and pulse input lines. Users can connect 16 modules to a twisted-pair wire up to 5,000 ft long, and each supports up to 100 I/O lines. Transmission up to 19.2 kb/s is supported and software for host compatibility is available. Pricing and delivery for the Netpac 2 have not yet been set, but the product will be available in both NEMA 2 and NEMA 4 enclosures for use in harsh industrial environments. Acurex Corp., Autodata Division, P. O. Box 7042, Mountain View, Calif. 94039. Phone (415) 967-9100 [Circle 466] IMAGE CARDS WORK IN VMEBUS SYSTEMS A series of boards for image-processing applications are designed for VMEbus computers and compatible machines. The RTI-400 series includes an intelligent pipeline pixel processor, an analog subsystem for video I/O, and two memory boards. Used together, they form a complete vision-engine OUR 3D GRAPHICS WORKSTATION IS SO FAST IT'S AMAZING THESE PICTURES AREN'T BLURRED. Until now, if you wanted realtime 3D graphics on a powerful workstation, you had only two choices. Both of them a compromise. You could purchase a specialized graphics workstation but sacrifice a compatible product family, software and application choice. Or you could opt for a reputable engineering workstation and settle for color graphics that could only be labeled dull. Introducing Apollo's new DN580. It asks no sacrifice. Instead, this powerful 32-bit graphics workstation overwhelms anything that claims to be either graphic or a workstation. Its powerful graphics pipeline processor performs 3D graphics at an amazing 100,000-plus vectors per second, a record well beyond the reach of any other computer or workstation supplier. Yet despite such speed, the DN580 complies with PHIGS and supports UNIX™, dispelling the myth that performance and standards are contradictory terms. Of course the DN580 offers more than alacrity. As part of an advanced distributed processing network, its speed is greatly enhanced by an ability to access other Apollo workstations. As well as systems made by DEC® and IBM®. And with Apollo's open architecture, you can perform all these stunts with a single system view. The Apollo DN580. The only thing faster is the rate at which our competitors will have to work to catch up. Did You Know That Last Year's U.S. Savings Bonds Average Rate Exceeded 9%? Surprise! U.S. Savings Bonds give you market-based interest rates—like the money markets—plus a guaranteed return. What's more, Savings Bonds give you big tax advantages, cost as little as $25 and are easy to buy. For more information, call toll-free 1-800-US-BONDS. U.S. SAVINGS BONDS Paying Better Than Ever Market-based rates apply to Bonds purchased on and after 11/1/82 and held at least five years. Bonds purchased before 11/1/82 earn market-based rates when held beyond 11/31/82. Bonds held less than five years earn lower rates. A public service of this publication. For those of you whose pursuit of workstation power remains interminable, Apollo has some very good news. Your quest is over. Introducing the DSP9000 parallel processor. A new compute server that will quell even the most intense passion for power. Accessible from any Apollo workstation, the DSP9000 is unarguably the fastest server in its class. One machine can concentrate up to eight, high-performance 32-bit processors on a single task. Combined they perform at 99.4 million floating point operations per second, making the DSP9000 ten times faster than DEC's® comparably priced 8600 series. Some computer makers would consider such power sufficient for multiple programs. The DSP9000, on the other hand, was intelligently designed to concentrate all of its authority on a single application. Allowing it to overwhelm the most demanding electrical, mechanical and artificial intelligence programs. Technical sophisticates will also be pleased to know that the DSP9000 is endowed with more than sheer potency. It's the only parallel processor that doesn't force you to rewrite software already created to run on vector and scalar computers. Instead, it's equipped with a Fortran compiler so advanced it automatically schedules your code for the most efficient programming sequence. Since such performance is only as useful as it is accessible, the DSP9000 adheres to Apollo's philosophy of open environments. It works with virtually all of your existing computer investments, supporting SNA, UNIX™ and other popular standards. Apollo's new DSP9000. It's an unparalleled processor. MEETINGS ACM Conference on Lisp and Functional Programming, Association for Computing Machinery (Robert Halstead, Massachusetts Institute of Technology, 419 Technology Sq., Cambridge, Mass. 02139), MIT, Cambridge, Aug. 4-6. National Conference on Artificial Intelligence, American Association for Artificial Intelligence (445 Burgess Dr., Menlo Park, Calif. 94025-3496), University of Pennsylvania, Civic Center, and Franklin Plaza Hotel, Philadelphia, Aug. 11-15. 30th International Technical Symposium on Optical and Optoelectronic Engineering, Society of Photo-Optical Instrumentation Engineers (P.O. Box 10, Bellingham, Wash. 98227-0010), Town and Country Hotel, San Diego, Aug. 17-22. Siggraph '86, Association for Computing Machinery and IEEE (Siggraph Conference Management, 111 E. Wacker Dr., Chicago, Ill. 60601), Dallas Convention Center, Dallas, Aug. 18-22. International Conference on Parallel Processing, IEEE Computer Society and Pennsylvania State University (IEEE Computer Society, 1730 Massachusetts Ave., N. W., Washington, D.C. 20036-1903), Pheasant Run Resort, St. Charles, Ill., Aug. 19-22. ICSSDM: 1986 International Conference on Solid State Devices and Materials, the Japan Society of Applied Physics (1986 ICSSDM, c/o Japan Academic Societies Center, 4-16, Yayoi 2-chome, Bunky-ku, Tokyo 113, Japan), Tokyo Prince Hotel, Tokyo, Aug. 20-22. 3rd International Congress on Advances in Non-Impact Printing Technologies, Society of Photographic Scientists and Engineers (Samuel W. Ing, Xerox Corp., 800 Phillips Rd., Webster, N.Y. 14580), Fairmont Hotel, San Francisco, Aug. 24-28. 8th Quartz Devices Conference and Exhibition, EIA (2001 Eye St., N.W., Washington, D.C. 20006), Westin Crown Center, Kansas City, Kan., Aug. 26-28. Interconnect '86, United States Telecommunications Suppliers Association (333 N. Michigan Ave., Suite 1618, Chicago, Ill. 60601), San Mateo Expo Center, San Mateo, Calif., Aug. 26-28. IFIP Congress '86: International Federation for Information Processing (Philip H. Dorn, Dorn Computer Consultants Inc., 25 E. 86th St., New York, N.Y. 10028), Trinity College, Dublin, Sept. 1-5. NCC-Telecommunications Conference, IEEE Computer Society et al. (Mike Sherman, 1899 Preston White Dr., Reston, Va. 22091), Civic Center, Philadelphia, Sept. 8-10. International Test Conference, IEEE at al. (Peter Bottorff, International Test Conference, P.O. Box 264, Mt. Freedom, N.J.), Sheraton Washington Hotel, Washington, Sept. 8-11. 16th European Microwave Conference, Royal Irish Academy et al. (Microwave Exhibitions and Publishers Ltd., Convex House, 43 Dudley Rd., Tunbridge Wells, Kent TN1 1LE, England), National Concert Hall, Dublin, Ireland, Sept. 8-11. Symposium on Optical Fiber Measurements, IEEE and Optical Society of America (D.L. Franzien, National Bureau of Standards, Division 724.02, 325 Broadway, Boulder, Colo., 80303), NBS Laboratories, Boulder, Sept. 9-10. Swissdata '86, Swiss Industries Fair (Secretariat, Swissdata 86, Postfach, CH-4021, Basel, Switzerland), Fairgrounds, Basel, Sept. 9-13. Symposium on Optical Communication Primary Standards, EIA (G.P. Watkins, Corning Glass Works MP-BH-03, Corning, N.Y. 14831), National Bureau of Standards Laboratories, Boulder, Colo., Sept. 11. Midcon 86, IEEE et al. (Electronic Conventions Management, 8110 Airport Blvd., Los Angeles, Calif. 90045), Dallas Convention Center, Dallas, Sept. 9-11. Mid-Atlantic Electronics Design and Production Conference, International Marketing Services Ltd. (1030 S. LaGrange Rd., LaGrange, Ill. 60525), Valley Forge Convention Center, King of Prussia, Pa., Sept. 11-12. Intercomm '86: International Communications Exposition and Conference for Science and Technology, Cahners Exposition Group (P.O. Box 70007, Washington, D.C. 20088), Exposition Center, Beijing, China, Sept. 15-20. IEEE International Symposium on Electromagnetic Compatibility, IEEE (George Ufen, GRU Associates, 1105 E. Commonwealth Ave., Fullerton, Calif. 92631), Town & Country Hotel, San Diego, Sept. 16-18. Fall National Design Engineering Show & Conference, Cahners Exposition Group (999 Summer St., Stamford, Conn. 06905), Jacob Javits Convention Center, New York, Sept. 16-18. International Broadcasting Convention, Electronica Engineering Association et al. (IBC Secretariat, Institution of Electrical Engineers, Savoy Place, London WC2R 0BL, England), Metropole Conference Centre, Brighton, England, Sept. 19-23. Before the chips are down, turn to Collins & Aikman Static-Control Products. Electrostatic discharge is costing manufacturers like you billions of dollars a year in product loss and degradation. You can stop it. With a new line of static-control products so dramatically superior in construction and design you won't believe your eyes. Call toll-free for a free demonstration today. Collins & Aikman Static-Control Products P.O. Box 1447, 1702 Kimberly Park Drive Dalton, GA 30722 (404)259-9711 • Ext. 410 Telex No. 810-759-4404 Toll-Free Customer Service 1-800-235-8817 Circle 143 on reader service card Ever notice how some people always seem to have the right answer? No matter what the situation, they always seem to be a step ahead. And they're successful—the first in line to lead an important project or the first in line for a promotion. It's certainly not magic. Usually it's a combination of hard work, brains, guts and desire. It probably means they read Electronics ... regularly. Electronics has helped propel many a manager and designer to the front lines—to the "leading edge" of the industry. And just as we've been providing many of your colleagues with the right information, the important information—when they need it most—so too can we provide it for you. Every week, a Electronics subscriber receives an issue crammed with the latest information on new products, developments, concerns and trends in the worldwide electronics industry. When you become a subscriber, we'll supply you with the intelligence you need to make the right decision, or the everyday decision to further your career and reap profits for your company. For a taste, just browse through the issue you're holding. But, you really don't have to believe us. Just ask the person who's always a step ahead. To become a subscriber, just complete and mail the insert card in this magazine. If subscription card is missing, write: Circulation Manager, Electronics, P.O. Box 511, Hightstown, N.J. 08520. Electronics The Voice of the industry CAE for Analog Circuit Designers 50 TIMES FASTER than your HP AC Circuit Analysis for your 300, 217, 9836, 9816, 9920, 9845 and 9020 PLUS: • OPTIMIZATION • 98 NODES • TIME DOMAIN • GROUP DELAY • RELATIVE PHASE • NEGATIVE COMPONENTS • COMPONENT SENSITIVITY • OUTPUT WAVEFORM for any INPUT • IMPROVED ALGORITHMS • COMPATIBLE with HP DATA FILES • INTEGRATED with FFT and MEASUREMENTS with HPIB • 6 YEAR TRACK RECORD • 30 DAY TRIAL ALSO: FFT WAVEFORM ANALYSIS • 4 TIMES FASTER than HP • INTEGRATED into ONE FILE THREE INTEGRATED MODULES: AC-CAP AC Circuit Analysis with OPTIMIZATION $750.00 S-WAVE FFT Waveform Analysis for Time Domain $750.00 PLOTTT Time Domain Data Acquisition $750.00 Double all prices for 9020 Computer version. DEMO DISK AVAILABLE COMTRAN INTEGRATED SOFTWARE FROM jensen transformers INCORPORATED 10735 Burbank Boulevard • North Hollywood, California 91601 Contact Deane Jensen (213) 876-0059 • Closed Fridays Circle 166 on reader service card ## Advertisers Index ### Advertising Sales Staff Atlanta, Ga. 30319: Joseph Milroy 4170 Ashford-Dunwoody Road N.E. [404] 252-0262 Boston, Mass. 02116: M.E. McKibben - McKibben, Jr. 575 Boylston St. [617] 262-1162 633-A Boylston Phone Chicago, Ill. 60611: William J. Walker [312] 751-3738 645 North Michigan Avenue Cleveland, Ohio 44113: [216] 496-3800 Costa Mesa, Calif. 92626: Fran Cowen 3001 Red Hill Ave. Bldg. # 1 Suite 222 [714] 557-6292 Dallas, Texas 75240: Harry B. Doyle, Jr. 5161 Bullis Road, Suite 907 [214] 458-2400 Englewood, Co. 80112: Harry B. Doyle, Jr. 7400 S. Alton Court Suite 111 [303] 740-4630 Houston, Texas 77040: Harry B. Doyle, Jr. 7600 West Tidwell Suite, 500 [713] 742-0757 Los Angeles, Calif. 90010: Chuck Crowe 3333 Wilshire Blvd. [213] 480-5210 New York, N.Y. 10020: Matthew J. Reseska [212] 512-3617 John Gallie [212] 512-4420 1221 Avenue of the Americas Stanford, CA. 94302: Albert J. Luce 777 Long Ridge Road, Bldg. A [603] 949-7116 San Mateo, Ca 94404: Larry Goldstein, Jeffrey C. Hoopes, William H. Sleight 3rd Floor 951 Minerva's Island Blvd. [415] 349-4100 Philadelphia, Pa. 19102: Joseph Milroy Three Parkway, 16th Floor Pittsburgh, Pa. 15222: Joseph Milroy Suite 215, 6 Gateway Center, [215] 496-3800 Southfield, Michigan 48075: 4000 Town Center, Suite 770, Tower 2 [313] 362-9760 San Francisco, Calif. 94111: Larry Goldstein, Jeffrey C. Hoopes, William H. Sleight 425 Battery Street [415] 362-4600 Frankfurt/Main: Fritz Krusebecker, Dieter Rothenbach 19 Liebigstrasse, Germany Tel: 72-01-81 Milan: Maria Capuano 1 via Baracchini, Italy Tel: 86-90-656 Paris: Jean Christian Acis, Alan Faure 128 Faubourg Saint Honore, 75008 Paris, France Tel: [1] 42-89-0381 Scandinavia: Anders Karrig Finnvadavagen S-131 31 Nacka Sweden Tel: 46-8-240-04 Tele: 7391 AKA S Tokyo: Hirokazu Morita McGraw-Hill Publications Overseas Corporation, Kasumigaseki Building 2-5, 3-chome, Kasumigaseki, Chiyoda-Ku, Tokyo, Japan [581] 8611 United Kingdom: Art Scheffer 34 Dover Street, London W1 Tel: 01-493-1451 ### Business Department Thomas E. Vazsony Director of Operations [212] 512-2627 Leon Irgang Circuits Director [800] 426-5542 Roseann Lehmann Office Administrator [212] 512-2409 Customer Service [212] 512-6643 Frank M. Vallone Mgr./Mktg. Admin. [212] 512-6058 Patricia Parks Billing Specialist [212] 512-2569 Thomas M. Egan Production Director [212] 512-2409 Carol Gallagher Production Manager [212] 512-2045 Evelyn Diller Production Manager Related Products [212] 512-2044 ### Classified and Employment Advertising [212] 512-2556 --- ### Abbott Transistor 104 ### Matrox Electronic Systems Ltd. 4th C ### Advanced Micro Devices 6-7 ### Mitel Corp. 2 ### Altera Corp. 78-79 ### Mitsubishi 115 ### AMP Inc. 62-63 ### Multiwire 100-101 ### Analog Devices Inc. 132-133 ### Murata Mfg. Co. 46 ### Apollo Computer Inc. 145, 147, 149, 151 ### National Microelectronics (UMC Corp.) 112-113 ### Arium Corporation 56-57 ### NEC Corporation 43 ### ATT Communications Int'l 13 ### NEC Electronics (Europe) GMBH 31 ### ATT Information Systems 3rd C ### Nicolet Scientific 26 ### ATT Information Technologies 71, 74 ### OKI Semiconductor 16A, 16B ### California Instrument 31 ### PEP GmbH 90-91 ### Catalyst Semiconductor 16 ### Philips T&M 88-89 ### China External Trade 128 ### Precicontact 25 ### Collins & Aikman 153 ### Rantec 55 ### CTS Corporation 99 ### RCA Solid State 123-127 ### Data Translation 22 ### Sanyo 114-115 ### Datum Inc. 128 ### Seiko Instrument 43 ### Digital Equipment Corp. 14-15 ### Sequent Computer Systems 44-45 ### DSP Associates 33 ### Siemens AG 9, 14-15 ### ENI 17 ### Symbolics Inc. 64-65 ### Forth Inc. 87 ### Techmasexport 33 ### Fujitsu Ltd. 153 ### Teledyne Relays 92 ### Galacticomm 25 ### Thomson Semiconductor 134-135 ### GENRAD 34-35 ### Toshiba America Inc. (memory Div.) 118-119 ### Gould AMI 37, 39, 41 ### Toshiba Semiconductor 141 ### Harris Semiconductor 2nd C ### VME Microsystems International 120 ### Hitachi America Inc. Ltd. 9 ### VLSI Technology 102-103 ### Hughes Aircraft Co. (Semiconductor Division) 46-47 ### Wandel & Goltermann 119 ### Image Data Corp. 61 ### Wintek Corporation 48 ### Industrial Computer Design 8 ### Zenith Data Systems 18 ### Infotek Systems 1 ### Inmos Corp. 10-11, 21 ### Interactive Circuits and Systems 150 ### Jensen Transformers 153 ### Kennedy Technology 134-135 ### Laserpath 33 ### LeCroy Corp. 66 ### Malco 12 ### Compuaker 155 ### Dekker Ltd. 155 ### Intergraph Inc. 155 ### J&R Software 155 ### NFI Electronics Ltd. 152 ### Ovation Inc. 155 ### T-Cubed Systems 155 ### ZTEC 155 --- For more information of complete product line see advertisement in the latest Electronics Buyers Guide * Advertisers in Electronics International ‡ Advertisers in Electronics domestic edition Electronics / July 10, 1986 COMPUTER SOFTWARE FREE DEMO DISK SCHEMA is a complete, integrated schematic drawing software package for IBM Personal Computers. Use SCHEMA with your PC to draw schematics and automatically generate design documentation such as Wire and Pin Lists, Bills of Materials, Design Rule Checks, etc. SCHEMA is $495 and supports most common PC hardware configurations. Call or write today for a free demo disk and brochure. OMATION, INC. 1701 N. Greenville Ave., Suite 809 Richardson, TX 75081 (214) 231-5167 HOT LINE To place your Computer Software Ad in Electronics call Ilene Fader at 212/512-2984 PROJECT MANAGEMENT SOFTWARE The Choice of Leading Aerospace & DOD Contractors Easy to use • Precedence and IJ with PERT/CPM • Baseline schedule control • Subprojects with automatic roll-up/down • Creates Network on screen • Tracks hours and dollars per activity • Prints: Network, Gantt and Bar Charts Labor and Cost Time Flow Charts • Runs on IBM PC, XT, AT and compatibles, and Wang PC • Hotline 800 telephone support TASK MONITOR from MONITOR SOFTWARE 960 North San Antonio Rd. Suite 210 Los Altos, CA 94022 415-949-1688 TMS320 Development Software runs on PCI/XT/AT SPL Signal Processing Compiler $560 TAS20 Assembler for TMS32020 $325 TASSM Assembler for TMS32010 $325 S320 Simulator for TMS32010 $325 COMPUTALKER 1730 21st St., Santa Monica, CA 90404 (213) 828-6546 RELIABILITY PREDICTION SOFTWARE ARE YOUR PRODUCTS RELIABLE? RelCalc 2 automates MIL-HDBK-217D on your IBM PC! Very easy to use. Demo package $50. T-CUBED SYSTEMS (818) 991-0057 Westlake Village, CA 91362 SPICE up your PC Z/SPICE professional circuit simulation Full version $245 Student version $79 ZTEC box 737, college place, wa 99362 (509)-529-7025 FREE CATALOG NEW comprehensive SOFTWARE CATALOG for ELECTRONIC ENGINEERS. Over 100 programs listed. Call or write for your free copy: Interleave, Inc. 2940 Noble Road, Cleveland, OH 44121 (216) 291-1001 ENGINEERS, PROFESSORS, STUDENTS. LINCAD, CALCAD, FFTSA—Linear and Logic electronic circuit analysis and design programs. SOFCAD ELECTRONICS, INC. P.O. Box 21845 Columbus, Ohio 43221 MOTOROLA SETS TRANSFER PLANS... Paving the way for 45-year-old George Fisher to take over Motorola Inc. as president and chief executive officer on Jan. 1, 1988, is a series of moves intended to ensure a smooth transition. After 22 years at the company's helm, chairman Robert Galvin, 63, is relinquishing his second title of chief executive officer to vice chairman and chief operating officer William Weisz, 59. Weisz will hold the post until Fisher, who now becomes senior executive vice president and deputy to the chief executive, takes over in 1988. Previously, Fisher was a senior vice president in Motorola's Communications Sector. John Mitchell, 58, president and assistant chief operating officer, will take sole responsibility for operations until 1988, when he will join Weisz as vice chairman and surrender his operations role to Gary Tooker, 47. For now, Tooker will become senior executive vice president and chief corporate staff officer. ... WHILE RCA PAIR ARE NAMED GE VPs General Electric Co. is making moves to consolidate in the wake of its merger with RCA Corp. by naming Richard Miller, 45, senior vice president for GE's Consumer Electronics Business, including RCA Consumer Electronics. Miller had been executive vice president for consumer products and entertainment at RCA. Carl Turner, 58, was named vice president of the Solid State Division of GE's Semiconductor Business. Turner continues as division vice president and general manager of RCA Solid State and assumes "some responsibility" at GE Semiconductor. MATSUSHITA TO MAKE TVs IN MEXICO Matsushita Electric Industrial Co. plans to decrease color TV exports to the U.S. from its Japanese plant by shifting production to a new facility in Mexico. The plant will produce 250,000 large-screen sets in the first year of operations and is scheduled to ramp up to 400,000 within two years. The company's annual U.S. sales now total 400,000 sets of all sizes. NTT COULD ADOPT ISDN STANDARDS A private advisory panel to Japan's Ministry of Posts and Telecommunications is urging the ministry to ask Nippon Telegraph and Telephone Corp. to adopt international standards for its integrated services digital network. NTT had been developing its own standards, but with the expected completion of recommended international standards this month, the panel urged NTT to conform. NTT is expected to agree to the request. TWO TEAMS VIE FOR MIL CONTRACT Two industry teams are competing for final development of the joint Air Force/Navy Integrated Electronic Warfare System. TRW and Westinghouse will square off against Sanders Associates and General Electric in a $93.6 million contract for the 31-month Phase 1B, the design and development of components and subsystems. Inews will integrate a number of electronic systems in combat aircraft. One team will be picked to develop the next-generation Inews in Phase 2, and in Phase 3 the teammates will split to compete for production rights. CMI DROPS DISK DRIVES Having lost its major customer—IBM Corp.—Computer Memories Inc. is dropping out of the disk-drive business. The Chatsworth, Calif., firm [Electronics, Aug. 19, 1985, p. 19] hasn't managed to promote sales of its 20-megabyte 5¼-in. drive to other customers; nor could it introduce higher-performance products. Despite a loss of $18 million on $116 million sales for the year ended March 31, CMI still has about $30 million in capital, officials say. Now two likely scenarios loom: either the company will seek an acquisition or be acquired itself. ELECTRONICS SHARE OF U.S. GNP SLIPS With $229 billion in sales, the U.S. electronics industry accounted for 5.7% of the U.S. gross national product in 1985—down from its 6.1% share in 1984, according to the American Electronics Association. The AEA attributes the decline to the slump the industry is suffering and calls the dip "a short-term break" in a long-term upward trend. Electronics is the nation's third largest industry, ranking behind only transportation equipment, at 7.8% of the GNP, and food and related items, at 7.4%. ROLM REALIGNS MANUFACTURING ... Rolm Corp. is realigning its manufacturing operations. Focusing activities of the newly formed Telephone Products Division in Austin, Texas, and the ChX Division in Santa Clara, Calif., the company is keeping only final assembly and test at the Santa Clara headquarters. Board-level activity will be moved to sites in Colorado and Texas. Analysts say the changes, which put like activities in the same locations, give Rolm a chance to achieve higher efficiency in production. ... WHILE HP SPLITS SYSTEMS SECTOR In an effort to give greater emphasis to its 3000 line of business computers and to its personal computer line, Hewlett-Packard Co. is splitting its Information Systems and Network Sector in two. The new Business Systems and Personal Computation Sector is headed by senior vice president Douglas C. Chance, 44, a 20-year HP veteran. Executive vice president John L. Doyle will run the new Systems Technology Sector and maintain responsibility for peripherals and computer manufacturing. AT&T SCALES DOWN OFFICE PBX Gunning for a bigger share in what it thinks is the fastest growing segment of the PBX market—small-to-medium-sized offices—AT&T Co. introduced the System 25, a scaled-down version of its System 75 digital switch that will replace its analog Horizon system. AT&T has not secured as large a share in this niche as in others, because of competition from ITT, Mitel, and Rolm. The company is positioning the 50-to-200-line switch between two successful systems—Merlin, which can handle up to 80 users, and the System 75, which supports up to 800 stations. 18 NATIONS BLESS EUREKA PROJECT Research ministers from 18 European countries have approved 62 joint research and development projects worth more than $2 billion under the pan-European Eureka project. Eureka is a French-proposed alternative to the U.S. Strategic Defense Initiative. French companies are the most widely represented, participating in 40 of the projects, while British and West German firms will take part in 29 and 15 of the projects, respectively. With life spans of 2 to 10 years, most of the projects concentrate on software engineering, microelectronics, data processing, and telecommunications. If the card below has already been used, you may obtain the needed information by writing directly to the manufacturer, or by sending your name and address, plus the Reader Service card number and issue date, to Electronics Reader Service Department, P.O. Box 2713, Clinton, Iowa 52735. Electronics July 10, 1986 This Reader Service Card expires October 10, 1986 Name ____________________________________________ Title ____________________________________________ Phone ____________________________________________ Company ____________________________________________ Street address (company) \| at home \| check one) ____________________________________________ City ____________________________________________ State ____________________________________________ Zip ______________ Signature ____________________________________________ Date _______________________ Was this magazine personally addressed to you? \| Yes \| No Industry classification (check one) A. Computers Computer Systems and Equipment B. Communications Communications Systems and Equipment C. Military/Space Systems and Equipment D. Industrial Systems Controls and Equipment E. Electronics Subassemblies Components and Materials F. Test and Measurement Equipment G. Consumer Products H. Medical Systems and Equipment I. Software J. Research & Development K. Educational 2-4 Year College University L. Other_________________________ Please specify Your principal job responsibility (check one) A. Corporate/Operating and General Management B. Design & Development Engineering C. Engineering Services D. Basic Research E. Manufacturing & Production F. Other Estimate number of employees (at this location): 1. under 20 2. 20-99 3. 100-999 4. over 1000 \| 1 \| 16 \| 31 \| 46 \| \|----\|-----\|-----\|-----\| \| 2 \| 17 \| 32 \| 47 \| \| 3 \| 18 \| 33 \| 48 \| \| 4 \| 19 \| 34 \| 49 \| \| 5 \| 20 \| 35 \| 50 \| \| 6 \| 21 \| 36 \| 51 \| \| 7 \| 22 \| 37 \| 52 \| \| 8 \| 23 \| 38 \| 53 \| \| 9 \| 24 \| 39 \| 54 \| \| 10 \| 25 \| 40 \| 55 \| \| 11 \| 26 \| 41 \| 56 \| \| 12 \| 27 \| 42 \| 57 \| \| 13 \| 28 \| 43 \| 58 \| \| 14 \| 29 \| 44 \| 59 \| \| 15 \| 30 \| 45 \| 60 \| \| 61 \| 76 \| 91 106 \| \|----\|-----\|-------\| \| 62 \| 77 \| 92 107 \| \| 63 \| 78 \| 93 108 \| \| 64 \| 79 \| 94 109 \| \| 65 \| 80 \| 95 110 \| \| 66 \| 81 \| 96 111 \| \| 67 \| 82 \| 97 112 \| \| 68 \| 83 \| 98 113 \| \| 69 \| 84 \| 99 114 \| \| 70 \| 85 \| 100 115 \| \| 121\| 136\| 151 166 \| \| 122\| 137\| 152 167 \| \| 123\| 138\| 153 168 \| \| 124\| 139\| 154 169 \| \| 125\| 140\| 155 170 \| \| 181\| 196\| 211 226 \| \| 182\| 197\| 212 227 \| \| 183\| 198\| 213 228 \| \| 184\| 199\| 214 229 \| \| 185\| 205\| 215 230 \| \| 241\| 256\| 271 348 \| \| 242\| 257\| 272 349 \| \| 243\| 258\| 273 350 \| \| 244\| 259\| 274 351 \| \| 245\| 260\| 275 352 \| \| 363\| 378\| 393 408 \| \| 364\| 379\| 394 409 \| \| 365\| 380\| 395 410 \| \| 366\| 381\| 396 411 \| \| 367\| 382\| 397 412 \| \| 423\| 438\| 453 468 \| \| 424\| 439\| 454 469 \| \| 425\| 440\| 455 470 \| \| 426\| 441\| 456 471 \| \| 427\| 442\| 457 472 \| \| 483\| 498\| 703 718 \| \| 484\| 499\| 704 719 \| \| 485\| 500\| 705 720 \| \| 486\| 501\| 706 900 \| \| 487\| 502\| 707 901 \| \| 488\| 503\| 708 902 \| \| 489\| 504\| 709 951 \| \| 490\| 505\| 710 952 \| \| 491\| 506\| 711 953 \| \| 492\| 507\| 712 954 \| \| 493\| 508\| 713 956 \| \| 494\| 509\| 714 957 \| \| 495\| 510\| 715 958 \| \| 496\| 701\| 716 959 \| \| 497\| 702\| 717 960 \| YES please enter my subscription to Electronics for the term checked below: Name ____________________________________________ Title ____________________________________________ Company ____________________________________________ Div or Dept. ____________________________________________ Co. address ____________________________________________ City ____________________________________________ State ____________________________________________ Zip ______________ Check here if you wish publication to be sent to home address: ____________________________________________ Street ____________________________________________ City ____________________________________________ State ____________________________________________ Zip ______________ Those not qualifying may pay higher than basic price. US CANADA MEXICO One Year: $32 $ 51 CDN 17,850 Pesos Two Years: $55 $ 86 CDN 29,925 Pesos Three Years: $76 $117 CDN 40,950 Pesos Payment enclosed Bill me Charge my subscription to: AMEX DC VISA MC-Bank# Acct# ____________________________ Exp. ______________ Signature ____________________________________________ Subscriptions are normally entered within 2 weeks, but please allow 4 weeks for shipment. Please check which best describes your company's business at your location: Manufacturing Distribution Retailing Other Indicate the primary product manufactured or services performed at your plant: Space applicable A. Computer hardware processing and peripheral equipment office and business machines B. Communications, data communications, telecommunications systems and equipment C. Navigation, guidance, flight control, launch and missile systems or equipment (aerospace) D. Test and measurement equipment E. Consumer products (audio, video, television, home computers, appliances) F. Medical, surgical, medical controls, systems or equipment G. Semiconductor production equipment (component insertion, coil winding, etc.) H. Electronic components, materials, semiconductors (passive devices, discrete devices, IC's, diodes, hybrids, power supplies) I. Other manufacturing using electronics equipment as part of their product (machine tools, chemicals, metals, plastics, pharmaceuticals, etc.) J. Government K. Independent research and development laboratory or consultant L. Research and development organizations which are part of an educational institution M. Independent testing laboratories N. Operators of communications equipment (utilities, railroads, police and airlines broadcasting, etc.) O. Educational 2-4 year college, university P. Other Indicate that your business functions occur in more than one of the above categories: A. Corporate Management (Owners, Partners, Presidents, VP's, etc.) B. Operating Management (General Managers, Group Managers, Division Heads, etc.) C. Engineering Management (Project Managers, Engineering Executives, Division Heads, VP Engineering) D. V.P. Research and Development V.P Quality Control, etc. E. Sales F. Systems Engineering/Integrators G. Design Engineering H. Marketing (Sales, Market Lab Assistants, etc.) I. Test Engineering (Materials, Test Evaluation) J. Field Service Engineering K. Research and Development (Scientists, Chemists, Physicists) L. Manufacturing and Production M. Purchasing and Procurement N. Marketing and Sales O. Operations and Director at ____________________________ P. Senior Manager at ____________________________ Q. Graduate Student at ____________________________ Indicate your principal job responsibility. (space the appropriate number in box) 1. Manufacturing 2. Engineering Management 3. Engineering Estimated number of employees at this location (check one) 1 to 49 50 to 249 250 to 999 over 1,000 Date ____________________________ Signature ____________________________ All information is required for reentry of subscription 5571 For additional information on products advertised, new products or new literature, use these business reply cards. Complete entire card. Please print or type. Circle the number on the Reader Service postcard that corresponds to the number at the bottom of the advertisement, new product item, or new literature in which you are interested. To aid the manufacturer in filling your request, please answer the three questions. All inquiries from outside the U.S. that cannot reach Electronics before the expiration date noted on the Reader Service postcard must be mailed directly to the manufacturer. The manufacturer assumes all responsibilities for responding to inquiries. Subscriptions & Renewals Fill in the subscription card adjoining this card. Electronics will bill you at the address indicated on the card. Electronics P.O. BOX NO. 2713 CLINTON, IOWA 52735 BUSINESS REPLY MAIL FIRST CLASS PERMIT NO. 55 MARTINSVILLE, NJ Postage will be paid by Addressee Electronics CN 807 MARTINSVILLE, NJ 08836 NO POSTAGE NECESSARY IF MAILED IN THE UNITED STATES Hello, AT&T's Small Business Connection? Listen, my small business lives and dies by the telephone... and with these "bargain" phones I bought, it's mostly dying... Sure, I'd like reliable AT&T phones, but won't that take a big bite out of my bottom line? You're kidding! I can afford the Merlin system with AT&T's Push-Button Savings Plan? That's great... Hey, can you speak up? This so-called telephone is really going to the dogs... The AT&T MERLIN® System, America's favorite small business phone, can cost you less than you think. If your business has been bitten by bad phones, you're probably ready for the MERLIN Communications System. It offers all of today's standard features—speed dialing, conference calling, an intercom. Plus it lets you add extra features (or extra telephones) as your business needs change. Which means the MERLIN System is totally adaptable. And it's also totally affordable, thanks to AT&T's Push-Button Savings Plan. Call AT&T's Small Business Connection (1 800 247-7000) for more information about financing the MERLIN System. Perhaps the ultimate telephone for today's dog-eat-dog world. Powerful, single board imaging and graphics systems for the VMEbus. A requirement for high performance graphics or imaging is no longer a difficult one for designers who are building on strengths of the VMEbus. Matrox has brought the video-based expertise to the VMEbus with two powerful new single board products. You'll know it when you see the masterstroke. Call us, and we'll understand why. VG-340 Color Display Processor - Via-VME high level 2D/3D command set C - 40 x 40 x 8 bit resolution - 16 colors from a palette of 32K 1:4 - 128x128 bit display list processor - 20,000 vectors/second drawing speed - Single VME dual height board VIP1024 Real-Time Image Digitizer - Single board input/output system - 128x128 x 8 bit frame buffer - 1024 x 1024 x 8 bit display resolution - Interlaced grabbing capability - 8-bit flash frame grabber - 256 colors from a palette of 16.7 million matrox 1055 St. Régis, Dorval, Quebec, Canada H9P 2T4 Tel.: (514) 685-2630 Telex: 05-822798 Toll-free no.: 1-800-361-4903
To Live With Christ A priest friend of mine visited a family with several sons thinking about the priestly vocation. During his visit he celebrated Mass for them at the local parish. The Gospel reading was the same as today: “Whoever loves father or mother more than me is not worthy of me....” My friend thought twice before reading it, afraid that the father who wasn’t open to his sons’ vocations would think he picked this passage on purpose. He read it anyway and, yes, got the reaction he suspected he’d get. Most words in the Gospels are challenging. Jesus calls us to love him above all things. This means he has to come first, and sometimes other loves, as good as they are, cannot be the ultimate criteria. He tells us to “lose our life” to find it, and sometimes that loss is real and painful. He demands this because he himself experienced its truth. In his public ministry, he put his mission and love of souls ahead of spending more time with family and friends. When religious leaders, government officials, and others criticized and persecuted him, he didn’t back down, but he didn’t run from even if it affected his loved ones. He was faithful in charity. He didn’t harm them, either. He avoided unnecessary conflict when he could, but he completely obeyed God’s will and embraced the hardships of loving until the end. This kind of sacrifice brings great joy if endured with acceptance. As Romans tells us, “If, then, we have died with Christ, …we shall also live with him.” —Fr. Mark Haydu, LC For Reflection How can I live my life as though nothing is more important than Jesus? Do I regularly witness to my love for Jesus, using words when necessary? “…Whoever does not take up his cross and follow after me is not worthy of me” (Matthew 10:37). THIRTEENTH SUNDAY IN ORDINARY TIME – JULY 2, 2017 The Sanctuary Lamp Burns In Memory of Adeline Villano ★ ★ MASS INTENTIONS ★ ★ SAT 5:15 PM Angelo Bascelli Family SUN 7:30 AM Living & Deceased Parishioners 9:00 AM Eleanor Painter – 16th Anniversary 11:00 AM Joseph, Mary, Kathleen & Rosalie Ruby MON No Mass TUES 9:00 AM Daniela Serpico WED 7:30 AM Lois Wingle – 4th Anniversary THURS 7:00 PM Donald Wheeler FRI 7:30 AM Paul Kakalecik SAT 5:15 PM Living & Deceased Parishioners SUN 7:30 AM John Bogus 9:00 AM Annette & Raymond Brennan 11:00 AM Bill Flanagan NEXT WEEK AT A GLANCE SCRIP: Sponsor: Dowling 5:15 Costanzo 7:30 Kasperowicz 9:00 Carvalho 11:00 Citrano DAY MASS LECTOR COMMENTATOR ALTAR SERVERS Sat 5:15PM Jason Mariano Ann Marie Arbo Vinny Ferrara, Aiden Reber, Luke Voelker Sun 7:30AM Jim Kaucher Toni Kaucher Mason and Morgan Cambria, Samantha Reitnour 9:00AM Isabelita Santos Gretchen Markel Adam and Carlina Davis, Ally Santos 11:00AM Bob Newman Mary Jo Simon Lilia and Simon Crew, Sidney Wolf Second Collection Next Week is Monthly Collection 2017 Vacation Bible School "Mission Discovery" July 17 through July 21, 6-8pm Open to children ages 5 to 11 Adult and teen volunteers are always welcome. Join the mission to discover what God is calling you to do with your life! Contact Mrs. O'Donnell at 610-582-3880 or email@example.com with questions. Remember in your prayers the sick of our parish: Marco Fiato, John & Margaret Kakas, Delores Woy, Alice Arantowicz, Yolanda Marino, Ann Schiott, Kathy Patrick, Joseph & Susan McCorristin, Mary & Phil Morici, Dan Casciano, Jake Maccarone, Madge Butto, Joan Cronin, Rita Conrad, Rose Osorio, Steve Mannarella, Joseph Kolenda, Dominic Strock, Bob Hocker, Mike DiScala, Mildred Mattassa, Deb Petras, Ronald Krymnan, Anna Swiezak, Margarie & Walter Hansen, Robert Thomas, Rita Dougherty, Norma Nanni, Lou Franco, Lucille Schiel, John Fritz, Lucy Doney, Mary Okuniewski, Joe Weyand, Mark McDonald, Stephen Spitko, Virginia Marks, Ken Runler, Helen McDonald, Mark Vignone, George Sohn, Justine Marie Mattassa, Paul Milshaw, Maria Houck, Barbara Reznick, Darlene Strock, Irene Wade, Harry Wood, Tony Guilliams, Fannie Hunter, Jacob Hafer, John LeFosse, Eileen Ridall, Lucia Scotto, Anthony Cossa and TJ Doyle. Pray for the recently departed: Maryanne Prentice HOSPITAL VISITATIONS: Monsignor McCann, Deacon Hiryak, Deacon Bardi and Deacon Giordano visit our parishioners who are hospitalized, in nursing homes or homebound regularly. If you or a member of your family is hospitalized, please call the parish office at 610-582-2411. On Independence Day, Tuesday, July 4, 2017, the Holy Sacrifice of the Mass will be offered at 9am. We invite you to pray for our nation and for peace and justice throughout the world. Summer Office Hours Immaculate Conception Church Tuesday through Friday, 9:30am to 4:30pm Immaculate Conception Academy Tuesday through Thursday, 9:00am to 1:00pm Immaculate Conception Academy has openings in its Full Day Kindergarten program for the 2017-18 school year! In addition to providing a strong foundation in mathematics and language arts, our program features S.T.E.M. (Science, Technology, Engineering and Math) education as well as art, music, and physical education. Before and after school care options are available. For more information or to schedule a tour, please contact Adele Reilly, Advancement Director at 610-404-8645 or firstname.lastname@example.org. Openings are still available in Preschool, Pre-K and grades 1-8 as well! Contact us today!!! Space is limited in some classes Mission Trip: Please pray for two of our young parishioners Aubrey and Faith Ouimet who departed on Friday for a Mission Trip to Antigua and Guatemala where they will participate in a variety of service activities including building homes for families who are in desperate need and distributing food and clothing. Through your generosity to our Charity Fund Collection, our parish family has contributed to this most worthwhile endeavor. Please accompany Aubrey and Faith with your prayers this week. "With a profound sense of gratitude for God’s many gifts, we would like to thank Monsignor McCann and the parish of Immaculate Conception for their prayers and support for our mission trip to Guatemala. The thoughtful donations provided by our parish will help us serve those in need in Guatemala to provide food and clothing as well as improved housing. Thank you for your generosity to help us bring God’s love to those less fortunate. We are blessed to be a part of this loving and generous parish." Resorts AC Trip – Come along and enjoy the fabulous Jersey ocean! Friday, July 21st – leave church promptly at 9:30am. Package deal – Pay $32, get $25 back. Call Carol J if interested at 610-775-4447 Position Opening Principal 2017-2018 Academic Year Immaculate Conception Academy Immaculate Conception Academy (Grades K-8), Douglassville, PA is seeking a dynamic and innovative educational leader with a strong commitment to academic excellence and Catholic values to serve as Principal. The ideal candidate must be experienced in and committed to the mission of Catholic education. Candidates should possess the acumen and experience to work effectively with academic and student affairs, the application of instructional technology, strategic planning, enrollment management, institutional advancement and fiscal management. Applicants should possess excellent oral and written communication skills and a collaborative leadership style capable of working with a Board of Limited Jurisdiction. Those interested in the position should submit their letter of interest, resume, background clearances, and appropriate certifications with personal and professional references to ICA Search Committee at email@example.com or by mail to ICA Search Committee c/o 310 Washington Street, Birdsboro, PA 19508. The Diocesan application packet may be obtained by calling 610-866-0581, ext. 3032. Application deadline: July 10, 2017 No phone calls please. Immaculate Conception Academy is an alumni school of the Healey Education Foundation. EOE. Love, Learning, Leadership Immaculate Conception Academy Positions Available Immaculate Conception Academy is accepting applications for teacher and classroom aide openings, including a part time music teacher for the 2017-18 school year. Teacher candidates must hold or be eligible for PA certification. Teaching candidates, please send letter of interest, resume, copies of certification and transcripts, along with clearances and references. Classroom aide candidates, please send letter of interest and resume to Mrs. Christine Foley, 903 Chestnut Street, Douglassville, PA 19518. Immaculate Conception Academy Transfer Grants Any public, private, or charter school student who transfers into Immaculate Conception Academy, grades 1 through 7, will receive a Bishop’s Tuition Transfer Grant of $1,500 over two school years ($1,000 the first year; $500 the second). Call Adele Reilly at 610-404-8645. CCD Registration Forms are available in the Narthex, on icbvmcod.org, at the parish office, and at the Office of Religious Education. Classes are filling up so register now! Volunteers are always welcome! If you would like to volunteer an hour of your time once a week to bring the Good News of Christ to the children of our parish by educating, please call Mrs. O’Donnell at (610) 582-3880. Dear Padre, If my fiancé and I abstain from sex, why can’t we live together before we get married? It seems to me there are two reasons living with your future spouse before marriage is not a good idea. One is that you are setting each other up for some serious temptations, or an “occasion of sin.” It would be similar to putting a bottle of gin on the desk of an alcoholic or a carton of cigarettes in front of someone who has succeeded in giving up smoking. I’m not saying you have some sort of addiction, but the natural pull of sex is going to be particularly strong, especially when you’re young, in love, and ready to commit to marriage. You will be free and even encouraged in marriage to pursue the greatest intimacy. But for the moment, it is asking too much to put you both in the same circumstances now that you will be in after you wed. The second reason for not living together before marriage is that the living situation could give scandal to others. If you are active in your church and vocal about practicing your Catholic faith but then start living with your future spouse before marriage, it can raise questions (and eyebrows) about what you really believe and how you’re living your faith. There may be some good-sounding reasons for living together. You can cut down on costs, get to know your future spouse better, and so on. But these reasons aren’t worth the risk. My advice is that you help each other live a virtuous life now, by sacrificing the convenience of living together outside of marriage, as the best practice for your life together in the sacrament of matrimony. Fr. Gary Lauenstein, CSsR firstname.lastname@example.org A Word From Pope Francis “Commitment to mission is not something added on to the Christian life as a kind of decoration, but is instead an essential element of faith itself. A relationship with the Lord entails being sent out into the world as prophets of his word and witnesses of his love. Even if at times we are conscious of our weaknesses and tempted to discouragement, we need to turn with God with confidence. We must overcome a sense of our own inadequacy and not yield to pessimism, which merely turns us into passive spectators of a dreary and monotonous life. There is no room for fear! God himself comes to cleanse our ‘unclean lips’ and equip us for the mission.” —Message, Fifty-fourth World Day of Prayer for Vocations, November 27, 2016
Electrically charged strange stars with an interacting quark matter equation of state V. P. Gonçalves and L. Lazzari High and Medium Energy Group, Instituto de Física e Matemática, Universidade Federal de Pelotas (UFPel) Caixa Postal 354, 96010-900, Pelotas, RS, Brazil. The properties of electrically charged strange quark stars predicted by an interacting quark matter equation of state (EoS) based on cold and dense perturbative quantum chromodynamics (pQCD) are investigated. The stability of strange stars is analyzed considering different models for the electric charge distribution inside the star as well as for distinct values for the total electric charge. A comparison with the predictions derived using the MIT bag model is also presented. We show that the presence of a net electric charge increases the stars masses and decreases their mass in comparison to their neutral counterparts. Moreover, we demonstrate that the pQCD EoS implies larger values for the maximum mass of charged strange stars, with very heavy charged stars being stable systems against radial oscillations. For an electric charge distribution given by $q(r) = \beta r^3$, the pQCD EoS implies unstable configurations for large values of the renormalization scale as well as for large values of $\beta$, in contrast to the MIT bag model predictions. I. INTRODUCTION The description of matter at high densities and/or high temperatures is one of the main challenges of the strong interactions theory – the Quantum Chromodynamics (QCD) (For a recent review see, e.g. Ref. [1]). While the regime of high temperatures and vanishing baryon density have been explored in heavy ion collisions at RHIC and LHC, the QCD at high baryon densities and low temperatures is fundamental to determine the properties of compact stars [2], where the density of the matter is predicted to exceed the nuclear density matter and the system is expected to be described in terms of deconfined quark degrees of freedom. According to the the Bodmer-Witten hypothesis [3, 4], the absolute ground state for the hadronic matter is the strange quark matter (SQM) rather than $^{56}$Fe, which implies the possible existence of compact stars entirely made of deconfined up, down and strange quarks, usually denoted strange quark stars (SQS) [5–8]. Another possibility is the presence of deconfined quark matter inside neutron stars (NSs), forming hybrid stars. In this case, SQM is expected to be formed through nucleation processes, converting a previously hadronic star (HS) into a SQS with the liberation of a large amount of energy that generates a neutrino burst and an intense gravitational waves emission. In fact, both HSs and SQSs can exist in Nature, in the so called two-families scenario, where neutron stars with masses up to $1.5\,M_\odot$ are HSs and the those with mass above this threshold are SQSs [10]. These distinct scenarios have been investigated in detail by several authors over the last years. Its predictions compared with recent experimental data constrain the equation of state (EoS) that describes matter in NSs (for recent reviews see, e.g. [2, 9, 10]). Robust astrophysical constraints on the EoS came from the two solar mass limit of the pulsars PSRJ1614 − 2230 and PSR J0348 + 0432 [11–13] and from the LIGO/Virgo detection of gravitational waves originating from the NS merger event GW170817 [14]. In particular, the electromagnetic and gravitational wave information from the GW170817 event have been used to constrain the radii $R$, maximum mass $M_{max}$ and tidal properties of NSs. In addition, a recent result [15] pointed out that dense matter in the interior of massive NSs ($M \approx 2M_\odot$) exhibits characteristics of a deconfined quark phase in the core of the star. Even so, the interpretation of these recent results still is a theme of intense debate and the existence of hybrid and/or strange quark stars remains an open question. As pointed out before, the discovery of pulsars with large masses ($M \gtrsim 2\,M_\odot$) [16] has put strong constraints on the equation of state (EoS) of dense stellar matter and has challenged the description of these objects as being quark stars, which were predicted to have smaller masses by models based on the phenomenological MIT bag model EoS. However, the main properties of these compact objects strongly depend on a precise description of the matter that composes the star [8]. In particular, it is the EoS that defines the magnitude of the internal pressure that competes with gravity and, consequently, that establishes the stability of the star. During the last years, several phenomenological models have been proposed to describe the EoS for the deconfined quark system, considering different assumptions and approximations for the description of the interaction between quarks, as well as for the treatment of the running quark masses and coupling constant (See, e.g. Refs. [17–28]). In particular, in Ref. [19], the authors have derived an EoS based on cold and dense perturbative QCD (pQCD). They were able to estimate the pressure at non-zero density at order $\alpha_s^2$, where $\alpha_s$ is the strong coupling constant, assuming a non-zero value of the strange quark mass. One advantage of this EoS, is that it allows to estimate the systematic uncertainty present in a perturbative calculation. The results presented in Refs. [19, 20] demonstrated that strange quark stars with masses larger than $2\,M_\odot$ can be reached for large values of the renormalization scale. In addition to the EoS, the maximum mass of the star can be modified by the presence of electric charge in its interior. Neutron stars, so as SQSs, are electrically neuHowever, to achieve electrical neutrality in quark matter one must allow the possibility of leptons being present. In particular, for the latter, the presence of electrons is necessary to form a chemically equilibrated system, where $\beta$-equilibrium is achieved. Such electrons are distributed in a layer on the surface of the star, which is separated from strange matter by several hundred Fermi. These two systems interact through electrostatic force, resulting in a charge distribution inside the star. In general, stars can remain in equilibrium under their own gravity and electric repulsion, with the Coulombian force acting in addition to the gradient pressure to counterbalance the gravitational attraction, which implies that the charged SQS can be more stable than the neutral one. As pointed out in Refs. [29–31], the surface of a SQS has a high electric field, which can reach about $1 \times 10^{21}$ V/m. The results presented in [30, 31] and [32] using the massless version of the MIT bag model and a non-linear EoS, respectively, have demonstrated that charged stars are heavier than their neutral counterparts, with the increasing in the maximum mass being dependent of the magnitude of the electric charge. As a consequence, the presence of the electric charge can modify the star compactness, given by the ratio between its mass and radii $(M/R)$, which determines tidal deformability. Such property have been determined using the recent data for the GW170817 event, which have also been used to constrain the EoS (See e.g. Refs. [33–35]). Our goal in this paper is to present, for the first time, the predictions from pQCD for charged SQSs. In particular, we will investigate the hydrostatic equilibrium considering the interacting quark matter EoS derived in Refs. [19, 20] and different models for the electric charge distribution inside the star, as well as for different values of the total charge. In addition, we will investigate the impact of radial oscillations on the stability of charged SQS. Our study is strongly motivated by the possibility to use its predictions and experimental tests to improve our understanding of the inner structure of compact objects as well as to constrain the EoS of the system (See e.g. Ref. [36]). Previous calculations for radial modes in neutral and charged SQS were performed in Refs. [30, 31, 37–39] considering different models for the EoS. Our aim is to investigate the dependence on the EoS of the dynamical stability of charged SQS against radial perturbations considering different assumptions for the electric charge distribution. This paper is organized as follows. In the next section, we will present a brief review of the formalism used to describe charged SQSs and their radial oscillations. The different models used to describe the charge distribution in the star will be discussed and the pQCD EoS presented. In section III, we present our results for the mass-radius profile and for the fundamental mode of oscillation considering the pQCD EoS and distinct models for the treatment of charge inside the star. The predictions derived using the MIT bag model considering massive quarks are also presented, which improve the analysis performed in Ref. [31] and allow a detailed comparison with the pQCD results. Finally, in section IV we summarize our main conclusions. In what follows, we use the units $c = 1 = G$. II. FORMALISM The presence of charge in a SQS implies that the system should be described by the Einstein-Maxwell field equations, with the energy density associated to the electric field being present into the energy - momentum tensor. One has that due to the high value of the electric field on the surface of the star, the electric energy density is of the same order as the energy density of the strange quark matter. As a consequence, the high electric field expected on the surface of a SQS, affects the space-time metric and the associated energy density contributes to its own gravitational mass. Moreover, the presence of the Coulomb interaction modifies the structure equations that describe the relativistic hydrostatic equilibrium (See Ref. [38]). In order to describe a spherically symmetric static charged star, we will assume a line element given by $$ds^2 = e^{2\nu(r)}dr^2 - e^{2\lambda(r)}dr^2 - r^2(d\theta^2 + \sin^2\theta d\phi^2), \quad (1)$$ where $\lambda$ and $\nu$ are known as the metric functions. Such metric implies that a charged star constituted by a perfect fluid will satisfy the following stellar structure equations $$\frac{dq}{dr} = 4\pi r^2 \rho_e e^\lambda, \quad (2)$$ $$\frac{dm}{dr} = 4\pi r^2 \epsilon + \frac{q}{r} \frac{dq}{dr}, \quad (3)$$ $$\frac{dP}{dr} = -(\epsilon + P)\left(4\pi rP + \frac{m}{r^2} - \frac{q^2}{r^3}\right)e^{2\lambda}$$ $$+ \frac{q}{4\pi r^4} \frac{dq}{dr}, \quad (4)$$ $$\frac{d\nu}{dr} = -\frac{1}{\epsilon + P}\left(\frac{dP}{dr} - \frac{q}{4\pi r^4} \frac{dq}{dr}\right), \quad (5)$$ where $\rho_e(r)$ is the electric charge density, $q(r)$ and $m(r)$ represent the charge and mass within radius $r$, respectively. The metric potential $e^{-2\lambda}$ has the Reisser-Nordström form $$e^{-2\lambda(r)} = 1 - \frac{2m(r)}{r} + \frac{q(r)^2}{r^2}. \quad (6)$$ For neutral stars one has that $q(r) = 0$ and Eq. (4) reduces to the usual Tolman-Oppenheimer-Volkoff equation. To solve such a system of equations, one needs to establish the boundary conditions. At the center of the star we assume that $$q(0) = m(0) = 0, \quad \epsilon(0) = \epsilon_c, \quad \nu(0) = \nu_c. \quad (7)$$ Moreover, we assume that the solutions on the surface of the star $(r = R)$ satisfy the following conditions \[ p(R) = 0, \tag{8} \] \[ m(R) = M, \tag{9} \] \[ q(R) = Q, \tag{10} \] \[ \nu(R) = -\lambda(R), \tag{11} \] where $M$ and $Q$ are the total mass and electric charge of the stellar system, respectively. In order to investigate the stability of the charged SQS against radial oscillations, we will consider the approach proposed by Chandrasekhar [40] many years ago, which demonstrated that perturbing the fluid and space-time variables in a manner that maintains the spherical symmetry of the system, it is possible to derive an equation for infinitesimal radial oscillations of a spherical object: the pulsation equation. Such equation is given by \[ \frac{\mathrm{d}}{\mathrm{dr}} \left[ P \frac{\mathrm{du}}{\mathrm{dr}} \right] + [\mathcal{Q} + \omega^2 \mathcal{W}] u = 0 \tag{12} \] where $u$ is the renormalized displacement function and for a charged star we have that [38] \[ P = e^{\lambda + 3\nu} r^{-2} \gamma P, \tag{13} \] \[ \mathcal{Q} = (\epsilon + P) r^{-2} e^{\lambda + 3\nu} \left[ \nu' \left( \nu' - 4r^{-1} \right) - (8\pi P + r^{-4} q^2) e^{2\lambda} \right], \tag{14} \] \[ \mathcal{W} = e^{3\lambda + \nu} r^{-2} (\epsilon + P), \tag{15} \] where $\gamma$ is the adiabatic index. The pulsation equation constitutes a Sturm-Liouville eigenvalue problem, which allows to obtain the eigenvalues and eigenfunctions of the radial perturbation. Defining the auxiliary variable $\eta \equiv P \frac{\mathrm{du}}{\mathrm{dr}}$ we can transform the above second-order differential equation into two first order differential equations given by \[ \frac{\mathrm{du}}{\mathrm{dr}} = \frac{\eta}{P}, \tag{16} \] and \[ \frac{\mathrm{d}\eta}{\mathrm{dr}} = -[\mathcal{Q} + \omega^2 \mathcal{W}] u. \tag{17} \] In Ref. [36], the authors demonstrated that for $\eta(0) = 1$ one has $u(0) = r^3/(3P(0))$, which are the initial conditions for the integration of the equations from the origin to the surface of the star. As in Refs. [31, 36] we will use the shooting method to obtain the values of $\omega^2$ that satisfies the boundary condition given by \[ \left. \frac{\mathrm{du}}{\mathrm{dr}} \right\|_{r=R} = \eta(R) = 0. \tag{18} \] The structure equations are solved using a Runge-Kutta-Cash-Karp method with adaptive step size. Starting from a trial value for $\omega^2$, we obtained the values that satisfy the boundary conditions using the Newton-Raphson method. This values are the eigenfrequencies of the pulsation equation. One important aspect is that for charged SQS the condition $\partial M/\partial \epsilon > 0$ is not sufficient to determine the stability of the star [31]. Therefore, in order to investigate the stability of such objects, one has to perform the analysis of its radial perturbation modes. Since $\mathcal{Q}$ is real, the next eigenfrequency is always larger than the previous one, i.e., \[ \omega_0^2 < \omega_1^2 < \omega_2^2 \cdots < \omega_n^2 < \cdots . \] Consequently, to determine the stability of the star it is sufficient to analyze the sign of the fundamental mode. For $\omega_0^2 < 0$ the star is unstable. In order to solve the structure equations, we must specify the charge distribution and the EoS that describes the matter inside the star. In our analysis, motivated by the studies performed in Refs. [31, 41], we will consider the following models to describe the charge in the SQS: - Model A: The charge is proportional to the third power of the radial coordinate as follows \[ q(r) = Q \left( \frac{r}{R} \right)^3 \equiv \beta r^3, \tag{19} \] where $\beta \equiv Q/R^3$; - Model B: The charge density is proportional to the energy density, i.e., \[ \rho_c = \alpha \epsilon, \tag{20} \] where, in geometric units, $\alpha$ is a dimensionless proportionality constant; - Model C: The star has a fixed total charge $Q$. The main difference between models A and B is associated to the region inside the star where the charge is expected to be larger. In model A, the charge is almost totally concentrated on the surface of the star. While in model B there is a relevant amount of charge in the intermediate region between the center of the star and its surface. In contrast, model C is independent of the model assumed for the electric charge distribution inside the star. Regarding the EoS, the simplest model and more frequently used to describe the interior of a quark star is the MIT bag model [42], which characterizes a degenerate Fermi gas of up, down and strange quarks. In such model, the main properties only depend on the bag constant $B$. However, the MIT bag model is a naive approximation, which is not sufficiently powerful to characterize a system with interacting quarks or more complex structures. In our analysis, we will consider the pQCD EoS calculated in Ref. [19] at order $\alpha_s^2$ and a non-zero value of the strange quark mass. This description was put in a simple to use formula in Ref. [20], being given by \[ P = P_{SB}(\mu_B) \left( c_1 - \frac{a(X)}{\mu_B - b(X)} \right), \] where \[ P_{SB}(\mu_B) = \frac{3}{4\pi^2} \left( \frac{\mu_B}{3} \right)^4 \] corresponds to the pressure of a gas composed by three massless non-interacting quarks, also called a Stephan-Boltzmann (SB) gas, and the functions $ a(X) $ and $ b(X) $ are auxiliary functions (for details, see Ref. [20]). The dimensionless parameter $ X $ is proportional to the renormalization scale parameter $ \Lambda $ that arises in the perturbative expansion and is expressed as $ X = 3\Lambda/\mu_B $. Fixing $ X $, the energy density comes from the following relation \[ \epsilon = -P + \mu_B n_B, \] where $ n_B $ is the baryon number density obtained from the thermodynamical relation \[ n_B = \frac{\partial P}{\partial \mu_B}. \] In the next section, we will present our predictions for the mass-radius profile of the SQS as well as for the fundamental mode considering the pQCD EoS and the distinct models for the distribution of charge discussed above. ### III. RESULTS In Fig. 1 (a), we present a comparison between the EoSs for the quark matter predicted by the MIT bag model and the by the cold and dense pQCD calculation performed in Ref. [19]. For comparison, we present the MIT bag model predictions derived assuming that the bag pressure is 60 MeV fm$^{-3}$ and that the strange quark has a mass of $ m_s = 150 $ MeV. For the pQCD EoS we present the predictions derived assuming different values for the renormalization scale $ X $. We have that the pQCD EoS is strongly dependent on the renormalization scale, with the band representing the uncertainty associated to this scale. For completeness, let us initially solve the structure equations for neutral SQS, in which $ q(r) = 0 $, for the distinct EoSs discussed above. Our results for the mass-radius profile are presented in Fig. 1(b). We can see that the SQSs maximum masses depend strongly on the value of $ X $, increasing with $ X $ and reaching values larger than 2 M$_\odot$ for $ X \gtrsim 3 $ [20]. We have verified that the MIT bag model predictions are similar to those derived using the pQCD EoS for $ X \approx 2.8 $, which predicts values of maximum SQS masses smaller than 2 M$_\odot$. As already pointed out in Ref. [39], only values of $ X $ in the range between 3 and 3.2 satisfy simultaneously the GW170817 constraints of mass and radius [14]. The studies performed in Refs. [43–45] suggested that the upper limit on the neutron star mass is 2.17 M$_\odot$ (90% credibility). We present in Fig. 1 (c) the comparison between the fundamental eigenfrequencies obtained from pQCD (for different values of $ X $) and the MIT bag model as a function of the central energy density. For convenience, we are presenting results for the linear frequency associated to the eigenfrequency by $ f = \omega/2\pi $. The results indicate that the configurations for the distinct values of $ X $ are stable against radial oscillations ($ \omega_0^2 > 0 $), in agreement with the results presented in Ref. [39]. In what follows, we will present our results for charged SQS considering the pQCD EoS for $ X = 3 $, which satisfies the GW170817 constraints, and the distinct models for the electric charge distribution discussed above. For comparison, we will also present the predictions derived using the massive MIT bag model EoS with $ B = 60 $ MeV fm$^{-3}$ for the bag pressure. First, we present in Fig. 2 our predictions for model A, in which the charge distribution charge is given by $ q(r) = \beta r^3 $, assuming different values of $ \beta $, with $ \beta = 0 $ corresponding to the neutral SQS. Our results show that for both EoSs the presence of charge increases the maximum mass and radius of the star in comparison to their neutral counterpart. Such result agrees with those derived in Ref. [30], which pointed out that increasing the charge allows the star to sustain larger masses. However, the predictions derived using the pQCD EoS are even more sensitive to the presence of charge in the star, since the same variation of $ \beta $ leads to a considerable larger modification in the mass and radius of the respective stellar configurations in comparison to the MIT bag model predictions. The full circles and squares in the figures indicate, respectively, the points where the maximum mass configuration occurs and where the fundamental eigenfrequency is zero. For $ \beta = 0 $ we have that these two points coincide. Therefore, for neutral SQS, $ \partial M/\partial \epsilon_c > 0 $ is a necessary and sufficient condition to determine the configurations of stable equilibrium [31]. In contrast, for charged SQS, these two points are not coincident, with the zero eigenfrequency configuration occurring for larger central densities. Consequently, in order to determine the stability of a charged SQS the signal of the fundamental mode should also be analyzed [31]. In our analysis, using the MIT bag model EoS, we have obtained that increasing $ \beta $ up to $ 9 \times 10^{-4} $ M$_\odot$ km$^{-3}$ always produces stable configurations in which $ \partial M/\partial \epsilon_c > 0 $ and $ \omega_0^2 > 0 $. In contrast, for the pQCD EoS, the largest value of $ \beta $ considered, $ \beta = 9 \times 10^{-4} $ M$_\odot$ km$^{-3}$, gives an unstable solution for all values of $ \epsilon_c $. This explains why the associated predictions is not presented on the right panels of Fig. 2. It is important to emphasize that we have verified that a similar instability occurs in the MIT bag model predictions for $ \beta \gtrsim 5 \times 10^{-3} $ M$_\odot$ km$^{-3}$. In Fig. 3, we present our predictions using model B, in which the charge density is proportional to the energy density $ \rho_e = \alpha \epsilon $, considering different values for the diFIG. 1. (a) Comparison between the MIT bag model EoS and the pQCD one considering different values of the renormalization scale $X$; (b) Mass – radius profile and (c) linear fundamental frequency for a neutral SQS derived assuming the MIT bag model and pQCD EoS’s. dimensionless constant $\alpha$. Similarly to the results obtained for model A, the charge density proportional to the energy density also implies in SQSs with larger masses and radii. Again, the impact of increasing the values of $\alpha$ is larger on the pQCD EoS. The main difference between models A and B is associated to the fact that model B is stable for all values of $\alpha$ under analysis, independently of the EoS being considered. In particular, for large values of $\alpha$, model B predicts stable configurations with very high masses and radii. This difference occurs because, in model B, the changes in mass and charge depend on the energy density, which implies that the increasing of the electric field throughout the star is exactly counterbalanced by gravity. In contrast, in model A, the increasing of the electric field is very fast near to the surface of the star. At large values of the electric charge, the outward pressure associated to the electric field cannot be sustained by gravity, and the associated configuration is unstable. Even though stable charged SQSs with very large masses and radii are predicted by model B, it is important to emphasize that the largest mass pulsar ever observed has a mass of $2.14^{+0.10}_{-0.09} \, M_\odot$ for the 68.3% credibility interval and $2.14^{+0.20}_{-0.18} \, M_\odot$ for the 95.4% credibility interval [16]. Therefore, charged SQSs with highest mass values predicted by model B, although mathematically stable, probably do not exist in Nature. Finally, let’s consider model C, in which we assume a star with a fixed total charge $Q$. From Gauss’ law for electric fields, we have that such assumption implies that the predictions are independent of the charge distribution inside the star. Our results are presented in Fig. 4 for different values of the total charge $Q$. As in the previous models, model C also predicts that the presence of charge increases the maximum mass of the star. Our results also indicate that the point where the configuration with maximum mass occurs is slightly distinct from the position where the frequency of the fundamental mode is zero, with the distance between these points increasing with the value of the total charge $Q$, in contrast with the results obtained in Ref. [31]. Two comments are in order. First, in our analysis, we only have presented the predictions derived using the pQCD EoS for $X = 3$. However, we also have performed the analysis for other values of $X$. For values of $X < 3$, the results are similar to those presented above, with the main difference being that the predicted values of the maximum masses of the neutral and charged SQS are smaller. On the other hand, for $X = 4$ and using model A for the electric charge distribution, we have found that the presence of charge implies in unstable configurations, independent of the value of $\beta$. We have verified that for $\beta = 5 \times 10^{-4} \, M_\odot \, \text{km}^{-3}$, the configurations outside equilibrium start to appear when $X \geq 3.6$. Such result demonstrate the strong dependence of the predictions for the charged SQS on the EoS considered. Second, we have estimated the compactness $(M/R)$ of the star for the different values of the charge $Q$ and verified that it increases for larger values of $Q$. The increasing is expected to modify the tidal deformability of the star and, therefore, can be used to discriminate between the neutral and charged scenarios. Such result motivates a more detailed study, which we intend to perform in the near future. IV. SUMMARY In this paper we have investigated, for the first time, the equilibrium and stability of charged strange stars considering the EoS derived in Ref. [19], which takes into account the interaction between quarks and have been derived using cold and dense perturbative QCD. The predictions from the pQCD EoS for SQSs are expected to be more realistic in comparison to those derive e.g. using the MIT bag model, because it takes into account quark interactions in a more systematic way. We have considered three models for the treatment of charge in the SQS and have performed a detailed comparison between the predictions derived using the pQCD and MIT bag model. FIG. 2. Predictions from the MIT bag model (left panels) and pQCD (right panels) EoSs for the total gravitational mass as a function of the central energy density (upper panels) and radius (central panels), as well as for the linear fundamental frequency as a function of the mass (lower panels) considering Model A and different values of $\beta$, where $\beta = 0$ (black solid line) corresponds to a neutral strange star. The full squares represent the configurations for which $\omega_0^2 = 0$ and the full circles represent the maximum mass configurations. EoSs. For both EoSs, we have verified that the presence of a net electric charge implies in SQSs with larger maximum masses in comparison to their neutral counterparts. However, the pQCD EoS leads to larger values for the maximum mass of the charged SQS, with very heavy charged stars being stable systems against radial oscillations. In addition, our results also demonstrated that for a distribution of electric charge inside the star given by $q(r) = \beta r^3$, the pQCD EoS implies unstable configurations for large values of the renormalization scale $X$ as well as for large values of $\beta$, in contrast to the MIT bag model predictions. FIG. 3. Predictions from the MIT bag model (left panels) and pQCD (right panels) EoSs for the total gravitational mass as a function of the central energy density (upper panels) and radius (central panels), as well as for the linear fundamental frequency as a function of the mass (lower panels) considering Model B and different values of $\alpha$, where $\alpha = 0$ (black solid line) corresponds to a neutral strange star. The full squares represent configurations for which $\omega_0^2 = 0$ and the full circles represent the maximum mass configurations. ACKNOWLEDGEMENTS This work was partially financed by the Brazilian funding agencies CNPq, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) – Finance Code 001, FAPERGS and INCT-FNA (process number 464898/2014-5). FIG. 4. Predictions from the MIT bag model (left panels) and pQCD (right panels) EoSs for the total gravitational mass as a function of the central energy density (upper panels) and the radius (central panels), as well as for the linear fundamental frequency as a function of the mass (lower panels) considering Model C and different values of the total charge $Q$, where $Q = 0$ (black solid line) corresponds to a neutral strange star. The full squares represent the configurations for which $\omega_0^2 = 0$ and the full circles represent the maximum mass configurations. [1] R. Pasechnik and M. Sumbera, Universe 3, no.1, 7 (2017) [2] G. Baym, T. Hatsuda, T. Kojo, P. D. Powell, Y. Song and T. Takatsuka, Rept. Prog. Phys. 81, no.5, 056902 (2018). [3] A. Bodmer, Phys. Rev. D 4, 1601 (1971). [4] E. Witten, Phys. Rev. D 30, 272 (1984) [5] C. Alcock, E. Farhi and A. Olinto, Astrophys. J. 310, 261 (1986) [6] C. Alcock and A. Olinto, Ann. Rev. Nucl. Part. Sci. 38, 161 (1988) [7] N. K. Glendenning, Compact stars (Springer Science & Business Media, New York, 1997). [8] F. Weber, Pulsars as astrophysical laboratories for nuclear and particle physics (Institute of Physics, London, 1999). [9] I. Bombaci, D. Logoteta, I. Vidaña and C. Providência Eur. Phys. J. A 52, 58 (2016) [10] A. Drago and G. Pagliara, Eur. Phys. J. A 52, 41 (2016) [11] P. Demorest, T. Pennucci, S. Ransom, M. Roberts and J. Hessels, Nature 467, 1081-1083 (2010) [12] J. Antoniadis, P. A. C. Freire, N. Wex, T. M. Tauris, R. S. Lynch, M. H. van Kerkwijk, M. Kramer, C. Bassa, V. S. Dhillon, T. Driebe, J. W. T. Hessels, V. M. Kaspi, V. I. Kondratiev, N. Langer, T. R. Marsh, M. A. McLaughlin, T. T. Pennucci, S. M. Ransom, I. H. Stairs, J. van Leeuwen, J. P. W. Verbiest and D. G. Whelan, Science 340, 6131 (2013) [13] E. Fonseca, T. T. Pennucci, J. A. Ellis, I. H. Stairs, D. J. Nice, S. M. Ransom, P. B. Demorest, Z. Arzoumanian, K. Crowter, T. Dolch, R. D. Ferdman, M. E. Gonzalez, G. Jones, M. L. Jones, M. T. Lam, L. Levin, M. A. McLaughlin, K. Stovall, J. K. Swiggum and W. Zhu, Astrophys. J. 832, no.2, 167 (2016) [14] B. Abbott et al., [LIGO Scientific and Virgo], Phys. Rev. Lett. 119, no.16, 161101 (2017) [15] E. Annala, T. Gorda, A. Kurkela, J. Nattila and A. Vuorinen, Nature Phys. (2020), [arXiv:1903.09121 [astro-ph.HE]] [16] H. F. Cromartie, E. Fonseca, S. M. Ransom, P. B. Demorest, Z. Arzoumanian, H. Blumer, P. R. Brook, M. E. Deller, T. Dolch, J. A. Ellis, R. D. Ferdman, E. C. Ferrara, N. Garver-Daniels, P. A. Gentile, M. L. Jones, M. T. Lam, D. R. Lorimer, R. S. Lynch, M. A. McLaughlin, C. Ng, D. J. Nice, T. T. Pennucci, R. Spiewak, I. H. Stairs, K. Stovall, J. K. Swiggum and W. Zhu, Nat. Astron. 4, no.1, 72 (2019) [17] G. Lugones and J. E. Horvath, Phys. Rev. D 66, 074017 (2002) [18] Y. Ivanov, A. Khorovostukhin, E. Kolomeitsev, V. Skokov, V. Toneev and D. N. Voskresensky, Phys. Rev. C 72, 025804 (2005) [19] A. Kurkela, P. Romatschke and A. Vuorinen, Phys. Rev. D 81, 105021 (2010) [20] E. S. Fraga, A. Kurkela and A. Vuorinen, Astrophys. J. Lett. 781, no.2, L25 (2014) [21] J. Torres and D. Menezes, EPL 101, no.4, 42003 (2013) [22] A. Zacchi, R. Steele and J. Schaffner-Bielich, Phys. Rev. D 92, no.4, 045022 (2015) [23] T. Kojo, P. D. Powell, Y. Song and G. Baym, Phys. Rev. D 91, no.4, 045003 (2015) [24] S. Benic, D. Blaschke, D. E. Alvarez-Castillo, T. Fischer and S. Typel, Astron. Astrophys. 577, A40 (2015) [25] J. F. Xu, G. X. Peng, F. Liu, D. F. Hou and L. W. Chen, Phys. Rev. D 92, no.2, 025025 (2015) [26] H. Chen, J. B. Wei and H. J. Schulze, Eur. Phys. J. A 52, no.9, 291 (2016) [27] M. Dutra, O. Loureiro and D. P. Menezes, Phys. Rev. C 93, no.2, 025806 (2016) [28] C. Y. Shen and S. G. Zhou, Nucl. Phys. B 916, 669 (2017) [29] V. V. Usov, Phys. Rev. D 70, 067301 (2004); V. Usov, T. Harko and K. Cheng, Astrophys. J. 620, 915-921 (2005) [30] R. P. Negreiros, F. Weber, M. Malheiro and V. Usov, Phys. Rev. D 80, 083006 (2009) [31] J. D. V. Arbanii and M. Malheiro, Phys. Rev. D 92, 084009 (2015) [32] G. Panotopoulos and A. Rincon, Eur. Phys. J. C 79, no.6, 524 (2019) [33] M. Sieniawska, W. Turczanski, M. Bejger and J. L. Zdunik, Astron. Astrophys. 622, A174 (2019) [34] J. E. Christian, A. Zacchi and J. Schaffner-Bielich, Phys. Rev. D 99, no.2, 023009 (2019) [35] Q. w. Wang, C. Shi, Y. Yan and H. S. Zong, [arXiv:1912.02312 [hep-ph]]. [36] K. Kokkotas and J. Ruoff, Astron. Astrophys. 366, 565 (2001) [37] G. Lopez Flores and G. Lugones, Phys. Rev. D 82, 063006 (2010) [38] A. Brüllante and I. N. Mishustin, EPL 105, no.3, 39001 (2014) [39] J. C. Jimenez and E. S. Fraga, Phys. Rev. D 100, no.11, 114041 (2019) [40] S. Chandrasekhar, Phys. Rev. Lett. 12, 114 (1964) [41] D. Deb, S. V. Ketov, M. Khlopov and S. Ray, JCAP 10, 070 (2019) [42] A. Chodos, R. Jaffe, K. Johnson, C. B. Thorn and V. Weisskopf, Phys. Rev. D 9, 3471 (1974) [43] B. Margalit, D. Metzger, Astrophys. J. Lett. 850 (2017) no.2, L19 [44] M. Shibata, S. Fujibayashi, K. Hotokezaka, K. Kiuchi, K. Kyutoku, Y. Sekiguchi and M. Tanaka, Phys. Rev. D 96 (2017) no.12, 123012 [45] L. Rezzolla, E. R. Most and L. R. Weih, Astrophys. J. Lett. 852 (2018), L25
Estimating Use-Dependent Synaptic Gain in Autonomic Ganglia by Computational Simulation and Dynamic-Clamp Analysis Dick W. Wheeler, Paul H. M. Kullmann and John P. Horn JN 92:2659-2671, 2004. First published Jun 22, 2004; doi:10.1152/jn.00470.2004 You might find this additional information useful... Supplemental material for this article can be found at: http://jn.physiology.org/cgi/content/full/00470.2004/DC1 This article cites 40 articles, 20 of which you can access free at: http://jn.physiology.org/cgi/content/full/92/5/2659#BIBL This article has been cited by 2 other HighWire hosted articles: Physiological Classification of Sympathetic Neurons in the Rat Superior Cervical Ganglion C. Li and J. P. Horn J Neurophysiol, January 1, 2006; 95 (1): 187-195. [Abstract] [Full Text] [PDF] Post-stimulus potentiation of transmission in pelvic ganglia enhances sympathetic dilatation of guinea-pig uterine artery in vitro J. L. Morris, I. L. Gibbins and P. Jobling J. Physiol., July 1, 2005; 566 (1): 189-203. [Abstract] [Full Text] [PDF] Updated information and services including high-resolution figures, can be found at: http://jn.physiology.org/cgi/content/full/92/5/2659 Additional material and information about Journal of Neurophysiology can be found at: http://www.the-aps.org/publications/jn This information is current as of January 13, 2006. Estimating Use-Dependent Synaptic Gain in Autonomic Ganglia by Computational Simulation and Dynamic-Clamp Analysis Diek W. Wheeler, Paul H. M. Kullmann, and John P. Horn Department of Neurobiology and Center for the Neural Basis of Cognition, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261 Submitted 6 May 2004; accepted in final form 16 June 2004 Wheeler, Diek W., Paul H. M. Kullmann, and John P. Horn. Estimating use-dependent synaptic gain in autonomic ganglia by computational simulation and dynamic-clamp analysis. J Neurophysiol 92: 2659–2671, 2004. First published June 22, 2004; 10.1152/jn.00470.2004. Biological gain mechanisms regulate the sensitivity and dynamics of signaling pathways at the systemic, cellular, and molecular levels. In the sympathetic nervous system, gain in sensory-motor feedback loops is essential for homeostatic regulation of blood pressure and body temperature. This study shows how synaptic convergence and plasticity can interact to generate synaptic gain in autonomic ganglia and thereby enhance homeostatic control. Using a conductance-based computational model of an idealized sympathetic neuron, we simulated the postganglionic response to noisy patterns of presynaptic activity and found that a threefold amplification in postsynaptic spike output can arise in ganglia, depending on the number and strength of nicotinic synapses, the presynaptic firing rate, the extent of presynaptic facilitation, and the expression of muscarinic and peptidergic excitation. The simulations also showed that postsynaptic refractory periods serve to limit synaptic gain and alter postsynaptic spike timing. Synaptic gain was measured by stimulating dissociated bullfrog sympathetic neurons with 1–10 virtual synapses using a dynamic clamp. As in simulations, the threshold synaptic conductance for nicotinic excitation of firing was typically 10–15 nS, and synaptic gain increased with higher levels of nicotinic convergence. Unlike the model, gain in neurons sometimes declined during stimulation. This postsynaptic effect was partially blocked by 10 μM Cd^{2+}, which inhibits voltage-dependent calcium currents. These results support a general model in which the circuit variations observed in parasympathetic and sympathetic ganglia, as well as other neural relays, can enable functional subsets of neurons to behave either as 1:1 relays, variable amplifiers, or switches. INTRODUCTION The synaptic organization of sympathetic and parasympathetic ganglia enables spinal preganglionic neurons to control autonomic behavior by driving postganglionic neurons. The resulting transformation of motor activity depends on the intrinsic ganglionic circuitry, whose anatomy incorporates presynaptic divergence and postsynaptic convergence, while also allowing for different forms of short-term plasticity and for interplay between the ionotropic and metabotropic actions of acetylcholine and peptidergic co-transmitters (Akasu and Nishimura 1995; Gibbins et al. 2000; Kuba and Koketsu 1978; Simmons 1985; Smith 1994). Functional descriptions of autonomic ganglia often stress their role as simple relays in which divergence helps to provide peripheral targets with an adequate level of neurally mediated excitation or inhibition (Iversen et al. 2000; Jänig 1995; McLachlan 2003; Powley 2003). In contrast, the consequences of convergence, plasticity, and metabotropic modulation remain relatively unclear. To resolve this issue, we have sought to develop a general method for exploring ganglionic integration by combining theory and simulations with cell physiology. Such an approach may illuminate the variations in function that have arisen through phenotypic specialization in subsets of autonomic neurons and through the evolutionary adaptations observed in ganglia from different vertebrate species (Gibbins 1995; Gibbins et al. 2000; Purves et al. 1986). To develop a generic view of ganglionic organization, we began by considering paravertebral sympathetic ganglia 9 and 10 of the bullfrog. This model system has many features that facilitate experimental study (Smith 1994). These advantages include 1) low levels of synaptic convergence that can be readily assessed, 2) functional criteria that permit the physiological identification and independent analysis of B and C neurons, which selectively innervate cutaneous glands and blood vessels, 3) well-defined modulatory mechanisms of muscarinic and peptidergic co-transmission, and 4) established methods that permit the study of fully differentiated neurons from dissociated ganglia in adult animals. In B-type neurons, Karila and Horn (2000) have identified an $n + 1$ convergence pattern of weak and strong nicotinic synapses, where every postganglionic neuron receives one very strong, always suprathreshold, synapse and a variable number ($n$) of weak synapses. Arguing that a similar pattern occurs in the paravertebral ganglia of all vertebrates, they derived a mathematical theory of ganglionic integration based on the principle of stochastic coincidence detection. This approach led to the broad prediction that, paravertebral sympathetic ganglia, in general, behave as use-dependent synaptic amplifiers of preganglionic activity. In this scheme, gain is regulated by the number of converging nicotinic synapses and by their strength and plasticity. Although the coincidence detection theory of Karila and Horn (2000) provided a simple means for understanding how seemingly unrelated features of ganglionic organization could interact to produce amplification, it was limited by a lack of biophysical details describing ionic conductances and determinants of synaptic plasticity. Schobesberger et al. (1999, 2000) partially addressed this problem by adapting an earlier conductance-based model of the sympathetic B-type neuron (Yamada... et al. 1989) and examining the interaction between slow metabotropic excitation and subthreshold nicotinic excitatory postsynaptic potentials (EPSPs). The resulting analysis showed that muscarinic and peptidergic slow EPSPs could act to strengthen nicotinic synapses, but did not directly test the consequences for synaptic gain. This study examines ganglionic integration by combining simulations of ganglionic activity with dynamic-clamp experiments. The simulations reveal how different patterns of preganglionic activity, in conjunction with synaptic mechanisms, control synaptic gain. In addition to clarifying certain limitations of the more minimal coincidence detection theory (Karila and Horn 2000), the present simulations establish the range of synaptic transformations of activity that may be implemented by different populations of autonomic neurons. In whole cell recordings from dissociated sympathetic B neurons, the dynamic-clamp method has been used to drive postsynaptic firing with different patterns of virtual synaptic activity. By taking this approach, one has precise control over the number, strength, and occurrence of each synaptic event and can compare the response of different cells to identical stimulation, an experimental procedure that would not be possible with living synapses. The experimental results confirm key predictions of the simulations and identify a postsynaptic role for calcium in regulating synaptic gain. METHODS Theoretical basis for simulations The part of the autonomic system we are studying—sympathetic and parasympathetic ganglia—consists of a population of spinal, preganglionic presynaptic neurons connected via a simple network of synapses to a population of purely postsynaptic ganglionic neurons. As in previous work (Karila and Horn 2000), synaptic gain ($G$) has been defined as the multiplicative factor that relates the system’s spike output to its input, where each is represented by their average firing rates ($f_{\text{pre}}, f_{\text{post}}$). Gain in this system has a simple physical interpretation. When $G > 1$, the postsynaptic neurons, on average, are firing faster than the presynaptic neurons $$f_{\text{post}} = Gf_{\text{pre}} \quad (1)$$ The multiplicative definition of gain (Eq. 1) is standard in control theory and has application to many problems including autonomic control of blood pressure (Khoo 2000). We refer to ganglionic gain as synaptic gain because it arises through synapses. To simplify the task of simulating ganglionic integration, three central postulates were adopted from earlier work (Karila and Horn 2000). Each postulate represents an idealization of experimental observations. POSTULATE 1—VARIATIONS IN CONVERGENCE FOLLOW AN $N + 1$ RULE IN SYMPATHETIC GANGLIA. Under this rule, every postganglionic neuron receives one very strong, always suprathreshold, nicotinic synapse denoted as primary and a variable number ($n$) of weak nicotinic synapses denoted as secondary. In many instances, summation between pairs of secondary EPSPs is sufficient to trigger an action potential. Due to this synaptic arrangement, the spike output of postganglionic neurons reflects the activity of primary synapses combined with that driven by suprathreshold coincidences between subthreshold secondary EPSPs. The $n + 1$ rule describes nicotinic convergence in bullfrog sympathetic ganglia (Dodd and Horn 1983; Karila and Horn 2000) and also seems to operate in mammalian sympathetic ganglia (Hirst and McLachlan 1986; Jänig and McLachlan 1992; Skok and Ivanov 1983). POSTULATE 2—ENTIRE POPULATIONS OF AUTONOMIC NEURONS ARE UNIFORM IN THEIR SYNAPTIC ORGANIZATION. Functional subsets of ganglionic neurons can be differentiated by their levels of secondary synaptic convergence ($n$) and by their selective expression of neuromodulatory mechanisms. For example, bullfrog secretomotor B-type sympathetic neurons have between one and three secondary synapses with a mean of 1.2 (Karila and Horn 2000), and they selectively express an excitatory muscarinic synaptic mechanism (Smith and Weight 1986; Tosaka et al. 1968; Weight and Padjen 1973). As a consequence of uniformity within the B-cell population and the general lack of lateral interactions between postganglionic neurons, one can infer population behavior of this cell type from that of a single neuron, thus simplifying the design of simulations and experiments. Using this approach, it also becomes possible to compare integration in subpopulations of sympathetic and parasympathetic neurons that differ in their synaptic organization. Such variations appear to reflect the phenotypic specialization of ganglionic neuronal subtypes that control different target tissues and the evolutionary adaptation of homologous ganglia from different vertebrate species (Gibbins 1995; Gibbins et al. 2000; Purves et al. 1986). POSTULATE 3—PREGANGLIONIC NEURONAL ACTIVITY IS GENERATED BY A POISSON PROCESS IN WHICH THE TIME INTERVALS BETWEEN SYNAPTIC EVENTS ARE EXPONENTIALLY DISTRIBUTED. Postganglionic firing in anesthetized animals and awake humans is characteristically very irregular (Macefield et al. 2002; McAllen and Malpas 1997). In cases where the pattern of presynaptic activity has been recorded intracellularly from postganglionic neurons in vivo, the noisy intervals between synaptic events can be approximated by an exponential distribution (see Fig. 5A2 in McLachlan et al. 1998 and Fig. 4A in Häbler et al. 1999). By idealizing presynaptic activity as a Poisson process, one can mathematically model the occurrence of synaptic events. Computational modeling All simulations of synaptic activity followed the multi-step process outlined in Fig. 1A. In this procedure, templates of synaptic conductance were constructed by first determining the timing of primary and secondary nicotinic events. For each event, an appropriately sized postsynaptic conductance waveform was added to the template. Finally, the completed template was used to drive the conductance-based model sympathetic neuron by numerically integrating the underlying system of ordinary differential equations. Timing of synaptic events was determined by repeatedly drawing a random number (0–1) and comparing it with $P_{\text{syn}}$, the probability of an event occurring during a specified time window ($t_{\text{win}}$). If the random number was $\leq P_{\text{syn}}$, an event was added to the template. For a Poisson process governing the timing of $n$ identical synapses, $P_{\text{syn}}(t \leq t_{\text{win}}) = 1 - \exp(-nf_{\text{pre}}t_{\text{win}})$ (Colquhoun 1971). The value of $t_{\text{win}}$, which also signifies the temporal resolution of the saved data, was set to 0.25 ms for numerical simulations and to 0.05 ms for templates used in dynamic-clamp experiments. The equation for $P_{\text{syn}}$ permitted us to simulate the generation of synaptic events by nonintegral as well as integral numbers of synapses. For example, in Fig. 4A, synaptic gain was calculated for $n$’s of 1.2, 3, and 9 to model the average convergence levels associated with cellular populations of bullfrog B and C neurons and rat sympathetic neurons. The time course of the fast nicotinic synaptic conductance ($g_{\text{syn}}$) was derived by fitting a synaptic current, recorded from a B neuron under two-electrode voltage clamp, to the sum of two exponentials and scaling it to a peak amplitude of 1 (see Fig. 2A in Schoesberger et al. 2000). The time constants were 1 ms for the rise and 5 ms for the decay. Unless stated otherwise, the strengths of nicotinic synapses were adjusted by scaling the unitary template to $\bar{g}_{\text{syn}} = 100$ nS for primary synapses and 9.61 nS for secondary synapses. The latter represents 90% of threshold-$g_{\text{syn}}$, the nicotinic conductance required... to depolarize the resting model to threshold for generating an action potential (Schobesberger et al. 2000). Other conductances in the model sympathetic B neuron have already been described (Schobesberger et al. 2000). They include a fast inactivating sodium conductance ($g_{\text{Na}}$), a noninactivating delayed-rectifier potassium conductance ($g_K$), an M-type potassium conductance ($g_M$), a cyclic nucleotide-gated cation leak conductance ($g_{\text{CNG}}$), and a voltage-insensitive background leak conductance ($g_{\text{leak}}$). These are sufficient to produce a realistic resting potential, steady-state current-voltage relation, and spike threshold. For all simulations, $\bar{g}_{\text{Na}} = 800 \, \text{nS}$, $\bar{g}_K = 2,000 \, \text{nS}$, $g_{\text{leak}} = 3 \, \text{nS}$, $E_{\text{Na}} = 60 \, \text{mV}$, $E_K = -90 \, \text{mV}$, $E_{\text{leak}} = -40 \, \text{mV}$, $E_M = -90 \, \text{mV}$, $E_{\text{CNG}} = 0 \, \text{mV}$, $E_{\text{syn}} = 0 \, \text{mV}$, and the membrane capacitance was 100 pF. For simulations of muscarinic and peptidergic excitation, $\bar{g}_M$ and $\bar{g}_{\text{CNG}}$ were adjusted from resting values of 40 and 0 to 30 and 4 nS. See the supplemental material$^1$ for a full listing of the equations describing each ionic conductance. In one set of simulations, presynaptic facilitation of nicotinic synapses was modeled as an additive process followed by exponential decay. In the presence of facilitation, each presynaptic event transiently increased the amplitudes ($A$) of all successive events by a factor having the following form: $A = 1 + 2 \exp(-t/\tau_{\text{decay}})$. The magnitude of facilitation (2) and the range of $\tau_{\text{decay}}$ (12–50 ms) were chosen to reproduce experimental data from bullfrog B neurons (Shen and Horn 1995). Software implementation of simulations Computational simulations employed programs written in MATLAB R11–13 (MathWorks, Natick, MA) using ode15s, a stiff variable-step differential-equation solver, and a personal computer. Our software, collectively called NEUROSIM, is freely available and can be obtained by contacting the authors. The voltage data and activation/inactivation parameter values from each simulation were saved every 0.25 ms. Simulating 20–4,000 s of neural activity for each measurement of synaptic gain required between 2 and 100 h of real time. Illustrations were drafted with Igor Pro 4.0 (PC Edition, WaveMetrics, Lake Oswego, OR). Cell culture Paravertebral sympathetic ganglia 8–10 were dissected from pithed adult bullfrogs (Rana catesbeiana, 4–6 in) using a protocol approved by the University of Pittsburgh’s Institutional Animal Care and Use Committee. The ganglia were desheathed, minced, and dissociated by incubation in Ringer containing 300 $\mu g/ml$ Liberase Blendzyme 3 (Roche Diagnostics, Indianapolis, IN) for 30–60 min at 32°C, followed by mechanical trituration with graded fire-polished Pasteur pipettes. Ringer contained (in mM) 115 NaCl, 2 KCl, 1.8 CaCl$_2$, and 4 NaHEPES, adjusted to pH 7.3. After centrifugation and resuspension in culture medium, the dissociated cells were plated onto glass coverslips coated with poly-d-lysine (BD Biosciences, Bedford, MA). Cell cultures were maintained for $\leq 2$ wk in diluted L-15 medium (65%) supplemented with 5 mM dextrose, 2 mM l-glutamine, 1 mM CaCl$_2$, 5% fetal bovine serum, 100 U/ml penicillin, and 100 $\mu g/ml$ streptomycin. Perforated-patch clamp recording and dynamic clamp Whole cell perforated-patch recordings were made at room temperature from neurons visualized on an inverted microscope. Fire-polished patch pipettes were filled with internal solution containing (in mM) 110 K$^+$-gluconate, 10 NaCl, and 5 NaHEPES, adjusted to pH 7.1 with HCl. Stock solution of amphotericin-B (Sigma; 25 mg/ml in dimethyl sulfoxide) was added to the internal solution to a final concentration of 250 $\mu g/ml$. With pipette resistances of 2–5 M$\Omega$, this amphotericin-B concentration typically resulted in final access resistances of 5–10 M$\Omega$, after 30–60 min of recording. Access resistance was monitored throughout the experiment and compensated using the bridge circuitry of the recording amplifier. The external recording solution consisted of Ringer as described above. The dynamic current-clamp method was used to record membrane potential and create virtual nicotinic synapses. For this purpose, we used G-clamp (Kullmann et al. 2004), which includes an AxoClamp 2B amplifier (Axon Instruments, Union City, CA), G-clamp software written in the LabVIEW-RT programming environment (National Instruments, Austin, TX), a standard Windows-based computer, an embedded controller running under a real-time operating system (National Instruments), and a multifunction I/O board for data acquisition and signal generation (PXI-6052E, National Instruments). This system has the ability to read data files containing conductance templates of synaptic activity identical to those used in the numerical simulations (Fig. 1A). These templates were used to stimulate neurons while measuring and storing their membrane potential responses and the currents injected by the clamp. The dynamic clamp’s real-time loop speed and data sampling rate were both 20 kHz. Recordings were made from the largest cells because they are generally B neurons (Dodd and Horn 1983; Jan and Jan 1982). Once patch perforation by amphotericin-B had reduced access resistance to 15 M$\Omega$ or less, we began measuring threshold-$g_{\text{syn}}$. When a stable estimate was obtained, this value was used to scale a synaptic conductance template for measuring synaptic gain. Because execution of a synaptic conductance template could transiently alter postsynaptic excitability for several minutes, we continually measured threshold-$g_{\text{syn}}$ between trials, waiting until it stabilized again before execution of the next synaptic conductance template. Steady-state $I-V$ relations were measured repeatedly throughout each experiment to monitor leak resistance. Analysis of synaptic gain and the associated errors Synaptic gain ($G$) was calculated by dividing the number of postsynaptic action potentials in a given trial by the average number of presynaptic events per synapse. To estimate the errors associated with counts of postsynaptic action potentials and with estimates of $G$, we assumed that postsynaptic activity was generated by an inhomogeneous Poisson process (Berry and Meister 1998). The inhomogeneity arose from filtering of short interspike intervals at high stimulus frequencies (Fig. 3A). For all Poisson processes, the mean number of events ($\mu$) equals the total number of events generated, and the CV = $1/\sqrt{\mu}$ (Colquhoun 1971). We therefore adjusted the duration of simulations to produce $\sim 400$ postsynaptic action potentials to obtain a CV or “counting error” of $\sim 5\%$. The error in $G$, which depends on the count of presynaptic and postsynaptic spikes, was estimated as $$\text{CV}_G = \sqrt{\text{CV}_{\text{post}}^2 + \text{CV}_{\text{pre}}^2}$$ (Colquhoun 1971). The SD of $G$ can be calculated as $\sigma_G = \text{CV}_G G$. Since a single simulation was used to calculate each value of $G$, the SE equals $\sigma_G$. All errors in $G$ are plotted or reported as $\pm \text{SE}$. In dynamic-clamp experiments, shorter templates of 200 events per synapse were employed so that the data could be collected from a stable recording in <1 h. Grouped data from these experiments are reported as the means $\pm \text{SE}$. In statistical comparisons, the criterion for significance was $P < 0.05$. RESULTS Frequency dependence of synaptic gain To examine how the presynaptic firing frequency controls gain, templates of synaptic activity were constructed and used to stimulate a model sympathetic neuron (Fig. 1A). Each synaptic template incorporated information about the number of converging nicotinic synapses, their strength, and their temporal pattern of activity. We began by considering a case with one primary synapse and nine secondary synapses, an $n + 1$ arrangement similar to that observed in the rat superior cervical ganglion (Purves et al. 1986). The strength of each secondary nicotinic synapse was set to 9.61 nS, which is 90% of threshold-$g_{\text{syn}}$. The temporal pattern of synaptic events was created by assuming presynaptic activity arises from a Poisson process. By combining the timing information with appropriately scaled synaptic conductance waveforms, a complete template was constructed that described the time course of total synaptic conductance over an entire simulation. The membrane potential response to synaptic stimulation was computed by numerically integrating the Hodgkin-Huxley system of equations (see supplemental material for a full listing) that describe the model sympathetic neuron (Fig. 1, B and C). The efficacy of secondary synapses and their ability to generate gain varied with the presynaptic frequency. In response to stimulation at 0.3 Hz, most postsynaptic action potentials were driven in one-to-one fashion by primary EPSPs (Fig. 1B). In contrast, setting the presynaptic frequency to 3 Hz changed the outcome so that most postsynaptic firing was driven by secondary synapses (Fig. 1C). Inspection of the records revealed that virtually all of the spikes triggered by secondary EPSPs resulted from binary coincidences. At 3 Hz, there were occasional triple and quadruple coincidences, but they had little impact on the overall rate of postsynaptic firing (Fig. 1C). These effects of activity on synaptic gain can be understood more easily by considering a simpler case with only three synapses (Fig. 2A). In this schematic illustration, all three presynaptic inputs fire an average of four times, and the postsynaptic cell responds by firing seven times, thus resulting in a gain of 1.75. Four of the action potentials (dark shading) originate from the primary synapse, while the remaining three (light shading) arise from temporal summation between secondary EPSPs. If the presynaptic frequency were increased in this example, the probability of secondary synaptic coincidences would also increase, eventually accounting for the majority of postsynaptic action potentials and a further increase in gain. On the other hand, reducing the presynaptic frequency would lessen the influence of secondary synapses, thereby causing gain to drop toward one. To measure synaptic gain systematically as a function of mean presynaptic frequency, we drove the $9 + 1$ model over a range of frequencies (0.1–20 Hz). The durations of these simulations were each adjusted to produce $\geq 400$ postsynaptic spikes to reduce counting errors arising from the stochastic nature of the synaptic activity. The resulting relation shows that gain was a biphasic function of the mean presynaptic frequency (Fig. 2B), with a peak in the range of 2–5 Hz. The overall shape of the synaptic-gain relation constructed from the simulations was similar to that predicted by the more sparsely framed coincidence detection theory (Karila and Horn 2000), especially when the presynaptic frequency was <1 Hz (Fig. 2B). The main differences were observed at higher frequencies where the theory overestimated the peak gain and predicted an overly steep postpeak decline. We suspected these disparities were due to the different ways in which the two approaches introduce refractory periods that limit the rate of postsynaptic firing. For this purpose, the coincidence detection theory of Karila and Horn (2000) incorporates the arbitrary assumption... that secondary EPSPs can only drive postsynaptic firing at a maximum rate of 10 Hz. In contrast, refractory properties of the model neuron emerge from its conductances and are not limited to spikes triggered by secondary EPSPs. Lowering the refractory cut-off frequency in the Karila and Horn (2000) theory reduced the peak of the gain relation but did not improve the overall fit to the simulation results (data not shown). To better understand the behavior of the conductance-based model, we therefore examined how primary and secondary EPSPs affect one another. Interactions between primary and secondary synapses have consequences for gain In addition to limiting the excitation produced by the summation of secondary EPSPs, one would expect refractory periods to introduce interactions between primary and secondary synapses. If, for example, a pair of secondary EPSPs coincided during or shortly after an action potential driven by a primary EPSP, the first action potential would prevent the second. To measure the extent of such effects, we first constructed separate gain relations to describe a single primary synapse and a group of nine secondary synapses (Fig. 2C). The gain of the solitary primary synapse was unitary at low stimulation rates, but decreased by ~17% when the presynaptic frequency was >10 Hz. This high-frequency roll-off suggests that refractory interactions between primary synaptic events can limit postsynaptic firing. In contrast, the gain relation for nine secondary synapses was biphasic, with a baseline that approached zero at lower frequencies, a peak of ~2.1 at 3–10 Hz, and a roll-off at higher rates. The high-frequency roll-off in this relation suggests that refractory mechanisms can also limit the efficacy of secondary synapses in the absence of a primary synapse. To uncover interactions between primary and secondary synapses, we added the gain responses for isolated primary and secondary synapses and compared their sum with the gain relation describing combined $n + 1$ convergence (Fig. 2D). Although the two curves were similar when $f_{\text{pre}} < 1$ Hz, the summed gain response was greater than the combined gain response at higher frequencies. This indicates that interactions between primary and secondary synapses are minimal at low $f_{\text{pre}}$, but limit gain at higher stimulation rates. Refractory periods alter postsynaptic spike timing If postsynaptic refractoriness is responsible for the roll-off in synaptic gain at higher presynaptic frequencies (Fig. 2, B–D), frequency-dependent effects should be evident in the patterns of postsynaptic action potentials. This prediction was tested by constructing cumulative probability distributions of postsynaptic interspike intervals (Fig. 3A). When the presynaptic frequency was 0.6 Hz and gain was low, the distribution of postsynaptic interspike intervals was exponential, just like the distribution of presynaptic events. However, at higher presynaptic frequencies, the distributions of postsynaptic intervals fell short of the expected exponential relations. The disparity was clearest when the presynaptic firing rate was 20 Hz, which produced a large shortfall in postsynaptic interspike intervals <30 ms, with none <8 ms in duration. The lack of intervals <8 ms is also evident at the other presynaptic frequencies when the time scale of the distributions is expanded (Fig. 3A, insets). To test whether refractory effects could account for the postsynaptic filtering effect, we measured two types of interactions. For pairs of primary EPSPs, the refractory period for production of an overshooting action potential was 8 ms (Fig. 3B). When the trailing stimulus was created by a pair of... simultaneous secondary EPSPs, which was much weaker than a primary EPSP, the refractory period was 26 ms (Fig. 3C). The brief refractory period between primary EPSPs explains the total absence of postsynaptic interspike intervals <8 ms (Fig. 3A), and it accounts for the roll-off in the gain relation of the isolated primary synapse (Fig. 2C). When the primary synapse was driven at 20 Hz, 13% of the intervals between EPSPs were <8 ms, which agrees well with the observed 17% reduction in gain. In contrast, the longer refractory period between primary and secondary EPSPs explains the partial filtering of postsynaptic interspike intervals <30 ms (Fig. 3A) and the roll-off of the gain relation for secondary synapses (Fig. 2C). In the absence of refractory mechanisms, one would expect gain to increase as a monotonic function of the mean presynaptic frequency, saturating when gain is $n + 1$ (Karila and Horn 2000). We tested this prediction by counting the number of EPSPs separated by intervals that were <26 ms in the secondaries-only 20-Hz data. If all of these EPSPs had successfully triggered spikes, the gain would have been 9 rather than the observed 1.2. This agrees with the theoretical maximum for gain in the secondaries-only configuration, thereby supporting the conclusion that the refractory occlusion by secondary EPSPs causes most of the roll-off in the synaptic gain relation. In comparison, the effects of primary-primary synaptic occlusion (Fig. 2C) and primary-secondary synaptic occlusion (Fig. 2D) were relatively small. Synaptic gain regulation by convergence and synaptic strength Lowering secondary synaptic convergence to mimic the levels seen in sympathetic ganglia of amphibians and mice ($n = 1, 2, 3$) (Karila and Horn 2000; Purves et al. 1986) caused a proportional reduction in maximum gain (Fig. 4A). Reducing secondary synaptic strength from 98 to 50% threshold-$g_{syn}$ while holding convergence constant at $n = 9$, showed that the peak of the gain relation also varied as a function of secondary synaptic strength (Fig. 4B). Although the effects of convergence and synaptic strength on gain were generally consistent with the predictions of coincidence detection theory (Karila and Horn 2000), it was surprising to find $G > 1$ when secondary synaptic strength was set to 50% threshold-$g_{syn}$ (Fig. 4B). This result contradicts the theory, which would require perfect coincidences between pairs of secondary EPSPs of this strength to generate action potentials. Since perfect coincidences are rare, the theory predicts there should be little or no gain above the baseline of 1 (Fig. 4B, dashed line). We examined this further by using even weaker synapses. For $f_{pre} = 5$ Hz and $n = 9$, $G = 1.31 \pm 0.06$ when secondary synaptic strength was 49% threshold-$g_{syn}$ and $G = 1.17 \pm 0.05$ when secondary synaptic strength was 40% threshold-$g_{syn}$. The ability of sub-50% secondary EPSPs to trigger action potentials suggested it was very unlikely that binary coincidences were causing these effects. By examining the simulation records for triple, quadruple, and quintuple coincidences, we found that such higher order events accounted for 97–99% of the spikes driven by secondary synapses at 50, 49, and 40% threshold-$g_{syn}$. Metabotropic modulation of synaptic gain Metabotropic muscarinic excitation of sympathetic neurons operates through a branched signaling pathway that controls two ionic conductances. The better known is the M-type potassium conductance $g_M$, whose suppression durPresynaptic facilitation increases synaptic gain Mechanisms of presynaptic facilitation are widespread in autonomic ganglia. When facilitation was incorporated into synaptic templates, it increased synaptic gain at frequencies of 1–10 Hz (Fig. 4D). The exponential rate of decay of the presynaptic facilitation factor directly affected the magnitude of the gain response. Experimental tests of computational predictions The simulation results in the first part of this paper show how synaptic gain can arise as a use-dependent consequence of nicotinic synaptic convergence and be modulated by synaptic strength, metabotropic excitation, and presynaptic facilitation (Fig. 4). Evaluating the physiological significance of these findings requires that appropriate experimental tests be devised. Ideally, one would want to activate independently each of the synaptic inputs to an autonomic neuron with trains of asynchronous stimuli. Due to the impracticality of such an experiment, we adopted an alternative strategy of using the dynamic-clamp method to implement virtual nicotinic synapses on dissociated bullfrog B neurons. With this approach, one can compare how real B neurons and the model B neuron respond to identical patterns of synaptic stimulation. We began by measuring synaptic strength and the gain produced by $ n + 1 $ convergence. Dynamic-clamp estimates of synaptic strength and gain An automated binary search paradigm was used to measure the magnitude of virtual nicotinic synaptic conductance required to depolarize B neurons to threshold (Fig. 5). The procedure begins by testing boundary values that straddle... threshold and homes in on a stable result within 8–12 trials (Kullmann et al. 2004). In a group of 12 neurons that were subsequently used for measurements of gain, we observed threshold-$g_{syn} = 14.7 \pm 1.7$ nS (range, 7.0–23.7 nS), $V_{rest} = -54.7 \pm 3.0$ mV (range, −40 to −70 mV), and $R_{leak} = 739 \pm 111$ MΩ (range, 153 MΩ–1.48 GΩ). All of these cells had overshooting actions potentials. In comparison, the computational model assumed $R_{leak} = 333$ MΩ and predicted that threshold-$g_{syn} = 10.68$ nS and $V_{rest} = -62.7$ mV. To measure the influence of secondary nicotinic EPSPs on synaptic gain, we constructed 5-Hz templates that were 40 s in duration and contained one primary synapse and zero, three, or nine secondary synapses. Five hertz was chosen as the presynaptic frequency because it generated maximal responses in gain simulations (Fig. 4A). After recording a value of threshold-$g_{syn}$ that remained stable with repeated tests, the strength of secondary synapses in the stimulus templates was scaled to 90% threshold-$g_{syn}$, and the strength of the primary synapse was set at 10 times that of the secondary synapses. In cases where conditions drifted slowly over time, synaptic conductances were adjusted to maintain constant strength. Only cells in which at least two different levels of secondary convergence could be compared were included in the analysis. Figure 6 shows an example comparing convergence levels of three and nine secondary synapses. The top traces show dynamic-clamp recordings from one cell, and the bottom traces show simulated responses to the same synaptic templates using the computational model. From this example, three key points emerge. First, secondary synapses contributed to the spike output of the cell at both levels of convergence. As consequence of this effect, the gain produced in both cases was >1. Second, the contribution of secondary synapses was greater when $n = 9$ than when $n = 3$. This effect corresponds to greater convergence producing higher gain. Third, the pattern of spike output recorded from the B neuron was very well predicted by the computational simulations. In this example, only one spike is missing from the simulated output when $n = 3$ (Fig. 6A), and one spike is missing from the cellular output when $n = 9$ (Fig. 6B). After calculating the gain produced by different levels of convergence, we made comparisons within cells and between groups of cells. Figure 7A shows an experiment in which it was possible to measure gain at three levels of $n$ during 1 h of stable recording. Although stimulation with virtual synaptic activity caused transient changes in the resting potential and threshold-$g_{syn}$, these effects could not account for the fact that higher levels of convergence consistently and repeatedly produced higher levels of gain. In paired comparisons within individual cells, the gain produced by $n = 3$ ($1.10 \pm 0.03$) was significantly greater than with $n = 0$ ($G = 0.93 \pm 0.01$) in eight of eight cases ($P < 0.01$, nonparametric 2-tailed sign test). Similarly, the gain produced by $n = 9$ ($1.42 \pm 0.08$) was significantly greater than with $n = 3$ in 11 of 11 cells ($P < 0.001$). FIG. 6. Virtual synaptic activity elicits similar firing patterns in sympathetic neurons and the computational model. Templates incorporating 3 secondary synapses (A) and 9 secondary synapses (B) were used to stimulate a neuron and the computational model. In each case, $f_{pre} = 5$ Hz, secondary synaptic strength = 90% threshold-$g_{syn}$, and primary synaptic strength = 900% threshold-$g_{syn}$. Note that with only 1 exception () for each template, the cell and model generated identical patterns of firing. The presence of action potentials driven by secondary synapses (●) in both cases suggests gain >1. The greater contribution of secondary synapses to firing when $n = 9$ (B) than when $n = 3$ (A) indicates that higher convergence enhances gain. The magnitude of gain produced by one primary synapse alone ($n = 0$) and by three secondary synapses was very similar to that predicted by our simulations. In contrast, the gain observed with $n = 9$ was lower than predicted. To identify the possible origin of this disparity, we divided the cells into three groups based on their resting potentials ($<-60$, $-60$ to $-50$, and $>-50$ mV). Plotting the data in this way (Fig. 7B) revealed a voltage-dependence in gain that was most evident when $n = 9$. In the most hyperpolarized group of cells with $n = 9$, the disparity between cellular responses and simulations using the same templates was reduced but still evident (Fig. 7C). Activity and calcium dependence of synaptic gain* Finding that gain could vary with resting potential (Fig. 7B) suggested that it might also change with activity. To test for such effects, the responses to 5-Hz stimulation were analyzed as four 10-s segments. This revealed that synaptic gain decreased over the course of individual trials when $n = 9$, but not when $n = 3$, and not in simulations. The decline was reversible when cells were given a few minutes rest between trials. When dynamic-clamp records and simulations from the last 10 s of stimulus trials were compared, secondary synapses failed to drive many of the spikes that were predicted by the simulations. This outcome differed from the close correspondence between recorded and simulated firing patterns at the beginning of the stimulus trial (Fig. 6B). To document the use-dependent decline in gain more carefully and to examine its possible origin, we designed an experiment using a 40-s synaptic template ($f_{\text{pre}} = 5$ Hz, $n = 9$) consisting of 10 identical 4-s repeats. In a group of seven neurons ($V_{\text{rest}} = -63.1 \pm 3.1$ mV, threshold-$g_{\text{syn}} = 10.6 \pm 2.1$ nS, $R_{\text{leak}} = 732 \pm 84$ M$\Omega$), the repeating template yielded an average gain of $1.64 \pm 0.13$ over the entire 40-s period. This was similar to the gain observed with the nonrepeating template (1.42), but lower than the gain (2.40 $\pm$ 0.19) produced in a simulation with the same template. The gain generated during each repeat declined gradually from a high of $2.00 \pm 0.21$ in the first 4 s to a low of $1.37 \pm 0.09$ in the last 4 s (7 cells). The time course of the decline in gain was similar when viewed for one cell (Fig. 8A) and for the group (Fig. 8B). By contrast, simulated voltage responses to the first and last segments of the repeating template were identical, with a gain of $2.40 \pm 0.26$. We hypothesized that calcium entry was responsible for the observed reduction in synaptic gain. Indeed, the absence of calcium and calcium-activated potassium conductances in the computational model might account for the steady gain response seen in simulations with the 5-Hz repeating template. To test whether calcium entry mediates a reduction in synaptic gain, voltage-dependent calcium channels were blocked in the same group of seven cells using 10 $\mu$M Cd$^{2+}$ (Jones and Marks 1989). Cadmium did not affect $V_{\text{rest}} (-63.0 \pm 3.3$ mV; $P = 0.86$, paired $t$-test) or $R_{\text{leak}} (761 \pm 220$ M$\Omega$; $P = 0.72$, paired $t$-test), but caused significant increases in threshold-$g_{\text{syn}} (13.3 \pm 2.6$ nS; $P = 0.01$, paired $t$-test) and in the average gain over the entire 40-s template ($1.94 \pm 0.17$; $P < 0.05$, paired $t$-test; Fig. 8C). The overall changes in gain were also reflected in the 4-s repeats. Cadmium increased gain in the first 4-s FIG. 8. Activity dependence and cadmium sensitivity of synaptic gain. A group of 7 cells was stimulated with a template containing 10 4-s repeats of identical virtual synaptic activity ($f_{pre} = 5$ Hz; $n = 9$). Each cell was stimulated several times with the template in normal Ringer and in Ringer containing 10 $\mu$M CdCl$_2$. $A$: averages of 3 trials from 1 cell. In normal Ringer, synaptic gain declined throughout the time course of stimulation. Adding cadmium to block voltage-dependent Ca$^{2+}$ currents increased the initial value of gain and seems to attenuate the use-dependent decline. $B$: grouped data from 7 cells showing the same trends as in $A$. $C$: averaging gain data across the entire duration of the synaptic template shows that cadmium significantly increased the overall synaptic gain (7 neurons, $P < 0.05$, paired $t$-test). period to $2.32 \pm 0.19$ and in the last 4-s period to $1.76 \pm 0.16$. These observations suggest that increased intracellular [Ca$^{2+}$] may be partially responsible for the reduction in synaptic gain seen during prolonged synaptic stimulation. DISCUSSION Simulations define the general form of ganglionic gain in sympathetic ganglia By simulating ganglionic integration with a conductance-based model, we have shown how the stereotypic $n + 1$ convergence pattern of fast nicotinic synapses found in paravertebral sympathetic ganglia can serve to create a low-pass filter with unitary baseline gain and a high-frequency boost (Figs. 1, 2, and 9A). Plastic changes in several factors including convergence (Fig. 4A), secondary synaptic strength (Fig. 4B), metabotropic excitation (Fig. 4C), and presynaptic facilitation (Fig. 4D) are all capable of enhancing the filter’s high-frequency boost. These findings show how the diversity in synaptic organization reported for subclasses of sympathetic neurons (Jänig 1995; Smith 1994) may provide a basis for specialized regulation of ganglionic gain. For example, vasmotor, secretomotor, and pilomotor sympathetic neurons may each have unique synaptic-gain functions that are optimized to drive their specific targets. Quantitative limitations of coincidence-detection theory The architecture of the conductance-based simulations followed from a much simpler coincidence-detection theory (Karila and Horn 2000) that rests on two basic ideas—preganglionic activity is noisy and pairs of secondary EPSPs can sum to reach threshold. Perhaps it is not surprising then that the theory successfully predicted the general form of gain relations produced by simulating $n + 1$ convergence and by simulating modulation through different types of synaptic plasticity (Figs. 2, $B$ and $C$, and 4). Nonetheless, the theory consistently erred by predicting an overly sharp peak in the gain relation and an overly steep roll-off at high presynaptic frequencies. These effects arise from the theory’s assumption that secondary synapses cannot drive firing at rates $>10$ Hz. In contrast to this hard cut-off, more complex refractory interactions emerge in the conductance-based model to control the peak value of gain and its high-frequency roll-off (Fig. 3). The net effect of these refractory interactions is to produce a more gradual filtering of secondary synaptic events as a function of the presynaptic firing rate. In addition, the simulations with weak secondary synapses have shown that higher-order coincidences can influence firing (Fig. 4B). Using refractory filters having arbitrary sigmoidal shapes rather than step functions, we were able to fit closely the gain curves generated by simulations of the conductance-based model (data not shown). However, we were unable to modify the theory to incorporate these mechanisms in a form that could be derived on mechanistic grounds and that ![FIG. 9. Three generic forms of integration that arise from the innervation patterns observed in autonomic ganglia. $A$: $n + 1$ pattern of synaptic convergence seen in paravertebral sympathetic ganglia produces a variable synaptic amplifier whose gain function can be modulated by changes in the number and strength of secondary synapses, excitatory metabotropic synapses, and presynaptic facilitation. $B$: pattern of single innervation by 1 primary synapse enables some parasympathetic ganglia to behave as 1-to-1 relays. $C$: $n$ pattern of synaptic convergence produces a synaptic gain function that behaves as a high-pass filter or switch. The ganglion only transmits preganglionic activity that is above a critical frequency. This pattern is found in some parasympathetic ganglia and in some prevertebral sympathetic ganglia. Given the nature of secondary synapses, the efficacy of this switch would also be subject to modulation.] would readily generalize to different conditions. One might expect such limitations to emerge, given that the underlying ordinary differential equations, based on the Hodgkin-Huxley formalism, are rarely susceptible to simple analytical solutions. In other words, coincidence-detection theory concisely captures the basic form, but not the exact form, of simulation results from the conductance-based model. Insights from dynamic-clamp experiments Even though the conductance-based model was designed to include many biophysical mechanisms, it omitted some details to achieve computational efficiency and to minimize the inclusion of arbitrary assumptions. For this reason, it was important to compare the responses of living neurons to the computational results. By taking such an approach, one can critically evaluate the model and focus attention on mechanisms that need further scrutiny. The dynamic-clamp data show that the model correctly predicted the magnitude threshold-$g_{syn}$ for nicotinic excitation and the effect of secondary synaptic convergence on synaptic gain during 5-Hz stimulation. In contrast, the model did not predict the activity dependence of synaptic gain, as revealed at $n = 9$ by the apparent correlation with resting potential (Fig. 7B) and by the decline in gain during the course of single stimulus trials (Fig. 8, A and B). Treating cells with 10 $\mu$M Cd$^{2+}$ raised the level of gain during the first 4 s of stimulation ($f_{pre} = 5$ Hz, $n = 9$) from $2.00 \pm 0.21$ to $2.32 \pm 0.19$, which is close to the level of $2.40 \pm 0.26$ predicted by the model. Cadmium also appeared to decrease, but not eliminate, the decline of gain (Fig. 8A). Although these results indicate that calcium-dependent processes play a role in the regulation of gain, they cannot fully account for the use-dependent decline observed during 5-Hz stimulation and should therefore be interpreted with caution. Within our records, we also looked for, but were unable to detect obvious changes in spike threshold and morphology (upstroke, peak, repolarization) that could account for the decrease in gain during stimulation. Solving this problem will require further experimental analysis. Function follows form: other autonomic ganglia and brain relays Outside of the paravertebral sympathetic system, the different patterns of convergence seen in other ganglia can now be interpreted, based on our results. In the simplest pattern, postganglionic neurons are innervated by one primary nicotinic synapse. This arrangement occurs in the rat parasympathetic submandibular ganglion (Lichtman 1980), which controls salivation, and in the frog parasympathetic cardiac ganglion (Dennis et al. 1971), which modulates contractions of the heart. In our model, neurons with this innervation pattern behave as 1:1 relays that roll off at higher presynaptic frequencies (Figs. 2C and 9B). In another motif, characterized by the absence of a primary synapse and the presence of multiple secondary synapses, autonomic neurons behave as a band-pass filter (Figs. 2C and 9C). This organization enables the parasympathetic pelvic ganglion in the cat to perform as an on-off switch during the micturition reflex (DeGroat and Saum 1976), which gates filling and emptying of the bladder. A similar convergence pattern also exists in a subpopulation of neurons in the guinea-pig celiac ganglion (Gibbins et al. 2003; McLachlan and Meckler 1989). These prevertebral sympathetic neurons, which receive both preganglionic and intestinofugal inputs (Gibbins et al. 2003), exert a modulatory gating influence on gastrointestinal motility, secretions, and blood flow (Furness and Costa 1987). The levels of convergence seen in autonomic ganglia are sparse compared with the massive convergence found on cortical pyramidal cells, cerebellar Purkinje cells, and spinal motoneurons. However, sparse convergence operates in many brain nuclei that are traditionally viewed, like autonomic ganglia, as synaptic relays. Perhaps the best example is the lateral geniculate nucleus (LGN). Simultaneous recordings between retinal ganglion cells and the LGN have revealed that LGN neurons receive excitatory input from fewer than six ganglion cells (Mastronarde 1992), and these connections are characterized as strong or weak (Usrey et al. 1999). Our approach may therefore have relevance to other circuits with sparse convergence. Importance of gain in autonomic negative-feedback loops Negative-feedback loops are critical elements of physiological control processes, including those mediated by the neuroendocrine and autonomic systems (Khoo 2000). The functional gain that emerges as an inherent component of a negative-feedback loop leads to faster and more accurate homeostatic control (Scher 1989; Scher et al. 1991). An important example of this mechanism arises in the baroreceptor reflex, an important focal point for neural control of the circulation (Scher et al. 1991). The reflex allows blood pressure to be maintained within a relatively narrow range despite changes in body posture and cardiac output. In the process, afferent feedback from sensory receptors regulates the output of motor activity in the sympathetic and parasympathetic systems, ultimately leading to changes in cardiac function and vascular tone. Despite extensive study of the baroreceptor reflex, the neural origin of its gain remains unclear. Much of the focus has centered on synapses in the nucleus tractus solitarius and the rostral ventral lateral medulla (Dampney 1994; Sun 1995), but relatively little attention has been paid to the potential contributions of synapses in the spinal cord, in sympathetic ganglia, and at neuroeffectors to the overall gain of the reflex. We propose that synaptic gain in the baroreceptor reflex may be distributed throughout much of the circuit, including sympathetic ganglia, rather than being confined to one or two anatomical loci. In vivo evidence of synaptic gain in paravertebral sympathetic ganglia Our results indicate that synaptic gain rises above one when secondary synapses help to drive postganglionic firing. Evidence for such behavior under physiological conditions comes from in vivo intracellular recordings in the mouse, guinea pig, hamster, rabbit, rat, and bullfrog (Häbler et al. 1999; Ivanoff and Smith 1995; Ivanov and Purves 1989; McLachlan et al. 1997, 1998; Skok and Ivanov 1983). The data from these studies of paravertebral sympathetic neurons provide evidence that primary and secondary nicotinic synapses generally form an $n + 1$ pattern of convergence. In several instances, the morphology of the spike afterhyperpolarization has been used to identify which class of synapses triggers individual action potentials. Using this approach, it is clear that secondary synapses play a role in driving postganglionic activity. For examples, see Fig. 1B in Skok and Ivanov (1983), Fig. 1A in McLachlan et al. (1998), Fig. 4 in Ivanoff and Smith (1995), and Fig. 3 in McLachlan (2003). Additional evidence for synaptic gain comes from the observation that average rates of subthreshold and suprathreshold synaptic events are very similar in a range of mammals (see Table 2 in Ivanov and Purves 1989). When convergence is taken into account, these data indicate that individual postganglionic neurons fire at higher frequencies than preganglionic neurons. When viewed in the context of our model (Fig. 2A), this means that amplification is a common integrative feature of sympathetic ganglia. ACKNOWLEDGMENTS We thank Drs. S. Iyengar and A. Saul for helpful discussions of statistics. Present address of D. W. Wheeler: Max-Planck Institute for Brain Research, Department of Neurophysiology, Deutschordensstrasse 46, D-60528 Frankfurt/Main, Germany (E-mail: email@example.com). GRANTS This work was supported by National Institute of Neurological Disorders and Stroke Grant NS-21065 to J. P. Horn and by a postdoctoral fellowship from the American Heart Association, Pennsylvania/Delaware Affiliate, to D. W. Wheeler. REFERENCES Adams PR, Brown DA, and Constanti A. M-currents and other potassium currents in bullfrog sympathetic neurones. J Physiol 330: 537–572, 1982. Akasu T and Nishimura T. Synaptic transmission and function of parasympathetic ganglia. Prog Neurobiol 45: 459–522, 1995. Berry MJ and Meister M. Refractoriness and neural precision. J Neurosci 18: 2200–2211, 1998. Colquhoun D. Lectures on Biostatistics—An Introduction to Statistics with Applications in Biology and Medicine. Oxford, UK: Clarendon Press, 1971. Dampney RA. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 74: 323–364, 1994. DeGroat WC and Saum WR. Synaptic transmission in parasympathetic ganglia in the urinary bladder of the cat. J Physiol 256: 137–158, 1976. Dennis MJ, Harris AJ, and Kuffler SW. Synaptic transmission and its duplication by focally applied acetylcholine in parasympathetic neurons in the heart of the frog. Proc R Soc Lond B Biol Sci 177: 509–539, 1971. Dodd J and Horn JP. A reclassification of B and C neurones in the ninth and tenth paravertebral sympathetic ganglia of the bullfrog. J Physiol 334: 255–269, 1983. Furness JB and Costa M. The Enteric Nervous System. Edinburgh: Churchill Livingstone, 1987. Gibbins I. Chemical neuroanatomy of sympathetic ganglia. In: Autonomic Ganglia, edited by McLachlan EM. Australia: Harwood Academic Publishers, 1995, p. 73–121. Gibbins IL, Jobling P, Messenger JP, Teo EH, and Morris JL. Neuronal morphology and the synaptic organisation of sympathetic ganglia. J Auton Nerv Syst 81: 104–109, 2000. Gibbins IL, Teo EH, Jobling P, and Morris JL. Synaptic density, convergence, and dendritic complexity of prevertebral sympathetic neurons. J Comp Neurol 455: 285–298, 2003. Häbler HJ, McLachlan EM, Jamieson J, and Davies PJ. Synaptic responses evoked by lower urinary tract stimulation in superior cervical ganglion cells in the rat. J Urol 161: 1666–1671, 1999. Hirst GD and McLachlan EM. Development of dendritic calcium currents in ganglion cells of the rat lower lumbar sympathetic chain. J Physiol 377: 349–368, 1986. Ivanoff AY and Smith PA. In vivo activity of B- and C-neurones in the paravertebral sympathetic ganglia of the bullfrog. J Physiol 485: 797–815, 1995. Ivanov A and Purves MA. Ongoing electrical activity of superior cervical ganglion cells in mammals of different size. J Comp Neurol 284: 398–404, 1989. Iversen S, Iversen L, and Saper CB. The autonomic nervous system and the hypothalamus. In: Principles of Neural Science (4th ed.), edited by Kandel ER, Schwartz JH, and Jessell TM. New York: McGraw-Hill, 2000, p. 962–981. Jan LY and Jan YN. Peptidergic transmission in sympathetic ganglia of the frog. J Physiol 327: 219–246, 1982. Jänig W and McLachlan EM. Characteristics of function-specific pathways in the sympathetic nervous system. Trends Neurosci 15: 475–481, 1992. Jänig WM. Ganglion transmission in vivo. In: Autonomic Ganglia, edited by McLachlan EM. Luxembourg City, Luxembourg: Harwood, 1995, p. 349–395. Jones SW and Marks TN. Calcium currents in bullfrog sympathetic neurons. I. Activation kinetics and pharmacology. J Gen Physiol 94: 151–167, 1989. Karila P and Horn JP. Secondary nicotinic synapses on sympathetic B neurons and their putative role in ganglionic amplification of activity. J Neurosci 20: 908–918, 2000. Khoo MCK. Physiological Control Systems. Piscataway, NJ: IEEE Press, 2000. Kuba K and Koketsu K. Synaptic events in sympathetic ganglia. Prog Neurobiol 11: 77–169, 1978. Kullmann PHM, Wheeler DW, Beacom J, and Horn JP. Implementation of a fast 16-bit dynamic clamp using LabVIEW-RT. J Neurophysiol 91: 542–554, 2004. Lichtman JW. On the predominantly single innervation of submandibular ganglion cells in the rat. J Physiol 302: 121–130, 1980. Macefield VG, Elam M, and Wallin BG. Firing properties of single post-ganglionic sympathetic neurones recorded in awake human subjects. Auton Neurosci 95: 146–159, 2002. Mastronarde DN. Nonlagged relay cells and interneurons in the cat lateral geniculate nucleus: receptive-field properties and retinal inputs. Vis Neurosci 8: 407–441, 1992. McAllen RM and Malpas SC. Sympathetic burst activity: characteristics and significance. Clin Exp Pharmacol Physiol 24: 791–799, 1997. McLachlan EM. Transmission of signals through sympathetic ganglia—modulation, integration or simply distribution? Acta Physiol Scand 177: 227–235, 2003. McLachlan EM, Davies PJ, Häbler HJ, and Jamieson J. On-going and reflex synaptic events in rat superior cervical ganglion cells. J Physiol 501: 165–181, 1997. McLachlan EM, Häbler H-J, Jamieson J, and Davies PJ. Analysis of the periodicity of synaptic events in neurones in the superior cervical ganglion of anaesthetized rats. J Physiol 511: 461–478, 1998. McLachlan EM and Meckler RL. Characteristics of synaptic input to three classes of sympathetic neurone in the coeliac ganglion of the guinea-pig. J Physiol 415: 109–129, 1989. Powley TL. Central control of autonomic functions: organization of the autonomic nervous system. In: Fundamental Neuroscience (2nd ed.), edited by Squire LR, Bloom FE, McConnell SK, Roberts JL, Spitzer NC, and Zigmund MJ. San Diego: Academic Press, 2003, p. 911–933. Purves D, Rubin E, Snider WD, and Lichtman J. Relation of animal size to convergence, divergence, and neuronal number in peripheral sympathetic pathways. J Neurosci 6: 158–163, 1986. Scher AM. Cardiovascular control. In: Textbook of Physiology: Circulation, Respiration, Body Fluids, Metabolism, and Endocrinology, edited by Patton HD, Fuchs AF, Hille B, Scher AM, and Steiner R. Philadelphia, PA: Saunders Company, 1989, p. 972–990. Scher AM, O’Leary DS, and Sheriff DD. Arterial baroreceptor regulation of peripheral resistance and of cardiac performance. In: Baroreceptor Reflexes: Integrative Functions and Clinical Aspects, edited by Persson PB and Kirchheim HR. Berlin: Springer-Verlag, 1991, p. 75–125. Schobesberger H, Gutkin BS, and Horn JP. A minimal model for metabotropic modulation of fast synaptic transmission and firing properties in bullfrog sympathetic B neurons. Neurocomputing 26–27: 255–262, 1999. Schobesberger H, Wheeler DW, and Horn JP. A model for pleiotropic muscarinic potentiation of fast synaptic transmission. J Neurophysiol 83: 1912–1923, 2000. Shen WX and Horn JP. A presynaptic mechanism accounts for the differential block of nicotinic synapses on sympathetic B and C neurons by d-Tubocurarine. J Neurosci 15: 5025–5035, 1995. Simmons MA. The complexity and diversity of synaptic transmission in the prevertebral sympathetic ganglia. Prog Neurobiol 24: 43–93, 1985. Skok VI and Ivanov AY. What is the ongoing activity of sympathetic neurons? J Auton Nerv Syst 7: 263–270, 1983. Smith PA. Amphibian sympathetic ganglia: an owner’s and operator’s manual. Prog Neurobiol 43: 439–510, 1994. Smith PA and Weight FF. The pathway for the slow inhibitory postsynaptic potential in bullfrog sympathetic ganglia. J Neurophysiol 56: 823–834, 1986. Sun M-K. Central neural organization and control of sympathetic nervous system in mammals. Prog Neurobiol 47: 157–233, 1995. Tosaka T, Chichibu S, and Libet B. Intracellular analysis of slow inhibitors and excitatory postsynaptic potentials in sympathetic ganglia of the frog. J Neurophysiol 31: 396–409, 1968. Tsujii S and Kuba K. Muscarinic regulation of two ionic currents in the bullfrog sympathetic neurone. Pfluegers 411: 361–370, 1988. Usrey WM, Reppas JB, and Reid RC. Specificity and strength of retinogeniculate connections. J Neurophysiol 82: 3527–3540, 1999. Weight FF and Padjen A. Acetylcholine and slow synaptic inhibition in frog sympathetic ganglion cells. Brain Res 55: 225–228, 1973. Yamada WM, Koch C, and Adams PR. Multiple channels and calcium dynamics—modeling bullfrog sympathetic ganglion cells. In: Methods in Neuronal Modeling from Synapses to Networks, edited by Koch C and Segev I. Cambridge, MA: MIT Press, 1989, p. 97–133.
Lightweight User Grouping with Flexible Degrees of Freedom in Virtual MIMO Ouldooz Baghban Karimi, Milad Amir Toutouchnian†, Jiangchuan Liu, Chonggang Wang‡ School of Computing Science, Simon Fraser University †School of Engineering Science, Simon Fraser University ‡InterDigital Communications Abstract—Virtual MIMO (Multiple Input Multiple Output) groups multiple single-antenna mobile devices to form an antenna array, offering higher degrees of freedom and improved spatial diversity gain as a real MIMO does, yet with much lower costs. In this paper, we focus on the user grouping problem in uplink transmission from multiple single-antenna users to one multiple-antenna base station. State-of-the-art solutions mostly target two single-antenna users, solving a pairing problem. Having more than two single-antenna users in a grouping has yet to be addressed. Intuitively, a higher number of users in a VMIMO group enables higher spectrum efficiency and thus higher throughput gains; the group dynamics however becomes higher too, making fairness harder to be achieved with reasonable computation overhead. To address these challenges, we present a novel solution that decomposes the VMIMO user grouping into two steps. We lighten the computations in user grouping by using instantaneous signal to noise ratio (SNR) as selection criteria, and combining it with proportional fairness for larger groups of users. Lightweight computation in determining the SNR in our solution allows faster grouping and feasible scheduling for a large number of users, as well as fast decision on the efficiency of the number of users in each group. We have evaluated our solution under different network configurations, and the results demonstrate that it achieves much higher data throughput as compared to existing solutions and also well preserves fairness. Keywords: Wireless, VMIMO, uplink scheduling, Virtual MIMO, OFDM, Proportional Fair I. INTRODUCTION The demands for higher data rates over longer distances, scarcity of mobile wireless resources, and need for efficiency of spectrum usage [1] [2] have motivated the development of Multi Input Multi Output (MIMO) antenna over orthogonal frequency division multiplexing (OFDM) communication systems. MIMO antenna systems exploit diversity and spatial multiplexing using more than one antenna at the sending and receiving ends of a transmission, which enhances data throughput in the presence of interference and fading, with minimum additional overhead on bandwidth or transmit power. Although the advantages of MIMO systems are highly appreciated, the complexities of implementing MIMO antenna, including physical barriers in implementing multiple antennas in small hand-held devices and driving multiple radio frequency chains, have challenged the practicality of such systems. MIMO antenna implementation requires multiple antennas on both the receiver’s and sender’s sides. However, the limited size of hand-held mobile devices and the limited number of antennas in each user device (often to a maximum of two) have severely defeated the MIMO gains for such users. Even if the physical challenges of implementing multiple antennas without increasing the weight and size of hand-held devices are met, spatial correlation may frequently appear on compact mobile devices, which degrades the efficiency of spatial multiplexing. An alternative approach to provide the MIMO benefits without the physical impediments of having more antennas on one device is Virtual MIMO. Virtual MIMO creates a wide-area MIMO system by allowing multiple users, or access points, to form a virtual antenna array using the existing physical antennas on each of the user devices. Specifically, multiple single-antenna mobile devices are grouped to create an array of single antennas to act as a multiple antenna device, offering higher degrees of freedom and improved spatial diversity gain as a real MIMO does. This reduces the complexity at each device, but obviously shifts the complexity to forming the multiple antenna groups. Grouping antennas is one of the most important problems in virtual MIMO, both in uplink and downlink scenarios. In downlink, grouping is performed by choosing the access points that can provide service to a user with multiple antennas. In uplink, grouping is performed by choosing users with similar channel qualities to connect to the access point with multiple antennas. Efficient grouping will increase the connected users’ throughput. However, making the decisions only on the quality of channel may starve users with lower channel qualities. Therefore, trade-off should also be made for all of the grouped users in scheduling the uplink traffic, between the overall throughput and the individual’s fairness. In this paper we focus on the user grouping problem in uplink transmission from multiple single-antenna users to one multiple-antenna base station. The current solutions mostly target two single-antenna users, solving a pairing problem [3]. The possibility of having more than two users in a grouping has yet to be addressed, particularly with the throughput and fairness constraints. This is worthwhile because the higher the number of users in a VMIMO group, the higher the spectrum efficiency of the transmission and the MIMO throughput gains of the system. On the other hand, the wireless channel condition changes drastically for a group with multiple mobile users. The higher the number of users in a group, the higher the group dynamics is, making it harder to meet the fairness and throughput criterion together. To address these challenges, we present a novel solution that decomposes the VMIMO user grouping into two steps. We first ensure the proportional fairness in assigning a scheduling time slot to a group of users. Then, we choose the uplink grouping of the users within each large user group and a round robin process among the smaller groups with variable number of users, using the instantaneous Signal to Noise Ratio (SNR) values. Lightweight computation in using instantaneous SNR in our solution allows faster grouping and feasible scheduling for a large number of users, as well as fast decision on the efficiency of the number of users in each group. The fairness and quality decisions made in different steps of the solution also enables extendibility to larger number of users. We have evaluated our proposed solution under different network configurations to examine practicality of the solution. We check diverse wireless networks using MIMO antenna systems in different bands, with varying user dynamics. The simulation results demonstrate that our solution achieves much higher data throughput as compared to existing solutions. It also follows fairness criteria to make sure the users are not starved in different grouping scenarios. The rest of this paper is organized as follows. In section II we quickly review the uplink virtual MIMO and multi-cell MIMO solutions, then we discuss the strengths and shortcomings of state-of-the-art uplink user grouping solutions. Section III offers an overview of our VMIMO system model, followed by our grouping problem formulation. Our VMIMO clustering, grouping, and scheduling algorithms are discussed in IV. We evaluate our proposal in V through simulations and analyze the performance improvement. Section VI concludes the paper and outlines the proposed future work. II. Virtual MIMO Systems: An Overview Grouping multiple streams and forming virtual groups for higher transmission efficiency have been addressed in different types of communication systems in the literature, including wireless systems with code division multiple access sharing [4], wireless local area networks [5], wireless sensor networks [6], cognitive radio networks [7] [8], long term evolution (LTE) of 3GPP macrocells [9] [10] [11], and LTE femtocells [12]. The gains of such virtual multiple antenna systems, even with early simple implementations and idealized analysis, has motivated their consideration as the future of the communication systems [13]. As the future is towards using virtual multi-user MIMO, solving problems associated with multiple independent users has attracted significant interest. Examples include square error minimization on downlink multiuser MIMO using duality between the downlink and uplink and mean square error feasible regions [14], mean square minimization for each receiver branch [15], duality for multiuser MIMO beamforming [16], queueing model for multi-rate multi-user MIMO systems [17], increasing the number of antennas sending at a given time by zero forcing current transmissions [18], and solving the problem of limited feedback in multi-user MIMO systems using alternating codebooks [19]. A great amount of research has also been performed on analytical modelling [20] [21] and measurements [22] [23] on virtual MIMO systems. The impact of a random distribution of nodes on the overall performance of virtual MIMO [21], and traffic load, distance and throughput tradeoffs among the single user and multi-user MIMO systems [24] are analyzed. Multi-user MIMO measurements have been conducted in an urban macrocellular environment [23] and a $4 \times 4$ virtual MIMO [22]. Virtual MIMO systems need cooperation among the users with distributed antennas to re-use the data on the receiver side. This means different designs are needed for uplink and downlink as the grouping is on different set of users. In downlink MIMO, multiple access points are providing access to a single user, which will still need to have multiple antenna to use this functionality. However, each of the antennas are used for downloading a different stream from the associated access point. Virtual MIMO at the uplink is provided by grouping single antenna user devices into groups that together act like a multiple antenna transmitter. The same grouping is not as efficient in downlink as the users will have to communicate to find each others’ channels to use the transmitted data. Figures 1, 2, and 3 illustrate the communication among multiple antennas in a MIMO, an uplink virtual MIMO, and a multi-cell MIMO system, respectively. Tx and Rx denote the transmit and receive antennas, respectively. We next categorize the researches into downlink and uplink VMIMO in more detail. A. Downlink Most of the works on downlink VMIMO have focused on grouping the access points to provide access to a single user. It is also referred to as multi-cell MIMO [25]. Gesbert et al. [25] provided a comprehensive review of all downlink multi-cell MIMO theories, techniques, and solutions incorporated in current practical systems. Ramp et al. [26] also examined it in cooperative cellular networks. Solutions in the literature vary from opportunistic medium access control design [27], to joint solutions for increasing throughput and minimizing power [28] [29], interference, or mean square error [14]. An important problem in multi-cell MIMO is the complexity in synchronization among the cells, as well as complexities on the user side to use the downloaded data. Badic et al. [9] investigated the impact of feedback and user pairing schemes on receiver performance in long term evolution of 3gpp (LTE) systems, emphasizing the tradeoff between receiver performance and complexity. The complexity is even higher when the users also have to be grouped to sending antenna array groups. An example research on this topic is user pairing control for reducing the interference in the cell-edge in multi-cell MIMO [30]. Lan et al. [31] proposed a group competition-based selection. The first user is chosen based on having the largest channel gain, and the rest of the group members are selected using an orthogonality threshold. The best group having the highest overall rate is then selected as the final user set. Fairness is not considered in their work. The details of how the users collect the channel information in the user side, which is the significant challenge for grouping users for downlink VMIMO, is not discussed either. Other solutions for downlink VMIMO include precoding techniques for reducing the signal to interference and noise ratio of multiple data streams [32], and beamforming design solutions with different design constraints [29]. B. Uplink In uplink virtual MIMO, multiple users each with a single antenna operate at the same time to increase the aggregate uplink throughput, which can also be viewed as a form of spatial division multiple access. Researches on uplink MIMO range from power allocation [33] to multi-cell uplink coordination [34]. Particular focuses have been on grouping the users and scheduling the grouped users for higher throughput, while keeping such promises as fairness and power constraints. Saad et al. [35] proposed a game theoretic approach for distributed clustering to form uplink coalitions by collaborating users in time division multiple access, and designed the signalling among the users for such decision making. They considered fairness in their utility function and payoff, but the throughput is not explicitly considered or measured. Such distributed decision for group forming also introduces a high complexity and message passing requirements to the system. The need for such distribution is not justifiable given the low number of users in uplink coalitions as well as the cost and complexity of distribution. Most of the research in grouping users only considered two users, and discussed pairing and scheduling for the users [3] [36] [37]. Traditionally the uplink scheduling was done by a proportional fair scheduler, which prevents starving users by choosing only those with high quality connections. The same approach has been adopted in user pairings, known as double proportional fair scheduling. Chen et al. [3] presented the most famous pairing algorithm. The algorithm uses the proportional fairness to decide the first user, and a modified proportional fairness criterion to choose the pairing user. This pioneer work was followed by a series of pairing algorithms [38], including combined optimization of user pairing and spectrum allocation [39], and robust pairing for changing channel conditions [37]. Li et al. [36] proposed a pairing scheduling that combines the advantages of the proportional fair and maximum rate rules, using successive interference cancellation to meet the tradeoff between aggregate throughput and user fairness. Wang et al. [40] proposed a pairing scheduling based on selecting the first user with proportional fairness and pairing users with a fairness adjustable mechanism that also considers the channel orthogonality and system capacity. The focus on pairing in the state-of-the-art solutions, rather than considering higher number of users, is the high complexity of grouping and scheduling, making them impractical with more users. Our solution, while ensuring the fairness and providing higher throughput, incorporates a lightweight decision making scheme that makes it practical for higher number of users. This will increase the efficiency in spectrum usage, as the higher number of antennas directly translates into higher throughput. III. SYSTEM MODEL AND PROBLEM STATEMENT We consider a single access point scenario, where there is no cooperation or interference from the neighbouring access points anywhere in the region covered by the access point of interest. There are $U$ users in the sensing range of the access point, and all have uplink data to send to the access point. Each user has only one transmit antenna. The access point is equipped with $k$ receiver antennas. An access point scheduler chooses $U_g \leq k$ users to share the same time-frequency resource blocks for their uplink transmission to the access point as if there are $U_g$ transmit antennas in the uplink. ![Fig. 4. Uplink VMIMO User Grouping Problem](image) Figure 4 illustrates the conceptual model of our uplink VMIMO system. Tx and Rx denote the transmitting and receiving antennas, respectively. The different set of arrows show the transmissions for each set of transmitting and receiving antennas. We assume perfect channel information, as well as perfect phase, symbol, and frame synchronization. These perfect assumptions are common practice when their estimation is not the main subject of interest. In the uplink virtual MIMO, one resource block, which is traditionally allocated to one users, is scheduled for concurrent uplink transmission of more than one user to improve the spectral efficiency. The scheduled users transmit their signals TABLE I DEFINITION OF PARAMETERS \| Symbol \| Description \| \|--------\|-------------\| \| $k$ \| Number of antennas in the access point \| \| $\mu$ \| Number of transmit antennas \| \| $\nu$ \| Number of receive antennas \| \| $u$ \| Uplink user \| \| $n$ \| Transmission subcarrier \| \| $N$ \| Number of subcarriers \| \| $N_0$ \| White noise \| \| $U$ \| Set of users in the system \| \| $U_g$ \| Cluster of users to be scheduled \| \| $U_s$ \| Group of users selected for transmission \| \| $p_u$ \| Path loss for user $u$ \| \| $\Delta t$ \| Scheduling period \| \| $x^t_{U_s}$ \| Scheduling variable \| \| $b^t_{U_s}$ \| Bandwidth share variable \| \| $\alpha^i$ \| Weight of clustering parameter $i$ \| \| $q^i_u$ \| Clustering criteria parameter $i$ for user $u$ \| \| $\gamma_{U_s}$ \| Current mean of parameters in $U_s$ \| \| $C(U_s)$ \| Estimated capacity for group $U_s$ \| independently over different antennas to achieve the spatial multiplexing of an MIMO system. Therefore, the access point should employ advanced receiver structures to differentiate the data streams being transmitted from different users. For this purpose, we consider using a semi-definite relaxation decoder (SDR) at the receiver [41]. The SDR decoder is a recent solution for multiuser detection, which can provide a very competitive performance in approximating maximum likelihood detection in MIMO systems at low computational cost. Our goal is to choose the $U_g$ users out of $U$ possibilities while scheduling each resource block (RB) to have the highest uplink throughput over the $N$ available subcarriers: $$\argmax_{U_g \subseteq U} \sum_{U_s \subseteq eqU} \log(C(U_g))$$ \hspace{1cm} (1) where $C(U_g)$ is the throughout of the simultaneous transmission to group $U_g$. An efficient grouping algorithm should maximize the total throughput for all users at the same time without starving some users with lower quality connections. We maximize $\log(C(U_g))$ to achieve the proportional fairness, and to reduce the aggressiveness of the grouping towards high bandwidth. This means serving each user in a group with a fair portion of throughput. Thus, while achieving the aforementioned goal in equation (1), the system will be fair among the users. For easier reference, table I summarizes the variables used in our system model and formulation. IV. LIGHTWEIGHT FAIR MULTI-USER GROUPING Our lightweight fair multi-user algorithm groups $\|U_g\|$ users for each scheduling time $\Delta t$. The selection criteria should meet the fairness requirements among the users while maximizing the uplink throughput of the VMIMO system for them. We meet these requirements with first clustering the users to smaller groups of $U_s$ in the sensing region of the access point. This is specially important in cellular networks. We use a simple k-mean clustering algorithm [42], for clustering the users: $$\argmin_{u_i \in U_s} \sum_{i=1}^{\|U\|/\|U_s\|} \sum_{u \in U_s} \\| \alpha^i q^i_u - \gamma_{U_s} \\|$$ \hspace{1cm} (2) where $\|U\|$, and $\|U_s\|$ denote the total number of users and users within the selected group respectively. These numbers are available to access points based on their connected users. $q^i$ shows the $i$th clustering criteria and $\alpha^i$ shows it’s weight in the clustering algorithm. $\gamma_{U_s}$ is the current mean of the cluster. Our criteria are received signal power, location, and velocity information of the users, all available at the access point. A. Proportional Fairness: Clusters After we cluster the users, we assign each set of users in a cluster to a specific scheduling period using a proportional fair scheduling algorithm. We define the proportional fairness for a group of users as the following: $$\max \sum_{U_s \in U_s} \log(w_{U_s} x^t_{U_s} b^t_{U_s})$$ $$\sum_{U_s \in U_s} x^t_{U_s} = 1$$ $$b^t_{U_s} = \frac{C_{U_s}}{U_s \in U_s x^t_{U_s}}$$ $$x^t_{U_s} \in \{0, 1\}$$ \hspace{1cm} (3) where $x^t_{U_s}$ is the selection variable that shows the cluster $U_s$ will be scheduled within the scheduling period $t$. $C_{U_s}$ is an estimated capacity of transmission for group $U_s$. $w_{U_s}$ is the weight assigned to group $U_s$, showing the importance of group to the scheduling algorithm. Because $x^t_{U_s}$ takes integer values, optimization (3) is an integer linear program, which is NP-hard in general. We can approximate the solution by relaxing $x^t_{U_s}$ which yields in a 2-approximation of the proportional fair scheduling that can be solved within polynomial time. The next step is to use a lightweight physical layer grouping algorithm, and to schedule each group of VMIMO users to transmit on the same channel to the receiver antenna array. B. Throughput Maximization: Grouping Within Clusters We schedule the selected users within each cluster in a round robin basis in the resource blocks within the given scheduling period. We need the round robin procedure because we have flexible number of users chosen to transmit at each resource block. Therefore, if we have a higher number of users selected for transmission, we will have a few resource blocks left at the end of the scheduling period. Other reasons for using a simple round robin are our prior application of proportional fairness at cluster level, and keeping it simple for group level. We re-schedule the $U_s$ members to fill out the unused resource blocks in a round robin scheduler. The physical layer of uplink virtual MIMO model is represented as a $k \times k$ MIMO-OFDM system where up to $\|U_g\| \leq k$ selected users form a group of decentralized transmitters. The throughput of such a system is given by: $$C = \frac{1}{N} \sum_{n \in N} \sum_{u \in U_g} \log_2[1 + \frac{E_x p_u \lambda_u (\mathbf{H}(n))}{\|U_g\| N_0}]$$ \hspace{1cm} (4) where $\mu$ is the number of transmit antennas after $k$ user selection, $p_u$ is the path loss of the sender $u$, and $E_s$ is the average transmitting signal energy at one symbol time which is a constant value, being identical for all of the users. $N_0$ is the single side spectral density of additive white Gaussian noise (AWGN). $\lambda_n(\mathbf{H}(n))$ is the link capacity of the $n$th subcarrier with an optimal receiver. The group of the selected users is reflected in the usage of the $n$th subcarrier, and $N$ is the total number of subcarriers. The users are each equipped with a single transmit antenna, and the access point is equipped with $k$ receive antennas. The received vector $\mathbf{y}(n) \in \mathbb{C}^{k \times 1}$ at $n$th subcarrier can be written as: $$\mathbf{y}(n) = \mathbf{H}(n)\mathbf{x}(n) + \mathbf{Dw}(n)$$ (5) where $\mu$ denotes the transmit, and $\nu$ denotes the receive antenna. In our uplink VMIMO model, we assume that $\nu \in \{1, \cdots, k\}$ and $\mu$ is any possible combination of $\binom{k+1}{k}$. $\mathbf{D}$ is a diagonal matrix with diagonal elements representing the variables of the system including the effect of path loss, transmit power, transmit antenna gain, receive antenna gain, and receiver noise figure. With a perfect power control (PPC) mechanism through the access point, all the elements of the diagonal matrix $\mathbf{D}$ will be identical. $\mathbf{w}(n)$ is the noise in the frequency domain. $\mathbf{H}(n) \in \mathbb{C}^{k \times k}$ and its entries, $H_{\nu,\mu}$, contain the frequency response of $h_{\nu,\mu}$: $$H_{\nu,\mu}(n) = \sum_{l=0}^{L} b_{\nu,\mu}(l)e^{-2\pi jl n/N}$$ (6) where $l$ is channel tap, and $L+1$ denotes the number of paths between the transmitter and receiver antennas. For decoding the data, the access point uses the SDR decoder for multiuser detection. Recently, quasi-maximum-likelihood detection based on semidefinite relaxation (SDR) shows near-maximum likelihood (ML) performance with a worst case complexity of $O(k^{3.5})$ [41]. $$\min_{\mathbf{x}(n)} \\| \mathbf{y}(n) - \mathbf{H}(n)\mathbf{x}(n) \\|^2$$ s. t. $\mathbf{x}(n) \in \mathcal{A}_x^k$, (7) where $\mathbf{x}(n)$ belongs to the finite alphabet constellation set $\mathcal{A}_x^k$. Define $\mathbf{X} \triangleq \mathbf{x}(n)\mathbf{x}^T(n)$, so rank$(\mathbf{X}) = 1$. It is proved that if rank constraint can be relaxed in (7) then (7) is solved by semidefinite programming (SDP) [41], but at the second step the rank-1 constraint is satisfied through the randomized rounding procedure applied to the SDP results. The main advantage of using SDR is its polynomial complexity [41]. The decision metric here is the instantaneous SNR over a single subcarrier, as given by the second term in equation (4), or: $$\text{SNR}(n) = \frac{\text{tr}(\mathbf{H}(n)\mathbf{H}^H(n))}{N_0 \text{tr}(\mathbf{D})}$$ (8) 1) Maximum Summation of Instantaneous SNR: The maximum summation of instantaneous SNR can be defined as a metric for user selection within a given cluster. The following problem $\mathcal{G}$ formulates it: $$\mathcal{G}: \max_{k, \mu \in \binom{[U_g]}{k}} \sum_{n=1}^{N} \text{SNR}(n)$$ (9) This provides a low complexity, low feedback grouping algorithm that is ideal for wireless networks with lower user dynamics, including wireless local area networks. 2) Individual Values of Instantaneous SNR: Solving problem $\mathcal{G}$ for grouping $k$ users can effectively, and fairly, improve the capacity of the virtual MIMO system over user pairing. Being very lightweight, however, comes with a few drawbacks, including fluctuations in the SNR values after making the decision, and potentially less fairness among the users. This is particularly important for cellular networks with higher user dynamics within long scheduling periods. We use individual values of instantaneous SNR for grouping algorithm in such situations. Algorithm 1: Lightweight Fair Grouping Algorithm* Inputs: Set of $U$ users; Path loss values for $U$ users; $N$ number of subcarriers; $\Delta t$ scheduling frame; Output: Scheduled groups of users within the given $\Delta t$ scheduling time consisting of $N_{\Delta t}$ resources blocks Begin Cluster the users in the $U$ to smaller groups $U_g$, using received signal power, location, and velocity information and k-mean clustering; $U_s(\Delta t) = U_g$, using Proportional Fair scheduling; for each scheduling duration $\Delta t$, and group $U_s(\Delta t)$ do Start Round scheduling for $U_s(\Delta t)$ as follows; $U_s = U_s(\Delta t)$ for $i = 0$ to $N_{\Delta t}$ do Perform grouping algorithm within the cluster as follows: Pick $U_g \leq U_s$ based on equation (10): $$\max_{k, \mu \in \binom{[U_g]}{k}} \min_{n=1,\cdots,N} \text{SNR}(n);$$ if $U_g \geq U_s$ then $U_g = U_s$; Assign $RB_i$ for group $U_g$; Mark $RB_i$ as used; $U_s - = U_g$; if $U_s == 0$ then Perform Round Robin: $U_s = U_s(\Delta t)$; End Achieving the highest throughput over the $N$ subcarriers translates into the largest minimum instantaneous SNR over $N$ subcarriers. Our goal is to choose $k$ and $\mu$ to have the highest minimum $\text{SNR}(n)$, for $n = 1, \cdots N$, where $N$ is the number of subcarriers. We find the $k^$ and $\mu^$ are the solution of the following problem: $$\mathcal{P}: \max_{k, \mu \in \binom{[U_g]}{k}} \min_{n=1,\cdots,N} \text{SNR}(n)$$ (10) For example if we assume the values of $k = 2$ and $k = 3$, meaning $2 \times 2$ and $3 \times 3$ VMIMO, we can easily show that with PPC, where all received signals have the same received signal values, $D = \sigma^2 \mathbf{I}$, $$\max_{\mu \in \binom{[U_g]}{g}} \min_{n=1,\cdots,N} \text{SNR}(n) > \max_{\mu \in \binom{[U_g]}{g}} \min_{n=1,\cdots,N} \text{SNR}(n)$$ (11) However, without PPC, different channel gains for the users may cause degradation in the higher values of $k$ compared to lower ones. This happens when one of the users in the group has a significantly lower channel quality, or higher velocity. When we have different average received SNR values, e.g. for $k = 2$ or $k = 3$ at each time-slot, the user selection is performed by solving the problem $\mathcal{P}$, and the number of grouped users is variable by the channel conditions of the users. Solving equation (10) improves the throughput or minimizes the bit error rate of the system significantly as it receives the new values of the SNR while performing the round robin scheduling on the previous users. Since the calculations are made at each step of the round robin for choosing the next $U_g$ out of the given $U_s$, these updates do not add to the computational complexity of the algorithm. On the other hand, this algorithm requires computing all of the combinations of number of the users in $U_s$ to be considered as candidates for $U_g$. First, since the $U_s$ is small, and this computation is done on the $U_s$ members to determine $U_g$ members, it will introduce a very high time complexity to the system. Second, the size of $U_s$ and $U_g$ are fairly adjustable and if there is a need for larger sets of $U_s$, an approximation on grouping different levels of SNR threshold will solve the problem with a high precision. Algorithm 1 summarizes our lightweight fair grouping algorithm, briefly presenting our clustering and grouping steps of VMIMO user grouping. ### Table II SIMULATION SETTINGS - SAME FOR BOTH SCENARIOS \| Parameter \| Sim (1) \| Sim (2) \| \|----------------------------\|-----------\|-----------\| \| FFT Size \| 16 \| \| \| User Antennas \| 1 Tx \| \| \| AP Antennas \| 3 Rx \| \| \| Receiver \| SDR \| \| \| Power control \| without PPC\| \| \| Channel estimation \| Ideal \| \| \| $\|U\|$ \| 120 \| \| \| $\|U_g\|$ \| $\leq 12$ \| \| \| Access point range \| 500m \| 135m \| \| Carrier frequency \| 2.0GHz \| 2.4GHz \| \| User velocity \| 0 - 60km/hr\| 0 \| ### V. PERFORMANCE EVALUATION In this section we evaluate our solution and also compare it with proportional fair pairing algorithm. The first part of our simulator consists of a detailed yet simplified model of user dynamics in wireless networks. This part simulates our users movement and its effect on the grouping. We use random movement pattern for users within the access point sensing region. Therefore, when a user is going out of the region with the current speed and direction, we change the direction of movement so that the user stays within the sensing range of the access point. It computes the physical layer SNR, path loss, and frequency selective channel quality index feedback values for every single user separately and registers them in a file. We use the time and frequency selective channel frequency index and path loss feedback values for the users from the first part of the code output and run the scheduling part with MATLAB. Scheduling is over the given values from the first part of the simulation, so although scheduling is not real-time over the given trace of user movements and their channel values, it performs it within real-time implementation deadlines. It has two parts of user clustering and proportional fair scheduling, and then applying the VMIMO transmission grouping as proposed in our algorithm. Accordingly, our simulation consists of these two steps implemented in the MATLAB where the scheduling for a random data bits are done using the user information from the C file. #### A. Simulation Settings Virtual MIMO can be used for a wide range of wireless networks. We provide two different set of parameters to simulate both cellular networks and wireless local area networks using the virtual MIMO antenna system given the proposed scheduling. We evaluate the bandwidth and fairness for both of the simulation scenarios. Users will be fairly static in the wireless local network scenario, but they are highly dynamic within the cellular network. We Table II summarizes our simulation settings. As we discussed earlier, our physical layer VMIMO user grouping algorithm can provide different degrees of freedom, and also based on the number of users in the cell, we can choose different sizes for $U_g$ and $U_s$ in the clustering algorithm. and grouping algorithm to facilitate the scheduling. It is intuitive, and has been proven by the MIMO theory [18], that the higher the number of users grouped in the VMIMO transmission $U_g$, the higher the capacity and throughput of the transmission. Therefore, we only compare the results of our grouping with $k = 3$ and $\|U_g\| \leq 3$ which means choosing 2 or 3 users to transmit at the same time. Throughout the simulations we will monitor the throughput and fairness of this scenario and compare it to the double proportional fair pairing algorithm. B. Simulation Results We first compare the results of the physical layer grouping module only. Figures 6 and 5 illustrate the SNR(n) and capacity over the VMIMO channels our grouping algorithm for $U_s = 12$ and $U_g \leq 3$, and compare it to double proportional fair pairing [3]. Our method provides degree of freedom, but we focus on $U_g \leq 3$ as selecting the larger number of user makes the system more complex for real-time applications. The results are averaged over 10 different simulations with different combination of randomly placed users. Figure 8 also provides a more detailed look at the SNR(n) values for lower values of SNR. We observe that our flexible grouping algorithm at the physical layer, which gathers the users three-user or two-user groups, provides a higher VMIMO capacity, throughput, and higher shared signal strength than the pairing algorithm. Another observation is the nodes with higher instantaneous SNR values benefit more throughput increase compared to the pairing algorithm. However, the users with very low instantaneous SNR values may experience an slight drop in throughput compared to the pairing algorithm. This means that the algorithm favours the users with higher instantaneous SNR values. This is because in the trade-off between fairness and throughput, our physical layer flexible grouping algorithm is more aggressive towards getting higher throughput for the VMIMO system. This is on the physical layer grouping algorithm only, and as we will see in the next simulations, can be solved in the proportional fair cluster scheduling. We observe that the throughput of the users have been increased on average. This increase has been mostly on the low SINR users. On the other hand, we know that our grouping algorithm favours the users with higher instantaneous SNR values. Therefore, this is the result of the proportional fairness in the cluster level. Figures 9 and 10 also illustrate the throughput of users in a cellular settings and a wireless local network setting as done in simulation settings (1) and (2), respectively. We can see that our grouping algorithm fairly treats the users while increasing the bandwidth. We can argue that this is because of applying the proportional fairness at the cluster level, in contrast to finding each user using proportional fairness for a pairing as it is done in the pairing algorithm. This way, all of the poor signal users get the same chance to transmit. Also, when scheduled within the same cluster, the round robin scheduling within a cluster may bring chances of more scheduled resource blocks for a user of poor signal reception. VI. CONCLUSION In this paper, we examined the implementation of Virtual MIMO antenna systems with varying scales of users. We expanded the traditional pairing and scheduling that have been largely based on two users towards larger group of users. We developed a light-weight fair scheduling algorithm to organize the users to achieve VMIMO functionalities. With minimum overhead, our solutions ensures that the higher the number of users in each group, the higher the efficiency of the transmission, and the higher the throughput of the uplink VMIMO system is. We evaluated the performance of proposed solution through extensive simulations. The results suggest that it achieves up to 30% increase in throughput and, with lower delay owing to grouping a higher number of users together, while keeping the fairness criteria. This enables checking the higher degrees of freedom for more transmission capacity. There are a number of possible future avenues toward improving our design including energy aware scheduling, and reducing the computation even further by feedback from the clustering algorithm. In particular, we are currently working on further improvement on the approximation algorithm for scheduling to exclude a number of decisions between degrees of freedom based on the feedback from clusters. Moreover, it is worth noting that the proposed algorithm, although initially for for multiple users with a single antenna, works for higher number of antennas on each user. This is achieved by assuming different streams being sent on each antenna of a specific user. In that case assuming each of the multiple antennas on a user device can be assumed as a single user with single antenna to use the same algorithm for grouping and uplink scheduling. REFERENCES [1] H. Zhu and J. Wang, “Chunk-based resource allocation in OFDMA systems: part I: chunk allocation,” IEEE Transactions Communications, vol. 57, no. 9, pp. 2734–2744, 2009. [2] ——, “Chunk-based resource allocation in OFDMA systems: part II: Joint chunk, power and bit allocation,” IEEE Transactions Communications, vol. 60, no. 2, pp. 499–509, 2012. [3] X. Chen, L. Hu, B. Wang, H. Iwai Chen, and M. Guizani, “Double proportional fair user pairing algorithm for uplink virtual MIMO systems,” IEEE Transactions on Wireless Communications, vol. 7, no. 7, pp. 2425–2429, 2008. [4] S. Sfar and K. Ben Letaief, “Layered group detection for multiuser MIMO wireless networks,” IEEE Transactions on Wireless Communications, vol. 5, no. 9, pp. 2208–2213, 2006. [5] Y. Hua, Q. Zhang, and Z. Niu, “A cooperative MAC protocol with virtual-antenna array support in a multi-AP WLAN system,” IEEE Transactions on Wireless Communications, vol. 8, no. 9, pp. 4806–4814, 2009. [6] J.-S. Chung, J. Kim, and D. Han, “Multihop hybrid virtual MIMO scheme for wireless sensor networks,” IEEE Transactions on Vehicular Technology, no. 99, p. 1, 2012. [7] C.-X. Wang, X. Hong, H.-H. Chen, and J. Thompson, “On capacity of cognitive radio networks with average interference power constraints,” IEEE Transactions on Wireless Communications, vol. 8, no. 4, pp. 1620–1625, 2009. [8] X. Ban, P. Martins, T. Song, and L. Shen, “Capacity of hybrid cognitive network with outage constraints,” IET Communications, vol. 5, no. 18, pp. 2712–2720, 2011. [9] B. Badic, R. Balraj, T. Scholzard, Z. Bai, and S. Iwelski, “Impact of feedback and user pairing schemes on receiver performance in MU-MIMO LTE systems,” in IEEE Wireless Communications and Networking Conference, 2012, pp. 399–403. [10] C. Wang, Q. Cui, S. Li, X. Tao, and X. Xu, “Multiuser pairing in uplink CoMP MU-MIMO systems using particle swarm optimization,” in IEEE Vehicular Technology Conference, 2011, pp. 1–5. [11] P. Chou, S. Narayanan, A. Chandra, I. F. Akyildiz, T. Derham, G. Li, X. Shen, P. Zorzi, and J. Yang, “LTE-advanced: an operator perspective,” IEEE Communications Magazine, vol. 50, no. 2, pp. 104–114, 2012. [12] L. Yu, J. Liu, and K. Long, “Analysis of virtual MIMO-based cooperative communication in broadcast networks,” in IEEE Global Telecommunications Conference, 2011, pp. 1–6. [13] A. Kuree, “Multi-user MIMO systems: the future in the making,” IEEE Potentials, vol. 28, no. 6, pp. 37–42, 2009. [14] S. Shi, M. Schubert, and H. Boche, “Downlink MMSE transceiver optimization for multiuser MIMO Systems: Duality and sum-MSE minimization,” IEEE Transactions on Signal Processing, vol. 55, no. 11, pp. 5436–5449, 2007. [15] J. Wang, “Recursive combiner bank for MIMO system under multi-user co-channel interference,” IEEE Communications Letters, vol. 16, no. 3, pp. 328–330, 2012. [16] B. Song, R. Cruz, and B. Rao, “Network duality for multiuser MIMO beamforming in cellular networks,” IEEE Communications, IEEE Transactions on, vol. 55, no. 3, pp. 618–630, 2007. [17] B. Bellalta, V. Daza, and M. Oliver, “An approximate queuing model for multi-rate multi-user MIMO systems,” IEEE Communications Letters, vol. 15, no. 4, pp. 392–394, 2011. [18] K. Lin, S. Mukkula, and D. Katabi, “Random access heterogeneous MIMO networks,” in ACM SIGCOMM. New York, NY, USA: ACM, 2011, pp. 146–157. [19] C. Jiang, M. Wang, F. Shu, J. Wang, W. Sheng, and Q. Chen, “Multi-user MIMO with limited feedback using alternating codebooks,” IEEE Transactions on Communications, vol. 60, no. 2, pp. 333–338, 2012. [20] N. Czink, B. Bandemer, C. Oestges, T. Zemen, and A. Paulraj, “Analytical multi-user MIMO channel modeling: Subspace alignment matters,” IEEE Transactions on Wireless Communications, vol. 11, no. 1, pp. 367–377, 2012. [21] C. Buratti and A. Zanella, “Multihop virtual MIMO systems with channel reuse in a poisson field of nodes,” IEEE Transactions on Vehicular Technology, vol. 60, no. 5, pp. 2060–2069, 2011. [22] J. Wang, M. Yu, and M. Beach, “Propagation characteristics, metrics, and statistics for virtual mimo performance in a measured outdoor cell,” IEEE Transactions on Antennas and Propagation, vol. 59, no. 1, pp. 236–244, 2011. [23] B. K. Lau, M. Jensen, J. Medbo, and J. Furuskog, “Single and multi-user cooperative MIMO in a measured urban macrocellular environment,” IEEE Transactions on Antennas and Propagation, vol. 60, no. 2, pp. 624–632, 2012. [24] H. Jin, B. C. Jung, and D. K. Sung, “A tradeoff between single-user and multi-user MIMO schemes in multi-rate uplink WLANs,” IEEE Transactions on Wireless Communications, vol. 10, no. 10, pp. 3332–3342, 2011. [25] D. Gesbert, S. Hanly, H. Huang, S. Shamai Shitz, O. Simeone, and W. Yu, “Multi-cell MIMO cooperative networks: A new look at interference,” IEEE Journal on Selected Areas in Communications, vol. 28, no. 9, pp. 1380–1408, 2010. [26] S. Rangarajan, H. Papadopoulos, A. Benjebbour, Y. Kishiyama, N. Jain, and G. Caire, “Cooperative cellular networks using multi-user MIMO: trade-offs, overheads, and interference control across architectures,” IEEE Communications Magazine, vol. 49, no. 5, pp. 70–77, 2011. [27] D. Jung and H. Lim, “Appropriate MAC protocol for coordinating simultaneous transmissions in multi-user MIMO based WLANs,” IEEE Communications Letters, vol. 15, no. 1, pp. 201–203, 2011. [28] G. Zheng, K.-K. Wong, and T. S. Ng, “Throughput maximization in linear multiantenna MIMO-OFDM downlink systems,” IEEE Transactions on Vehicular Technology, vol. 57, no. 3, pp. 1993–1998, 2008. [29] M. Razaviyayn, H. Baligh, A. Callard, and Z. Luo, “Joint transceiver design and user grouping in a MIMO interfering broadcast channel,” in 45th Annual Conference on Information Sciences and Systems, 2011, pp. 1–6. [30] T. Fujii and Y. Houmura, “Multi-cell shared multi-user MIMO relay with user pairing control,” in Fourth International Conference on Ubiquitous and Future Networks, 2012, pp. 363–368. [31] Y. Lai, J. Liu, and H. V. Poor, “Efficient group competition-based user selection scheme for multiuser beamforming in high order MIMO broadcast systems,” in IEEE Vehicular Technology Conference, 2010, pp. 1–5. [32] P. Chen, M. Tao, and W. Zhang, “A new SLNR-based linear precoding for downlink multi-user multi-stream MIMO systems,” IEEE Communications Letters, vol. 14, no. 11, pp. 1008–1010, 2010. [33] A. Tenenbaum and R. Adve, “Minimizing sum-MSE implies identical downlink and dual uplink power allocations,” IEEE Transactions on Communications, vol. 59, no. 3, pp. 686–688, 2011. [34] S. Li, Q. Cui, C. Wang, and X. Tao, “Coordinated cell threshold and scheduling criteria for uplink CoMP MU-MIMO,” in International Conference on Computer Science and Network Technology, vol. 3, 2011, pp. 1740–1744. [35] W. Saad, Z. Han, M. Debbah, and A. Hjorungnes, “A distributed coalition formation framework for fair user cooperation in wireless networks,” IEEE Transactions on Wireless Communications, vol. 8, no. 9, pp. 4580–4593, 2009. [36] Y. Li, W. Wang, X. Zhang, and M. Peng, “Combined proportional fair and maximum rate scheduling for virtual MIMO,” in Vehicular Technology Conference, 2008, pp. 1–5. [37] Y. Rui, J. Hu, and Y. Li, “A robust user pairing algorithm under channel estimation errors for uplink virtual multiple-input multiple-output systems,” IET Communications, vol. 6, no. 3, pp. 318–323, 2012. [38] Y. Song, G. Su, S. Wang, and Y. Xie, “Group-based user pairing of virtual MIMO for uplink of LTE system,” in International Conference on Consumer Electronics, Communications and Networks, 2012, pp. 346–349. [39] M. Ruder, D. Ding, U. I. Dang, and W. Gerstacker, “Combined user pairing and spectrum allocation for multiuser SC-FDMA transmission,” in IEEE International Conference on Communications, 2011, pp. 1–6. [40] X. Wang, W. Wang, Z. Zhao, and Y. Zhu, “Fairness adjustable grouping for uplink virtual MAC with MMSE-SIC receiver,” in IEEE GLOBECOM Workshops, 2009, pp. 1–5. [41] M. Kaisioulis, X. Luo, and Z. Luo, “Efficient implementation of quasi-maximum-likelihood detection based on semidefinite relaxation,” IEEE Transactions on Signal Processing, vol. 57, no. 12, pp. 4811–4822, 2009. [42] T. Kamphans, D. M. Mount, N. S. Netanyahu, C. D. Piatko, R. Silverman, and A. Y. Wu, “Approximate k-nearest neighbor clustering algorithm: Analysis and implementation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 881–892, 2002.
Appendix A from J. Usinowicz, “Limited Dispersal Drives Clustering and Reduces Coexistence by the Storage Effect” (Am. Nat., vol. 186, no. 5, p. 000) Derivation of Mean-Field, Local-Dispersal (LD), and Pairwise Approximations Mean-Field Approximation The mean-field approximation uses the minimum amount of information to characterize the global state of each species. It assumes that every potential recruit can reach any open site on the lattice. Thus, the mean-field approximation ignores local interactions created by limited dispersal. The global density of each species is represented by the variables $p_i(t)$ and $p_0(t)$, which count the proportion of sites occupied by the invader and resident species, respectively, so that $p_i(t) + p_0(t) = 1$. This means that $p_1(t)$ can be treated as the probability of finding a site that is occupied by species 1. The law of total probability then makes it possible to write $p_1(t+1)$ as a function of two conditional probabilities: the probability that a resident site becomes an invader and the probability that an invader site becomes an invader, which I write as $p_1(t+1) = p_0(t)P[0 \rightarrow 1] + p_1(t)P[1 \rightarrow 1]$, where the arrows indicate a conditional probability (Hiebeler 2000; Ives et al. 1998). Writing out expressions and simplifying leads to $$p_1(t+1) = \gamma_{\text{mf}}(t)\delta + p_1(t)(1 - \delta). \quad (A1)$$ The quantity $\gamma_{\text{mf}}(t)$ represents the colonization probability for species 1 under global dispersal $$\gamma_{\text{mf}}(t) = \frac{p_1(t)R_1(t)}{p_1(t)R_1(t) + (1 - p_1(t))R_0(t)}. \quad (A2)$$ Although the notation here is slightly different, the mean-field approximation of the lottery model with limited dispersal is identical to the classic lottery model (Chesson and Warner 1981). The invasion growth rate for this approximation is given by equation (3) in the text. LD Approximation The simplest approximation that incorporates limited dispersal is an LD mean-field approximation (Hiebeler 1997). This approach calculates the probability that a site $z_v$ will be occupied by either species 1 or species 0 at the next time step based on the identity of $z_v$ and the average proportion of either species in the interaction neighborhood: $$p_1(t+1) = \gamma_{\text{ld}}(t)\delta + p_1(t)(1 - \delta), \quad (A3)$$ where the colonization probability for species 1, $\gamma_{\text{ld}}(t)$, is given as $$\gamma_{\text{ld}}(t) = \sum_{k=0}^{\Phi} \left(\frac{\Phi}{k}\right) p_1(t)^k p_0(t)^{(\Phi-k)} \frac{R_1(t)k}{R_1(t)k + R_0(t)(\Phi - k)}, \quad (A4)$$ The LD approximation and the classic lottery model differ in their colonization probabilities, given by equations (A4) and (2), respectively. In the LD approximation, $\gamma_{\text{ld}}(t)$ is based on the average proportion of both species in $N_v$. It is a weighted average where possible values of $R_1(tk)/[R_1(tk) + R_0(t)(\Phi - k)]$ are each weighted by the likelihood of seeing exactly $k$ invaders and $\Phi - k$ residents in $N_v$. Specifying this likelihood as $p_1(t)^k p_0(t)^{(\Phi-k)}$ assumes that the cooccurrence of invader and resident individuals is uncorrelated. This approach also implicitly assumes that $\gamma_{\text{ld}}(t)$ is the same regardless of location. The derivation of equations follows Hiebeler (2000). In the language of cellular automata, each possible arrangement of $\Phi + z_v$ represents a “preimage.” A preimage determines the value of $z_v$ on the next time-step. Using the definition of $p_1(t)$ as a probability of seeing an individual of species 1, $p_1(t+1)$ can be found by summing over all possible preimages that lead to a site being in state 1. If $g$ represents an individual preimage and $G$ is the set of all possible preimages, then $$p_i(t + 1) = \sum_{g \in G} P[g] \times P[g \rightarrow 1],$$ where $P[g]$ refers to the probability of seeing a given preimage, and $P[g \rightarrow 1]$ is the (conditional) probability that a preimage produces a 1. In this approximation, every site is assumed to be statistically independent, which means that the actual physical arrangement of each preimage does not matter. Then preimages can be grouped together on the basis of proportions of species 1 and species 0. Independence also means that the probability of finding a particular preimage is $P[g] = p_0(t)^{\#0(g)} p_1(t)^{\#1(g)}$ where $\#0(g), \#1(g)$ refers to the number of sites in state 0, state 1 in that preimage. For example, when $g$ is all preimages with three sites occupied by species 1 and the remaining six occupied by species 0, then $P[3,6] = p_0(t)^6 p_1(t)^3$. Grouping terms and using $p_0(t) + p_1(t) = 1$ leads to equation (A3), and $\gamma ld(t)$ represents a sum over relevant preimages. The first grouping of terms, $p_1(t)^k p_0(t)^{\Phi(1-k)}$, corresponds to the probability of finding preimage $g = [k, (\Phi + 1 - k)]$. The second grouping of terms gives the probability that species 1 can capture the open site given the particular arrangement of the two species specified by $[k, (\Phi + 1 - k)]$ based on the species-specific recruitment rates at the given time step. The invasion criterion can be derived for the LD approximation in the same way as for the classic lottery model (Chesson and Warner 1981), yielding $$E\left[\log\left(\frac{p_1(t + 1)}{p_1(t)}\right)\right] = E\left[\log\left(1 + \delta\left(\frac{R_1(t)}{R_0(t)} \frac{\Phi}{\Phi + (R_1(t)/R_0(t) - 1)} - 1\right)\right)\right].$$ A damping coefficient can then be defined, as in the text, as the terms that multiply $R_1(t)/R_0(t)$: $$D_{LD} = \frac{\Phi}{\Phi + (R_1(t)/R_0(t) - 1)}.$$ Pair-Wise Approximation The pair-wise approximation considers “block probabilities,” or the likelihood of seeing pairs of sites; specifically, the pairs $p_{00}(t), \ p_{01}(t), \ p_{11}(t)$. Although the blocks are joint probabilities, they can be treated as population densities in an equivalent way to $p_0(t)$ and $p_1(t)$. The formalism is developed more extensively in other sources; see Gutowitz et al. (1987) and Hiebeler (2000), for example, for discussions of block probabilities in the context of cellular automata and Ives et al. (1998) for more general treatment in terms of conditional probabilities. An important property of the block probabilities is that (Gutowitz et al. 1987; Hiebeler 2000): $$p_1(t + 1) = p_{11}(t + 1) + p_{01}(t + 1).$$ The global state of the invader population can be determined from the populations of all blocks with a 1 in the right position. Because $p_0(t) + p_1(t) = 1$, it is only necessary to track $p_1(t)$ directly. The dynamics of $p_{00}(t + 1)$ and $p_{11}(t + 1)$ are determined by the rates at which other blocks transition into these states: $$p_{00}(t + 1) = p_{00}(t)P[00 \rightarrow 01] + p_{01}(t)P[01 \rightarrow 01] + p_{10}(t)P[10 \rightarrow 01] + p_{11}(t)P[11 \rightarrow 01],$$ $$p_{11}(t + 1) = p_{00}(t)P[00 \rightarrow 11] + 2p_{01}(t)P[01 \rightarrow 11] + p_{11}(t)P[11 \rightarrow 11].$$ For $p_{00}(t + 1)$, the first term gives the current population of $p_{00}(t)$, multiplied by the transition rate $P[00 \rightarrow 10]$. In the transition $P[00 \rightarrow 01]$, there are two things that must occur: the right site becomes occupied by the invader, species 1, and the left site stays occupied by the resident, species 0. There is only one way that a site can switch from a 0 to a 1: the resident species 0 must die, allowing the open site to be colonized by the invader. This happens with probability $\delta \gamma_{00}(t)$, where $\gamma_{00}(t)$ is the colonization probability in the pair-wise approximations. There are two ways that a site occupied by the resident can stay in that state: the current resident can survive, or the current resident can die and be replaced by a new member of species 0. This probability is $(1 - \delta) + \delta(1 - \gamma_{ps}(t))$. Putting these together gives \[ P[00 \rightarrow 01] = [(1 - \delta) + \delta(1 - \gamma_{ps}(t))] [\delta \gamma_{ps}(t)]. \] (A10) The remaining block transitions can be specified following the same logic to give \[ p_{01}(t + 1) = \] \[ p_{00}(t)[(1 - \delta) + \delta(1 - \gamma_{ps}(t))] [\delta \gamma_{ps}(t)] + \] \[ p_{01}(t)[(1 - \delta) + \delta \gamma_{ps}(t)][(1 - \delta) + \delta(1 - \gamma_{ps}(t))] + \] \[ p_{10}(t)[(1 - \delta)(1 - \gamma_{ps}(t))][(1 - \delta)\delta \gamma_{ps}(t)] + \] \[ p_{11}(t)[(1 - \delta) + \delta(1 - \gamma_{ps}(t))]^2, \] (A11) \[ p_{11}(t + 1) = \] \[ p_{00}(t)[\delta \gamma_{ps}(t)]^2 + \] \[ 2p_{01}(t)[\delta \gamma_{ps}(t)][(1 - \delta) + \delta \gamma_{ps}(t)] + \] \[ p_{11}(t)[(1 - \delta) + \delta(1 - \gamma_{ps}(t))]^2. \] (A12) Combining these two equations and simplifying leads to \[ p_1(t + 1) = p_{00}(t)[\delta \gamma_{ps}] + p_{01}(t)[(1 - \delta) + \delta \gamma_{ps}] + p_{10}(t)[\delta \gamma_{ps}] + p_{11}(t)[\delta(\gamma_{ps} - 1) + 1]. \] (A13) The colonization probability, $\gamma_{ps}(t)$, incorporates the likelihood of finding a site with a given arrangement of neighbors and the likelihood that any neighborhood configuration leads to a site capture by either species. For each block, one neighbor is already known; at least one neighbor is of species 0. Because pairs of sites are assumed to be correlated, the conditional probability of finding a site among the remaining unknown neighbors that is occupied by species 1 is $P(01\|0) = p_{01}(t)/p_0(t)$. The assumption that only neighbors are correlated leads to statistical independence between blocks. That is, if site $z_a$ and $z_b$ constitute a block, and $z_c$ is a third site in the neighborhood of both sites, then the probability of finding $[z_a, z_b]$ in a given state is independent of the probability of finding either $[z_a, z_c]$ or $[z_b, z_c]$ in any given state. This leads to the general form of $\gamma_{ps}(t)$, \[ \lambda_{ps} = \sum_{k=0}^{s} \binom{\Phi}{k} \left( \frac{p_{01}(t)}{p_0(t)} \right)^k \left( 1 - \frac{p_{01}(t)}{p_0(t)} \right)^{\Phi-k} \frac{R_1(t)k}{R_1(t)k + R_0(t)(\Phi + 1 - k)}. \] (A14) The form of $\gamma_{ps}(t)$ is similar to $\gamma_{ds}(t)$ for the LD approximation. The primary difference is in the terms describing the likelihood of finding a particular neighborhood configuration. It is also worth noting that this form is not directly generalized to neighborhoods that include sites beyond nearest neighbors only. Larger neighborhoods require reformulating the conditional probabilities in $\gamma_{ps}(t)$ (Gutowitz et al. 1987; Hiebeler 2000). The invasion growth rate for this model is approximately \[ E \left[ \log \left( \frac{p_1(t+1)}{p_1(t)} \right) \right] = E \left[ \log \left( 1 + \delta \left( \frac{R_1(t)}{R_0(t)} \frac{(\Phi - 1)}{\Phi + (R_1(t)/R_0(t)) - 2} - 1 \right) \right) \right]. \] (A15) This derivation is based on the observations that, during invasion, the probability of seeing anything other than a 00 pair approaches zero and that, due to the independent likelihood of invasion occurring at any site, conditional probabilities become independent. A damping coefficient $D_{PW}$ can be defined for equation (A15) as well: $$D_{PW} = \frac{(\Phi - 1)}{\Phi + [(R_1(t)/R_0(t)) - 2]}. \quad (A16)$$ Approximation Performance Neither the LD nor pairwise approximations perform well against actual simulations, although both are closer to matching invasion growth rates than the mean-field model. This was true regardless of parameter values. Figures A1.4 and A1.6 demonstrate this result for the specific case when $\delta = 0.9$, $R_i\theta = 1$, $R_0 = 1.01$, $\sigma_0 = \sigma_1 = 1$, $\rho = 0$, $\beta = 10^{-4}$, and $L = 256$. In figure A1.4, the average invasion growth rate was $0.0092 \pm 0.0011$ (mean ± SD) for the limited-dispersal simulation model calculated from 3,000 runs. The invasion growth rate was 0.0676 for the lottery model. The LD and pairwise approximations give invasion rates of 0.0312 and 0.0292, respectively. The low-density growth rates for the nucleation approximations were all closer. Avrami’s law slightly underpredicted at 0.0088, and the difference approximation slightly overpredicted at 0.0108. This pattern is repeated when considering the strength of the storage effect, shown as the bounds on coexistence as a function of fitness differences. Each of the spatial approximations is closer to the limited-dispersal model than the classic lottery model, but the nucleation approximations give the best predictions; Avrami’s law and the difference approximation gave indistinguishable results. Literature Cited Only in Appendix A Gutowitz, Howard A., Jonathan D. Victor, and Bruce W. Knight. 1987. Local structure theory for cellular automata. Physica D: Nonlinear Phenomena 28:18–48. Figure A1: Invasion growth rates (A) and coexistence bounds (B) as a function of the fitness difference $\overline{R}_i / \overline{R}_c$ and interspecific correlation in reproduction rates $\rho$. A, Three example runs of the simulation are plotted against an average log-linear slope fit from 3,000 runs ($\pm$ SD), giving $0.0092 \pm .0011$. The invasion growth rate predicted by the classic lottery model (mean-field approximation) is much larger (0.0676). The local dispersal (0.0312) and pairwise (0.0292) approximations also predict higher invasion rates. The nucleation approximations provide the best match to the actual invasion rate, with Avrami’s law slightly underpredicting (0.0088) and the difference approximation slightly overpredicting (0.0108). B, Each pair of curves gives the upper and lower bounds on $\overline{R}_i / \overline{R}_c$ that permit coexistence. The distance between upper and lower bounds measures the total strength of the storage effect as a function of $\rho$ and other key parameters of the model ($\delta$, $\sigma_i$, $\sigma_c$). Coexistence is reduced in the limited-dispersal model, relative to the classic lottery model. See the text and figure 5 for additional interpretation. The local dispersal and pair-wise approximations to the limited-dispersal model incorrectly give a broader range of $\overline{R}_i / \overline{R}_c$ allowing coexistence. The nucleation approximations (Avrami’s law) provide the most accurate prediction of species coexistence. The results of Avrami’s law and the difference approximation were indistinguishable and are thus presented together. Parameter values are $\delta = 0.9$, $\overline{R}_i = 1$, $\overline{R}_c = 1.01$, $\sigma_i = \sigma_c = 1$, $\rho = 0$, $L = 256$, and $\beta = 10^{-4}$.
THIRD SUNDAY OF EASTER TODAY’S WORSHIP Divine Service III LSB pages 184-202 L = Liturgist * = Stand C = Congregation PRELUDE 8:00 & 10:30 Bells of Peace Joyous Spirit Valerie Stephenson WELCOME AND ANNOUNCEMENTS OPENING HYMN The Lord, My God, Be Praised LSB 794 1. The Lord, my God, be praised, My light, my life from heaven; My maker, who to me Has soul and body given; My Father, who does shield And keep me day by day, And make each moment yield New blessings on my way. 2. The Lord, my God, be praised, My trust, my life from heaven, The Father’s own dear Son, Whose life for me was given; Who for my sin atoned With His most precious blood And gives to me by faith The highest heav’nly good. 3. The Lord, my God, be praised, My God, the ever-living To whom the heav’ny host, Their laud and praise are giving. The Lord, my God, be praised, In whose great name I boast, God Father, God the Son, And God the Holy Ghost. INVOCATION AND GENERAL CONFESSON L: In the name of the Father and of the Son and of the Holy Spirit. C: ♫ Amen. L: In the name of God the Father, source of all hope and light and life: C: Who has loved us with an everlasting love, who is faithful to His word and to His promises. L: In the name of Jesus Christ, C: The living word of God, the one who is the Way, the Truth and the Life. L: In the name of God, the Holy Spirit, C: Who brings us into the way, teaches the truth, and works faith within us. L: Let us confess our sins before God and in the presence of one another. C: Merciful Father, we pray for pardon, peace and forgiveness for our sins, because of the person and work of Jesus Christ, who suffered and died to pay the penalty of physical and eternal death which we deserve. Forgive us and release us from our guilt. Amen. THE ABSOLUTION C: ♫ Amen. THE GLORIA PATRI (Taken from LSB, page 186) C: ♫ Glory be to the Father and to the Son and to the Holy Ghost; as it was in the beginning, is now, and ever shall be, world without end. Amen. THE KYRIE C: ♫ Lord, have mercy upon us. Christ, have mercy upon us. Lord, have mercy upon us. THE GLORIA IN EXCELSIS Taken from Lutheran Service Book pages 187-189 (8:00) ✠ Glory be to God on high: and on earth peace, good will toward men. We praise Thee, we bless Thee, we worship Thee, we glorify Thee, we give thanks to Thee, for Thy great glory. O Lord God, heav’ny King, God the Father Almighty. O Lord, the only-begotten Son, Jesus Christ; O Lord God, Lamb of God, Son of the Father, that takest away the sin of the world, have mercy upon us. Thou that takest away the sin of the world, receive our prayer. Thou that sittest at the right hand of God the Father, have mercy upon us. For Thou only art holy; Thou only art the Lord. Thou only, O Christ, with the Holy Ghost, art most high in the glory of God the Father. Amen. THE COLLECT OF THE DAY L: The Lord be with you. C: ✠ And with thy spirit. L: Let us pray. C: ✠ Amen. THE FIRST READING for the third Sunday of Easter is recorded in Acts 9:1-6, page 897: 1 But Saul, still breathing threats and murder against the disciples of the Lord, went to the high priest 2 and asked him for letters to the synagogues at Damascus, so that if he found any belonging to the Way, men or women, he might bring them bound to Jerusalem. 3 Now as he went on his way, he approached Damascus, and suddenly a light from heaven flashed around him. 4 And falling to the ground he heard a voice saying to him, "Saul, Saul, why are you persecuting me?" 5 And he said, "Who are you, Lord?" And he said, "I am Jesus, whom you are persecuting. "But rise and enter the city, and you will be told what you are to do." L: This is the Word of the Lord. C: ✠ Thanks be to God. THE SECOND READING is from Revelation 5:8-14, page 1031: 8 And when he had taken the scroll, the four living creatures and the twenty-four elders fell down before the Lamb, each holding a harp, and golden bowls full of incense, which are the prayers of the saints. 9 And they sang a new song, saying, "Worthy are you to take the scroll and to open its seals, for you were slain, and by your blood you ransomed people for God from every tribe and language and people and nation, 10 and you have made them a kingdom and priests to our God, and they shall reign on the earth." 11 Then I looked, and I heard around the throne and the living creatures and the elders the voice of many angels, numbering myriads of myriads and thousands of thousands, 12 saying with a loud voice, "Worthy is the Lamb who was slain, to receive power and wealth and wisdom and might and honor and glory and blessing!" 13 And I heard every creature in heaven and on earth and under the earth and in the sea, and all that is in them, saying, "To him who sits on the throne and to the Lamb be blessing and honor and glory and might forever and ever!" 14 And the four living creatures said, "Amen!" and the elders fell down and worshiped. L: This is the Word of the Lord. ALLELUIA AND VERSE C: ✠ Alleluia. Alleluia. Alleluia. THE HOLY GOSPEL - according to St. John, the twenty-first chapter, verses 1-14, page 907: C: ✠ Glory be to Thee, O Lord. 1 After this Jesus revealed himself again to the disciples by the Sea of Tiberias, and he revealed himself in this way. 2 Simon Peter, Thomas (called the Twin), Nathanael of Cana in Galilee, the sons of Zebedee, and two others of his disciples were together. 3 Simon Peter said to them, "I am going fishing." They said to him, "We will go with you." They went out and got into the boat, but that night they caught nothing. 4 Just as day was breaking, Jesus stood on the shore; yet the disciples did not know that it was Jesus. 5 Jesus said to them, "Children, do you have any fish?" They answered him, "No." 6 He said to them, "Cast the net on the right side of the boat, and you will find some." So they cast it, and now they were not able to haul it in, because of the quantity of fish. 7 That disciple whom Jesus loved therefore said to Peter, "It is the Lord!" When Simon Peter heard that it was the Lord, he put on his outer garment, for he was stripped for work, and threw himself into the sea. 8 The other disciples came in the boat, dragging the net full of fish, for they were not far from the land, but about a hundred yards off. 9 When they got out on land, they saw a charcoal fire in place, with fish laid out on it, and bread. 10 Jesus said to them, "Bring some of the fish that you have just caught." 11 So Simon Peter went aboard and hauled the net ashore, full of large fish, 153 of them. And although there were so many, the net was not torn. 12 Jesus said to them, "Come and have breakfast." Now none of the disciples dared ask him, "Who are you?" They knew it was the Lord. 13 Jesus came and took the bread and gave it to them, and so with the fish. 14 This was now the third time that Jesus was revealed to the disciples after he was raised from the dead. L: This is the Gospel of the Lord. C: Praise be to Thee, O Christ. THE APOSTLES' CREED I believe in God, the Father Almighty, maker of heaven and earth. And in Jesus Christ, His only Son, our Lord, who was conceived by the Holy Spirit, born of the virgin Mary, suffered under Pontius Pilate, was crucified, died and was buried. He descended into hell. The third day He rose again from the dead. He ascended into heaven and sits at the right hand of God the Father Almighty. From thence He will come to judge the living and the dead. I believe in the Holy Spirit, the holy Christian Church, the communion of saints, the forgiveness of sins, the resurrection of the body, and the life everlasting. Amen. CHILDREN'S SONG (Sung as the children come forward) (8:00 & 10:30) 1. Jesus loves me this I know, For the Bible tells me so; Little ones to Him belong; They are weak but He is strong. (Chorus) Chorus: Yes, Jesus loves me, Yes, Jesus loves me, Yes, Jesus loves me, The Bible tells me so. 2. Jesus loves me, He who died, Heaven's gates to open wide. He will wash away my sin, Let His little child come in. (Chorus) CHILDREN'S MESSAGE (8:00 & 10:30) Christina Stackle HYMN OF THE DAY In Thee Is Gladness LSB 818 1. In Thee is gladness, Amid all sadness, Jesus, sunshine of my heart. By Thee are given, The gifts of heaven, Thou the true Redeemer art. Our souls Thou wakest, Our bonds Thou breakest; Who trusts Thee surely, Has built securely; He stands forever: Alleluia! Our hearts are pining, To see Thy shining. Dying or living, To Thee are cleaving; Naught can us sever; Alleluia! 2. Since He is ours,—We fear no powers, Not of earth nor sin nor death. He sees and blesses In worst distresses; He can change them with a breath. Wherefore the story Tell of His glory, With hearts and voices; All heav'n rejoices In Him forever: Alleluia! We shout for gladness, Triumph o'er sadness, Love Him and praise Him, And still shall raise Him, Glad hymns forever: Alleluia! THE SERMON “Conversion and Forgiveness” (Acts 9:1-6) THE OFFERING - During the offering all worshipers are asked to fill out the blue or ivory cards found in the pew rack. Please pass cards down to center aisle. They will be collected immediately after the offering. VOLUNTARY 8:00 Bells of Peace Peace Joshua Kramer 10:30 Chimes of Peace Hymn Duet Arranged: Burnell Hackman THE OFFERTORY LSB page 192 ℟ Create in me a clean heart, O God, and renew a right spirit within me. Cast me not away from Thy presence, and take not Thy Holy Spirit from me. Restore unto me the joy of Thy salvation, and uphold me with Thy free spirit. Amen. PRAYERS OF THE CHURCH LORD’S PRAYER (Found on the inside back cover of the hymnal.) THE BENEDICTION C: ℟ Amen, amen, amen. THE CLOSING HYMN Abide, O Dearest Jesus LSB 919 1. Abide, O dearest Jesus, Among us with your grace That Satan may not harm us Nor we to sin give place. 2. Abide, O dear Redeemer, Among us with Your Word And thus now and hereafter True peace and joy afford. 3. Abide with heav’nly brightness Among us, precious Light; Your truth direct and keep us From error’s gloomy night. SILENT PRAYER POSTLUDE 8:00 & 10:30 Bells of Peace Danza Tielman Susato SERVING IN OUR WORSHIP TODAY: Preacher: Pastor Jon C. Furgeson Liturgist: Pastor Dennis A. Kastens Organist: 5:00 & 9:20: Alice Kastens; 8:00 & 10:30: Burnell Hackman Choir Dir.: Burnell Hackman, DMM KFUO 850 AM Broadcaster/Engineer: Rick Goodman/Jeanette Mattingly NEWS AND NOTES: THIRD SUNDAY OF EASTER MAY 5, 2019 TODAY’S EDUCATION HOUR: 9:15 - 10:15 a.m. Adult Bible Class (Lower Commons): The Gospel of Luke Women’s Bible Class (Office Conf. Rm.): led by Janice McCreary Youth Bible Class: led by Christina Stackle, DCE 8th Gr: Rick Goodman & Mike Renner 7th Gr: Greg Wirtel & Louise Oster ALTAR FLOWERS are from Ladies Guild. NEW MEMBER CLASSES: Adult Inquiry series on Sundays, 6:00-8:00 p.m., starts today May 5, with completion for membership reception Sunday, June 2. Sign up is posted on the bulletin board in the Commons. ALASKA MISSION BAKE SALE Don’t forget to stop by the Welcome Center in the Commons and order your gooey butter cake or pies for Mother’s Day. We have apple and cherry pies as well as the gooey butter cake, $8 each. All these items are freshly made here at Peace. Orders can be picked up before and after services Mother’s Day weekend (Sat., May 11 and Sun., May 12). Thank you for supporting this mission work in rural Alaska! RETIREMENT CELEBRATION for Teacher Debra Prange who completes 32 years as classroom instructor, reading and math specialist, pianist, and choir director at Green Park Lutheran School at the end of the current academic year, will be Sunday, May 19, with “Open House” at Peace 9:00 - 9:15 a.m. and 11:30 - 1:30 p.m. in the gymnasium. Light refreshments will be served as we thank Mrs. Prange for her dedicated service to God’s children at GPLS and Peace. GROCERY BAGS are being distributed today for the annual Feed My People Pantry drive. Please return food items Mothers’ Day weekend to the Commons. INTERESTED IN BEING A PRAYER PARTNER? Are you interested in being a prayer partner for the High School Youth and adults attending the LCMS National Youth Gathering (Real. Present. God.) in Minneapolis this July? Please talk with Christina about how you and your family can be a part of this exciting week. YOUTH NEWS: The 9th Grade Welcome event is today from 3:00-6:00 p.m. High School Seniors Recognition will be held on Sunday, May 19 at 9:15 a.m. in the Youth Room. Please RSVP to Christina if you are able to attend. Jr. Youth will be helping to lead the worship service at Bethesda on May 19 from 10:30-11:30. Please sign up on the Youth Ministry Bulletin Board if you are able to help. ATHENS VACATION BIBLE SCHOOL is right around the corner, June 24-28. Please sign your children and grandchildren (ages 3 - 6th grade) up for VBS by June 1 to assure them a theme t-shirt. Registration can be found on our church website: peacelutheranstl.org. GRADUATING? Names of those graduating from college, high school, elementary and other schools needed for future acknowledgement. SURVIVING SPOUSES BREAKFAST at 9:00 a.m. this Thurs. at Denny’s. Contact Joyce Schutte for more info. MIDWEEK BIBLE STUDY RESUMED this past Wednesday, May 1, at 7:00 p.m. in the Lower Commons with Pastor Furgeson. You are welcome! LUTHERAN SOUTH SUMMER CAMPS registration is now open for 2019 camps. Visit lhssstl.org for more information and to sign up. REV. RICHARD LABORE, long time religion instructor at Lutheran High School South has announced his retirement from the associate pastoral ministry at the Lutheran Church of the Resurrection. He will be saluted at a special celebration at Resurrection, this afternoon, Sunday, May 5, 3:00 p.m. to which the public is invited. ATTENTION VETERANS If you are a new member or have not previously given us your service information, please provide it to us. Put your service information (branch of service, dates, rank) in Edie Turner’s mailbox (#836) or turn it in to the church office. Thank you for helping to keep our records current. ADULT VOLLEYBALL has concluded for the Spring. We look forward to having you join us again in the Fall. CALENDAR FOR WEEK OF MAY 5, 2019: Today: 11:30 Youth Board Meeting 11:30 Life Ministry Team 3:00 9th Grade Youth Welcoming 6:00 Adult Inquiry Class Mon: 5:00 Chimes of Peace Rehearsal 6:30 Council Tue: 9:00 Quilters 9:00 Plastic Mat Making Wed: 11:15 Preschool Chapel 7:00 Bells of Peace Rehearsal 7:00 Wednesday Bible Study Thu: 9:00 Mary/Martha Bible Study 6:00 St. Trinity Cemetery Board 6:30 ESL Classes 6:30 College Girl’s Bible Study 7:00 Chancel Choir Rehearsal Sat: 5:00 Worship Service w/Comm Sun: 6:00 Adult Inquiry
Neural networks can be FLOP-efficient integrators of 1D oscillatory integrands Anshuman Sinha School of Computational Science & Engineering, Georgia Institute of Technology, Atlanta, GA email@example.com Spencer H. Bryngelson School of Computational Science & Engineering, Georgia Institute of Technology, Atlanta, GA Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA Guggenheim School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA firstname.lastname@example.org Reviewed on OpenReview: https://openreview.net/forum?id=5psgQEHn6t Abstract We demonstrate that neural networks can be FLOP-efficient integrators of one-dimensional oscillatory integrands. We train a feed-forward neural network to compute integrals of highly oscillatory 1D functions. The training set is a parametric combination of functions with varying characters and oscillatory behavior degrees. Numerical examples show that these networks are FLOP-efficient for sufficiently oscillatory integrands with an average FLOP gain of $10^4$ FLOPs. The network calculates oscillatory integrals better than traditional quadrature methods under the same computational budget or number of floating point operations. We find that feed-forward networks of 5 hidden layers are satisfactory for a relative accuracy of $10^{-3}$. The computational burden of inference of the neural network is relatively small, even compared to inner-product pattern quadrature rules. We postulate that our result follows from learning latent patterns in the oscillatory integrands that are otherwise opaque to traditional numerical integrators. 1 Introduction Numerical integration of highly-oscillatory functions is required for problems in fluid dynamics, nonlinear optics, Bose–Einstein condensates, celestial mechanics, computer tomography, plasma transport, and more (Connor & Curtis, 1982; Islerles, 2005). Classical numerical integration schemes are based on quadrature rules, like those of Newton–Cotes type (e.g., trapezoidal or Simpson’s rule), Romberg integration, or Gauss quadrature (Davis & Rabinowitz, 2007; Milne, 2015; Hildebrand, 1987). These are unsuited for highly oscillatory integrands, requiring many quadrature points before reaching their asymptotic convergence rates. This work uses feed-forward, fully connected neural networks as approximate integrators for highly oscillatory integrands. Focus is paid to the floating point operation (FLOP)-efficiency of different integrators: Can a feed-forward neural network outperform classical integration schemes for a given FLOP budget? 2 Background Numerical integral methods crafted for highly oscillatory integrands have been developed. These are, for example, based on the stationary phase approximations (Filon, 1930; Levin & Sidi, 1981; Levin, 1996; Islerles & Norssett, 2005; Evans & Chung, 2007; Hascelik, 2009). Each method is powerful when used appropriately but operates under relatively strict conditions, including the type of oscillatory features (e.g., sine and cosine). These methods can be generalized when more sophisticated algorithms are added to the integration technique to identify a suitable basis for integration. Though the methods for this identification are not well documented in the literature or any open-source software, their use or evaluation of the additional computational cost associated with them is challenging to align with our objective. More general techniques for dealing with highly oscillatory integrands are desirable. Neural networks are one such possibility for generalization. Indeed, neural networks have been used to approximate integrals before. Previous works used single hidden layer neural networks for PID controller applications (Zhe-Zhao et al., 2006), and dual neural networks with application to material modeling (Li et al., 2019). These neural networks were particularly useful in many-query settings but do not appear to generalize beyond their specific applications. Some works have taken a tailored approach to neural network design for integration. Ying Xu & Jun Li (2007) used oscillatory basis functions for interpolation and integration, though the cosine activation function prevents approximation to broader function classes (Wu, 2009). Lloyd et al. (2020) trained single-layer networks for multidimensional integrals, focusing on parameterized, many-query problems, and showed promising results. The main limitation of Lloyd et al. (2020) is the restriction to high-dimensional many-query problems for FLOP-efficiency. This work investigates the FLOP-efficiency, or integration accuracy per required floating point operation, for integrating 1D functions via a feed-forward fully connected neural network. We aim to provide a computationally efficient solution for problems that require repeated integration of an integrand with varying parameters. ### 2.1 Problem definition The formulation is to compute the integral $I$ of any function $f(x)$ in a bounded domain $[s_1, s_2]$ which is expressed as follows, $$I = \int_{s_1}^{s_2} f(x) dx.$$ (1) The method of fig. 1, where the inputs of the network are shown as $f(x_i)$ for a fixed set of $x_i \in [s_1, s_2]$, and the outputs of the network are the integral values $I$ of the input function. We train the neural network model with parametrically varying samples. These neural network integrators are compared with the classical numerical integration methods like Newton–Cotes quadrature rule, as shown in eq. (2). We do not consider more complex integration schemes like Gauss quadrature as they do not reach asymptotic convergence behaviors for realistic numbers of quadrature points and typically become unstable before this. The input domain $[s_1, s_2]$ is divided into $n_q$ quadrature points. The integral is approximated as a weighted sum of function values at those quadrature points $(x_i)$. \[ \int_{s_1}^{s_2} f(x) \, dx \approx \sum_{i=1}^{n_q} w_i f(x_i) \] (2) Common Newton–Cotes formulations include the second-order accurate trapezoidal method \[ \int_{s_1}^{s_2} f(x) \, dx \approx \sum_{k=1}^{n_q} \frac{f(x_k) + f(x_{k+1})}{2} \Delta x, \] (3) where $x_k$ are uniformly spaced points in the domain $x \in [s_1, s_2]$ and $\Delta x = (s_2 - s_1)/n_q$, the first-order accurate is the mid-point method \[ \int_{s_1}^{s_2} f(x) \, dx \approx \sum_{k=1}^{n_q} f\left(s_1 + \left(k - \frac{1}{2}\right) \Delta x\right) \Delta x. \] (4) and the third-order accurate Simpson’s method \[ \int_{s_1}^{s_2} f(x) \, dx \approx \sum_{k=1}^{n_q/2} \frac{(f(x_{2k-2}) + 4f(x_{2k-1}) + f(x_{2k}))}{3} \Delta x, \] (5) where, again, $f(\cdot)$ is the integrand of interest. ### 2.2 Neural network method The network’s learned weights reduce computational costs. With an optimized architecture, we seek accurate integration using few input (quadrature) points. Improved accuracy for a given number of quadrature points is feasible as the neural network has trainable weights for approximation, but classical techniques rely on fixed interpolating functions. For such methods, the more oscillatory a function, the more quadrature points are required to achieve integration of the same accuracy. In our notation, the values $y_i$ of an integrand $f(x)$ at abscissa $x_i$, $(y_i = f(x_i) \ i \in 1, \ldots, n)$ are shown in fig. 1. The parameters \[ a_{ij} = \sigma\left(\sum_{i=1}^{n} w_{ij} x_{i,j} + b_i\right). \] (6) form each layer of fig. 1, $x_{ij}$ represent the information from previous layers (i.e, $a_{(i-1)j}$) and $b_i$ is the bias term in each layer. In the final layer of fig. 1, the sum of each input to a final node ($\Sigma$) and the output $\tilde{I}$ to train the network via back-propagation. ReLU is used as the activation function. The number of hidden layers and the number of neurons are optimized during hyperparameter optimization. Section 3.4 shows the results for the hyperparameter optimization for the model. ### 3 Experiments We evaluate the proposed neural network method with several example cases. We present the employed dataset, metric of evaluation, and other details on the hyperparameter study, then the results. #### 3.1 Experimental dataset ##### 3.1.1 Training data The model inputs the function values and the baseline integral value as output in the current work. The training is done by varying the function’s parameters while keeping its nature the same. While training for one network configuration of the model, we keep the number of quadrature points fixed. We trained separate models for numbers of inputs ranging from $2^9$ to $2^{13}$ and evaluated their performance on separate test data. These inputs (with varying quadrature points) were tested on the specific network configuration, yielding results with varying validation accuracy. For a fixed number of quadrature points, the input values of the model are the function’s values $f(x_i)$ at those quadrature points $x_i$. ### 3.1.2 Testing data The model is tested by calculating the integral of a function that is the same kind as the training functions but parametrically different and unseen. The integral is calculated within the same input space, i.e., within the same limits of the input domain $[s_1, s_2]$. The training data are divided into disjoint subsets, with an 80–10–10 split for train, test, and validation. ### 3.2 Evaluation We evaluate the performance of the neural network model using the normalized mean squared error (Normalized MSE) metrics on the test set. The error in the model’s result ($\hat{I}_k$) is computed against a representative exact solution to the integral ($I_k$), which is evaluated using the trapezoidal method with $2^{13}$ quadrature points. This strategy is sufficiently accurate for our purposes, which is checked against more and fewer such baseline quadrature points. We use the surrogate truth integrand $I_k$ for the $k$’th sample in the test set as $$I_k = \sum_{j=1}^{n_q=2^{13}} \left( \frac{f(x_j) + f(x_{j+1})}{2} \right) \Delta x,$$ where $n_q = 2^{13}$ for $\Delta x = (s_2 - s_1)/2^{13}$ while calculating the integral of $f(x)$ (the integrand) between $[s_1, s_2]$ using the trapezoidal rule in eq. (3). The normalized mean-square error (MSE) for the test set is calculated using the $m$ samples of functions is $$\text{Normalized MSE} = \frac{1}{m} \sum_{k=1}^{m} \frac{(I_k - \hat{I}_k)^2}{I_k^2}.$$ These $m$ samples are generated by parametric variations of the same function, as described in table 1. We evaluate the performance of the neural network model by comparing it to standard numerical integration methods, such as the trapezoidal and midpoint methods. ### 3.3 Hyperparameter optimization The neural network’s architecture was optimized according to a test-set error value below $10^{-3}$ and minimizing a fixed number of FLOPs. The validation data are used for hyperparameter optimization. Table 1 shows the optimization results. Table 1: Result of the neural network model on various oscillatory functions. The normalized FLOP gain for a given accuracy is $\alpha$, which serves as a measure of performance (higher is better) and is computed with respect to the classical quadrature-based model as shown in eq. (9). All functions are 1D in $x$. \| Function Type \| Equation \| Parameter space \| $\alpha$ \| Result \| \|------------------------\|-----------------------------------------------\|----------------------------------\|----------\|--------------\| \| Bessel($k, \nu$) \| $\cos(kx)J_0(\nu, x)$ \| $\nu \in [125, 175], k \in [75, 125]$ \| 6.01 \| Figure 4 (a) \| \| Evan–Webster-1($k_1, k_2$) \| $\cos(k_1x^3)\sin(k_2x)$ \| $k_1 \in [5, 15], k_2 \in [25, 75]$ \| 17.72 \| Figure 4 (b) \| \| Rayleigh–Plesset($\rho$) \| See appendix A.3 \| $\rho \in [500, 1000)$ \| 23.46 \| Figure 4 (c) \| \| Evan–Webster-2($k$) \| $\exp(x)\sin(k\cosh(x))$ \| $k \in [25, 75]$ \| 19.60 \| Figure 4 (d) \| \| Sine($k$) \| $\sin(kx)$ \| $k \in [5, 15]$ \| 0.91 \| Figure 6 (a) \| \| Exponential($k$) \| $\exp(kx)$ \| $k \in [1, 5]$ \| 0.60 \| Figure 6 (b) \| ![Bessel-type](image1.png) ![Evan–Webster-2](image2.png) Figure 3: Example canonical oscillatory test functions as labeled. The architecture was optimized heuristically with an iterative increment of network hidden layers and neurons in each layer. Each configuration was executed using the model until the training normalized MSE stagnated. An example of this convergence is shown in fig. 2, along with the testing and training performance for two separate network configurations. The performance of each network configuration was evaluated on a validation dataset, which was separate from the training data. The iterative tests were done with increasing samples as the parameters of the network increased (Bishop, 1995). The upper limit complexity of the network is obtained via the performance of the classical integrators. With the increasing complexity of the network, we observe diminishing returns against the FLOP burden. A deeper neural network can still achieve smaller integration errors. We aim to achieve the highest accuracy for a given computational expense while avoiding overfitting. ### 3.4 Results The results of this study present the current model’s applicability for efficiently computing and predicting the integrals of oscillating functions. The model is tested on oscillatory and non-oscillatory functions. Table 1 compares results for various functions with different oscillatory features. The degree of oscillatoriness is defined by the function’s parameters, which are denoted in parentheses in the “Function Type” column of table 1. A larger coefficient corresponds to a more oscillatory function. The parameter space column is the range in which the function’s parameters vary to create the training data. The gain in the number of FLOPs, shown in eq. (9), defines the computational advantage obtained for the number of required floating point operations while implementing the current method over the other numerical-based methods of computing the integral. The term FLOPS$_{\text{NN}}$ represents the number of FLOPs required by the neural network model, and FLOPS$_{\text{QM}}$ represents the number of FLOPs required by a classical integration quadrature based model. The number of FLOPs for the neural network method is calculated by enumerating the floating-point operations involved in computing the output of each neuron in every layer. This computation follows the formula given in eq. (6). For an $H$ layer fully-connected feed-forward neural network with each layer having $N$ neurons, each activation $a_{ij}$ comprises $N$ multiplications and addition operations. Given $N$ operations per layer and $H$ layers, the total FLOP count is $(4N + 2)N^2H(1 + n_q)$. The number of FLOPs associated with the trapezoidal, mid-point, and Simpson’s methods are $2n_q + 1$, $3n_q + 1$ and $4.5n_q + 2$, following eqs. (2), (4) and (5). To study the influence of the oscillatory nature of the input function on the results, we performed integral calculations for the function while gradually increasing its oscillatory behavior. We tested the model on sinusoids with varying frequencies (increasingly oscillatory). The measure of neural network integration efficiency is the normalized ratio of FLOP gain for a neural network model over traditional integration for the same MSE error, $\alpha$: $$\alpha = \frac{\|FLOP_{SNN} - FLOP_{SQM}\|}{FLOP_{SNN}}.$$ A larger $\alpha$ corresponds to a higher normalized gain by the neural network model. fig. 3 shows the results of the model’s performance by providing examples of individual functions. fig. 4 shows the test set’s normalized mean square error (NMSE) in integral computation from the model as a function of the number of FLOPs required to compute the individual integral. Figure 4 shows the results of normalized MSE loss values for different numbers of quadrature points as input training data points for the model. Table 2 shows the results for the two best-performing network configurations (neurons × hidden layers) achieved through parametric optimization. The number of FLOPs required for computing the integral increases as the number of quadrature points increases. Increasing quadrature points also increases the accuracy of the integral computation for both methods. For a normalized MSE of $10^{-3}$, the neural network strategy computes the integral using fewer FLOPs than traditional quadrature methods, making it FLOP-efficient. We observe the opposite result for the less oscillatory functions of fig. 5. Figure 6 shows that the neural network model requires more FLOPs to compute the same integral for a given normalized MSE. This result Figure 5: Example test functions tested as labeled. Figure 6: Results of the NN model with traditional numerical integration methods for non-oscillatory (a) sine and (b) exponential functions shown in fig. 5. Table 2: Model hyperparameter selection space. The optimum values are chosen based on the normalized MSE. The sample number, learning rate, and test–train split are fixed. \| Hyperparameter \| Search space \| Optimum value \| \|---------------------------------\|--------------------\|---------------\| \| Number of hidden layer \| \{1, 2, 3, 4, 5\} \| 3 Hidden layers \| \| Neurons in each hidden layer \| \{1, 2, 3, 4, 5, 6, 7\} \| 5 Neurons \| \| Number of samples \| \{10^2, 10^3, 10^4\} \| 10^4 Samples \| \| Learning rate \| \{10^{-5}, 10^{-4}, 10^{-3}\} \| 10^{-4} \| \| Test-train split \| \{0.1, 0.15, 0.2\} \| 0.15 \| is expected, as these integrands can be computed accurately with only few quadrature points using low-cost Newton–Cotes methods. Figure 7 shows that the performance gain increases rapidly with the increasing oscillation of the integrand. This indicates that the numerical integral methods require more quadrature points to evaluate the same integral. After some level of oscillation has been reached, the performance gain curve starts to plateau. Even if one adds more oscillations to the domain (by increasing the oscillatory parameters), the integral value within the sub-domain is the same. Thus, no further gain in performance is expected. For less oscillatory integrands, the values of $\alpha$ decay to unity. The neural network integral then requires more FLOPs than a classical integrator. The summary of the comparative study on various functions is presented in table 1. The number of FLOPs for this comparison was calculated for a fixed Normalized MSE value of $10^{-3}$ for the loss function. This value was chosen to remain within the application limits for downstream integration use. Overall, section 3.4 shows a general trend of decreasing normalized MSE in the integral as the complexity of the network increases. The neural network model predicts the integral result with the same accuracy but exhibits nearly two orders of magnitude better efficiency. 4 Discussion and conclusion An approach for computing the integrals of 1D functions of highly oscillatory and non-oscillatory behaviors. We used a feed-forward neural network to estimate the integrals and evaluated their accuracy for a fixed FLOP budget. In a comparative study, several cases of oscillatory functions were examined. The approach of calculating integrals with a neural network model outperforms existing numerical methods for a fixed FLOP budget. The neural network model was an increasingly efficient integrator for functions with more oscillatory behavior. The proposed method does not extrapolate to qualitatively different integrands than it was trained on, mostly owing to the restriction of FLOP efficiency. Thus, our results are most applicable to cases where parametric variations of an integrand must be evaluated in a many-query setting, and one can forecast the integrand characteristics. Acknowledgement The authors appreciate discussion with Dr. Ethan Pickering at an early stage of this work. SHB acknowledges support of the US Office of Naval Research under grant no. N00014-22-12519 (PM Dr. Julie Young). References C. M. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, 1995. S. H. Bryngelson and T. Colonius. Simulation of humpback whale bubble-net feeding models. Journal of the Acoustical Society of America, 147(2):1126–1135, 2020. S. H. Bryngelson, R. O. Fox, and T. Colonius. Conditional moment methods for polydisperse cavitating flows. Journal of Computational Physics, 477:111917, 2023. J. N. L. Connor and P. R. Curtis. A method for the numerical evaluation of the oscillatory integrals associated with the cuspidal catastrophes: Application to Pearcey’s integral and its derivatives. Journal of Physics A: Mathematical and General, 15(4):1179, 1982. P. J. Davis and P. Rabinowitz. Methods of Numerical Integration. Courier Corporation, 2007. G. A. Evans and K. C. Chung. Evaluating infinite range oscillatory integrals using generalised quadrature methods. Applied Numerical Mathematics, 57(1):73–79, 2007. L. Filon. III. On a quadrature formula for trigonometric integrals. Proceedings of the Royal Society of Edinburgh, 49:38–47, 1930. I. Hascelik, A. On numerical computation of integrals with integrands of the form $f(x) \sin(w/xr)$ on $[0, 1]$. Journal of Computational and Applied Mathematics, 223(1):399–408, 2009. F. B. Hildebrand. Introduction to Numerical Analysis. Courier Corporation, 1987. A. Iserles. On the numerical quadrature of highly-oscillating integrals II: Irregular oscillators. IMA Journal of Numerical Analysis, 25(1):25–44, 2005. A. Iserles and S. P. Norsett. Efficient quadrature of highly oscillatory integrals using derivatives. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 461(2057):1383–1399, 2005. D. Levin. Fast integration of rapidly oscillatory functions. Journal of Computational and Applied Mathematics, 67(1):95–101, 1996. D. Levin and A. Sidi. Two new classes of nonlinear transformations for accelerating the convergence of infinite integrals and series. Applied Mathematics and Computation, 9(3):175–215, 1981. H. Li, Y. Li, and S. Li. Dual neural network method for solving multiple definite integrals. Neural Computation, 31(1):208–232, 2019. S. Lloyd, R. A. Irani, and M. Ahmadi. Using neural networks for fast numerical integration and optimization. IEEE Access, 8:84519–84531, 2020. W. E. Milne. Numerical Calculus. Princeton University Press, 2015. Milton S Plesset and Andrea Prosperetti. Bubble dynamics and cavitation. Annual review of fluid mechanics, 9(1):145–185, 1977. H. Wu. Global stability analysis of a general class of discontinuous neural networks with linear growth activation functions. Information Sciences, 179(19):3432–3441, 2009. L. Ying Xu and L. Jun Li. The new numerical integration algorithm based on neural network. In IEEE Third International Conference on Natural Computation (ICNC 2007), volume 1, pp. 325–328, 2007. Z. Zhe-Zhao, W. Yao-Nan, and W. Hui. Numerical integration based on a neural network algorithm. Computing in Science & Engineering, 8(4):42–48, 2006. A Appendix A.1 Neural network memory costs For a fully connected neural network with $L$ layers of $N$ neurons, each with a 4-byte floats parameterization, the memory footprint in bytes, ignoring input layer biases, is $$4[N^2(L - 1) + N(L - 2)], \tag{10}$$ for $N^2(L - 1)$ weights, $N(L - 2)$ biases (each neuron in the hidden and output layers, excluding the input layer), and 4 bytes per float. Thus, the memory footprint for a 5-layer, 3-neuron network is 180 bytes, which is negligible in almost any practical computing environment. A.2 Training dataset generation The general data generation process for training is $$y(x) = f(k, x), \tag{11}$$ where $y(x)$ is the function generated for a set of $x$ values with parameter $k$. The total number of $x$ for a simulation defines the total number of samples. Here, $k$ is the randomly selected parameter associated with the oscillatoriness of the integrated for each $x$, $k \in [k_1, k_2]$. \| Param. \| Value \| Description \| Units \| \|--------\|-------------\|---------------------------\|-------\| \| $\Delta p$ \| $-7670$ \| Ambient pressure difference \| Pa \| \| $p(t)$ \| $1.3 \times 10^6 \cos(53000\pi t)$ \| Driving pressure \| Pa \| \| $\sigma$ \| $0.0725$ \| Surface tension \| N/m \| \| $\mu$ \| $8.9 \times 10^{-4}$ \| Dynamic viscosity \| Pa s \| \| $k$ \| $1.33$ \| Polytropic exponent \| — \| Table 3: Parameterization of the Rayleigh–Plesset equation. A.3 Rayleigh–Plesset dynamics The Rayleigh–Plesset equation represents the oscillatory radial dynamics of gas bubbles suspended in a liquid. The equation is a second-order ODE and its solution is highly oscillatory and dominated by nonlinear behavior in many regimes. Specifically, it can be expressed as $$\rho \left( R \frac{d^2 R}{dt^2} + \frac{3}{2} \left( \frac{dR}{dt} \right)^2 \right) = \Delta p - p(t) - \frac{2\sigma}{R} - \frac{4\mu}{R} \frac{dR}{dt} + \left( \frac{2\sigma}{R_0} - \Delta p \right) \left( \frac{R_0}{R} \right)^{3k} \tag{12}$$ $$R(0) = R_0, \quad \frac{dR}{dt}(0) = 0,$$ where the bubble radius $R$ is the dependent variable and time $t \in [0, T]$ is the independent variable. Equation (12) is buttressed by initial conditions for bubble radius $R_0 = 2.6 \times 10^{-6}$ (all following quantities in SI units) and zero initial radial velocity of the spherical bubble. The coefficients in eq. (12) are expressed in table 3. Elaborating on their meaning: the ambient pressure difference represents the initial discontinuity between the pressure inside the bubble and the pressure in the suspending liquid, the driving pressure is how the pressure of the liquid surrounding the bubble changes in time, the surface tension coefficient is a standard quantity representing the proclivity for one fluid to want to adhere to another, the dynamic viscosity is the resistance of the suspending liquid to changes in motion due to a surrounding change in velocity, and the polytropic coefficient serves as a mediating rate for which the gas inside the bubble is able to expand or contract. We generate an arbitrarily varying set of oscillatory data by varying the ambient density of the suspending liquid as $\rho \in [500, 1000]$. Equation (12) is integrated over the time interval, which is accomplished using a traditional numerical differential equation solver with a suitably small time step that captures the short time scales occurring at solutions cusps (corresponding to bubble size decay to growth). The solution is oscillatory in the regimes tested, but integrals of it are required in the context of underwater hydrodynamics broadly (Bryngelson et al., 2023; Bryngelson & Colomus, 2020; Plesset & Prosperetti, 1977).
Leader’s Guides Table of Contents October 2006 - "The Pastor as Preacher" ................................................................. 2 November 2006 - "The Pastor as Missional Leader" .................................................. 7 December 2006 - "Worship: Stewards of the Mysteries" ........................................... 17 January 2007 - "Pastoral Integrity" ............................................................................. 28 February 2007 - "Confessing Pastors" ....................................................................... 34 March 2007 - "The Pastor’s Prayer and Devotional Life" .......................................... 45 April 2007 - "Balancing Family & Ministry" .............................................................. 53 May 2007 - "Equipping the Saints" ............................................................................ 59 The Fan into Flame Series Introduce the series and the series theme, referencing the study guide. Future sessions will focus on other aspects of the pastoral task, including missional leadership, worship, balance of family and ministry, integrity, confessing pastors, equipping the saints, and the pastor’s prayer and devotional life. Take a moment, too, to introduce the process – 30-45 minutes of guided total group Bible study followed by 30-45 minutes of conversation in small groups of 3-4. In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to remain faithful as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church-Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness. A Case in Point This “Case in Point” launching segment asks pastors to consider their own challenges or problems in preaching. It “primes the pump” for the study that follows. Read the little case study aloud along with the two questions that follow. Responses to the two discussion questions may be shared in the total group or already in small groups of 3-4. Try to keep this first conversation to 10 minutes. Pastor Bob takes his preaching seriously. He still translates his New Testament texts from the Greek. He works hard to get at the center of a text and make a connection with his listeners. He gets an array of positive comments at the door regarding his preaching. Like most preachers, he thinks he’s pretty good at it. Not long ago at a pastoral conference, pastors were asked to list their problems in preaching – aspects of preaching in which they thought they could improve. At first, Bob balked, thinking, “Hey, my reviews are good.” Then, as he thought about it, his list of problem areas in his preaching began to grow. Here’s what he wrote down at the conference: 1. I need more time for preparation. 2. I seem to use the same language for the gospel in every sermon. 3. My sermon structure is the same almost every time: law, then gospel, then sanctification. 4. I think I’m too tied to my manuscript when I preach. 5. I don’t preach enough on Old Testament texts. 6. My illustrations don’t seem to get to the heart of the listener. I’m more cognitive in my preaching than affective, preaching more to the head than the heart. 7. I’d like to preach more sermon series, still honoring the lectionary as my guide for texts. Once Bob had his list, he wasn’t sure what to do with it. After some thought, he did three things. First, he bought two recent books on homiletics. He hadn’t read a homiletics text since graduation from seminary. Second, he found two other pastors and began meeting every week for an hour and a half to talk about preaching and especially the coming Sunday’s sermon. Third, he began intentionally putting more time into preparing the presentation of his sermon, that is, the delivery of the sermon. He actually spends an hour a week now just rehearsing the sermon’s delivery. Getting the Conversation Going 1. We all need help with our preaching. Look at Pastor Bob’s list of seven challenges. Which of the seven stands out most as one of your challenges? Circle it. 2. Share one thing you include in your sermon preparation that really blesses your preaching. The Scriptures Speak “The Scriptures Speak” is designed as a guided plenary Bible study. The circuit counselor, host pastor, or another designated leader may present the study. Each participant should have a study guide. For each question, additional material is provided for the leader and is printed in bold italics. Because of electronic transmission issues, Greek words and phrases have been transliterated into English. 1. An Old Testament passage that gives a vivid picture of preaching is Nehemiah 8:1-12. The scene gives us God’s returned exiles gathered on the Feast of Trumpets to hear the Word of God and worship. Note the following about the preaching in this passage: - (Vss. 2-3) The reading and exposition of the Scriptures lasted 5-6 hours, during which the preachers and listeners (men, women, and children) stood (vs. 7). - (Vs. 4) The rudiments of a pulpit were present. - (Vs. 6) Preaching happens in the context of worship. - (Vs. 7) Preaching was a priestly task, here presented by Levites. - Vs. 8) The preaching included the following: reading aloud of the text, clarifying the text (perhaps translating it into a Targum or Aramaic idiom or perhaps a paraphrasing of the text), and an explanation of the text’s meaning, its application to the listeners’ lives. - (Vss. 9-12) Listeners responded, showing the impact of both law and good news. 2. The New Testament uses several words to capture the preaching task. Primary among them are the first two listed below (kerusso and euangelium). 95% of the English translations, “preach” and “proclaim,” are based on these two words. Romans 10:15 and Galatians 2:2 use both. Here are some of the root words used for the preaching task in the New Testament, each with its own emphasis. Circle the one or two that come the closest to your own practice of preaching: - Kerygma – heralding of a proclamation (Mark 1:38-39; 3:14; 1 Cor 1:17) - Euangelium -- announcement of Good News (Acts 16:10; Rom 1:15) - Homilia – conversation or interactive discussion, talk among friends (Luke 24:14, “talked with each other;” and Acts 20:11, “after talking until daylight”) - Logos – a reasoned message or saying (1 Cor 1:18) - Didache – teaching (Matthew 11:1, Acts 28:31, and Romans 2:21, and 1 Timothy 2:7 use a form of this word for “teaching” with “preaching” [kerusso]. So are they synonymous or different? It raises the issue of the role of teaching [today, catechesis] in our preaching. No doubt this will raise some issues among participants.) - Lalia – speaking; a simple word for verbal communication (Mark 2:2) 3. Jesus' ministry, according to Matthew, was one of "teaching, preaching, and healing" (Matthew 4:23). Turn to the Sermon on the Mount in Matthew 5-7. Just as his healing, Jesus' teaching and preaching were amazing to those who heard Him. As Matthew puts it, "...the crowds were amazed at his teaching because he taught as one who had authority, and not as their teachers of the law" (Matthew 7:28-29). We can learn much from Jesus' teaching for our own preaching. Find evidence of these qualities of Jesus' teaching in the Sermon on the Mount: - Using repeated phrases (not unlike the refrains of traditional African American call and response preaching or a preacher's repetition of a "hook" phrase): 5:3-11; 5:17, 21, 27, 31, 33 38, 43. - Engaging the hearer's imagination through the use of visual metaphor or narrative: 5:13-14; 6:19-24; 6:25-33; 7:6; 7:13-20; - Breaking open old truths from Scripture to extend or freshen their meaning and application: 5:17-47; - Using hyperbole or humorous exaggerated imagery to make a point: 5:29-30; 5:40; 7:3; 7:9. Are there other characteristics of Jesus' teaching that transfer well to our preaching task? Elsewhere in the Gospels, Jesus uses story or the extended metaphor of a parable to make a point (Lk 15), sometimes more than one to make the same point (Mt 13:44-46). Repeatedly Jesus uses the Scriptures in His preaching (Lk 4:16-21; 7:24-28). 4. Peter's Pentecost sermon in Acts 2 offers a brilliant example of the church's first preaching. At work in this sermon are at least these crucial moves: - Clear and pointed law (Vs. 23) - Repeated use of Scripture (Vss. 16-21; 25-28; 34-35) - Full and clear gospel presentation: the Christ event, the benefits of the Christ event, and where they can receive these benefits (vss. 38-39) 5. Paul's ministry of preaching is vividly recorded in the book of Acts. Sometimes we get a summary of his speaking and sometimes we get his very words. Note these insights on preaching from his visits in Thessalonica and Athens: - In Acts 17:1-4, Paul's messages on three different Sabbath days are summarized with an amazing variety of words, showing the multi-faceted nature of preaching: The words what he did in the synagogue include the following: 1. “Reasoned” 2. “Explaining” 3. “Proving” 4. “Proclaiming” 5. “Persuaded” - In Acts 17:16-34, Paul speaks to the Areopagus in Athens. While not technically a sermon, since it happens outside a worship setting, his evangelistic message offers these helps for preachers: 1. Paul is keenly aware of his audience, their values, longings, and limitations (vs. 22). 2. Paul breaks open secular, even pagan notions, recasting them from a Scriptural perspective (vs. 23), using well-known Greek poetry (vs. 28) for his language. 3. (Vss. 23-28) Paul builds on what his listeners already know (God as creator) to what they do not know (God of the resurrection). - In both incidents (vss. 4-9, 32-34), it matters to Luke to record those whom God brought to faith as well as other responses to the preached Word. There is an attention to remembering and recording those whom the Spirit graces with faith through the preaching of the Word. 6. Paul’s instruction to Timothy in 2 Timothy 2:15 is crucial for the preaching task. Note the following from this verse: - The Greek word translated here in the NIV here as “correctly handles” only appears outside the New Testament in the Septuagint’s Proverbs 3:6 and 11:5, both showing the word’s sense of cutting a straight path through a forest or desert, thus the KJV’s “rightly divides.” - Effective preaching is a matter of cutting a straight path through a text to Jesus Christ. The hermeneutic for that task? Law and Gospel. It is knowing how to rightly divide a text so that both law and gospel cut the way clearly to Jesus Christ. Words from the Wise on Preaching The three quotes included here can be helpful as resources for the conversations to follow. Use them selectively or read all three. - The Lutheran Confessions on the real-time effects of law/Gospel preaching: The word “real” here (“real terror, regret, and sorrow in their hearts”) shows the importance of the preached Word being in real time with read needs in mind. We do not talk about law and gospel. We do not aim for virtual judgment and grace. The Holy Spirit leads people to come under law and gospel in the very real moment of preaching. “Through these means (the preaching and hearing of the Word), God goes about his work and breaks our hearts and draws people, so that they recognize their sins and God’s wrath through the preaching of the law, and feel real terror, regret, and sorrow in their hearts. Through the preaching of the holy gospel of the gracious forgiveness of sins in Christ and through meditating upon it, a spark of faith is ignited in them, and they accept the forgiveness of sins for Christ’s sake and receive the comfort of the promise of the gospel. In this way, the Holy Spirit, who effects all of this, is sent into their hearts.” (Formula of Concord, Solid Declaration, Article II. The Book of Concord. Kolb/Wengert, [Minneapolis: Fortress Press, 2000], p. 554. - John Bradford, 16th Century English reformer and martyr, on the power of the Word to bind us and loose us with law and gospel: These words from Bradford reveal not only the influence of the Reformation in England but also the challenge of rightly dividing the Word of God in preaching. “He that is ignorant of [the division of the places of the Law and of the Gospel] cannot, though he were a great doctor of divinity, and could rehearse every text of the bible without book, but both be deceived, and deceive others; as the experience hereof (the more pity) hath taught, nay, seduced the whole world…Therefore, I say, take to thee the glass of God’s law; look therein, and thou shalt see thy just damnation, and God’s wrath for sin, which, if thou darestest, will drive thee not only to an amendment, but also to a sorrow and hatred of thy wickedness, and even to the brim of despair, out of which nothing can bring thee but the glad tidings of Christ, that is, the gospel: for as God's word doth bind thee, so can nothing but God's word unbind thee; and until thou comest to this point, thou knowest nothing of Christ." (John Bradford, The Writings of John Bradford, "Preface to: The Places of The Law and of the Gospel by Petrus Artopeus" [Cambridge: The Parker Society, 1848; orig. 1548], p. 5.) - C.F. Walther on the importance of the preacher’s own genuine experience of law and gospel: Walther’s words challenge the preacher to come under the judgment and consolation of the Word himself. That integrity of the applied Word is a crucial step in sermon preparation. "Now, since a person under the pedagogy of the Holy Spirit learns rightly to distinguish the Law from the Gospel and to divide both, it follows that genuine Christians, be they ever so feeble otherwise, as long as they have experienced the force of the Law and the consolation of the Gospel or the power of faith, are best prepared to apply to others what they have experienced in their own lives. Accordingly, ministers who may be classed among the poorest intellectually not infrequently are found to be the best preachers." (C.F. Walther, The Proper Distinction Between Law and Gospel [St. Louis, Concordia Publishing House], p. 51. Conversation (Groups of 3-4) You may want to ask a pastor in each group to serve as leader, whose primary task is to be sure everyone has an opportunity to speak. Participants may decide to focus on only 2-3 of the questions or to raise questions themselves. Be sure to encourage prayer for one another. 1. Which of the words for the preaching task in No. 2 above best captures the character of your preaching? Share the word with your group and explain why you chose it. 2. What is the relationship between preaching and teaching? Does teaching have a place in the preaching task? Share your responses. 3. Jesus engaged his hearer’s imagination through simile, metaphor, and the extended narrative metaphors of parables. Share a reference from one of your recent sermons when you did the same, 4. Is there a place for evangelistic preaching in the Lutheran tradition? If so, what cautions are attendant to such preaching? 5. In what sense(s) is preaching a matter of persuasion? If persuasion is part of preaching, just what is the goal of the preacher’s persuasion? Share your responses. 6. How much self-revelation is too much in preaching? Where are the boundaries here? 7. What can a preacher do to be sure he himself has come under the influence of the law and the gospel in a Biblical text? Note the Walther quote above. 8. As a group come up with a cultural icon waiting to be “broken open” with new meaning from the perspective of law and gospel. 9. Close your group with prayer for one another’s preaching ministries. Dean Nadasdy, Senior Pastor, Woodbury Lutheran Church, Woodbury, MN 3rd Vice President, LCMS email@example.com The Faithful and Afire Series Introduce the series and the series theme, referencing the study guide. Future sessions will focus on other aspects of the pastoral task, including missional leadership, worship, balance of family and ministry, integrity, confessing pastors, equipping the saints, and the pastor’s prayer and devotional life. Take a moment, too, to introduce the process – 30-45 minutes of guided total group Bible study followed by 30-45 minutes of conversation in small groups of 3-4. In 2 Timothy 1:6, the apostle Paul writes to remind Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” That little verse signals the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to motivate and grow us as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church-Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will look at key aspects of pastoral ministry, seeking fresh insights and continued growth. A Case in Point This “Case in Point” launching segment asks pastors to consider their own challenges as missional leaders. It “primes the pump” for the study that follows. Read the little case study aloud along with the questions that follow. Responses to the discussion questions may be shared in the total group or already in small groups of 3-4. Try to keep this first conversation to 10 minutes. The phone call that Pastor Bob just received has placed him into a pensive mood. His brother, Pastor Bill, recently accepted a call to a new congregation. He has phoned Bob to ask for assistance and a little brotherly insight. Bill sees his new congregation as stuck in a rut. Attendance at worship services is slowly declining. The Adult Bible Class has grown a little bit since “the new pastor” arrived, but numbers are dramatically down compared to the congregation’s heyday. The Board of Elders is looking to Pastor Bill for a new vision. Several new elders are young business executives who believe that their church needs a comprehensive strategic plan complete with a matrix of measurable objectives. Some have even suggested a complete restructure of the congregation’s governance system to reflect some of the new thinking going on in corporate America. Both Pastors Bob and Bill consider themselves to be strong leaders. Yet their formal training in the disciplines associated with leadership and organizational management has been minimal. What they know about leadership they have learned through experience. While experience has served them well, Pastors Bob and Bill would like to grow as both practically and theologically as leaders. Their list of questions includes the following: o How will we consider visionary & strategic planning from a Biblical perspective? o Where will we turn for help in understanding new models of church governance? o What types of matrix or measurement systems are appropriate in the church? o What can we in the church learn from corporate culture where leadership is concerned? To tackle the leadership issues both face, these brother pastors have agreed to hold each other accountable for three commitments. First, they will read six books over the next twelve months, each book deals addressing an aspect of leadership or organizational management. A couple of their selected books are written by authors within the churched world. They reflect thinking on leadership from a practical perspective within the church’s culture. Two books are from a secular worldview. They reflect some of the best leadership thinking from within the corporate culture. They hope that these can be applied to their churches without compromising their doctrine and values. The last two books are strongly theological in nature. These wrestle with the differences between worldly or corporate understandings of leadership and leadership understood from a Biblical worldview. Second, Bill and Bob have committed to subscribe to an audio club offered by a prominent church leader and former pastor. Each month, they will receive an audio resource on a leadership topic, listen to it, and talk about it. Third, the brothers have committed to conversation. They will set aside an hour every other week to apply what they’re reading to the congregations they serve. Both know that leadership is best learned through experience, yet they’re convinced they’ll benefit as leaders from their new commitments. Getting the Conversation Going: 1. Leadership in the church is difficult in our results-oriented culture today. Consider some of the challenges facing Pastor Bill had leadership challenges in his new congregation. What are some of yours? 2. Share with members of your group one or two leadership resources that you have found to be helpful in your ministry. The Scriptures Speak “The Scriptures Speak” is designed as a guided plenary Bible study. The circuit counselor, host pastor, or another designated leader may present the study. Each participant should have a study guide. For each question, additional material is provided for the leader and is printed in bold italics. Because of electronic transmission issues, Greek words and phrases have been transliterated into English. I. Leadership Credentials In 1984, Tom Peters and Nancy Austin co-authored a book on leadership. Titled A Passion for Excellence,¹ the book summarized research on the characteristics, traits and credentials of America’s most successful business leaders. Their book raises the question for the church: “What qualities and credentials does Scripture establish for the pastor as a leader?” A. 1 Timothy 3:1-7: Late in his ministry, St. Paul wrote to Timothy, instructing him regarding oversight of the church in Ephesus. In chapter 1 Timothy 3:1-7, Paul provides Timothy with a detailed list of the credentials necessary for one to serve as pastor. Note the following: o The attributes that make up this list of qualifications can be divided into two groups of seven (vs 2 and vss 3-6) with a concluding qualification in vs 7. The first qualification that Paul lists, anepilempton, means “irreproachable.” The term might be thought of as the over-arching introduction to the qualifications that Paul introduces here. The Greek word means, “not to be taken hold of” or “nothing left on his doorstep against him.” Paul is saying, “No failure at the negative qualities listed here should be able to be charged against an overseer of God’s church!” (vs 2) “Husband of one wife”: Paul begins with an emphasis upon sexual purity. The episkopos in the apostolic era came out of a culture rife with sexual immorality. Gross sexual immorality accompanied pagan temple rites. Homosexuality and promiscuity saturated the world in which the overseer served. (vs 2) “Temperate” and “sophrona” (of sober-mind). Paul indicates the need that for spiritual temperateness. The pastor must have a theological soundness of mind and judgment. (vs 2) “kosmion” (orderly): A pastor must be orderly in his mental life and in his habits. Note there is a quality of discipline here. In order to shepherd others, the pastor himself must be pastored or led by the Good Shepherd. Often this occurs as the pastor engages in a disciplined use of the Word in his own life. (vs 2) “philoxeron” (hospitable): In apostolic times this often meant two things. First, the pastor’s home was often the first stop for weary Christians seeking refuge from persecution. Second, most of the on-going people or community needs landed at the pastor’s doorstep. Those in poverty, sickness and need made their appeal to the pastor who is here being charged with the call to serve them. (vs 2) “didaktikon” (apt to teach): 1 Corinthians 12:29 suggest that not all Christians are gifted to teach. Here Paul suggests that pastors should have competence in their ability to lead others through the teaching of God’s Word. (vs 3) “Not violent but gentle”: Paul states that the pastoral office requires a man who is even-tempered. Martin Luther used the term, “gelinde” (yielding). This qualification might seem out of place in the cut-throat American marketplace, but it is essential to the office of the pastoral ministry. (vs 3) “Not a lover of money”: In his book, The Shepherd’s Covenant, H.B. London points out that too often pastors become fixated on the “4 P’s”: prestige, power, paychecks, and pensions. (vs 4) “prohistamenon” (managing his household): The participle has the sense of superintending. The pastor guides and leads his own marriage and family spiritually. Failure to invest in these primary relationships will spell trouble not only for the pastor, but for the church he serves. (vs 7) “A good reputation with outsiders, so that he will not fall into disgrace and into the devil’s snare”: God’s adversary, the devil, works with extra diligence to ensnare pastors in a sin-trap. One terrible result of their falling into public sin is the damage to their congregation. Another is stumbling block they present to those outside the church. B. Titus 1:1-9: Titus served as Paul’s companion during his third missionary journey. In this epistle, Paul addresses the need of the church in Crete. Titus has been left in Crete to “straighten out what was left unfinished” (1:5), namely, the appointment or placing into office of “presbuteros” (pastors). In vss. 1-9, Paul presents a list of qualifications for the office of pastor very similar to those found in 1 Timothy 3. Note the following: (vs 5) “presbuteros”: Note that this term is virtually synonymous with the word that Paul uses in 1 Timothy 3:1, ‘episkopos’. Both terms refer to the office of the pastoral ministry. “Overseer” indicates the task to be done, giving oversight and leading. The term, “elder,” carries the dignity attendant to the office. (vs 6) “Anegkletos”: An elder must be blameless. Note the similarity between this opening and that in 1 Timothy. In both cases Paul indicates that the pastor must be above reproach in the qualities identified, from within and outside of the church. - (vs 7) Note the similarities between the list of qualifications here and those given in 1 Timothy. Both texts address such issues as wine, temper, money, sexuality, judgment, and discipline. - (vs7) Paul includes the pastoral task of teaching, “didaskalia.” The pastor must be competent in teaching and capable in convicting and refuting those who oppose sound doctrine. II. The Servant Character of the Pastor as Leader: In 1995, author Laurie Beth Jones wrote Jesus, CEO,$^4$ a book that captured significant attention in the secular workplace. The idea behind this book seems good: to move Biblical leadership principles from the church to the marketplace. The title, however, seems greatly out of place! If Jesus were to hang a title on His office door, it would not resemble those we see in corporate America today. Instead of titles like CEO, CFO, CAO, etc., one title represents Jesus’ view on leadership best, the title, Servant! Without question, some of the most pointed Biblical texts on leadership emphasize the servant nature of the pastor/overseer in his congregation. Note the emphasis placed upon servanthood in the passages below: A. Matthew 20:20-28: In this account the issue of leadership is apparent. With Peter, James and John comprise Jesus’ inner circle. Without question, their expectation is toward the establishment of an earthly Messianic Kingdom. Here their mother, Salome, states their request for the highest place of leadership in Jesus’ Kingdom, to sit at His right and left. Instead of granting their request, Jesus redefines the very nature of leadership as it will unfold in His kingdom. 1. (vs 25) “hoi archontes”: Jesus desires to contrast the concept of leadership in His Kingdom with that exercised in the realm of the Gentile politic. 2. (vs 25) “katakuriuevousin” (lord over) + “katevousiazousin” (exercise authority over). In both verbs, the prefix, “kata,” indicates a form of abuse, a coming down on or up against those subject to them. 3. (vs 26) “It is not so with you!”: Leadership in God’s Kingdom is set in opposition to leadership of lording over, which characterized the corrupted secular leadership of the Roman government. 4. (vs 26) “diakonos” (servant): Leadership in God’s Kingdom is qualitatively described by the word “servant”. A diakonos focuses not on self but on the one being served. 5. (vs 27) “doulos” (slave): Here the concept of leadership as servanthood goes deeper. The slave is the humblest and lowest of all servants. On this matter, C. Gene Wilkes writes, “Being a slave involves several things: the loss of property, separation from roots, loss of individuality, and the freedom to choose. Being a slave means giving up personal rights.”$^5$ The implication is clear: leadership in the pastoral realm is qualitatively different than leadership in the secular realm. B. Philippians 2:1-11: This section of Paul’s exhortation immediately follows his call, “Do nothing out of selfish ambition or vain conceit, but in humility consider others better than yourselves,” (vs 3). Again, the essence of servant leadership in Christ’s kingdom is qualitatively described in relation to the self-humiliation of Jesus, Himself. 1. (vs 5) “phroneite” (Be minded): This exhortation encourages the pastoral leader to set his mind, attitude and intention on a servant orientation. 2. (vs 7) “ekenasin” (emptied): Jesus considers not Himself but the mission and work for which he has become human. He emptied himself in order to fill us, that we might be made the righteousness of God in Him. The servant nature of leadership is again underscored! Note 1: In his book, Descending into Greatness, author Bill Hybels writes of this text, “The message of Philippians is this: If you want to be truly great, then the direction you must go is down. You must descend into greatness… Greatness is not a measure of self-will, but rather self-abandonment.” Note 2: J. Oswald Sanders writes, “True leadership is achieved not by reducing men to one’s service but in giving oneself in selfless service to them…” III. Missional Framework for Leadership In his book, The 21 Irrefutable Laws of Leadership, author John C. Maxwell suggests that one of the most important roles a leader can play is that of helping an organization navigate well. “Above all,” Maxwell suggests, “leaders must have a firm grasp of the mission they are on and the business that they are in.” In similar fashion Pastor Rick Warren writes, “Nothing precedes purpose. The starting point for every church should be the question, ‘Why do we exist?’ Until you know what your church exists for, you have no foundation, no motivation or direction for ministry.” Without question, one of the key tasks for the pastor as a leader is the establishment of a clear, Biblical sense of the church’s mission. How do we, as Lutheran theologians, understand the mission of the church? A. In 1986 the LCMS, through its synodical convention asked the Commission on Theology and Church Relations (CTCR) to ‘initiate, facilitate, and develop a theological statement of mission.’ The problem at hand was confusion over the term, mission, both in the secular and theological domain. B. The CTCR document, Theological Statement of Mission, (1990) provides both a broad and narrow definition of the church’s mission. Here is a summary: 1. Mission understood broadly, is “God at work seeking to restore His creation to Himself.” 2. In the narrow sense, mission is to be understood Christologically, Mission is “what God has done through the work of Jesus Christ to restore fallen creation to Himself.” C. The creation of the church – a theology of Mission: Lutheran theologians teach that God has chosen to create the church in order to administer His means of grace through which He carries out His mission. 1. Pastors (properly called) carry out the mission of the church when they publicly (for and in behalf of the congregation) proclaim the word of God and administer the sacraments. 2. God involves individual saints in carrying out His mission in a variety of ways including: - Vocation - Service to neighbors - Use of the service gifts D. This understanding of mission is certainly evident in the Bible’s “Mission Magna Carta,” Matthew 28:18-20. (vs. 19) “poreuthentes” (going): Note that this participle, going, accompanies the imperative, “make disciples,” the primary verb in the sentence. “matheeteusate” (turn into disciples): This aorist imperative active verb forms the heart of Jesus’ charge to the disciples. As they go forth, they are to turn people of all nations into disciples of Jesus Christ. The pastor as leader does well to keep the intentional mission of making disciples at the forefront of the church’s mission. Note that (vs. 19) “baptizontes” (baptizing) and “didaskontes” (teaching): These supportive participles indicate the “how” of disciple-making. Both point to the reality of God working through His established means to make disciples. Indeed, it is through His Word and the Sacraments that God subjectively applies to people the objective benefits of Christ’s life, death, and resurrection. As leaders, pastors must keep Word and Sacrament ministry at the center of all that happens in the church. IV. Effective Leadership: Doing the Right Things! A. The New Testament utilizes several key words to describe the office of pastoral ministry. In Part I, two such words, episkopos and presbuteros, were identified. The first of these (overseer) provides some degree of insight into what the office of pastor was created to do. The overseer serves as a steward over Christ’s congregation. A third word, poimen (shepherd), is also useful in describing the leadership role of a pastor. Consider the following: 1. Jesus is depicted as the Good Shepherd. - The shepherd knows his flock and calls each by name. (John 10:3,14) - The shepherd seeks the lost sheep and rejoices upon finding it. (Luke 15:3-7) - The shepherd is prepared to lay down his life to protect His sheep. (John 10:11-13) - The shepherd will be struck down for His sheep. (Mark 14:27-28/Zechariah 13:7) - The shepherd oversees souls. (1 Peter 2:25) 2. The title of “shepherd” is given to pastors in Ephesians 4:11. In this chapter Paul is making a distinction between various offices within the body of Christ. The words chosen to describe the office of the pastoral ministry are shepherd and teacher. In Pauline literature the office of pastor is to be understood in light of all the New Testament says about Jesus as the Good Shepherd. The pastor is to function as an under-shepherd to Jesus. What Jesus considers important for Him to do for His sheep is reflected in the office and work of the pastor. In a sense, pastoral leadership could be described not only as servant leadership but also as shepherd leadership. - Note: This metaphor is also expressed in 1 Peter 5:1-4. Here Peter echoes the mandate that he personally received from Jesus (John 21:16): “Shepherd (NIV, “Take care of..”) my sheep!” The emphasis is on both relationship and task. The pastor is called to have a caring, shepherd/sheep relationship with the flock he has been called to oversee. Additionally, he is to focus his energy and efforts upon shepherd tasks! Effectiveness in pastoral ministry, doing the right things, can be measured by looking at the shepherd metaphor. B. In his book, Good to Great, author Jim Collins provides insight into the question of what makes leaders “great.” Collins notes that one trait consistent among “great” leaders is the ability to be effective.$^{13}$ Effectivity is defined as the ability to do those right things necessary to help an organization meet its goals. Using the metaphor of pastor as shepherd leader, note these three “right-things” to be accomplished pastoral office: 1. The shepherd feeds: The New Testament consistently emphasizes the pastoral work of feeding. Two specific feeding tasks are preaching and teaching. - Mark 3:13-19: At the appointment of the twelve apostles the first emphasis for their work is upon preaching. - 1 Corinthians 1:17: Paul identifies preaching as primary to his call. - Romans 10:16-17: Conversion comes by preaching the Word. - Matthew 28:20: After baptizing, teaching is listed as a primary disciple-making activity. - 1 Timothy 4:11: Timothy is called to teach. - Titus 2:1-2: Titus is called to teach sound doctrine in a formative manner. 2. The shepherd meets needs: A second work essential to the pastoral office is caring for a flock’s spiritual (and sometimes physical) needs. A clear emphasis is placed upon providing for the Sacramental needs of the flock (baptismal and Eucharistic needs). Passages like Matthew 25:35, 43; Acts 15:36; and James 1:27 highlight the care-giving nature of the pastoral office. 3. The shepherd leads: Just as a shepherd is given charge to lead his flock, so the pastor as shepherd leader is called to give priority to leadership activities directly impact the spiritual life of God’s people. Of this C. Gene Wilkes writes, “The leader becomes servant to those who have joined him when he provides adequate vision, direction, correction, and resources to carry out the mission entrusted to the group.”$^{14}$ - Consider how John’s inspired words to the pastors and churches of Asia Minor in Revelation 1-3 identify these tasks of shepherd leadership: - Establish/re-establish the mission of the church. - Affirm the positive efforts of each congregation. - Offer specific “correctives” to each church. V. Efficient Leadership If effectivity can be described as the art of doing the right things, efficiency can be described as doing the right things right. In his best-selling book, The Seven Habits of Highly Effective People, author Stephen Covey points to the difficulty inherent in efficiency. Covey suggests that too often a leader’s quest for efficiency is undermined by distractions.$^{15}$ In pastoral ministry there are numerous variables that can negate workers’ efficiencies. Here are a few examples: - Tyranny of the Urgent (responding to “urgent” but unimportant items) - Failure of Structure (a poor organizational system capsizing efficiency) - Failure to Delegate - Improper Delegation A. Consider two scriptural texts that seem dedicated to the issue of efficiency through delegation: o Exodus 18:1-27: The example of Jethro and Moses. o Acts 6:1-7: The disciple’s selection of “The Seven”. B. Consider how the “gift” passages of Scripture relate to the concept of delegation and efficiency! (See Romans 12:3-8; 1 Corinthians 12:4-11, 27-31; Ephesians 4:7-13; and 1 Peter 4:7-11 for the spiritual gift lists, highlighting the varied ministries within the church.) VI. Leadership Style: Too often, the leadership literature of our culture today seems to suggest that in order to “succeed” in leadership, a Type A Dominant personality type is necessary! Scripture, on the other hand, provides descriptions of a variety of leaders, each with a different personality and style of leadership. Of this, C. Gene Wilkes writes, “No one temperament is superior for leadership among God’s people…as we search the Scriptures we see that God chooses people with (different) styles.”$^{15}$ Consider the different styles of leadership represented in the following texts: o Galatians 2:11-19: Saint Paul’s “dominant style” of leadership is apparent. Gary Smalley might call Paul a “lion.” (See Gary Smalley, The Two Sides of Love, Colorado Springs: Focus on the Family, 1990, 34-36). o Acts 9:23-28: The “influence style” of Barnabas is set forth in this story. Barnabas is more “otter” than “lion.” (Gary Smalley). o Genesis 12:11ff and Genesis 16:1-6 display Abraham’s “steadier style” of leadership. Abraham looks more like a “golden retriever” in Smalley’s nomenclature. o Exodus 2:11-20: Here we see Moses’ “conscientious style” of leadership. Moses, it seems is a “Biblical beaver.” (Smalley) Words from the Wise on Leadership The three quotes included here can be helpful as resources for the conversations to follow. Use them selectively or read all three. o Martin Luther on the high view of the pastoral office in Scripture: “The office of preaching the Gospel (Pastoral Office) is the highest among them all; for it is the true Apostolic office, laying the foundation of all other offices, on which it is proper to build all others, namely the offices of teacher, of prophet, of governors, and of those who have the gift of healing.” Luther quoted in Franz Pieper, Christian Dogmatics Vol. III (St. Louis: Concordia, 1953), 462. o C.F.W. Walther on the Mission of the Church: “…Dear Brethren, we are assembled here not for our own sake. We are in the faith, and by this faith we hope to be saved! But there are still many millions who have no faith. This is why we are here – so that we might bring salvation to as many people as we possible can… Only for this reason does our gracious God allow Christians to live on earth, that they might bring others to the saving faith.” C.F.W. Walther, “Essays for the Church” (St. Louis: Concordia Publishing House, 1992) II: 262. LCMS Church Growth Strategy Task Force on the use of “ministerial reason”: “The church is encouraged to be “in the world by making inquiry into the fields of sociology, psychology, and technology in so far as such might serve the church ‘mission.’” The Church Growth Strategy Task Force, Toward Theological Basis, Understanding and use of Church Growth Principles in the LCMS. (St. Louis: The Standing Committee Pastoral Ministry, 1991), Pt. II, p 7. Conversation (Groups of 3-4) You may want to ask a pastor in each group to serve as leader, whose primary task is to be sure everyone has an opportunity to speak. Participants may decide to focus on only 2-3 of the questions or to raise questions themselves. Be sure to encourage prayer for one another. 1. Make note of the Luther quote above, (a high view of pastoral office). How is this view of the pastoral office often placed into conflict in the Hire/Fire mentality practiced within the American leadership culture? 2. “In what ways does the American leadership culture place pressure upon the pastoral leader to perform?” 3. Consider the Walther quote above, (Mission of the Church). How does our LCMS theology regarding mission shape the following leadership tasks often assumed by pastors? - Establishing a Mission Statement - Setting Vision - Developing Strategic Intent 4. Consider the quote regarding the use of ministerial reason. What aspects of “secular” leadership teaching seem to you to be appropriate for use in the church? 5. Consider the lists of leadership Credentials in Part I – Scripture Speaks. Which characteristics would you highlight as being particularly challenging for pastors today? Explain what makes these qualities so challenging. 6. Consider what Scripture has to say regarding the pastor as servant leader (Part II). How has the American mega-church given rise to pastoral expectations that contrast with this image? Are there aspects of the “servant/slave” role that are difficult for you? Explain. 7. Consider the Pastoral Tasks that are essential to the metaphor of the pastor as shepherd leader. Is it difficult to maintain these tasks as priorities in your ministry? Explain your response. 8. Consider Part V, “Efficient Leadership. What causes inefficiencies in your pastoral work today? As a group, compile a list of these causes and consider ways to address them. 9. Review Part VI, “Leadership Style.” How would you characterize your leadership style? What are the “strengths/weaknesses” of this style? How might you go about capitalizing on your leadership strengths? How can you strengthen aspects of your leadership needing improvement? 10. Close your group with prayer for one another’s ministries as pastoral leaders. Endnotes 1. Peters and Austin, A Passion for Excellence (New York: Warner Books, 1985). 2. H.B. London, Jr. and Neil B. Wiseman, The Shepherd’s Covenant (Regal, 1985). 3. Laurie Beth Jones, Jesus, CEO (Laurie Beth Jones, 1995). 4. C. Gene Wilkes, Jesus on Leadership (Tyndale, 1998). 5. Bill Hybels and Rob Wilkins, Descending into Greatness (Zondervan, 1998). 6. J. Oswald Sanders, Spiritual Leadership (Moody Press, 1967), 13. 7. John C. Maxwell, The 21 Irrefutable Laws of Leadership (Atlanta: Maxwell Motivation Inc., 1991), 33. 8. Rick Warren, The Purpose Driven Church (Bouwer/avant 1995), 81-10). 9. Commission on Theology and Church Relations of the LCMS, A Theological Statement of Mission (St. Louis: The LCMS, 1990), 3. 10. Ibid., 7, 12. 11. Ibid., 10, 13. 12. Jim Collins, Good to Great (Harper Collins, 2001). 13. C. Gene Wilkes, Jesus on Leadership (Tyndale, 1998), 10. 14. Stephen Covey, The Seven Habits of Highly Effective People 15. C. Gene Wilkes, Jesus on Leadership (Tyndale 1998), 144. Author: Rev. Dr. Luther C. Biggs, Senior Pastor Prince of Peace Lutheran Church Carrollton, TX [firstname.lastname@example.org] Session 3 Worship: Stewards of the Mysteries LEADER’S GUIDE In 2 Corinthians 4:1-2 Paul tells the congregation at Corinth: “This is how one should regard us, as servants of Christ and stewards of the mysteries of God. Moreover, it is required of stewards that they be found trustworthy.” Understanding pastors as “stewards of the mysteries” is foundational for all ordination rites in the church catholic for it describes with clarity what pastors do: they dispense the gifts of Christ’s flesh through the means of grace, the gifts of forgiveness, life, and salvation. Pastors represent Christ, speak His voice and deliver His gifts. As the one absolution in our liturgy so clearly states: “Upon this your confession, I, by virtue of my office, as a called and ordained servant of the Word, announce the grace of God unto all of you, and in the stead and by the command of my Lord Jesus Christ I forgive you all your sins in the name of the Father and of the + Son and of the Holy Spirit.” A key aspect of our pastoral ministry, and therefore of our pastoral care, is to recognize that the primary place for standing as stewards of the mysteries is in the worship of our church. A Case in Point Nicholas was the best and brightest in Pr. Jacobson’s confirmation class. He was also a great athlete. One spring afternoon, while pitching his team to victory, Nick’s shoulder gave out. The diagnosis was bleak. Cancer, the ravenous kind, eventually took his shoulder, his arm, and finally his life. His young parents, in their early thirties, could hardly cope with the loss. Pr. Jacobson was with them all the way, even driving into New York City three times a week during that last month as Nick struggled at a famous cancer hospital. Nick’s parents leaned heavily upon him, and his pastoral care was well received. In the afternoon following Nick’s funeral, his father said to Pr. Jacobson: “Thank you for you have done for us this past year, especially these last months of Nick’s life. I know you meant well in your sermon today, about how Nick is in heaven, but frankly, Pastor, that is no comfort to me. It’s not fair. Nick should not be heaven – he should be here, pitching in little league, going to high school. It is no comfort to me that Nick is in heaven, for he belongs here with us, on earth.” Pr. Jacobson, recently ordained, did not know what to say. He left that afternoon completely devastated as to how he might comfort these young parents at the loss of their son. And somehow, in the next day of two, he discovered what our liturgy means when it says: “therefore with angels and archangels and will all the company of heaven we laud and magnify your glorious name, evermore praising you and saying . . .” He hurried back to those grieving parents and encouraged them to come to the Lord’s Supper that Sunday. He described the mystery of the sacrament, the mystery of what we believe about the person of Christ, always present according to his divine and human nature, that is, bodily, in his flesh, and that wherever Jesus is, there is heaven. And so there on the altar, at the mystery of the Lord’s Supper, are present in Christ the angels and archangels and all the company of heaven. And that includes Nick, who is present with us “in Christ,” singing praises to the Lamb in heaven as we sing them here below. That is why we can truly say that at the table of the Lord, heaven and earth come together in Christ. There is no greater mystery and no greater comfort than this. That Sunday, Nick’s parents came to church to receive Christ’s body broken in death, his blood poured out for the forgiveness of sins. As Pr. Jacobson gave them this mysterious gift of Christ’s very body and blood, tears flowed down his cheeks, and so it did for Nick’s parents and for many in the congregation who were also mourning his loss. As Nick’s father greeted him at the door after the service, he said words to Pr. Jacobson that every pastor longs to hear: “Pastor, now I am comforted to know that Nick is in heaven with Jesus.” Getting the Conversation Going Over twenty years ago, William Willimon wrote a book entitled Worship as Pastoral Care. - How many pastors see the liturgy as the primary means to provide pastoral care for their flock, e.g. the wedding service to counsel couples on what the Church believes about marriage, the funeral service about our understanding of death, life, and resurrection? - In what ways does the historic liturgy, represented in the official worship resources of our church, help us teach our congregations that their most intimate communion with Christ occurs through his bodily presence in Word and Sacrament? The Scriptures Speak 1. Jesus’ ministry of preaching and miracles in bringing in the new creation Jesus’ sermon in the synagogue in Nazareth is programmatic for His ministry and for the worship life of the church (Luke 4:16-30). In this sermon, the evangelist gives us the structure of the synagogue service which formed the basis for the liturgy of the Word in the Christian worship services: The reading of the Word Torah Psalm Prophets Psalm [Historical Writings] [Psalm] Interpretation of the Word Preaching/Teaching – (also known as Midrash) In citing Isaiah 61, His first public words after his baptism, Jesus begins His ministry in the name of the Father and of the Son and of the Holy Spirit, announcing that he is the Messiah/Christ: The Spirit [HS] of the Lord [Father] is upon me [Son] because of which he has anointed me To proclaim good news to the poor He sent me To proclaim release to the captives and recovery of sight to the blind, To send away in release the broken ones To proclaim the year of jubilee (Luke 4:18-19) By this Old Testament prophecy Jesus is telling His hometown hearers and us: - That He is the long-awaited Messiah come to set His people free; That the purpose of His incarnation is proclamation -- performative speech -- creating reality by His words, as He did in the creation in Genesis 1; That the purpose of His incarnation is to perform miracles of the new creation, miracles that testify to the presence of the Creator in His creation beginning in the new creation; That the result of His proclamation and miracles is release of those in bondage to demon possession (Lk 4:33-37), sickness (Lk 4:38-39), sin/sickness (Lk 5:17-26), and death (Lk 7.11-17; 8:40-56); That those included in this release are therefore those who are both in physical and spiritual bondage, namely, the poor, the captive, the blind, the broken ones; That His ministry as an ongoing release of the captives culminates on a cross where all sin, sickness, and death are placed upon Him and he finishes the work of the new creation; That the entire cosmos responds physically to this work of the new creation by the signs of darkness, earthquakes, and the dead rising in Jerusalem; That the resurrection is the great miracle that announces that in Jesus’ death and His resurrected body all of creation has been released from bondage; That the twelve (Lk 9:1-16) and the seventy-[two] (Lk 10:1-20) continue Jesus’ work of preaching and performing miracles during His ministry; That in Acts the apostles continue preaching and performing miracles, but that these are now focused in preaching (Word) and the new miracles of the new creation that testify to the bodily presence of the Creator in His creation bringing in the new creation (Baptism and the Lord’s Supper); That pastors as stewards of the mysteries continue the ministry of Jesus, the twelve, and the seventy-[two] through their administration of the Word and the Sacraments; That what Jesus promised the seventy-[two] he promises pastors today: “The one who hears you hears me, and the one who rejects you rejects me, and the one who rejects me rejects him who sent me.” (Lk 10:16); That this is what Paul was referring to when he announced to the Corinthians that apostles (and pastors) are “servants of Christ and stewards of the mysteries of God.” 2. Jesus’ table fellowship with sinners Jesus continued in His ministry the table fellowship of the Old Testament that was celebrated in its climactic form once a year at Passover, where the atonement was celebrated by Word and food, especially a sacrificed lamb, and every Sabbath evening seder that was a remembrance of Passover and the sacrifice of the Messiah promised in Genesis 3:15. Table fellowship at the time of Jesus always included two elements: teaching at the table and special food that climaxed the intimacy of the communion. With Jesus, however, there was a third element: His bodily presence during His ministry and His sacramental presence at the Last Supper, at Emmaus, and at Pentecost. During His Galilean ministry and His journey to Jerusalem, Jesus teaches and eats with His people, His local presence at the table; The climactic meal during this period is the feeding of the five thousand (Lk 9:10-17) where Jesus serves as host in feeding the people miraculous food; On the night in which He is betrayed, Jesus celebrates His Passover in the upper room (Lk 22:14-38) where He now interprets the entire Passover in terms of himself – His presence at the table is both local and sacramental – a great mystery; The Passover liturgy becomes the basis for our service of Holy Communion: Passover Seder I. The Preliminary Course Blessing -- CUP I Blessing -- Feast of Unleavened Bread (Passover) Teaching at the Table Setting the Table Interpretation of the Food Narration of the Exodus Blessing -- CUP II -- Cup of Redemption Blessing -- Bread Breaking the Bread of Affliction II. Meal III. The Grace After the Meal -- (Birkat Ha-Mazon) Blessing -- CUP III -- Cup of Blessing 1. Blessing for creation -- feeding 2. Thanksgiving for revelation -- land 3. Petition for redemption -- Jerusalem • After His passion and resurrection, Jesus teaches the disciples on the road to Emmaus. He creates burning hearts (the structure of our Word service – Lk 24:25-27) and opens their eyes in the breaking of the bread (structure of our Sacrament service – Lk 24:28-31). He is at table with them in both His glorified and His sacramental presence – another great mystery; • In Luke 24:35, the evangelist announces the structures of the liturgy -- Word and Sacrament: “And they were expounding the things in the way [teaching on the road – Word] and how he was know to them in the breaking of the bread [Sacrament];” • At Pentecost following Peter’s sermon in Acts 2:42, Luke reports that the apostles continued the table fellowship of Jesus with the same liturgical structures: “They were devoting themselves to the teaching of the apostles [Word] and [kai] the fellowship, that is, the breaking of the bread [Lord's Supper] and [kai] the prayers [Lord's Prayer – temple prayers];” • Through these structures of Word and Sacrament we now serve as “stewards of the mysteries;” • Jesus’ table fellowship is with sinners, i.e. it is an inclusive event; • Jesus’ table fellowship teaches about the kingdom; • Jesus’ table fellowship is itself an expression of the new era of salvation. 3. Jesus as the boundary between heaven and earth Every week pastors serve as stewards of the mysterious crossing of the threshold between heaven and earth in the Divine Service where heaven and earth come together through the bodily presence of Christ. This is truly a moment in which the church is “in betwixt and in between” heaven and earth. We move from our mundane lives of work and family into a space charged by the presence of the Creator who comes to us in both His divine and human natures as the crucified and resurrected Lord. As heaven and earth were joined together in Jesus on the mount of transfiguration (Lk 9:28-36) as his earthly disciples gathered alongside Moses and Elijah, so even now in Jesus heaven and earth are joined in our liturgy. - As Moses and Elijah broke in from heaven at the transfiguration, they were engaged in a heavenly conversation which, as the evangelist St. Luke records it, concerned Jesus’ exodus he was about to fulfill in Jerusalem (Lk 9:31). - This “exodus” is the great deliverance of the Israelites from slavery in Egypt through the shedding of blood. - The “exodus” is always about the blood from the Passover lamb that atones for sin and releases from slavery. - Israel’s “exodus” out of slavery into the promised land was imperfect, so that Jesus comes to do what Israel could not do. - That is why this exodus is none other than Jesus’ journey from heaven to earth and back to heaven through suffering, death, resurrection, and ascension. - As Moses and Elijah appear on the mount with Jesus and the disciples, their conversation from their heavenly place does not stop. They continue on the mount the conversation of heaven which, in the words of St. John’s Apocalypse, is about the Lamb who was slain and raised again. - This heavenly conversation of Moses and Elijah takes place around the eternal banquet table where a messianic feast celebrating the Lamb’s atonement and eternal victory is ongoing. - What John says in Revelation of the saints applies to Moses and Elijah and all who have died and risen in Christ, that is, that these are the ones whose names have been written “before the foundation of the world in the book of life of the Lamb that was slain” (Rev 13:8). - Even now we pray to our Father that he “gather us together from the ends of the earth to celebrate with all the faithful the marriage feast of the Lamb in his kingdom, which has no end.” - This ongoing liturgy is both in word and supper, where Christ is host and we are guests, and a lavish banquet of aged wine and choice marrow is spread out before us (Is 25:6). - Our participation even now in this liturgical feast is a participation in the ongoing proclamation of heaven that the Lamb is worthy of the new song he has given us to sing. - Therefore, the language of liturgy is heavenly even though it happens here on earth. 4. Jesus’ teaching about the apostolic administration of the mysteries Jesus’ final words to his disciples in Luke’s Gospel followed the institution of the Lord’s Supper (Lk 22:21-38). This final teaching is sometimes called Jesus’ “farewell discourse” as he gives his disciples instructions about their future life as apostles in the church. First, Jesus describes them as those who “have remained with me in my trials.” But when did that take place? Jesus, the omniscient Lord, likely has in view the entirety of His disciples’ lives, not just their lives up to this point. Moreover, especially in light of their imminent abandonment of Jesus at his arrest, it must be stressed that their faithfulness is only the product of his grace. He will forgive their unfaithfulness, and his own perfect obedience and faithfulness to his Father will be given to them through faith in Him. In light of this grace that forgives, restores, and strengthens disciples, they are viewed as loyal in their identification with Jesus even in His trials, and they will also inherit the kingdom he is about to receive. He appoints to them a kingdom now that is the same kingdom that is appointed for Him by the Father (Lk 22:29 -- “a kingdom,” is the object of both “I appoint,” and “appointed”). This is the same kingdom Jesus has preached and taught throughout His ministry. He instructed His disciples to proclaim that the kingdom is near (Lk 10:9, 11) and to pray for the kingdom to come (Lk 11:2). While the kingdom will come in all its fullness only at Christ’s return, Jesus will be enthroned in glory after His death and resurrection. The purpose for the establishment of that kingdom is communion with God, fellowship at the Lord’s table. Lk 22:30 is yet another reference to table fellowship in this scene: “in order that you may eat and drink at my table in my kingdom.” This dialog is closely associated with the Words of Institution, which came before. The eschatological dimension of this table fellowship of Jesus with His disciples links this section to the eschatological prospect of Lk 22:16, 18 and earlier words about eating in the kingdom of God (Lk 13:29–30 and the beatitude of Lk 14:15). This eating and drinking at His table in his kingdom already begins with Jesus’ post-resurrection meals, starting at Emmaus (Lk 24:28–35; see also John 21; Acts 1:4; 10:41). While Jesus will be seated at the right hand of the Father after His ascension, His kingdom is already inaugurated by His death and resurrection, since He rises in glory. The kingdom is present whenever and wherever the King is present, and so when Jesus reveals Himself in the breaking of the bread at Emmaus, the disciples realize that they have started to eat and drink at Jesus’ table in His kingdom. The apostles will be gathered, instructed, absolved, and commissioned into the Lord’s ministry. They will be appointed as stewards of the mysteries of God (1 Cor 4:1; cf. Lk 8:10; Eph 3:9), and they will serve to the church the eschatological Meal of the new era of salvation—the Lord’s Supper (Acts 2:42, 46; 20:7). At the Supper, the apostles—and those in subsequent generations in the apostolic ministry, as well as those to whom they minister—will dine at Jesus’ Table in God’s kingdom, anticipating the final end-time banquet (Is 25:6–9; Rev 19:6–9). Jesus’ use of the present tense “I appoint” (Lk 22:29) indicates his imminent action of appointing this kingdom for them. They are soon to be instructed in the nature of this kingdom on the road to Emmaus (Lk 24:13–27), and they will recognize the King when He breaks bread with them (Lk 24:30–31, 35). As stewards of the mysteries, pastors are responsible for continuing in their ministry the preaching and miracles of Jesus and the apostles. Jesus’ appointment of the twelve as stewards of this kingdom has great application to our role as stewards of the mysteries in the worship life of the church: - In the ministry of the apostles and of pastors today, the spirit of perseverance with Jesus in his death is apparent in preaching the scandal of his death as part of the proclamation of the kingdom. - The theology of the cross marks one who is appointed for the kingdom. - The scandal of the cross will become the center of the church’s proclamation in Lk 24:44–49 after Jesus clarifies the significance of His death to the Emmaus disciples (Lk 24:25–27). - Only after the resurrection do the disciples understand the kerygma. And only after Jesus teaches them and the Spirit comes upon them are they capable of proclaiming the kerygma and willing to suffer for it. - Only then will they start to be conformed to Jesus’ statement that they have continued with Him in his trials. - Suffering, preaching, the cross, eating and drinking at Jesus’ table—these all belong together for the apostles and us. - So also does “you will sit [future indicative] on thrones judging the twelve tribes of Israel” (Lk 22:30). This refers not to condemning old, unbelieving Israel, but to the responsibility of shepherding and oversight in the new Israel, the church. - This is the Lukan equivalent to Jesus giving to Peter the keys of the kingdom in Matthew (16:19) and His bestowal of the office of the keys upon the disciples in John (20:22–23). The OT judges were saviors and deliverers who led the people of God to repentance and faith (e.g., Judg 2:10–23; the LXX uses the verb “to judge,” for their leadership [e.g., Judg 10:2–3]). So too the apostles and pastors in the apostolic ministry will “judge” in the following way: “The Gospel requires of those who preside over the churches that they preach the Gospel, remit sins, administer the sacraments, and, in addition, exercise jurisdiction, that is, excommunicate those who are guilty of notorious crimes and absolve those who repent.” Entrusted with the responsibility to oversee the mystery of Christ’s real presence, to apply Law and Gospel, to absolve or retain sins, and to preside at the Supper, the apostles—and those who serve as their successors in Jesus’ ministry—will lead the new Israel to gather around the Table of the Lord in His Kingdom. Words from the Lutheran Confessions on Worship: Stewards of the Mysteries - The Lutheran Confessions establish the office of the ministry as the means by which faith is obtained through the Gospel and sacraments: “To obtain such faith God instituted the office of the ministry, that is, provided the Gospel and the sacraments. Through these, as through means, he gives the Holy Spirit, who works faith, when and where he pleases, in those who hear the Gospel. And the Gospel teaches that we have a gracious God, not by our own merits but by the merit of Christ, when we believe this.” (AC, V) - The Lutheran Confessions appeal to Luke 10:16 to affirm that pastors represent Christ in their office as stewards of the mysteries, and that this is not dependent on their worthiness for such an office: “According to divine right, therefore, it is the office of the bishop to preach the Gospel, forgive sins, judge doctrine and condemn doctrine that is contrary to the Gospel, and exclude from the Christian community the ungodly whose wicked conduct is manifest. All this is to be done not by human power but by God’s Word alone. On this account parish ministers and churches are bound to be obedient to the bishops according to the saying of Christ in Luke 10:16, ‘He who hears you hears me.’” (AC, XXVIII, 21-22) “In accordance with the Scriptures, therefore, we maintain that the church in the proper sense is the assembly of saints who truly believe the Gospel of Christ and who have the Holy Spirit. Nevertheless, we grant that the many hypocrites and evil men who are mingled with them in this life share an association in the outward marks, are members of the church according to this association in the outward marks, and therefore hold office in the church. When the sacraments are administered by unworthy men, this does not rob them of their efficacy. For they do not represent their own persons but the person of Christ, because of the church’s call, as Christ testifies (Luke 10:16), ‘He who hears you hears me.’ When they offer the Word of Christ or the sacraments, they do so in Christ’s place and stead. Christ’s statement teaches us this in order that we may not be offended by the unworthiness of ministers.” (Apol VII and VIII, 28) - The Lutheran Confessions also teach that, in pronouncing absolution, pastors are serving as stewards of the mysteries: “The power of the keys administers and offers the Gospel through absolution, which is the true voice of the Gospel. In speaking of faith, therefore, we also include absolution since --- 1 Paul uses the language of judgment when he warns the Corinthians that “any one who eat and drinks without discerning the body eats and drinks judgment on himself” (I Cor 11:29). 2 Treatise on the Power and Primacy of the Pope, 60, emphasis added. ‘faith comes from what is heard.’ as Paul says (Rom. 10:17). Hearing the Gospel and hearing absolution strengthens and consoles the conscience. Because God truly quickens through the Word, the keys truly forgive sin before him, according to the statement (Luke 10:16), ‘He who hears you, hears me.’ Therefore we must believe the voice of the one absolving no less than we would believe a voice coming from heaven. Absolution may properly be called a sacrament of penitence, as even the more learned of the scholastics say. Meanwhile this faith is nourished in many ways, amid temptations, through the proclamation of the Gospel and the use of the sacraments. These are signs of the new testament, that is, signs of the forgiveness of sins. Therefore they offer the forgiveness of sins, as the words in the Lord’s Supper clearly state, ‘This is my body which is given for you. This cup is the new testament’ (Luke 22:19, 20). So faith is conceived and confirmed through absolution, through the hearing of the Gospel, so that it may not succumb in its struggles against the terrors of sin and death. This understanding of penitence is plain and clear, it adds to the honor of the power of the keys and the sacraments, it illumines the blessing of Christ, and it teaches us to make use of Christ as our mediator and propitiator.” (Apol XII, 39-43) • The Lutheran Confessions affirm that the mystery of Christ’s presence in the Lord’s Supper is a Christological issue: “Thus our faith in this article concerning the true presence of the body and blood of Christ in the Holy Supper is built upon the truth and omnipotence of the true and eternal God, our Lord and Saviour Jesus Christ.” (FC SD VII, 106) “Therefore he is able and it is easy for him [Christ] to impart to us his true body and blood which are present in the Holy Supper, not according to the mode or property of the human nature but according to the mode and property of God’s right hand, as Dr. Luther says on the basis of our Christian faith as we teach this to our children. This presence is not mundane or Capernaitic although it is true and essential, as the words of Christ’s testament declare, “This is, is, is my body,” etc.” (FC Ep VIII, 17) “The one body of Christ has three different modes, or all three modes, of being at any given place. . . . The comprehensible, corporeal mode of presence, as when he walked bodily on earth and vacated or occupied space according to his size. . . . There is, secondly, the incomprehensible, spiritual mode of presence according to which he neither occupies nor vacates space but penetrates every creature, wherever he wills. To use some imperfect illustrations . . . light and heat go through air, water, glass, or crystal and exist without occupying or vacating space, and many more like these. He employed this mode of presence when he left the closed grave and came through locked doors, in the bread and wine in the Lord’s Supper, and, as people believe, when he was born of his mother . . . Thirdly, since he is one person with God, the divine, heavenly mode, according to which all creatures are indeed much more penetrable and present to him than they are according to the second mode. . . . I do not wish to have denied by the foregoing that God may have and know more modes whereby Christ’s body can be anywhere.” (Martin Luther’s Grosses Bekenntnis) (FC SD VII, 98-103) • The Lutheran Confessions affirm Ignatius of Antioch’s estimation that the Lord’s Supper is the “medicine of immortality,” for as stewards of the mysteries, the Lord’s Supper has tremendous applications for healing in our pastoral care: “We must never regard the sacrament as a harmful thing from which we should flee, but as a pure, wholesome, soothing medicine which aids and quickens us in both soul and body. For where the soul is healed, the body has benefited also.” (LC V, 68) “The Mass was instituted that the faith of those who use the Sacrament should remember what benefits it receives through Christ and cheer and comfort the anxious conscience. For to remember Christ is to remember His benefits, and to realize that they are truly offered to us . . . The Mass is to be used to this end, that there the Sacrament may be administered to them that have need of consolation, as St. Ambrose says: ‘Because I always sin, I am always bound to take the medicine.’” (AC XXIV, 30-33) Conversation (groups of 3-4) 1) Defend the statement: “The most important act of a pastor is to bring people into communion with the flesh of Jesus Christ.” How do we bring people into communion with Christ as stewards of the mysteries? 2) What is the benefit for our congregations to recognize that when they gather together for worship – entering into the holy presence of Jesus Christ, the creator of the universe, the Son of God crucified, raised, and ascended – they are entering into heaven even now? How might the notion that heaven is on earth affect the way we administer the mysteries of baptism and the Lord’s Supper? 3) What is the benefit for our congregations to recognize that they are not worshipping alone, but that through the presence of Christ in his divine and human natures, heaven itself is present, and so are all the saints, past, present, and future? 4) How can we help our congregations recognize the two structures in our worship through which Christ is present, corresponding to the means of grace: the living voice of Jesus in the service of the Word, and His body and blood in, with, and under bread and wine in the service of the Lord’s Supper? 5) How can we help our congregations understand that their salvation may never be separated from the flesh of Jesus Christ? 6) What are the dangers of misrepresenting the very Gospel itself, especially in our Protestant religious culture that does not hold to Chalcedonian Christology, so central to our understanding of Christ and our role as stewards of mysteries? See Paul in Galatians. 7) Agree or disagree: “It’s not that you preach or teach Jesus, it’s which Jesus you proclaim, the Jesus of our Protestant religious culture or the Jesus of the Scriptures, the ecumenical creeds, and Lutheran Confessions.” Defend your position in your group. 8) Why is it important, as stewards of the mysteries, that pastors affirm that when they absolve sinners, they stand in the stead and by the command of their Lord Jesus Christ, 9) Defend the statement: “At the Lord’s Supper, God in Christ is recreating the world and making all things new.” 10) How are pastors both stewards of the mysteries and, at the same time, stewards of the new creation? How does this impact our understanding of the gifts we receive in the creation itself? 11) What pastoral responsibilities today go with pastors fulfilling Jesus’ instructions to the twelve who were appointed a kingdom, present at every Lord’s Supper in Christ the king, that they “eat and drink at my table in my kingdom and sit on thrones judging the twelve tribes of Israel (Lk 22:30)? 12) What are the pastoral ramifications of teaching that there is both spiritual and physical healing present in the sacraments, especially in light of the church’s affirmation that the Lord’s Supper is the “medicine of immorality?” Prayer resources Since I am coming to that holy room Where with the choir of saints for evermore, I shall be made Thy music; as I come I tune the instrument here at the door, --- 3 Prayers are from Minister’s Prayer Book: An Order of Prayers and Readings edited by J. Doberstein (Philadelphia: Fortress Press, 1986) 127-145. And what I must do then, think here before. John Donne O merciful Lord, incline thine ears to our prayer and enlighten our hearts by the grace of thy Holy Spirit, that we may be worthy stewards of thy mysteries, and may love thee with an everlasting love; through Jesus Christ our Lord. Amen. Lord Jesus, we are not worthy that thou shouldst come under our roof, but in thy wondrous mercy come and be with us in this hour. Have mercy upon thy humble servant, and fill my soul with thy peace. Behold, I give heart and mouth and hand to thy service. By thy grace, use me as a lowly and unworthy instrument of thy mercy to the salvation of thy congregation. Amen. G. C. Dieffenbach O Lord, graciously accept my ministry at thine altar, that as a faithful steward of thy mysteries I may rightly administer thy body and blood to thy glory and the salvation of those who share in thy Supper. Amen. Pfarregebete zum Gottesdienst O most merciful God, incline thy loving ears to my prayers, and illuminate my heart with the grace of the Holy Spirit, that I may be enabled worthily to minister thy mysteries, and to love thee with an everlasting love, and to attain everlasting joys, through Jesus Christ our Lord. Amen. Gallican O Lord Jesus Christ, thou true and only High Priest, who on the altar of the cross has offered thyself to thy heavenly Father as a ransom for us poor sinners, and as a memorial of thy sacrifice has instituted this holy Sacrament, in which thou givest us thy Body to eat and thy Blood to drink: I beseech thee, for the sake of thy boundless love and mercy, to grant that I, thine unworthy servant, may celebrate this Sacrament with such devotion and fear as are acceptable to thee. I know indeed that on account of my sins and transgressions I am not worthy to approach thine altar, O Lord. But I know, too, and acknowledge and confess, that thou canst make me worthy, O thou who redeemest and savest sinners. Deliver me from all unholy, vain, and hurtful thoughts, that I may serve thee with a clean heart and in humble faith. Take this heart of stone, and give me a heart of flesh, that I may love and serve thee and have all my joy in thee. Bless unto me and to all who approach this holy Table, this salutary gift of thy Body and Blood, that it may be to us the strength of body and soul, keeping us in the paths of righteousness, so that we may finally be permitted to see thee face to face, and be admitted to the great Communion on high, when thou drinkest the fruit of the vine anew with thy disciples in thy Father’s kingdom. Amen. Lohe’s Liturgy Grant, O Lord, that what we have received with our mouth we may take and keep with believing hearts, that this temporal gift may help us to eternal salvation; through Jesus Christ our Lord. Amen. Leonine Sacramentary O Lord Jesus Christ, thou Everlasting Son of the Father, I give thanks unto thee, that thou hast given me, thine unworthy servant, grace to fulfill thy ministry and to distribute and receive the life-giving gift of thy Body and Blood. I beseech thee that this feast may not minister to the condemnation but to the salvation of those whom thou hast admitted to it. Grant that it may be unto me a shield of faith and the power of a right and holy life. Destroy in me all evil, and implant and nourish that which is good; subdue the passions and mortify the deeds of the flesh, so that I may ever cling to thee, and with an acceptable walk and conversation magnify thy name. And finally when my course on earth is ended, receive me into the habitations of light, to feast with the true Light and Joy of thine elect: O thou who livest and reignest with the Father and the Holy Ghost, one God, forever and ever. Amen. Lohe’s Liturgy Author: Rev. Dr. Arthur Just, Professor, Exegetical Theology, Dean of the Chapel Concordia Theological Seminary, Fort Wayne, IN email@example.com Bibliography and Resources Brunner, Peter. Worship in the Name of Jesus. Concordia Publishing House: St. Louis, MO, 1968. Danielou, J. The Bible and the Liturgy. University of Notre Dame Press: Notre Dame, Indiana, 1956. Doberstein, J. Minister's Prayer Book (Philadelphia: Fortress, 1959). Jenson, R. "How the World Lost Its Story," First Things 36 (October, 1993) 19-24. Just, A. A. Heaven on Earth: An Introduction to the Historic Liturgy of the Church (Upcoming in 2007, CPH). _________. "Liturgical Renewal in the Parish," Lutheran Worship: History and Tradition. Concordia Publishing House: St. Louis, MO, 1994. pp. 21-43. _________. "Structure, Culture, and Theology in Lutheran Liturgy," Worship in the Lutheran Congregation: The 90's and Beyond. Concordia Seminary Monograph Series, Symposium Papers, Number 1: Concordia Seminary, St. Louis, 1992, 9-31. Jeremias, Joachim. The Eucharistic Words of Jesus. SCM Press: London 3rd edition, 1966. An important interpretation of the anamnesis: note the revised edition. Kavanagh, Aidan. On Liturgical Theology. Pueblo: New York, 1984. Leon-Dufour, X. Sharing the Eucharistic Bread. New York: Paulist Press, 1987. Mannion, M. Francis. "Liturgy and the Present Crisis of Culture" in Worship 62 (1988). Peterson, E. Five Smooth Stones for Pastoral Work. Atlanta: John Knox Press, 1980. Pieper, Josef. In Tune with the World: A Theory of Festivity. Franciscan Herald Press: Chicago, 1965. Precht, F. L. Lutheran Worship: History and Practice. Concordia Publishing House: St. Louis, MO, 1994. Sasse, H. This is My Body: Luther's Contention for the Real Presence in the Sacrament of the Altar (Augsburg Publishing House: Minneapolis, MN, 1959). Schmemann, Alexander. Introduction to Liturgical Theology. St. Vladimir's Seminary Press: Crestwood, New York, 1966. _________. The Eucharist. St. Vladimir's Seminary Press: Crestwood, New York, 1988. Senn, F. C. Christian Liturgy: Catholic and Evangelical. Minneapolis, MN: Fortress Press, 1997. _________. New Creation: A Liturgical Worldview. Fortress Press: Minneapolis, 2000. Spinks, B. Luther's Liturgical Criteria and His Reform of the Canon of the Mass. Bramcote, Notts: Grove Books, 1982. Stevenson, K. The First Rites: Worship in the Early Church. The Liturgical Press: Collegeville, MN, 1989. Talley, Thomas J. The Origins of the Liturgical Year. Pueblo: New York, 1986. Wainwright, G. Eucharist and Eschatology. Oxford University Press, 1971. Webber, R. E. Ancient-Future Faith: Rethinking Evangelicalism for a Postmodern World. Baker Books: Grand Rapids, 1999. Willimon, Worship as Pastoral Care. Abingdon Press: Nashville, 1979. The Faithful & Afire Series In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to remain faithful as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church-Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness. A Case in Point “A Case in Point” is designed to get participants into a first conversation concerning an aspect of pastoral integrity. Read the following scenario and allow time for the total group to respond to the three questions that follow. Pastor Dan is in trouble. One of his members, Al, has spoken with him personally about a recent sermon. Al saw an almost identical sermon on a preaching resource website when he was looking for resources for a devotion he would lead. To Al’s shock it was almost word-for-word the sermon Pastor Dan had preached, only it was the work of another pastor. Al is an elder and a well respected leader of the congregation. He has been supportive of Pastor Dan’s ministry. Al went to Pastor Dan and expressed his disappointment that his pastor would preach another pastor’s sermon. Pastor Dan explained the tough week he had with two funerals, a wedding, and several folks hospitalized. Al was understanding and sympathetic. Now, though, it’s happened again -- a borrowed sermon and Al again catching his pastor doing it. Just about word for word, Pastor Dan preached a sermon on file at the same website. Al has met with Pastor Dan and suggested that he may bring it up before the elders, hoping they can address the matter in a positive, helpful way. Pastor Dan has expressed his remorse. Pastor Dan, as expected, isn’t sleeping too well in advance of the elders meeting. Truth is, he’s been borrowing sermons off and on for quite some time, sometimes not even paraphrasing or giving the sermon his own spin. He presented them as he found them. He has never given credit or referenced the sermon source. “What am I going to say?” he thinks to himself, “This sermon is compliments of sermoncentral.com?” He told himself (and still does) that the sermons are posted for that very reason, to be preached. Al’s comments, though, and the threat of being exposed to the elders have him worried. Not long ago, he heard about a pastor in Missouri who lost his ministry in a congregation for exactly this reason. Al put it this way: “Pastor Dan, I’m really disappointed. We called you to preach the Word from your heart, not someone else’s.” 1. What exactly is wrong about Pastor Dan’s recurring practice? 2. What goes on in a pastor’s psyche to allow preaching another pastor’s sermon? 3. If you were an elder in Dan’s church, what action would you recommend? The Scriptures Speak 1. 1 Samuel 2:12-36 tells the story of Eli the priest and his priestly family. Answer the following: - What were the sins of Eli’s sons, Hophni and Phineas? (See vss. 12-17, 22) Gluttony, greed, promiscuity, and a lack of respect for the ceremonial laws of God were among their sins. - How did Eli mismanage the sins of his sons? (See vss. 24-26) Though he rebuked them, he did not remove them from their priestly office. - What would be the outcome of Eli’s mismanagement of the priesthood and his family? (see 2:26-36 and 5:10-22) As predicted Eli’s sons would die and his priestly family would come to an end. 2. Jeremiah spoke God’s judgment on the prophets and priests prior to the exile. See Jeremiah 23:1-32. Where had these prophets gone wrong? They compromised their loyalty to the LORD by naming Baal in their prophecies. They led people astray. They created visions and prophecies in their own minds, claiming they were from the LORD. They failed to lead people to repentance. 3. Malachi 2 brings a strong rebuke on the post-exilic priests of Judah. In Chapter 1, the LORD called them down for offering less than pure animals for sacrifice. In Chapter 2:1-9, the rebuke continues. Look closely at these verses. In what various ways had the priests gone wrong, according to the LORD’s rebuke? Their heart was not set on honoring the LORD. In their teaching they caused many to stumble. They also showed partiality or favoritism in their judgments. 4. Read the lists of qualifications for an overseer (Greek, episcopos) in 1 Tim 3:1-7 and for an elder (Greek, presbyteros) in Titus 1:5-9. You may also want to look at Titus 2. In your large group share responses to the following: What specific word or phrase in either Biblical list leaves you feeling indicted as a pastor? Responses here will vary. Give a few pastors a chance to single out a word or phrase in the texts, indicting their own lack of integrity. - The office of the public ministry has certainly developed over the centuries. Here is a list of ethical hazards facing pastors today, often leading them into danger and failure where pastoral integrity is concerned. Note how the items often echo the concerns of the texts above: You may want to connect some lines between items on this list and the Biblical texts above. 1. Compromising the truth of God’s Word 2. Neglecting or mismanaging one’s own family 3. Laziness 4. Failing to take care of oneself, physically or spiritually (1 Cor 6:19) 5. Neglecting the primary mission of the church – making disciples 6. Using another’s words as one’s own (plagiarism) 7. Sexual immorality 8. Failing to equip the saints for ministry (Eph 4:11-13) 9. Neglecting the sick and shut-in 10. Refusing to do the work of an evangelist (2 Tim 4:5) 11. Breaking confidentiality in pastoral relationships 12. Lack of authenticity, a failing to practice what is preached 13. Neglecting catechesis: teaching the basic truths of Christianity 14. Bearing a grudge to the point of bitterness 15. Showing partiality or favoritism 16. Giving less than one’s best in the Lord’s service 17. Seeking honor for oneself rather than for the Lord (Jude 16) 18. Settling for less than excellence in ministry 19. Misusing/misallocating one’s time (Eph 5:15) 20. Causing or abetting chaos and disorder (1 Cor 14:40) Which three of the above seem to be most prevalent among pastors today? Responses will vary. Are any items missing from the list, in your opinion? What else threatens pastoral integrity today? Encourage adding other items to the pastoral integrity hazards list. 5. The Scriptures offer encouragement toward pastoral integrity. Identify pathways to greater pastoral integrity from these passages: You may wish to divide a larger group into smaller groups of 3-4 and assign one or two of these passage to each small group. Then ask each group to report on their findings regarding pathways to greater pastoral integrity. Another option is to read each passage aloud and ask participants in the large group to identify the pathway(s) to greater integrity in each passage. - Psalm 41:12 A close and constant walk with God builds and protects integrity. - Isaiah 6:1-8 In worship, in confession, in Word and Sacraments, we are cleansed and readied for ministry. - Acts 18:24-26 A willingness to grow and to be corrected by others saves us from false or incomplete teaching and practice. - Colossians 4:2-4 Seeking and asking for the prayer support of others supports and builds our integrity in ministry. - 2 Timothy 2:22-26 Intentionally seek Scriptural virtues for ministry, avoiding the latest self-help plans and opinions of others. - 2 Timothy 3:14-17 Actively use the Scriptures in ministry. - Titus 2:11-14 Lean on the empowering grace of God which allows us to say “No!” to ungodliness and worldly passions. - 3 John 1-8 Consistently seek the fellowship of others committed to faithful ministry. Words from the Wise - From "The Good Parson" in Geoffrey Chaucer's Canterbury Tales: To draw his flock to heaven with noble art, By good example, was his holy art. Nor less did he rebuke the obstinate Whether they were of high or low estate. For pomp or worldly show he did not care; Nor morbid conscience made his rule severe. The lore of Christ and his apostles twelve He taught, but first he followed it himself.$^1$ Chaucer recognized the power of authenticity or personal integrity in a spiritual leader. - From Luther's Preface to The Small Catechism: The deplorable, wretched deprivation that I recently encountered while I was a visitor has constrained and compelled me to prepare this catechism, or Christian instruction, in such a brief, plain, and simple version. Dear God, what misery I beheld! The ordinary person, especially in the villages, knows absolutely nothing about the Christian faith, and unfortunately many pastors are completely unskilled and incompetent teachers. Yet supposedly they all bear the name Christian, are baptized, and receive the holy sacrament, even though they do not know the Lord’s Prayer, the Creed, or the Ten Commandments! As a result they are like simple cattle or irrational pigs, and despite the fact that the gospel has returned, have mastered the fine art of misusing all their freedom. Shame on you forever! Therefore, my dear sirs and brothers, who are either pastors or preachers, I beg all of you for God’s sake to take up your office boldly, to have pity on your people who are entrusted to you, and to help us bring the catechism to the people, especially to the young.$^2$ Luther caught that sloppiness and neglect can have deep consequences for the church. The need for integrity is highlighted in the last paragraph as he speaks of “the people entrusted to you.” That makes pastoral ministry very much a matter of stewardship and accountability to God. - L.B. London, Jr. and Neil B. Wiseman on the hazard of sexual temptation and infidelity among pastors: Don’t miss this important reality: Pastors are especially vulnerable to outside emotional support during seasons of fatigue, frustration, and hopelessness. That’s why they must nourish every possible prevention dynamic that flows from a happy marriage. Whatever the tempting circumstances, infidelity is sinfully wrong and it completely sabotages the work of God. One morally bankrupt pastor ruins the credibility of a thousand and makes their work immeasurably more difficult. And those closest to the fallen minister in the family or the local church usually carry scars forever.$^3$ --- $^1$ Translated by H. C. Leonard as quoted in John Doberstein, Minister’s Prayer Book (Fortress), 344. $^2$ The Book of Concord, Robert Kolb and Timothy Wengert, eds. (Minneapolis: Fortress, 2000), 347-348. $^3$ From L.B. London, Jr. and Neil B. Wiseman, Pastors at Risk (Regal, 2003), 50. For London and Wiseman, the blessings of a “happy marriage” serve as preventative in times of stress or exhaustion, keeping pastors from sexual impurity. Note how seriously they take the consequences of a pastor’s sexual fall, the consciousness of which also works as a prevention tool. - From Eugene Peterson’s Working the Angles: The pastors of America have metamorphosed into a company of shopkeepers, and the shops they keep are churches. They are preoccupied with shopkeeper’s concerns – how to keep the customers happy, how to lure customers away from competitors down the street, how to package the goods so the customers will lay out more money. Some of them are very good shopkeepers. They attract a lot of customers, pull in great sums of money, develop splendid reputations. Yet it is still shopkeeping; religious shopkeeping, to be sure, but shopkeeping all the same. The marketing strategies of the fast-food franchise occupy the waking minds of these entrepreneurs; while asleep they dream of the kind of success that will get the attention of journalists. “A walloping great congregation is fine, and fun,” says Martin Thornton, “but what most communities really need is a couple of saints. The tragedy is that they may well be there in embryo, waiting to be discovered, waiting for sound training, waiting to be emancipated from the cult of the mediocre.” Some would say that American churches have moved from the commercial model (church as corporation or mall) described here to the entertainment model (church as rock concert or performance venue). Ask for pastors to share how the corporate model is still at work in their ministry and how it endangers their pastoral integrity. Conversation (Groups of 3-4) These conversations are designed for small groups of 3-4 pastors. Remind pastors of the need to honor confidentiality. Encourage them as well to offer absolution, if appropriate, for responses to Question No. 3. 1. Return to Pastor Dan’s situation as described in “A Case in Point.” When do we preachers cross the ethical line in regard to tapping into the sermons of another preacher? Preaching another’s sermon word-for-word neglects our pastoral responsibility and commits outright plagiarism. The good, hard work of living in the text ourselves, finding its law/gospel dynamic, and applying it to our people’s lives is crucial for authentic preaching ministry. 2. Still with Pastor Dan: What underlying issues are likely even more important to address than the plagiarism? Dan likely needs to do a time audit and reset his priorities, allowing adequate time and energy for sermon preparation. He may also need to read and take some continuing education courses in homiletics. 3. As you look at the list of 20 ethical hazards under No. 4 in “The Scriptures Speak,” choose one which has threatened or damaged your ministry in some way in the past. Share it with your group. Responses will vary. Encourage absolution if responses here take the form of confession. 4. No. 5 in “The Scriptures Speak” offers eight texts to encourage our pastoral integrity. Choose one and explain why it especially encourages you at this time in your ministry. Responses will vary. --- 4 Eugene Peterson, Working the Angles (Eerdmans, 1987), 2. The inside quote is from Martin Thornton, Spiritual Direction (Boston: Cowley, 1984), 7. 5. Look at the first quote in “Words from the Wise,” Chaucer’s description of “The Good Parson.” How important is authenticity to the pastoral ministry? Do we pastors today see ourselves as role models? Should we? Responses will vary. 6. Many of the sins attending the pastoral office are sins of omission as highlighted by Luther’s comments regarding what he saw lacking in the pastors of Reformation churches. What expectations of pastors (internally or externally stated) are unreasonable, setting pastors up for apparent failure? On the other hand, what basic responsibilities, neglected and unmet, point to our sinful omissions? Responses will vary. 7. LCMS district presidents as ecclesiastical supervisors have a no-tolerance policy toward a pastor’s sexual infidelity or immorality. Discuss why this is necessary. Do you agree that it is necessary? Is there a difference between being forgiven through repentance and faith and yet no longer qualified for the office of the public ministry? If so, on what does the difference turn? The crucial difference here turns on a pastor’s need to be “above reproach” (1 Timothy 3:7; Titus 1:7). Though forgiven for his sins, a pastor may no longer be eligible for ministry because his integrity in the office has been inalterably damaged. 8. Twenty years ago Eugene Peterson (in the last quote above) likened pastors to entrepreneurial shopkeepers chasing a misguided ecclesiastical version of commercial success. Do you still see this culturally driven model for success at work among pastors today? Exactly how is it misguided? In what ways have you bought into it? See the leader’s note with the quotation above. Responses will vary. 9. Share ways by which you stay morally centered in pastoral ministry. Where and to whom do you turn for greater pastoral integrity? Responses will vary. 10. Close with prayer in small groups for one another as individuals, tapping into comments made during the conversation. Encourage adequate time for prayer. Dean Nadasdy, Senior Pastor, Woodbury Lutheran Church, Woodbury, MN 3rd Vice President, LCMS firstname.lastname@example.org The Faithful & Afire Series The leader should make a copy of both the Participants’ Guide and the Leaders’ Guide for his use. The first 45 minutes should be spent in a guided Bible Study with the whole group, followed by 30-45 minutes of discussion in small groups of 3-4. There is more material here than can be covered in that time, however. This will allow the leader to choose what will be most appropriate for his group, but will give the participants more resources for further discussion. Please begin in prayer. In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue in the pastoral ministry and the temptation to compromise our confession. We need God’s grace daily to remain faithful as pastors. This study, the sixth in a series of Bible studies and conversations for pastoral circuits in the Lutheran Church – Missouri Synod, is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness, this month especially in confessing the faith. A Case in Point Read the paragraphs of the “Case in Point” included in the Participants’ Guide. Don’t spend too much time on this as it was designed simply to introduce the point that we are “confessing pastors” for the sake of the Gospel and so that our people are comforted in Christ. [See Participants’ Guide] Here is the reason we insist the pastor repeat his ordination vow in front of the people he will serve. He needs to be firmly grounded in what he will preach and teach. The people need to know what they should expect – nothing but God’s Word of Law and Gospel, centered in the cross of Jesus. Nothing less than faithful practice directing people to Jesus and His promises. Getting the Conversation Going Here we are dealing “up front” with the “yes, buts” some will have. “Yes, we are confessing pastors, but you don’t know my situation…” The purpose of this section is to emphasize that the confessing pastor is equally concerned for the truth and for his people. Our confession of the truth needs to draw people in rather than push them away. Choose 2 or 3 of the situations listed below to discuss, albeit briefly. You will not be able to discuss them all, but do emphasize the need to teach from the Word of God and to bring people along with love and care. If any of these topics are “too hot” to handle in your present group, set a time in the future when you can focus on them with Scripture and the Confessions. Yes, pastors are called to be faithful confessors of the truth, both within the congregation and before the world. Christ is the content of our confession. Scripture and the Lutheran Symbols form and shape our ministry. We belong to a Synod with a definite doctrinal position drawn from Scripture and our confession, and to all of that we are committed by our ordination vow. But what if the people aren’t looking for that, or don’t know what it is? What’s the best way to approach the situation should the pastor find that his new parish is weak in some area of doctrine or practice? Talk together in groups of 3 or 4 how you should respond if… 1. You find that the previous pastor had been practicing “open communion.” 2. You believe the worship practices of your new parish need to be changed. 3. The Baptismal Font is off in the corner somewhere where no one can see it. 4. Those who plan the contemporary service in your new parish consistently leave out the Creed and Lord’s Prayer. 5. Your congregation is a closed circle with little or no interest in outreach. What are the first three things you would do in these situations? How do we confess the truth in each case? There is, of course, a “ditch” on either side of the road. One school of thought would tell you – make your changes right away. Just do it and ask questions later. Present them with a fait accompli. Certainly, the Baptismal Font belongs front and center, so just move it! On the other side, pastors have been known to say, “Whatever the people want… It’s their church. I just have to live with the decision, even if it’s wrong. I’ll go along to get along.” Discuss together the Scriptural and confessional issues in the above cases. How does the pastor committed to Scripture and the confessions proceed? Here are some ideas to discuss: 1. Start on your knees in prayer. Ask good questions and then listen carefully. Find out what has been happening. What is the history? Did the problem develop because the people didn’t know any different and just went along with a previous pastor because they liked him? Discovering the reason why something happened will help you determine your best approach to resolving it under the Word of God. 2. Develop a plan and share that plan with your District President for his help and support. 3. Always teach, teach, and teach some more, ever patient and gentle, yet firm in the Word, before changes are made – even good ones! Otherwise, when you leave, they’ll just go back to what they were doing before. Remember, be patient. God’s Word works on God’s time-table, not ours. Not every problem is solved in two weeks. 4. Make sure your people understand the reason for the change. Much more can be accomplished when your people are with you in the change. 5. In every case, lead with the Word of God. The Scriptures Speak This segment is designed as a plenary guided Bible Study. You will not necessarily be able to cover all these Scriptures in this session. Perhaps choose those you believe are most applicable to the situation of your group. 1. Jesus gives several warnings and also some promises for confessors in Matthew 10:24-33. Warnings – verses 24, 25, 28 and 33 Promises – Verses 26, 30-30, 32 How do you take heed of the warnings? Of what comfort are Jesus’ promises? (In your discussion connect this Scripture to your own ordination vows.) You have taken a vow to confess Christ clearly before the world. What opposition do you experience? How do the promises of Christ help? 2. Read through 1 Timothy 4:1-16 together, taking special note of 6-7, 9-11, 14-16. How does being part of a confessional synod help you fulfill these instructions to pastors? The issue always is salvation (4:16). How does faithful teaching save? There is one way, and His name is Jesus! 3. Examine the qualities required of a pastor in Titus 1:5-9. Where do they come from? Verse 9 explains that a pastor “must hold firm to the sure word as taught, so that he may be able to give instruction in sound doctrine and also to confute those who contradict it” (1:9). How does your ordination vow and your training in Scripture and the confessions help you “hold firm” and “give instruction”? We have a definite body of teaching that constantly points to Jesus. What is the purpose of “sound teaching” anyway? Only Jesus saves, and we want to see all our hearers in heaven! 4. Galatians 6:1-2 gives both a command and a warning to all who have spiritual responsibility for others. Discuss how this applies to the pastor faced with situations similar to those detailed above. The sin against which Paul is warning is the sin of pride. We can never come to someone caught in a sin or a false teaching while we hold pride in our hearts. If we do, we are in grave spiritual danger ourselves. When our efforts to restore a brother or a congregation end badly, it is often because pride got in the way. 5. The Apostle Peter gives more instruction to pastors and to all Christians – for all believers are witnesses of what Christ has done. What are the admonitions and the promises in 1 Peter 3:13-18? Always be prepared… Christ died also for you! How are confessing pastors to live with their people according to 1 Peter 5:2-5? Not domineering over their people, but living in humility. What does such faithful humility look like? From where does it come? This comes only from the Lord, from a life lived in the Word of God and in prayer. See also Numbers 12:3. 6. Of all the issues raised in these Scriptures, which is the most urgent for you? Use this question as a wrap up for this section. The Purpose of the Lutheran Confessions Again, there are paragraphs (a personal account) in the Participants’ Guide that are not included here. The purpose of the account is to illustrate two points – 1) Lutheran theology and Biblical theology are one and the same. 2) The purpose of our confessional theology is to bring the greatest possible comfort to repentant sinners. Such a thesis, of course, begs for demonstration. Here take out your copy of the Book of Concord and refer to the Augsburg Confession for a basic list of Lutheran teachings. Take out the Book of Concord, or have ready copies of the Augsburg Confession, articles 1-13. In the Participants’ Guide, there is a paragraph describing briefly each of the following doctrines of the Lutheran confessions and how that doctrine brings comfort to the penitent sinner. If time is short, pick 2 or 3 to look at in greater detail. The Trinity [See Participants’ Guide] Original Sin [See Participants’ Guide] The Person of Christ [See Participants’ Guide] Justification [See Participants’ Guide] The Means of Grace and the Ministry [See Participants’ Guide] The New Obedience [See Participants’ Guide] The Church [See Participants’ Guide] Baptism [See Participants’ Guide] Confession and Absolution (and Repentance) [See Participants’ Guide] The Lord’s Supper [See Participants’ Guide] The Use of the Sacraments [See Participants’ Guide] Do you see how each article of Lutheran teaching focuses our attention on the Gospel’s comfort for sinners? Further Reflections Quickly work through the paragraphs included here in the Participants’ Guide. Emphasize the evangelistic nature of our confession. It is not to be hidden but confessed in every way possible before the world. Jesus alone gives life! [See Participants’ Guide] The Scriptures Speak Again Please be sure to cover these Scriptures. Here we have provided more comfort in Christ to “confessing pastors” who, if they are honest, will always feel inadequate to the task. 1. We have made much in recent years of Paul’s charge to Timothy (2 Timothy 1:6-7) that he “fan into flame the gift of God.” Check out also the corollaries that follow in 2 Timothy 1:8-14. List them here Verse 8 – Do not be ashamed… But share in suffering… Verse 9 – We are saved, not by our works, but by His grace in Jesus. Verse 10 (especially appropriate as we confess the resurrection by the grave side) – This grace is found only in Christ Jesus, crucified and raised from the dead for us. Verse 11 – For this Gospel Paul was appointed… Verse 12 – For the Gospel Paul suffers, but he is convinced Christ is able to guard until the last day what has been entrusted to him. What do you hear in the admonitions of 13-14? Follow the pattern… Guard the truth… Keep the confession pure, for the sake of the comfort it gives and so that you are reaching out with the truth. It must never be hidden, but always brought before the world. 2. Read 2 Corinthians 4:1-4. What comforts you? What motivates you and “gets you going”? We have this ministry by the mercy of God. That’s why we do not lose heart. In 2 Corinthians 4:5-6 Christ is the center of our proclamation. How do our Lutheran Confessions help us keep Him at the center? The goal of every teaching is that the light of Christ may shine for all those in the darkness of sin – us included! 3. What does Paul mean in 1 Corinthians 2:1-5? Obviously, the message of the cross must be central to everything. Here is an illustration: the cross is like the hub of a bicycle wheel, the axle around which everything turns. If you have spokes and rim but no hub, the wheel cannot hold together or turn. If the cross is pushed to the side, everything is out of balance. If we put something other than the cross at the center, we lose our comfort and our faith will begin to rest in “the wisdom of men” rather than the “power of God” in the cross of Christ. How is this done in your ministry? How do our confessions help you in this regard? The confessions are always cross centered and Christ focused – they always drive us back into the Scriptures. 4. It is, after all, a matter of life and death. Nothing is more urgent than confessing Christ clearly before the world. If you have any doubts about that, look at Galatians 1:6-12. What got Paul so “worked up” here? The Galatians were following those who added works to the grace of God for salvation. But such additions amount to a “different gospel,” a gospel that cannot save, and so is no gospel at all. Which do you think Paul would be more angry about? Someone who got the confession wrong or someone who failed to reach out with it in as many ways as possible?¹⁴ 5. Our confession is not just a sectarian screed, but our confession is “catholic” in the best sense of the term. We confess the heart of what the church has always believed and taught. How does Ephesians 4:1-6 form and shape this aspect of our confession? For 2000 years the church has confessed “one Lord, one faith, one Baptism,” etc. We Lutherans have always been part of that confession and the Lutheran confessors always saw themselves in the mainstream of that confession. 6. 2 Timothy 4:1-5 is familiar to us since these verses are part of nearly every ordination/installation service among us. Talk about this together. How do you take to heart Paul’s charge: “As for you, always be steady, endure suffering, do the work of an evangelist, fulfill your ministry” (4:5). Discussion here will reflect the needs of the group. 7. Are there other Scriptures to which you can point that inform our confession as Pastors? Talk about one of your favorites with your group. Words from the Wise Regarding the Purpose of Our Confession These quotes are provided as resources for the discussion to follow. - The [Smalcald Articles](#) on the various forms of the Gospel because, again, it’s all about the Gospel: We now want to return to the Gospel, which gives guidance and help against sin in more than one way, because God is extravagantly rich in his grace: first, through the spoken word, in which the forgiveness of sin is preached to the whole world (which is the proper function of the Gospel); second, through Baptism; third, through the holy Sacrament of the Altar; fourth, through the power of the keys and also through the mutual conversation and consolation of brothers and sisters. Matt. 18:20, “Where two or three are gathered,” etc.¹⁵ - The [Apology of the Augsburg Confession](#) on the nature of justifying faith This quote from the Apology illustrates our point that our confession is to give the greatest comfort for those who know their need for mercy. --- ¹⁴ Unfair question! Wouldn’t he be equally angry at both? ¹⁵ Smalcald Articles III.4, Book of Concord, Kolb/Wengert, p. 319. Therefore, whenever we speak about justifying faith, we must understand that these three elements belong together: the promise itself; the fact that the promise is free; and the merits of Christ as the payment and atoning sacrifice [propitiation]. The promise is received by faith; the word “free” excludes our merits and means that the blessing is offered only through mercy; the merits of Christ are the payment because there must be some definite atoning sacrifice for our sins. Scripture contains frequent pleas for mercy, and the holy Fathers often teach that we are saved through mercy. Therefore, every time mercy is mentioned, we must bear in mind that faith is also required, for it receives the promised mercy. Conversely, every time we speak about faith, we want the object [of faith] to be understood as well, namely, the promised mercy. For faith does not justify or save because it is a worthy work in and of itself, but only because it receives the promised mercy.$^{16}$ - The Preface to the Solid Declaration of the Formula of Concord explains the approach of the 16$^{th}$ century confessors to the Augsburg Confession. Ought not our desire to be to take the same approach? Our purpose here is to illustrate the attitude of our Lutheran forefathers toward the Augsburg Confession. Herewith we again whole-heartedly subscribe this Christian and thoroughly scriptural Augsburg Confession, and we abide by the plain, clear, and pure meaning of its words. We consider this Confession a genuinely Christian symbol which all true Christians ought to accept next to the Word of God, just as in ancient times Christian symbols and confessions were formulated in the church of God when great controversies broke out, and orthodox teachers and hearers pledged themselves to these symbols with heart and mouth. Similarly we are determined by the grace of the Almighty to abide until our end by this repeatedly cited Christian Confession as it was delivered to Emperor Charles in 1530. And we do not intend, either in this or in subsequent doctrinal statements, to depart from the aforementioned Confession or to set up a different and new confession.$^{17}$ - This is our church’s confession. It shows us how to read Scripture and it identifies what we believe about Scripture. It’s purpose is to help us keep the Gospel clear. That’s why we, also in these last days, desire to stand in the same line of faithful confessors and fathers in the faith who prepared this confession. As confessing pastors, we also “pledge ourselves to the prophetic and apostolic writings of the Old and New Testaments as the pure and clear fountain of Israel, which is the only true norm according to which all teachers and teachings are to be judged and evaluated.”$^{18}$ - Again, our fathers in the faith point to the Augsburg Confession: [This quote is not included in the Participants’ Guide but is offered here as a further reference.] We therefore declare our adherence to the first, unaltered Augsburg Confession (in the form in which it was set down in writing in the year 1530 and submitted to Emperor Charles V at Augsburg by a number of Christian electors, princes, and --- $^{16}$ Apology IV 53ff, Book of Concord, Kolb/Wengert, p. 128f. $^{17}$ Tappert, T. G. (2000, c1959). The book of concord : The confessions of the evangelical Lutheran church (The Formula of Concord: 2, 4-5). Philadelphia: Fortress Press, p. 502. $^{18}$ Tappert, pp. 503f. estates of the Roman Empire as the common confession of the reformed churches) as our symbol in this epoch, not because this confession was prepared by our theologians but because it is taken from the Word of God and solidly and well grounded therein. This symbol distinguishes our reformed churches from the papacy and from other condemned sects and heresies. We appeal to it just as in the ancient church it was traditional and customary for later synods and Christian bishops and teachers to appeal and confess adherence to the Nicene Creed.\textsuperscript{19} C.F.W. Walther insists on a \textit{quia} subscription to the confessions because any other approach will rob the church of its symbolical books: \textit{The following two quotes from C.F.W. Walther’s famous 1879 essay on the Duties of an Evangelical Lutheran Synod show the importance for the proclamation of the Gospel of a \textit{quia} subscription to the confessions – because they are in agreement with Scripture.} For if those who pledge themselves to the confessions still retain the freedom to choose what they like, then there is no point in saying, “yes, I profess the same faith you do.” For I still have no idea what he is professing, if he does not say, “from its first sentence to its last one, the Lutheran Confession is my confession.”\textsuperscript{20} He explains that our Synod requires our pastors to swear to the confessions and testify: “I have checked them against Scripture and have found that they are in perfect agreement with Scripture, that they do not claim a single doctrine that is not taken directly from Holy Scripture.” Such a person need not, because of conscience scruples have to swear [to the confessions] “in so far as,” because you see, He himself has determined that they are in full agreement. Therefore he vows, “Because they are in perfect agreement with Holy Scripture.”\textsuperscript{21} Eugene Peterson is a Presbyterian pastor, but I believe he’s got it right in this passage from Working the Angles on why we take a vow at ordination, and is worth quoting at length: \textit{Included here is the full quote from Peterson (the first two paragraphs are not in the Participants’ Guide) in order to illustrate that we subscribe to our confessions and vow loyalty to the same, for the sake of the Gospel and for the sake of the people.} [C]entury after century Christians continue to take certain persons in their communities, set them apart and say, “We want you to be responsible for saying and acting among us what we believe about God and kingdom and gospel. We believe that the Holy Spirit is among us and within us. We believe that God’s Spirit continues to hover over the chaos of the world’s evil and our sin, shaping a new creation and new creatures. We believe that God is not a spectator in turn amused and alarmed at the wreckage of world history but a participant in it. We believe that everything, especially everything that looks like wreckage, is material God is using to make a praising life. We \textit{believe} all this, but we don’t see it. We \textsuperscript{19}Tappert, p. 504. \textsuperscript{20}C.F.W. Walther, \textit{On the Primary Duties Incumbent on a Synod that Wants Rightly to Be Considered an Evangelical Lutheran Synod}, a doctrinal essay presented to the constituting convention of the Iowa District of the Missouri Synod in 1879, in \textit{Essays for the Church}, Volume 2 (St. Louis: Concordia Publishing House, 1992), p. 10. \textsuperscript{21}Walther, \textit{On the Primary Duties}, p. 18. see, like Ezekiel, dismembered skeletons whitened under a pitiless Babylonian sun. We see a lot of bones that once were laughing and dancing children, of adults who once made love and plans, of believers who once brought their doubts and sang their praises in church — and sinned. We don’t see the dancers or the lovers or the singers — at best we see only fleeting glimpses of them. What we see are bones. Dry bones. We see sin and judgment on the sin. That is what it looks like. It looked that way to Ezekiel; it looks that way to anyone with eyes to see and a brain to think; and it looks that way to us. “But we believe something else. We believe in the coming together of these bones into connected, sinewed, muscled human beings who speak and sing and laugh and work and believe and bless their God. We believe that it happened the way Ezekiel preached it and we believe that it still happens. We believe it happened in Israel and that it happens in the church. We believe that we are part of the happening as we sing our praises, listen believably to God’s word, receive the new life of Christ in the sacraments. We believe that the most significant thing that happens or can happen is that we are no longer dismembered but are remembered into the resurrection body of Christ. “We need help in keeping our beliefs sharp and accurate and intact. We don’t trust ourselves — our emotions seduce us into infidelities. We know that we are launched on a difficult and dangerous act of faith, and that there are strong influences intent on diluting or destroying it. We want you to help us: be our pastor, a minister of word and sacrament, in the middle of this world’s life. Minister with word and sacrament to us in all the different parts and stages of our lives — in our work and play, with our children and our parents, at birth and death, in our celebrations and sorrows, on those days when morning breaks over us in a wash of sunshine, and those other days that are all drizzle. This isn’t the only task in the life of faith, but it is your task. We will find someone else to do the other important and essential tasks. This is yours: word and sacrament. “One more thing: we are going to ordain you to this ministry and we want your vow that you will stick to it. This is not a temporary job assignment but a way of life that we need lived out in our community. We know that you are launched on the same difficult belief venture in the same dangerous world as we are. We know your emotions are as fickle as ours, and that your mind can play the same tricks on you as ours. That is why we are going to ordain you and why we are going to extract a vow from you. We know that there are going to be days and months, maybe even years, when we won’t feel like believing anything and won’t want to hear it from you. And we know that there will be days and weeks and maybe even years when you won’t feel like saying it. It doesn’t matter. Do it. You are ordained to this ministry. Vowed to it. There may be times when we come to you as a committee or delegation and demand that you tell us something else than what we are telling you now. Promise right now that you won’t give in to what we demand of you. You are not the minister of our changing desires, or our time-conditioned understanding of our needs, or our secularized hopes for something better. With these vows of ordination we are lashing you fast to the mast of word and sacrament so that you will be unable to respond to the siren voices. There are a lot of other things to be done in this wrecked world and we are going to be doing at least some of them, but if we don’t know the basic terms with which we are working, the foundational realities with which we are dealing — God, kingdom, gospel — we are going to end up living futile, fantasy lives. Your task is to keep telling the basic story, representing the presence of the Spirit, insisting on the priority of God, speaking the biblical words of command and promise and invitation.” That, or something very much like that, is what I understand the church to say to the people whom it ordains to be its pastors.\textsuperscript{22} We live only by the cross and resurrection of Jesus. And as pastors we also live and serve for the sake of this Gospel in Word and Sacrament. As Paul writes, “It is no longer I who live, but Christ lives in me, and the life I now live in the flesh I live by faith in the Son of God who loved me and gave himself for me” (Galatians 2:20). \textbf{Conversation (Groups of 3-4)} \textit{You may want to ask a pastor in each group to serve as leader, whose primary task is to be sure everyone has an opportunity to speak. Participants may decide to focus on only 2-3 of the questions or to raise questions themselves. Be sure to encourage prayer for one another. Be sure to emphasize that our confession is always for the sake of Christ’s mission.} 1. Share with one another. Why are you a \textit{Lutheran} pastor? How did you come to this conviction? What ignites your passion? 2. One homiletics professor told his seminary class that pastors should re-read \textit{Apology IV} every six weeks. This may be a bit of hyperbole, but do you see value in regular reading of the confessions as part of your study program? How would you begin? 3. Talk about the false dichotomy between “confession” and “outreach.” How does our confession form, shape and motivate our outreach? 4. Above all else, the \textit{Augsburg Confession} is concerned for the comfort of sinners. How does that help you in your preaching and pastoral care? 5. Respond to the Eugene Peterson quote. How well does this reflect your understanding of your own ordination vows? 6. Our Synod has experienced division in several areas of teaching and practice. What are the issues you struggle with in your circuit? In light of Walther’s comments, how might an honest look at our confession help resolve some of these? 7. Living in our confession, how would you approach as pastor a parish that you believe is deficient in its doctrine or practice?\textsuperscript{23} 8. Living in our confession, how do you and your people reach out with the Gospel faithfully? How do you think our confession can help your congregation live and serve more consciously as a “mission outpost”? See also the \textit{Apology} quote above. \textsuperscript{22} Eugene H. Peterson, \textit{Working the Angles: The Shape of Pastoral Integrity} (Grand Rapids, Michigan: William B. Eerdmans Publishing Company, 1987), pp. 23-25. \textsuperscript{23} One way of proceeding: One of the parishes I served had the reputation before I came of being involved in ecumenical worship events with pastors of other denominations and also the reputation of having “open communion.” Before accepting the call, I talked it over with the Board of Elders. I knew that we would need to work toward change in those areas. They knew there would be differences and that we would not always agree, but they were willing to listen and to be taught. They also agreed it would not be good to make me go against my conscience. So what was our approach? 1. I told them we could not do ecumenical services and explained the reasons why (truth and error cannot be given an equal billing in the same service). 2. There were several on-going situations in which people who were not members of the LCMS were being regularly communed. 3. We taught the reasons for a proper practice of the Lord’s Supper. 4. Over the course of nearly two years we patiently worked through each of the situations individually. Some were instructed and confirmed. Another was examined and received by affirmation of faith. Others needed to choose between the Lutheran altar and that of another denomination. 5. In every case, we were able to explain the Word of God and allow it to go to work. Over the course of time, the congregation was drawn by the Word of God into a more faithful confession and practice. 9. Run through the basic list of Lutheran doctrines. Discuss again how each one is designed to bring maximum comfort to penitent sinners. How about more difficult teachings such as “predestination”? (See Ephesians 1:3ff). 10. What keeps us from reaching out with this confession more than we do now? 11. What will you do differently as a result of this study? 12. Be sure to lift up one another in prayer. Seek to hold each other accountable in your circuit to your decisions as a result of #10. “Finally, brothers, pray for us, that the Word of the Lord may speed on and triumph, as it did among you … The Lord is faithful; He will strengthen you and guard you from evil … May the Lord direct your hearts to the love of God and to the steadfastness of Christ” (2 Thessalonians 3:1,3,5). Rev. Herbert C. Mueller, Jr. – Waterloo, Illinois President, Southern Illinois District, LC-MS email@example.com --- 24 If you think the question unfair, ask why some seem to “water down” the confession and others seem to “hide it” in some way – or are these both caricatures? The Faithful and Afire Series In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to remain faithful as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church – Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness. Your task as a leader is to facilitate conversation and to maintain enough safety so that people can speak their mind from their heart. Depending on the number of persons present a decision can be made in every section whether to conduct the session in plenary or in small groups. If small group discussion is used, it is helpful to have a time when brief summaries of the conversations can be given to the larger plenary group. If sensitive material is discussed in the small groups, only the summary of the content should be shared. The area of prayer and devotional life is an obviously important area. The weakening of one’s own personal spiritual life and behavior is an occupational hazard of ministry. Since pastors are always using the scriptures and always praying for others, personal application and resonance is harder. The issue here is not strength of faith, for faith is a gift of God through Christ. The issue here is developing practices of the faith that are wholesome, personally strengthening, and that give support to fighting the good fight. Feedback to the author concerning the use and helpfulness of this material is encouraged and requested. A Fundamental Point (Before the Case in Point) “A Fundamental Point” is designed to help participants engage in a conversation concerning the nature of one’s personal spiritual prayer and devotional life. Read the following essay and allow time for the total group to respond to the questions that follow. The fundamental point is that the question is not whether there will be temptation, but rather what the temptation is; not whether there will be spiritual vulnerability for the pastor, but rather what that vulnerability is. Prayer, meditation, and temptation are all expected parts of the development of a spiritual and devotional life. It is possible to make the case that the weakening of one’s own spiritual practices is part of the “tentatio” described. “What’s more, I would like to show you the right way of studying theology, for I myself have practiced it…This is the way that holy King David…teaches in Psalm 119. In it you will discover three rules, abundantly explained throughout the whole Psalm. They are prayer (oratio), meditation (meditation), temptation (tentatio). First you must realize that the Holy Scriptures are the kind of book that turns the wisdom of all other books into folly, because none of them can teach about eternal life except this alone. So you should immediately despair of your own reason and understanding. With them you will not attain eternal life. Instead, with arrogance like that, you will hurl yourself, and others with you, from heaven, like Lucifer, into the abyss of hell. Rather, kneel down in your room and pray to God, with true humility and earnestness, that through his dear Son, he would give you his Holy Spirit, to enlighten you, lead you, and give you understanding. You can see how David keeps on praying in Psalm 119, “Teach me, Lord, instruct me, lead me, show me” and so on. Even though he knew the text of the Pentateuch well, and many other books, heard and read them daily, he wanted to have the true teacher (master) of the Scriptures for himself as well, so that he would not tackle them with his understanding and become his own teacher (master). That produces spiritual rabble-rousers who fancy that the Scriptures are subject to them and are readily grasped with their own understanding, without the Holy Spirit and prayer, like the tales of Markolf, or Aesop’s fables. Secondly, you must meditate, not only with your heart, but also externally, by always studying and rubbing, reading and re-reading the spoken word and written text in the Bible, with diligent attention and reflection on what the Holy Spirit means in it. Take care that you do not become bored and think that if you have read, heard, and spoken it once or twice that is enough for you to understand it fully. You will never become much of a theologian like that, but will be like immature fruit that drops down before it is half-ripe… Thirdly, there is temptation (tentatio), ‘Anfechtung’. This is the touchstone that teaches you not only to know and understand, but also to experience how right and true, how sweet and lovely, how powerful and comforting God’s word is, wisdom above all wisdom. Thus you see how David, In Psalm 119, laments so often about all the different enemies, arrogant princes or tyrants, and about all the false spirits and hordes that he has to suffer just because he meditates; that is, because he deals with God’s word, as we have said, in many different ways. For as soon as God’s word shoots up and spreads through you, the devil persecutes you. He makes you a true doctor (of theology); through his temptations, he teaches you to seek and love God’s word.” (Martin Luther, “Preface to the Wittenberg Edition”, revised translation by John W. Kleinig based on LW 34, 285-287 in Preparing Lutheran Pastors For Today, Editora da ULBRA, 2006, pp. 33-35, italics mine) Getting the Conversation Going 4. How does this passage from Luther help us understand the particular vulnerabilities of a pastor related especially to his devotional and spiritual life? This question is phrased in a general way so that conversation can be more theoretical. 5. How does this passage from Luther help us understand our own personal, particular vulnerabilities as we consider our own devotional and spiritual life? This question is phrased in a specific and personal way so that conversation can be more personal and existential. 6. How do you understand Luther’s apparent point that, as God’s Word grows in us, our level of persecution by Satan also grows, and that through that persecution or temptation we even more passionately “seek and love” God’s Word? This question specifically addresses the nature of temptation, and the reality that temptation propels us to the cross of Christ. We rely on Christ alone. (The Bonhoeffer and Kleinig quotations in “Words From the Wise” might also be used here.) A Case in Point “A Case in Point” is designed to move from the somewhat more general discussion into a specific scenario with which it is hoped the participants can identify. Read the following scenario and allow time for the total group (or smaller groupings) to respond to the questions that follow. Pastor George sees himself as less spiritual fit than he saw himself a decade ago. He came to that conclusion as he looked at his open Bible for the fifth time. “I really try to read faithfully, and I often do read, but it seems not to go anywhere – at least anywhere for my own life. I get the exegesis and I get the points to preach, but somewhere increasingly my heart and my soul are being missed” he mused to himself. Slowly he closed the Bible, put it down, and joined his spouse for a night-time cup of tea. Susan, George’s spouse, could sense his weariness, but she herself was pretty weary. Couple devotions had become perfunctory; her own spiritual life seemed not very alive. Very different this is than the spiritually afire sense they had in their early years. They stared into each other’s eyes, each wondering if the words could be formed to begin what seemed like a necessary discussion. “Ah, it is late” Susan said, “and I am tired.” They both trudged off to bed. Getting this Conversation Going 4. What might have happened or be happening in the lives of Susan and George that developed this trajectory in their spiritual lives? This question gives participants an opportunity to speculate about causes of the spiritual malaise described. Be aware that some speculations in response to stories might have a bit of personal connection. 5. How might this be seen as a “tentatio”, i.e., that this is a piece of the persecution of Satan that eventually neutralizes the effectiveness of the pastor? If Satan had a game plan, would it not be to neutralize the effectiveness of a pastor and his spouse, and if this neutralization were to be accomplished, would it not be a reasonable strategy to attack the spiritual life of the pastor and his spouse. 6. If George and Susan came to you for pastoral care, what might you say to them? What might you encourage them to do? Here pastors can consider responses of pastoral care. These responses can be used later to discuss overall personal strategies for spiritual growth. The Scriptures Speak 2. Paul’s charge to Timothy is focused in 1 Timothy 6: 11-16. The exhortations of Paul to Timothy are clear. “Fighting the good fight of faith” is a central theme. To what extent is attention to the pastor’s personal spiritual life part of the “fighting the good fight of faith”? Consider the following questions: - What support is necessary for pastors to fight the good fight of faith? Answers should vary here. Certainly being in the Word and participation in the Eucharist are central. Work to dig deeper than these standard answers, though, so that, hopefully, ideas such as the mutual consolation and admonition of the brethren, small group Bible study, private confession and absolution, use of a spiritual director, and the like might emerge. - What helps are necessary for the pursuit of “righteousness, godliness, faith, love, endurance, and gentleness”? Some of the responses here might repeat and/or build on the previous ones. How does a pastor’s devotional and spiritual life support fighting the good fight of faith? Be specific about what devotional practices you have found useful to help in the fight? The assumption here is that a pastor’s devotional and spiritual life, when healthy, does support fighting the good fight. This is an opportunity for participants to share devotional techniques, strategies, and activities that have been helpful to them. 3. Among the many possible selections from Psalm 119, consider Psalm 119:105-112. This is an ideal for personal devotion and the development of one’s spiritual life. “I will follow your righteous laws” (119:106b), “I will not forget your law” (119:109b), “Your statutes are my heritage forever; they are the joy of my heart. My heart is set on keeping your decrees to the very end.” (119:111-112). Here the emphasis is on the following of God’s laws as a central theme of one’s spiritual life. Or, consider Psalm 119: 97-104. This active approach to personal devotion and the development of one’s spiritual life is clear: “Oh, how I love your law! I meditate on it all day long” (119: 97) “How sweet are your words to my taste, sweeter than honey to my mouth” (119: 103). Here the emphasis is on the meditation and study of God’s Word, with all the wonderful positive motivation involved. Consider the following questions: - What picture do these portions of Psalm 119 paint of the pastor as a follower of Yahweh? Psalm 119 was selected because of its connection to the earlier quotation from Martin Luther. The pastor is to have all the characteristics listed. - What are the ways that the picture painted is one of Law for the pastor as different than Gospel? For instance, have you had times in which the study of God’s Word has not been “sweet”? If the answer is “yes”, where can you turn? The reality is that no one can perfectly live as the picture suggests. This can lead us into despair. For instance, if a pastor is having difficulty being regular and enthusiastic about his personal Bible study, having a picture that he should be regular and enthusiastic is Law. In all things, we turn to Christ and through Christ to the members of His Body. 4. The place of the community is central to the devotional and spiritual life of the pastor. Consult James 5: 13-19. At times prayer and praise is an individual response. At other times, it is a response within the community. Mutual confession of sins and prayer for one another is considered to be the vehicle God in Christ uses for healing. Consider the following questions: - Under what conditions might you “call the elders,” as colleagues in ministry, or the circuit counselor, or others in the community of the Body of Christ? Answers should vary here. However, the central theme that is hoped for is that pastors need to develop others who can be helpful, supportive, and provide accountability. - How can bringing people together in spiritual care and community devotion aid the pastor’s devotional and spiritual life? The community provides support and accountability. It is important for the pastor to be able to take his internal struggles, concerns, satisfactions, etc. and make at least some of them external. In that way others participate in the development of the pastor’s prayer and devotional life, among other things. - How does the community develop both support and accountability? Communities do this by working to develop themselves as safe spaces where it is possible to bring prayer, meditation, and temptation. - How can circuit gatherings and conferences be places where prayer specifically for each other, confession to one another other, and songs of praise be part of the very fabric of the gathering? This is an opportunity for the participants to address directly the nature of the circuit gathering and how it can be used to foster and strengthen the spiritual lives of the pastors, their spouses, and their children. 5. Finally, read Psalm 67. The end of all of personal devotion and spiritual life is the praise of God. Even when we as pastors do not “feel” like praising, praise of God for God’s love and care in the sending of the Christ to redeem us is a core devotional response. Consider the following questions: - What interferes with our praise of God in Christ? Be specific. This question is phrased is a more general way (“a pastor’s praise …) so that conversation can be more theoretical. - What interferes with your praise of God in Christ? Be specific. This question is phrased in a more personal and specific way so that answers can be more personal and existential. Allow the conversation to deepen if that is its natural development. - What strategies can you develop to work through the interferences? Encourage specific behaviors that can be done, as contrasted with general statements of attitude. The more specific participant’s can get, and the more the circuit brothers can participate in a supportive way, the more likely, empowered by God’s Holy Spirit, an intentional strategy can become a healthy and growthful behavior. - How can your circuit coming together be a place where both tentatio and praise exist together as part of the general rhythm of the gathering? Here is an opportunity to make specific plans about the function of the circuit meetings. Words from the Wise “Temptation is a concrete happening which juts out from the course of life. For the physical man all life is a struggle, and for the moral man every hour is a time of temptation. The Christian knows hours of temptation, which differ from hours of gracious care and preservation from temptation as the devil is different from God. The saying that every moment of life is a time of decision is for him a meaningless abstraction. The Christian cannot see his life as a series of principles, but only in its relation to the living God. The God who causes day and night to be gives also seasons of thirst and seasons of refreshment; he gives storms and peace, times of grief and fear, and times of joy. ‘Weeping may tarry for the night, but joy cometh in the morning’ (Ps. 30.5). ‘To every thing there is a season, and a time to every purpose under the heaven: a time to be born, and a time to die; a time to plant, and a time to pluck up that which is planted; a time to kill, and a time to heal; a time to break down, and a time to build up; a time to weep, and a time to laugh . . . . He hath made every thing beautiful in its time’ (Eccles. 3.1-4, 11). It is not what life may be in itself, but how God now deals with me, which matters for the Christian. God rejects me, and he accepts me again. He destroys my work, and he builds it up again. ‘I am the Lord, and there is none else. I form the light, and create darkness; I make peace, and create evil’ (Isa. 45.7). So the Christian lives from the times of God, and not from his own idea of life. He does not say that he lives in constant temptation and constant testing, but in the time when he is preserved from temptation he prays that God may not let the time of temptation come over him.” (Dietrich Bonhoeffer, Temptation. London: SCM Press LTD, 1955, p. 11) “Spiritual growth and development takes discipline. The person who is ordained to the parish ministry is expected to be a spiritual giant and an expert on all spiritual matters. Yet more than likely, he entered seminary as part of a spiritual quest or journey. He was looking for some answers to life’s deepest questions. At seminary he was supposed to develop spiritually by studying Bible, theology, and church history. In reality, any spiritual maturity he gained probably happened between academic courses and chapel attendance. The final illusion was that parish work would continue that process of spiritual development. Instead parish work often drains one spiritually. At this point, we realize that we have not been given all the training and skills needed to take even the first steps on our spiritual journey. Spiritual maturity usually comes about as a result of some disciplined work in the field. Parishioners expect sermons to have spiritual depth to them, yet few agree to have their pastor spend significant time in personal reflection, prayer, Bible reading, and meditation. They assume that their pastor has done this kind of inner work already. Clergy generally find it difficult to find time for their own personal/spiritual development. Perhaps they themselves think it would be selfish to spend time in this area. “That’s not what I am paid to do,” they may think. Yet, in reality, a strong argument could be made that a clergyperson’s spiritual life should be his first job responsibility.” (Roy M Oswald, New Beginnings: A Pastorate Start Up Workbook, The Alban Institute, 1989, p. 61) “The Gospel should predominate in Christian preaching. Says Walther: ‘Worst of all a preacher sins against his duty to preach the whole counsel of God, and more than in any other way will his hearers be made to suffer, if he preaches more Law than Gospel, or, expressed in other words, if the Gospel does not predominate in his sermons, if the comforting doctrine of justification of a poor sinner by grace, through faith in Jesus Christ, without the works of the Law, is not the outstanding feature of all his sermons. A preacher should be able to say with Paul: “God also hath made us able ministers of the New Testament; not of the letter, but of the spirit; for the letter killeth, but the spirit giveth life”’ (2 Cor. 3:6). A true preacher should ‘do the work of an evangelist’ (2 Tim. 4:5), and should therefore consider it his chief duty to be a witness unto Christ (John 15:27; Acts 1:8). A preacher who chiefly moralizes and has not himself yet experienced the power of the Gospel, but is yet a servant under the Law, has not the courage to preach the Gospel in its fullness and cannot give all its rich comfort to his hearers; a preacher who is rather troubled that by giving the full comfort of the Gospel he will make people secure and lead them into hell and therefore so limit the Gospel that poor sinners do not risk boldly to take a hold; a preacher who, as often as he speaks of faith, at once adds all manner of warnings against self-deception and against believing too soon and is not wholly concerned so to preach that the heart of his hearers be filled with faith: such a preacher indeed may believe that his way is the sure way of not distorting the Word of God and of not neglecting the care of souls entrusted to him, but actually such a preacher is guilty of distorting the Word of God and of sadly and inexcusably neglecting the welfare of souls dearly bought by the blood of Christ. A preacher who only sparingly dispenses the comfort of the Gospel and lets the Law predominate in his preaching does not truly encourage faith and a real Christian life; he, moreover, is just in that way preventing it. A true preacher should be able to say with Luther: ‘In corde meo iste unus regnat articulus, scilicet fides Christi; ex quo, per quem, et in quem omnes meae diu noctuque fluant et refluent theologicae cogitationes’ (In my heart there is uppermost that one thought: faith in Christ, from whom, through whom, and into whom all my theological thoughts flow to and fro by day and by night).” (John H.C. Fritz, The Essentials of Preaching, St. Louis: Concordia Publishing House, 1948, pp. 18-20) “One of our parishioners once said to us, ‘You spiritual leaders have the perfect job. You get to think about God all day.’ We wish it were so! Just as you might suppose that professional athletes are in excellent physical condition, many people assume that spiritual leaders have perfect spiritual lives. As you know, this is not always the case. Many spiritual leaders are like hosts who run around catering to the needs of their guests. They make sure that everyone else is fed, even while they may themselves be hungry. As a spiritual leader you pray for others. You counsel, encourage, serve, and care for them. Yet unless you take time to care for your own spiritual needs, you damage your ability to care for others.” (Rochelle Melander and Harold Eppley, The Spiritual Leader’s Guide to Self-Care, The Alban Institute, 2002, p. 158) “First of all we need to recognize that ‘now we see in a mirror, dimly’ when it comes to spirituality (1 Cor. 13:12). We can conjecture, rationalize, and pontificate about spirituality, but without the spiritual gift of discernment we have only our own limited notions to guide us. When dealing with God’s higher thoughts and higher ways (Isa. 55:9) we need spiritual disciplines that open us to God’s thoughts and ways; we need the community of faith to critique and reinforce our understandings; and we need our own intentional fitness quest to make our spiritual experiences authentic and personal.” (G. Lloyd Rediger, Fit to be a Pastor: A Call to Physical, Mental, and Spiritual Fitness. Louisville, KY: Westminster John Knox Press, 2000, p. 125) “This discussion of spiritual fitness can be summarized as follows: A—Awareness: I open my heart and mind as widely as possible, recognizing that spirituality is not a simple context for my rational mind. I humbly note that God’s ways are higher than my ways. And I accept my responsibility to develop spiritual fitness through disciplined and joyous practice of spiritual disciplines. B—Basics: I give thanks and remind myself that I can and must relax-release-trust in order to allow the spiritual dimensions to override my bodily appetites, my mental anxieties and habits, and my theological certainties so that I may be open to God’s dynamic grace and purposes. Then the spiritual disciplines become a more natural part of my life instead of forced add-ons. C—Congruency: The union-communion with God and creation, the intimacy and shared spiritual energy of human relationships now add synergy to my spiritual fitness.” (G. Lloyd Rediger, Fit to be a Pastor: A Call to Physical, Mental, and Spiritual Fitness. Louisville, KY: Westminster John Knox Press, 2000, p. 141) “The greatest problem in the Church is the pastor. People are no better than their leaders. ‘Like people, like priest.’ Hos. 4, 9; Matt. 15, 14. Qualis rex, talis grez. Much depends upon the pastor’s spiritual leadership. Said Oecolampadius: ‘How much more would a few good and fervent men effect in the ministry than a multitude of lukewarm ones!’ Said William Reid: ‘The mere multiplying of men calling themselves ministers of Christ will avail little. They may be but “cumberers of the ground.” . . . Even when sound in faith, yet through unbelief, lukewarmness, and slothful formality they may do irreparable injury to the cause of Christ, freezing and withering up all spiritual life around them. The lukewarm ministry of one who is theoretically orthodox is often more extensively and fatally ruinous to souls than that of one grossly inconsistent or flagrantly heretical.’ Luther complained that some men in the ministry ‘ought to be more properly swineherds and dog-tenders than caretakers of souls and pastors’ (Triglot, 567.2.) All detailed instruction and good advice given to a pastor will avail little or nothing if the pastor himself is not the right kind of man — a sincere Christian and a faithful and an able worker. In reference to natural endowments, and acquired knowledge, and practical ability, and a highly developed personality, ministers cannot all be Pauls and Luthers, but they can and ought to be filled with the same ardent love for Christ and for souls and with the same consciousness of the privilege which the Lord is according them and the responsibility which He has placed upon them. A minister of Jesus Christ should seek to become great by way of service, Matt. 20, 25—28. That is a laudable ambition. A pastor should therefore keep himself spiritually, intellectually, and physically fit.” (John H.C. Fritz, Pastoral Theology. St. Louis: Concordia Publishing House, 1932, pp. 14-15) “The study of theology, then, is based on prayer for the gift of the Holy Spirit. The Holy Spirit turns would-be masters of theology, spiritual self-promoters, into humble, life-long students of the Scriptures. Apart from Spirit and his empowerment, people know nothing about eternal life. Without his illumination, the teaching of the Scriptures remains mere theory without any reality. Prayer for God’s ongoing bestowal of the Holy Spirit through Jesus and the ongoing reception of the Holy Spirit makes a theologian. In short, the Holy Spirit makes a theologian. And this is a life-long undertaking…Luther claims that in the study of theology, prayer for the gift of the Holy Spirit needs to be accompanied by continual meditation on the Scriptures…No word; no Spirit. Likewise, no prayer; no Spirit.” (John W. Kleinig, “Oratio, Meditatio, Tentatio: What Makes a Theologian,” Preparing Lutheran Pastors for Today, Editora da ULBRA, 2006, pp. 20-21) Further Conversations In addition to the conversations in small groups that were generated during the course of the discussion of Luther’s “Preface to the Wittenberg Edition, of Pastor George and spouse Susan, and of the scriptural study, further conversation and discussion could revolve around the following questions: - What is the place and role of a Father-Confessor or spiritual director my personal spiritual and devotional life? Is this an opportunity to develop this discipline? This question gives participants an opportunity to discuss how a regular pattern of spiritual conversation and a regular pattern of individual confession/absolution could become part of the routine of the pastor. - What is the place and role of my brother pastors? Is the gathering of pastors an opportunity for deeper sharing in the areas of prayer, meditation, and temptation? If so, how can that be structured and encouraged? Are there impediments to the development of such? Much of this question has likely already been addressed in previous sections. The additional specific question here, however, is the last one in this group. It gives participants an opportunity, if they have not already done so, to speak about barriers to sharing at more deep levels. Each barrier needs to be heard and understood. But the barrier should never be left unaddressed. Always a facilitating question such as, ‘if we want to address this barrier what ideas do you have to do so?’ is important. - According to Duane Bidwell (Short-Term Spiritual Guidance), the central dual focus to all spiritual direction is “identifying God’s action” and “appropriate response” (xii). At times, guided by the Holy Spirit, we are able to do this individually. But often the opportunity to study together, pray together, and discern together is crucial to the process of clearly identifying God’s action and developing an appropriate response. How can pastors more clearly discern God’s action in their lives and ministries? How can you work toward such greater discernment? This simple sequence of two central foci of spiritual direction opens the possibility of significant conversation. It is with the eyes of faith that followers of Christ look to what happens in life to find God’s action. As Christians we seek discernment about the will of God and how to respond faithfully. All conversations in the Body of Christ are to be bathed in prayer. Allow plenty of time for such opportunities. Prayer can be interspersed throughout the meeting. It is not necessary to save it until the end. Especially if deeper conversation emerges, taking a moment for prayer might be both helpful and appropriate. The Faithful and Afire Series Introduce the series and distribute the Study Guide. Other sessions focus on various aspects of pastoral ministry, including missional leadership, worship, balance of family and ministry, integrity, confessing pastors, equipping the saints and the pastor’s prayer and devotional life. Explain that there will be 30-45 minutes of group Bible study, followed by 30-45 minutes of small group conversation. In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the gift of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to remain faithful as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church – Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness. A Case in Point In this “Case in Point” pastors are invited to consider the challenge of balancing the time and energy which they commit to family and to ministry. Hopefully this will “prime the pump” for a lively discussion. Be alert to those whose lives have become unbalanced and are in need of encouragement and healing. Read the case study out loud, and use the questions below to generate conversation for about 10 minutes. Ralph, a young minister, is filled with excitement, and his “plate” is filled with many tasks. In the midst of sermon preparation, he receives word of a death in the church. As he rises from his desk to respond to this pressing need, he notices several letters on his desk, for which replies are long overdue. Before he can put on his coat, his secretary buzzes him that his next appointment is waiting in the outer office. His office door opens, and he is confronted by his wife, who is red faced and crying. “Please, dear,” he says quickly, continuing to put on his coat, “I’ve got to go. I’ve just learned that Mr. Smith has died, and I need to be there. Can it wait until supper?” “No,” she replies quickly, “it won’t wait because it never happens.” Ralph is now aware of some anger in her voice. “What’s wrong dear?” he asks cautiously. “I . . . I . . .” she stammers, “I think you’re having an affair, and I want a divorce!” she blurts out. She covers her face and sinks into a chair as a wave of tears engulfs her. Pastor Ralph is stunned. “Why . . . why . . . why . . .” Now it is his turn to stammer. “Honey, I assure you that you’re the only one for me. There has never been another woman in my life!” He begins to move toward the door. “Oh, I know there’s not another woman. But it’s . . . it’s . . . it’s this church. She takes all of your time. You ask, ‘Can’t it wait until supper?’ But supper never comes – or at least yours doesn’t. And I’m tired of it. Last week you were late for supper every night except Tuesday, and on Wednesday you didn’t make it home until after 10! You haven’t been to an activity at the kids’ school in months! Two days ago it was an emergency at the hospital. Yesterday someone called just as you left.” “You said they needed to talk! But sometimes I need to talk too,” she continues, “and by the time you get home every night neither of us has any energy left. It’s like you give all you have to this church, and there’s nothing left for me. I know there’s not another woman, but I still feel cheated. You don’t have time for me. But it’s just like you’re having an affair, and I want out!” Getting the Conversation Going 1. So now what? What would your next move be? 2. Would it be easier if he were having an actual affair? Explain your response. 3. How would you know whether or not your wife has ever felt like this? 4. You’ve got a lot on your plate, too. If confronted with a similar plea from your wife to talk right now, how would you respond, given other crises competing for your time? 5. How do you keep your priorities straight so conversations like the one above do not occur? The Scriptures Speak Use “The Scriptures Speak” as a guide for a large group Bible study. The circuit counselor, host pastor or another designated leader can present this section. Each participant should have a study guide. Additional material is provided in bold italics for the leader. Because of electronic transmission issues, Hebrew and Greek words and phrases have been transliterated into English. 1. Invite participants to read to themselves. As you read 1 Timothy chapter 3, note the Biblical requirements for and descriptions of being a pastor, (deacon or pastor’s wife). Make special note of the ones you feel are particularly important in maintaining balance between family and ministry. - v 1: kalos, good, valuable, a noble task - v 2: anepileptos above reproach, inviolable, unassailable - sophron, self-controlled, sound in mind - kosmio of good behavior, well mannered, balanced! - philoxenos, hospitable, fond of guests - didaktikos, able to teach, and to be taught! - v 3: epieikes, gentle, patient – opposite of violent! - plektes, not quarrelsome, or looking for a fight - aphilarguros, not a lover of money, one who loves ill-gotten gain - v 4: tekna . . . en hypotage children in submission - v 7: kalov marturia, a good reputation, witness, externally, with outsiders - pagis diabulos, trick or noose of the devil! 2. Provide brief background for Moses leadership of the children of Israel. What did Moses do for God and for the people he had been called to lead? How was his a noble task? What evidence do you have that his leadership was balanced? - Ex 12:31ff: Moses had led God’s people out of Israel. - Ex 16:4; 11-12, 17: 5-6: Moses was used by God to provide food and drink for God’s people in the wilderness. - Ex 17:9-10: Moses directed the defeat of the Amelekites. - Ex 17:15: Moses led God’s people in worshipping God for His protection, care and deliverance. 3. Like Pastor Ralph, Moses was confronted, here by his father-in-law, about the impossible job of handling every situation himself and trying to please everyone. Take turns reading verses from Exodus 18:13-27. - vss 13, 15-16: What was Moses doing? Moses served as judge, shaphat: one who defends or contends, from morning till night • v 17: Review Jethro’s observation, “What you are doing is not good!” What was good and what was not so good about Moses’ strategies? He speaks for God, but apparently tries to do everything himself! • v18: n addition to wearing himself out, who will also wear out? The people! • V19: kabed, too heavy. How could this “too heavy” be both ‘good’ and ‘bad?’ Bad – wear you out. Good – 1 Corinthians 4:1, “… entrusted with the secret things of God,” but not as a burden to borne alone! • What responsibilities does Paul suggest to those entrusted as God’s servants in his farewell to the Ephesian elders, Acts 20:28? episkopos: overseer, is like a superintendent, of other workers, not the owner, or sole proprietor! “With His own blood He bought us …” • What alternate strategy did Jethro suggest to Moses? v 19: Give advice so you know that God is with you. Remember you are not God, but God’s representative to His people! • Before selecting capable men, v 21, what was Moses to do? v 20: Teach them God’s will, show them how they are to live and the duties they are to perform. • v 24: What might have happened had Moses chosen not to follow his father-in-law’s advice? Wear himself out, become too sick to continue, or worse, died prematurely. • vv 25-26: Compare the structure of your current ministry setting to Moses’ “church.” Some church structures place inordinate responsibilities upon called/salaried workers, while others are more reasonable and recognize human limitations. Share various views. 4. Paul is certainly the best known and perhaps the hardest working of the apostles. Most pastors have quoted and/or misquoted this verse. Read 1 Corinthians 9:22. What do you think Paul meant? • Then ask: o What is realistic about Paul’s goal? He wants people to be saved. o What is unrealistic about his goal? Only Jesus “saves” people. We are ambassadors of God’s good will, 2 Corinthians 5:20 • Survey the context to better understand v 22. o v 1: Paul had seen the Lord. How have you seen the Lord? o v 1: The faith of his readers is a direct result of Paul’s ministry. Do faith-building activities consume more or less of your time? o v 3: anakrino, judgment, question, discernment: What kinds of judgments could have been leveled against Paul? You? o vss 4-14: Pastors are sometimes compared as having more or less physical things than others in the church, including money/salary. What have your personal experiences been? o v 23: What does it mean for you to do something “for the sake of the Gospel?” poieo dia to euaggelion According to 1 Corinthians 4:1, “pastors are entrusted with the secret things of God.” The Gospel isn’t to be kept secret, but the Gospel message is to be handled with special care. o v 24-27: Paul compares the rigorous training necessary for a physical contest to similar, rigorous spiritual training and the single-mindedness necessary for the work of ministry. How have you succeeded and where do you still need work here? Why is balance between family and ministry important? Note specifics listed. • How is Paul’s strategy “to be all things to all people,” realistic and healthy? Seize all means to bring people to Christ, and Christ to people. Ask, “Do all phone calls, letters, and objections need to be answered? How can you be more realistic about the tasks “on your plate,” and the expectations of others around you, both area pastors and people in your church?” • How is Paul’s strategy unrealistic and unhealthy? Humans have limited time and energy. How could you “re-word” this verse to be more realistic and reasonable for you? • “God gave Adam a wife, not a congregation!” What does this mean to you? We are to live with our wives and serve God’s people • With what safeguards can you surround yourself to prevent v 22 from becoming your motto? Keep a running list of their suggestions. Talk briefly about how the shed blood of Jesus and the forgiveness He freely offers all is of special help to the proclaimer of the Gospel! You may want to pronounce absolution at this point, if it is appropriate. 5. Together read the Third Commandment, Exodus 20:8-11. • What does the word Sabbath mean to you? Shabbath: intermission, the observance of the seventh day, a day of rest • How and when do you observe Sabbath? Prime the pump by sharing personally how this time or the event is most important to you? Is Sabbath for you or for God? Mark 2:27 How so? • When the many and varied demands of parish ministry are closing in around you, how does the Third Commandment help you maintain a balance between work and family? • Is it easy or hard for you to worship while you lead others in worship? Describe how you worship God. How do you keep a day holy? Describe how your life/Sabbath practice is a paraphrase of the word “holy,” or a time and a place that is “set apart” for God. 6. Study some other verses where the word “Sabbath,” occurs, and note additional meanings: Invite others to include their favorite verses, perhaps first. • Exodus 16:23: day of rest • Exodus 16:25-26: no manna on the Sabbath, faith that God will provide! • Exodus 20:9: no work on the Sabbath, allowing time for worship. Some will talk nostalgically about how “things used to be.” Ask if pastors take another day for their Sabbath and how they observe it? Ask if they take one or two days off? In modern Israel observant Jews prepare Sabbath/Saturday meals the day before/Friday! Ask, “How do you get ready for Sabbath?” Would you eat a cold meal on the Sabbath because you preferred not to allow work/effort to get in the way of your worship of God? • Exodus 20:10: no work • Exodus 31:16: a celebration, that is passed on to other generations • Lev 16:31: a day of rest, during which you deny yourself some ordinary physical pleasures, in order to focus on God. What are the good parts of monasticism? How were the isolation and withdrawal of the monastics dangerous for their faith and service to others? • Numbers 28:9: make an offering to God • Matthew 12:1-8, especially v 7: Which is more important, mercy or sacrifice, and why? • Matthew 12:12: What did Jesus mean, “It is lawful to do good on the Sabbath?” • John 5:10: According to some interpretations of the law, it was unlawful to carry, even a sleeping mat on the Sabbath! • Acts 1:12: a Sabbath’s day’s journey (Exodus 16:29, Numbers 35:5, Joshua 3:4), about ¾ mile • Acts 13:27: reading of God’s Word • Hebrews 4:9: sabbatismos - Sabbath rest Words from the Wise Use these quotes a spring-boards to conversations to follow. Use as many as time or interest dictates. Are you happy? How do others perceive you? How does your family perceive your happiness? How would you respond to the young man who asks: “I am 15 and . . . want to become a pastor . . . my pastor is a good pastor but . . . does not seem really happy . . . I wonder if I will be getting into a job like his and I will be unhappy?” Bruce Hartung, “Pressure Points,” Reporter (February, 2007), 8. Here is part of Bruce Hartung’s response: “Talk . . . frankly as you can. ‘Tell him that you are concerned the pastoral ministry might not be a happy place. Let him know your concerns . . . if a pastor is close to his people, he will hurt for them, sigh with them, and walk arm-in-arm with them. This is stressful, but a necessary and actually rewarding part of ministry. Ask him what his stressors and his joys are in ministry . . . There are pieces of my life as a pastor/seminary professor that are quite satisfying, and others that are not so satisfying, and even others that are actually troubling. But I really believe that life is like that.” Henri Nouwen was a Trappist priest, trained in Europe, who spent his life teaching at Yale and Harvard Divinity Schools, helping people respond to the universal “yearning for love, unity, and communion that doesn’t go away.” What do his words say to you? Some pastors and professional church workers handle the pain and stress of the unhappy or troubling parts of their jobs by isolating themselves from others, including their families. How do you cope with the temptation to isolate yourself from friends and especially from your family? In what other ways do you deal with the stress, pain, and unsatisfying parts of your job in a healthier, more constructive and balanced way? “We live in a society in which loneliness has become one of the most painful human wounds. The growing competition and rivalry which pervade our lives from birth have created in us an acute awareness of our isolation. This awareness has in turn left many with a heightened anxiety and an intense search for the experience of unity and community. It has also led people to ask anew how love, friendship, brotherhood and sisterhood can free them from isolation and offer them a sense of intimacy and belonging.” Henri J.M. Nouwen, The Wounded Healer: In Our Own Woundedness, We Can Become a Source of Life for Others (New York, Doubleday, 1979). Martin Luther often felt the loneliness and stress of ministry. But he also knew the joys of leading people to Jesus. Consider these words: “We can engage in no sublimer and greater work on earth than educating people by preaching and teaching . . . But no work is more difficult than making other people good. Yet this is the best service we can render God.” Luther’s Works (American Edition, Vol. 36) 216. Luther’s words could be both encouraging and discouraging to those seeking the office of the ministry. But imagine how your family members would respond to Luther’s comments. Suggest that pastors share these readings with their family. Caution them to simply listen, without defensiveness to the responses they receive. In a fascinating little book, What Matters Most: When No Is Better Than Yes, long-time youth worker, Doug Fields, has some interesting things to say about self-care, balance, and by extension, the care and feeding of relationships that matter most. “You may be at a crossroads in your life and ministry, and the challenge of saying no is exactly that you need. So I want to challenge you now: The “good way” is saying no – have the courage to walk in it and find rest for your soul. Is busyness really getting you what you want – or need? In the end, busyness makes us look important but cripples our relationships. Busyness feeds our egos but ultimately starves our souls. Busyness fills our calendars but fractures our families. And busyness props up our images and shrinks our hearts.” Doug Fields, What Matters Most: When No is Better Than Yes, (El Cajon, CA: Youth Specialties, 2006), 20-21. Some discomfort may be generated by reading this passage. Hopefully it will motivate some to purchase and read the entire book. Perhaps others will challenge the reasonableness of pastors even saying no to anyone and anything. Some have jokingly called this “donning the cape!” (as in Superman!) Consider another quote from later in the book. You could read it now, or use it as a closing thought for your study. “Jesus said “no” to good things. He said “no” to important people. Jesus left people unhealed. He didn’t answer every question, go to every event, or meet everyone’s needs. He needed time away from his disciples. He needed space. Yes, he even needed sleep. He was 100 percent God and 100 percent human; therefore he had human limits. He needed solitude. He needed time to reflect and pray and nourish his spiritual life. It was this life-giving time of connection to the Vine that gave Jesus spiritual power and energy for His ministry to others. And I believe it was during those times of solitude that He was able to hear God’s voice and know what matters most.” (pp. 80-81) Conversation Suggest a division into groups of 2, 3 or no more than 4. Ask for or appoint a leader. Suggest that the most important goal for this leader be that everyone has opportunity to share with no one dominating the discussion. Suggest that you will announce a time for prayer at which time all participants will pray for one another. 1. For you, what is the most important reason to maintain balance between your commitment to your family and your commitment to your calling to serve God as a pastor? 2. Considering the Scriptures we have studied, imagine how and why some pastors have adopted the following “drivers” for their ministry: a. Be perfect b. Be strong c. Hurry up d. Try Harder e. You’re OK if you please everybody! Which of these, in your opinion, is the hardest to maintain? Describe the precautions you take, so that these are not your first considerations for ministry. 3. In what position of importance does your family come in the order of your current set of priorities? When do you (and when don’t you) alter these priorities? 4. Just how do you determine priorities between family and ministry? 5. How do you “keep heat” under all the kettles at the same time? 6. What about yourself? Is there any time for you, yourself? 7. Has your wife ever confided feelings similar to Ralph’s wife? How did you handle the situation? What changes, if any, did you make to change your priorities? 8. How do you maintain a healthy balance between being a husband and father (if indeed you are) and being a pastor? 9. Why, in the end, is a healthy balance between ministry and family so important? 10. Spend time in prayer for one another and for one another’s families. Dr. Bruce Frederickson, Senior Pastor Messiah Lutheran Church, Mounds View, MN firstname.lastname@example.org The Faithful and Afire Series In 2 Timothy 1:6, the apostle Paul reminds Timothy to “fan into flame the fit of God which is in you through the laying on of my hands.” His inspired words signal the reality of fatigue and malaise in pastoral ministry and our need for God’s daily grace to remain faithful as pastors. This series of Bible studies and conversations for pastoral circuits in the Lutheran Church-Missouri Synod is a resource for fanning the flame of the pastoral gift. Fired by the Word of God, pastors will study key aspects of pastoral ministry, seeking growth and greater effectiveness. A Case In Point In this “Case in Point” pastors are invited to consider the relationship of clergy to laity and their respective roles in carrying out the mission of the Church. Hopefully this will “prime the pump” for a lively discussion. Read the case study aloud. Use the three questions that follow to generate conversation for about 10 minutes. This can be done in the total group or in smaller groups of 3-4. A group of several pastors stood around the coffee and rolls table prior to the official beginning of their monthly Winkel meeting, as was their custom. Pastor Don, who had been ordained and installed ten months before stepped up to the table. Before he filled his coffee cup he began talking about how Lent and Easter had gone in his congregation. He expressed his disappointment at how low attendance had been April 15 compared with Easter Sunday the week before. Pastor Dale, in his seventeenth year of ministry, responded, “That’s the way it is. You have to expect that. It’s been that way forever.” Pastor Don, while appreciating Pastor Dale’s attempt to make him feel better, was not satisfied with the answer. “There has to be more we can be doing to help people understand how important it is to be faithful and committed on a constant basis, not just at certain times of the year! And I’m not just talking about worship attendance. It seems like the people I serve aren’t too excited about doing much of anything in the parish.” Pastor Dale said, “Don, don’t be too hard on yourself. The truth is that there are some built-in frustrations in the ministry. I know I had to deal with these very same issues early in my ministry. I guess I’ve come to the point where I just expect that some things aren’t going to be the way I want them. People just want to have things done for them.” Pastor Tom interjected, “Well, I’ve been serving for about ten years and I still deal with frustrations like the one you are talking about, Don. About two months ago someone I would consider a strong member informed me that he was not getting spiritually fed enough and had joined a community church not far from where we are. He says they are helping him grow more in his faith, addressing specific life issues and giving him, as he said, some real handles on how to live life. I thought I was doing that. It really hurt me that he and his family left.” He was noticeably downcast. Pastor Jim, reluctant at first, ventured to speak. “Several of my members, all of whom have been very supportive of our ministry and demonstrated servants’ hearts, came to me and said they had organized a mission group to go to the New Orleans area to assist in clean-up and rebuilding efforts. I didn’t know that they were planning this nor was I involved in this outreach in any way. At first I was annoyed, but now I’m thankful they took the initiative.” Pastor John, who was usually quiet in groups, was eager to join in the conversation. He stated that he would be thrilled if anyone in his congregation took the initiative in any part of the ministry. He said that trying to get his people to do anything proactive was like trying to herd cats. “They all kinda just go their own way and do what they want.” It was time to begin the conference so they stopped their conversation and took their places in the meeting room. But each of their minds was still focused on the words they had spoken and heard. Had they said too much? Or had they said to little? Getting the Conversation Going 1. Which of these pastors might you most identify with? Responses will vary. 2. How do you think each of these pastors, in general, viewed the members of their congregations? Note that this is a critical element in determining how pastor and people relate to one another. Personal experience may well impact responses. We will address this issue in the Scripture study portion of this lesson. 3. What do you think these brothers were wanting from one another? Did they get it? Responses will vary and should be interesting. The Scriptures Speak 1. Read Ephesians 4: 1-13. The Lord has given us a number of pictures and descriptions of the church and what He has called her to be. One of those is in this passage. - Broadly speaking, what does Paul want his readers to understand? God has given His Church the gift of unity – “one Lord, one faith, one baptism; one God and father of us all.” This unity exists because He has created it. He also gives a variety of gifts to His children who have been called by name into His family. We celebrate the unity of the Body as well as the diversity in the Body. (cf. Romans 12:3-8; 1 Cor. 12:12ff) - Why does Paul stress the unity of the Body of Christ before he talks about what the Body of Christ is to do? He is laying common ground, the core reality of the nature of the holy Christian church, the Body of Christ. The variety of gifts emerges from our unity in Christ. Therefore the differences in peoples’ gifts do not separate us from one another or cause us to go in our own direction. Rather, rooted and grounded in Christ, we have both a common ground and beginning and an eternal connection to Him and to one another. - What are some implications if one really believes what Paul says in verse 7? Implications include but are not limited to the following: laity will be seen as gifted by God; their jealousy of another’s gift will dissipate; the challenges and burdens of ministry will be shared; more will be done to reveal the Lord’s kingdom as more people are involved in ministry. - To what degree does a pastor have a responsibility for seeing that members discover and utilize their God given gifts? To a complete degree. Such responsibility doesn’t mean the pastor has to do this all or do it by himself. He is to see that it is done. - What are some ways you can help people discover and express the “grace” they have been given, “as Christ apportioned it.?” There are resources and instruments available for this purpose through CPH and other publishers. Also, pastors can help greatly by insuring that the congregation has an increasingly clear vision of what the Lord is calling that congregation to do. This brings a greater alignment between mission and the gifts needed to accomplish that mission. This topic was also covered in an earlier lesson on the “Missional Leader.” Take three minutes to list as many works of service as you can. Let the brains storm! Should be interesting and, hopefully, give participants “works” they might not have considered. What is the relationship between works of service and building up the Body of Christ? According to this passage there is a direct relationship. As God’s people serve others, Jesus is served. As people extend and share the grace of God in Christ and their unique “graces” in specific situations, lives are impacted and strengthened. As faith is shared, faith is strengthened. What would you see as evidence that the Body of Christ is being built up? Answers may include baptisms; acts of love and compassion; people responding to the Holy Spirit’s call to faith; new ministries; a joyful spirit in the congregation; sacrificial giving of resources and time; and public stands that glorify God as opposed to sliding into the culture’s mold. At the same time, the Lord can be doing things in people’s lives in ways that we cannot see at a given moment. 2. How do the following passages speak to the Lord’s words in Ephesians 4:11-13? - Ephesians 2:8-10 The Lord graces or gifts us with faith. He also has created us, clergy and lay alike, as “works of art” to “do good works,” which we are also called to do in Ephesians 4:12. - 1 Peter 2:9-10 Christians are addressed at large and designated a “royal priesthood.” We are given the command — lay and clergy alike — to “declare the wonderful praises of Him who called us out of darkness into His marvelous light.” As clergy, we are to assist others in knowing what it is we praise God for. This is why our role as proclaimers of the Gospel is so important. PEOPLE CANNOT BE EQUIPPED IF THEY HAVEN’T HEARD THE GOSPEL. - John 15:5 If people are to be equipped for works of service (“bear much fruit”), they must be connected to Jesus. This is where our call as clergy is to be taken with all seriousness. By God’s grace, people are connected as they hear the Word and receive the sacraments. We have been granted the privilege of being the messengers and bearers of these graces from God. - 2 Timothy 3:16-17 Equipping God’s people for works of service involves them in growing in their understanding and living out of the truths of Scripture. As clergy we are to insure that those we serve are immersed in the Word. This is an essential ingredient in equipping the saints. - Hebrews 13:20-21 Again we hear that God desires His people be equipped to know and do His will. The relationship between pleasing Him and doing His will is an intimate one. They are equipped as they hear the Word, receive the sacrament, and offer themselves as living sacrifices (Romans 12:2) in service to others. 3. Read Ephesians 4:14-16 - What happens as God’s people carry out works of service? We become more grounded in the true and saving faith, more focused in our mission, more able to declare and take the Gospel to others in the spirit of Christ’s love, more united in our witness, and more committed to carrying out that which the Lord has prepared and equipped us to do. 4. What do the following passages say about how Jesus equipped his disciples? He equipped, or prepared them to serve God’s people by: - Matthew 4:19 Calling them to “follow Him” (here as in other verses). In following Him they were in a way becoming His apprentices. The apprentice watches, learns from, and imitates the master. He modeled service and sacrifice. - Matthew 5:2 Teaching and showing them the Father’s kingdom—His plan, reign, and power. In His teaching He revealed the heart and the will of God the Father. - Matthew 6:9a Teaching them to pray. - Matthew 10:5-9 Instructing them in how they were to take His message to others. He told them what the message was, namely, “The kingdom of heaven is near” (v.7). Also, note that He trusted them to do it. He delegated and trusted they would take His message to others in the way He instructed them. - Matthew 28:18-20 Giving them “marching orders” and promising that He, though unseen, would be with them constantly until that day when they will once again see Him face to face. - Mark 10:41-45 Emphasizing in the clearest way possible that they were to be servants. Also, giving the example of washing their feet (John 13: 1-17) - Luke 24:49 Promising and sending them His Holy Spirit. - John 17:13-19 Praying for them. - John 21:15-17 Showing the depth of His love and grace by giving Peter the opportunity to serve and feed the flock, in spite of his previous betrayal. Are there other ways in which Jesus equipped his disciples? If so, name some. Other ways Jesus equipped His disciples may be shared. What guidance can you receive from these descriptions of ways Jesus equipped His disciples? See how well participants can apply what Jesus did to their ministry settings and their roles as equippers. Words from the Wise Through holy baptism, every Christian has been consecrated, ordained, and installed into the ministry to teach, admonish, and comfort his neighbor. Through holy baptism each Christian has obtained not only the authority, power, and right, but also the high, holy obligation—under pain of losing the divine grace—of rousing himself to care and to help so that others may be brought to Christ. C.F.W. WALther (Walther, C.F.W. “Bringing Souls to Christ—Every Christian’s Desire and Duty,” Sermon for the 12th Sunday after Trinity. 1842. Published in Festklange, CPH, 1892. Translated by Bruce Cameron, July, 1993.) Walther emphasizes the high calling (Eph. 4:1) of all Christians and our responsibility to carry out the most important “work of service,” namely, taking the saving name and person of Jesus to others. Luther wrote that a believer’s life was characterized by three stages. First, all the works in a believer’s life flow freely and joyfully out of love for God, without thought of merit. A person who doubts and frets whether enough has been done is not at one, or at peace with God, fully trusting Him and His mercy. above all things. The second stage of faith is exemplified by the fact that a believer loves God even when He permits suffering, confident that the Lord has not forsaken him. This child of God also refrains from blaming those around him or the devils in hell. The third stage of faith, and the highest according to Luther, is that which persists and prevails when torments of conscience seek to overwhelm and destroy the child of God. Luther personally had gone through these stages. He believed they also portrayed each believer’s life. Luther had come to realize that there was no help other than the sweet Gospel that God proclaims in His Word. Faith and trust in the Gospel, worked in hearts by the Holy Spirit, was the firm foundation on which every Christian’s life must be based. Out of this faith flowed every good work in the believer’s life. EUGENE KLUG (Klug, Eugene, Lift High This Cross: The Theology of Martin Luther, St. Louis: CPH, 2003, pp.73-74.) We are reminded of the maturing process in the life of the redeemed, and that the good works we are allowed to perform result from the life of Christ that is in us. We then trust that God will work this in the lives of clergy and lay alike. The role of church leadership is to help its members to identify their gifts and to integrate them into appropriate ministries. When Christians serve in their area of giftedness, they generally function less in their own strength and more in the power of the Holy Spirit. Thus ordinary people can accomplish the extraordinary! C.A. SCHWARTZ (Schwartz, C. A., Natural Church Development. Carolstream, IL: Church Source Resources, 1998, p. 24.) This is a reminder of our call as clergy to enlist, encourage, affirm, and support the works of service carried out by those we have been called to serve. In Ghana, priests are regarded as chiefs and this certainly goes to their heads. In fact, the pastor becomes a boss and a village chief. He is the banker, the cashier, the school teacher, the judge, the preacher, the mechanic, the driver(s), the builder—you name it. The “chief” mentality creates problems. Similar thinking can be found here in the United States. The pastor is often seen as a boss who does everything in the congregation. He decides where the altar should be. He is the liturgist. He leads the worship. He reads the lessons. He practically does everything. There is one thing he does not do in the congregation, and that is take the offering. Maybe the congregation doesn’t trust him with the money. It is as if the congregation is in a boat. The pastor alone paddles the boat while the members sit in it, smoking a pipe and drinking coffee. If anything happens to the pastor while on the sea, the whole boat will sink and the people will perish. The pastor teaches them to be lazy and to sit. They then become comfortable Christians. PAUL KOFI FLYNN, PRESIDENT, EVANGELICAL LUTHERAN CHURCH, GHANA, WEST AFRICA (In Joerz, J.C. and McCain, P.T., Church and Ministry: The Collected Papers of the 150th Anniversary Theological Convocation of The Lutheran Church – Missouri Synod. St. Louis: CPH, 1998, pp.242-243.) These words paint a vivid picture of the pastor who thinks he must, for whatever reason, “go it alone” in the parish. He asks us to look at our motives for doing so. Could they be self-serving? If we should take religion seriously as was done in the early Christian church, the dull picture presented by so many contemporary churches would be radically altered. . . pastors would not be performing while others watch, but helping stir up the ministry of the ordinary members. ELTON TRUEBLOOD (Quoted in Luecke, D., Apostolic Style and Lutheran Substance, Lima, Ohio: Fairway Press, 1999, pp. 114-115.) He challenges us to think like first and second century Christians in terms of our appraisal, perception, and enlistment of the laity. Further Conversation 1. What are some “graces” you believe the Lord has given you? 2. What are some of the graces you see in the lives of your brother pastors? 3. Think of five people in the congregation you serve. Encourages pastors to see and name the giftedness in congregation members. It will be good for all involved to see and affirm these gifts more clearly. - What “graces” do you see in each of them? - What have these gifts meant to the life of the congregation? - How did you come to recognize these “graces?” 4. What factors might impede one in “equipping” or “preparing God’s people for works of service?” Responses could include low congregational morale; not having a system in place to do so; bad experiences at previous attempts; fear that the pastor will lose some authority or power in the congregation (This can be a big one at times and the reason for some deep reluctance to “equip the saints.”); fuzzy congregational vision; or lack of mission focus. 5. What have been some of your frustrations in equipping the saints for works of service? 6. What have been some of your joys as you have equipped the saints for works of service? 7. What have you found to be most effective in equipping the saints for works of service? 8. There are district and synodical resources available in this vital area of equipping the people of God for works of service. There is general agreement that the mechanics of equipping the saints includes the following: - Discovering/Finding Out what graces reside in which people. - Connecting the gifts of God’s people to specific opportunities to serve. - Inviting/Enlisting/Recruiting/Motivating people to use their gifts in light of the church’s mission and the congregation’s vision. - Supporting/Providing resources necessary for the saints to do works of service. - Encouraging/Acknowledging/Thanking those who carry out works of service. Is there anything you believe should be added to this list? 9. What do you believe the Lord is calling you to do in order to “prepare (equip) God’s people for works of service, so that the body of Christ may be built up?” 10. Close with prayer for one another. Share with one another whatever aspects of personal or pastoral life are most important at this time. Lift them to the Lord. May this be a time of blessing, “consolation and conversation of the brethren,” and a “work of service” to others that the Body of Christ “may be built up” in love. William B. Knippa Senior Pastor Bethany Lutheran Church Austin, Texas
Book Review: Philosophy, the Federalist, and the Constitution. by Morton White. Eugene F. Miller Follow this and additional works at: https://scholarship.law.umn.edu/concomm Part of the Law Commons Recommended Citation Miller, Eugene F., "Book Review: Philosophy, the Federalist, and the Constitution. by Morton White." (1989). Constitutional Commentary. 329. https://scholarship.law.umn.edu/concomm/329 This Article is brought to you for free and open access by the University of Minnesota Law School. It has been accepted for inclusion in Constitutional Commentary collection by an authorized administrator of the Scholarship Repository. For more information, please contact firstname.lastname@example.org. fulfill "the preconditions of republican virtue." Indeed, "[w]hen sustaining republican virtue is the theme [of debate] the treatment [of those supporting adoption of the Constitution] is muted and surprisingly incomplete." This bothers Lerner, but the justification, it seems to me, is quite plain: in a government intended to secure men in their natural rights, talk of inculcating virtue or molding manners, morals, and beliefs grates upon the ears. So, too, does the word "regime" which Lerner uses repeatedly to describe America. In the tradition of political thought that term denied a separate sphere of private activity. It is not my purpose, however, to quarrel with Lerner over his desire to instill some virtue into the people and to borrow some of the attributes of a regime in doing it. What concerns me most is his attempt to draw the judiciary into what the framers considered to be the domain of the "political departments" of government, and to cite the most thoughtful framers in support of this project. That was not necessary to achieve Lerner's objective and will, I fear, mis-educate his readers into believing that it is all right for judges to make policy in a good cause. My dispute with Lerner over the role of the Court should not obscure my admiration for this book. Lerner is well worth arguing with, and in the course of doing so I have furthered my own education. PHILOSOPHY, THE FEDERALIST, AND THE CONSTITUTION. By Morton White.¹ New York, N.Y.: Oxford University Press. 1987. Pp. xi, 273. $29.95. Eugene F. Miller² Professor Morton White is the author or editor of more than a dozen books on philosophy and intellectual history, including The Philosophy of the American Revolution. In the present work, he seeks to carry forward his study of the American founding by extracting a philosophy from The Federalist. Professor White thinks of himself as a pioneer in this endeavor, because although various scholars have dealt with individual philosophical topics treated in The Federalist, "no other philosopher" has yet presented a synoptic view of its major philosophical ideas. --- 1. Professor at the Institute for Advanced Study, Princeton, New Jersey. 2. Professor of Political Science, University of Georgia. White emphasizes that The Federalist cannot be approached as if it were a philosophical text, since it was written mainly with the practical aim of securing ratification of the Constitution. He maintains, however, that the authors of The Federalist, whose collective pseudonym is "Publius," make frequent use of philosophical terms and statements. White proposes to discover what these terms and statements meant to Publius by subjecting them to logical analysis and also by explicating the philosophical texts from which Publius drew his meanings. This project is more problematic than it may sound. In selecting terms and statements for analysis, White disregards the fact that Publius's essays contain carefully structured arguments. Aside from what he says about No. 10 and No. 51, White makes no attempt to reproduce those arguments or to view Publius's terms and statements in their contexts. The terms and statements themselves are selected in a way that is unsystematic and often arbitrary. Yet White moves confidently from his analysis of them to broad assertions about Publius's implicit philosophy—his theory of knowledge, his ethics, his psychology, his theory of action, and his metaphysics. Rather than turning directly to the text of The Federalist, White begins by summarizing the views of John Locke and David Hume on the character of human knowledge. In his Introduction, White had promised to be attentive to "any philosophical text" which influenced Publius. Why then does the book focus almost exclusively on Locke and Hume? Locke is not mentioned at all by Publius, and Hume only once. Of course, White can show from other sources that Locke and Hume were read by the authors of The Federalist, but what of the other philosophers whom they read? White's choice of Locke and Hume as the philosophical precursors of The Federalist is guided more by his own predispositions than by any testimony of the authors of that work. As White proceeds, we discover that he is attempting to prove a very distinctive thesis about the philosophy of The Federalist. According to this thesis, Publius's theory of knowledge is marked by an inner tension or inconsistency, which shapes the major conclusions of The Federalist. Publius's moral principles are supposedly based on a rationalist epistemology, derived from Locke, while his principles of psychology and political science are based on an empiricist epistemology, derived from Hume. The Federalist thus appears to represent "a philosophical hybrid, an offspring of Lockean rationalism in morals and Humean empiricism in politics." It is White's determination to establish this thesis that leads him to focus almost exclusively on Locke and Hume as the philosophical sources of The Federalist. This thesis about the bifurcation in Publius's theory of knowledge is crucial to White's overall interpretation of The Federalist, so it requires close scrutiny. To begin, White is probably correct in holding that Publius approached the study of politics along the lines of Hume's experimental method of reasoning. He should, however, have explored the possibility that Publius, like Hume, took the same approach to morals. White attributes to Hume a much sharper distinction than Hume himself made between "science" and ethics or morals. He is certainly wrong in suggesting that Hume "denied that ethics was an experimental science." White arbitrarily rules out the possibility that Publius followed Hume in deriving moral knowledge from experience. As he notes, Hume shared with other Scottish philosophers the view that moral knowledge is grounded on a peculiar sense of approbation or blame that we feel upon observing human conduct. White's thesis requires him to explore the possibility that the authors of The Federalist grounded morality the same way, especially since James Madison had studied Scottish moral philosophy at Princeton under John Witherspoon. Yet White simply assures the reader that there is "no hint" that Madison and the other authors of The Federalist subscribed to the moral sense theory. We cannot tell if White has made the explorations that would be needed to justify this conclusion. Was Publius's approach to morals and natural law based on Lockean rationalism, as White maintains? We need to consider two questions: Does White depict Locke's moral philosophy accurately? Is he able to show that Publius adopted a Lockean view of morals, as he has depicted it? I believe that both questions must be answered in the negative. White describes Locke's moral philosophy as if it were a kind of Christian Aristotelianism. We are told that Locke derived man's moral duties and rights from his God-given nature or essence, using abstract reason. Natural rights are strictly derivative from and subordinate to duties, since we have a right to do only that which it is our prior duty to do. White seems unaware of the view that Locke broke sharply with traditional natural-law teachings and followed instead the approach of Hobbes, who treated man's natural rights as an extension of his strong desire for self-preservation and as a limitation on any duties that one might have to others.³ Locke --- 3. For this interpretation of Locke, see Strauss, Locke's Doctrine of Natural Law, in WHAT IS POLITICAL PHILOSOPHY? AND OTHER STUDIES (1959); see also Strauss, Natural Law, in 11 INTERNATIONAL ENCYCLOPEDIA OF THE SOCIAL SCIENCES 80 (D. Sills ed. 1968); may not in fact have accepted the Hobbesian view of natural rights and natural law, but White never addresses this possibility or tells us why it should be ruled out. White's account of Lockean moral philosophy is correct on one important point: Locke does treat morals as a demonstrative science, like mathematics, that rests on self-evident relations of ideas. White fails to see, however, that Locke is required, by the very epistemological principles that permit him to establish morals as a demonstrative science, to view moral principles in a way quite different from traditional writers. Lockean moral science can have a demonstrative character only because the "ideas" of which it consists are "mixed modes," constructed freely by the mind without reference to real archetypes or standing patterns existing in nature. Thus, for example, we can understand the nature of "justice" and the other virtues simply by analyzing the ideas we hold of those things, since there is no real existence or being, beyond the ideas, that has to be taken into account. Traditional natural law theory had sought by reason or experience to find in nature a basis for the meaning of moral ideas, but the ideas that constitute Locke's moral science have no natural archetypes. White implies that "the essence of man" has the same status in Locke's moral philosophy as these other ideas. According to Locke, however, the idea of "man" falls under the heading of "substances" rather than mixed modes, which means that this idea is formed by reference to something external to the mind. The "essence" of man is something more than the idea we form of man, so that propositions about the human essence cannot be true in a strictly analytic or self-evident way. This does not mean, however, that Locke accepts the traditional view that man's essence is intelligible to the human understanding. According to Locke, the idea of "man" cannot truly represent the inner constitution or being that makes a man what he is, since "real essences" are completely unknown to us. The idea of "man" can stand for nothing more than a "nominal essence," which our minds assemble from their experience of coexisting qualities. Locke even grants that the human essence or species is variable, since people in fact form different ideas of "man." The species "man" is the workmanship of the understanding rather than of nature, and there is no natural or experimental basis for settling disagreements as to what a true man is.4 Coby, The Law of Nature in Locke's Second Treatise: Is Locke a Hobbesian?, 49 REVIEW OF POLITICS 3 (1987). 4. I have explored these features of Locke's epistemology in Locke on the Meaning of Political Language: The Teaching of The Essay Concerning Human Understanding, 9 Pol. Sci. REVIEWER 163 (1979). White's account of Lockean moral philosophy gives a misleading picture of both the method and the substance of Locke's natural-law teaching and blurs important differences between Locke and the Aristotelian tradition. Yet according to White, there is "no doubt" that Madison and Hamilton were influenced by Locke's epistemology of natural law, as thus described. White claims that Publius defined man's essence by a strictly a priori method and would not have appealed to experience. Reasoning from the attributes of this God-given essence, Publius supposedly derived man's duties to preserve his life, perfect his essence, and pursue happiness. These are also natural rights, "simply because the duty to perform an action implies the right to perform that action." White's book offers no substantial evidence that Publius actually understood natural duties and rights in this way. Having made this unsupported assumption, White proceeds to select out a few statements by the authors of The Federalist that relate in some vague way to Locke, but also to other political philosophers. These isolated statements are then put forward as proof that Publius was in full agreement with the so-called Lockean epistemology of natural law. White attempts to link Madison to the Lockean natural-law teaching by citing two brief passages from The Federalist and one from the 1785 Memorial and Remonstrance. In No. 43, Madison refers to "the transcendent law of nature and of nature's God." He is speaking in this context only of a law that governs societies, not individuals; the law in question encompasses "the great principle of self-preservation," which sounds more Hobbesian than traditional. Nevertheless, White leaps to the conclusion that Madison intends here to endorse not only the full body of self-evident truths of the Declaration of Independence, but also what he has described as "a Lockean moral philosophy of natural rights and natural duties." In No. 51, Madison speaks of "the state of nature" and of the danger there that leads the strong as well as the weak to seek the protection of government. This single reference makes it "especially evident" to White that Madison is endorsing the body of Locke's moral philosophy, even though Madison says nothing in No. 51 about Locke or natural rights and natural laws. In his Memorial and Remonstrance, Madison declares it to be "a fundamental and undeniable truth" that "[r]eligion or the duty which we owe to our Creator and the Manner of discharging it, can be directed only by reason and conviction, not by force or violence." White asserts that by "undeniable," Madison must have meant "self-evident," so that he must have shared Locke's view that Morality is a demonstrative science, like mathematics.\textsuperscript{5} Madison speaks here only of the right of religious freedom, and he traces this right to the impossibility of coercing opinions as well as to the duty that we owe to our Creator. Yet White takes this passage as evidence that the American colonists derived the full range of inalienable rights from our duty to God. Jefferson, Madison, and Hamilton, we are told, must have regarded statements about these rights not as empirical statements about men, but as “truths which could be axioms or theorems in a demonstrable science of morality that Locke thought was constructible but never constructed.” I have reviewed the full body of evidence, such as it is, that White offers for his conclusion that Madison was a “Lockean rationalist” in morals. As we have seen, it is doubtful that Locke himself adopted the method or the principles that White attributes to him. Yet White avers, on the basis of isolated references here and there to the state of nature, to the laws of nature, and to religious duties, that Madison must have followed the same method and held the same principles. Since White’s argument applies broadly to “Publius” and not just to Madison, he must bring Jay and Hamilton under the umbrella of Lockean rationalism. All that he gives us in Jay’s case is a statement from \textit{The Federalist} No. 2 to the effect that the people must cede some of their natural rights to government in order to vest it with requisite powers. On the surface at least, the evidence of Hamilton’s moral rationalism is more compelling. In writings of 1774-75, Hamilton employed the concepts of natural law and natural rights, and he observes that to deny the principles of natural law “will be not less absurd, than to deny the plainest axioms.” This passage can be read as asserting only an analogy or parallel between mathematical and natural-law principles, but White takes it to mean that Hamilton accepted “Locke’s theory that morality could become a demonstrative science.” White finds further support for this claim in No. 31, but he badly distorts Hamilton’s actual statement. According to White, Hamilton asserts here that there are primary truths or first principles in morals, and that these moral or ethical statements—to quote Hamilton—“contain an internal evidence, which antecedent to all reflection or combination commands \textsuperscript{5} White shows that Madison’s argument for freedom of conscience closely parallels an argument by Francis Hutcheson, but he dismisses summarily the possibility that Madison could have adopted Hutcheson’s doctrine or moral sense. I wonder, however, whether Madison used the term “undeniable” here to apply to the moral sense or to principles derived from this sense. In \textit{The Philosophy of the American Revolution}, White himself grants that an “undeniable” truth need not be strictly self-evident, i.e., it may be deducted from self-evident principles. the assent of the mind." What Hamilton actually says is that such primary truths or first principles are found "in disquisitions of every kind," not just in morals and geometry. He goes on, in fact, to mention politics specifically as one of the sciences that rest on primary truths or first principles. Hamilton is not saying here that morals or ethics is a demonstrative science like mathematics. He is suggesting instead that all sciences, including experimental sciences such as politics, are in a way analogous to geometry in having some first principles. Moreover, the examples that Hamilton gives here of first principles in ethics and politics have nothing to do with natural rights or natural law, but include such maxims as "there cannot be an effect without a cause" and "the means ought to be proportioned to the end." Hamilton's endorsement of the former maxim leads White to argue that Hamilton was not constant in his devotion to Hume. Yet Hume himself thought it proper for the moral or political scientist to assume the truth of the principle of causality, even though this maxim is drawn into question by deeper and more skeptical reflections. White's account of the epistemology of The Federalist leads him to offer distinctive interpretations of that work on two important points. The first concerns Publius's way of reconciling natural rights with the public interest. According to White, Publius regarded his statements about the natural rights of individuals as true in a self-evident or demonstrative way. They were meant to be tested by a logical analysis of man's essence and of the duties imposed on man by his Creator. Yet White interprets Publius's statements about the interest, good, or happiness of the public as having an empirical or experimental foundation. Statements about the public interest are thus separated by an "epistemological gulf" from statements about the natural rights of the individual. Both kinds of statements are moral or ethical in character, but their truth is tested in different ways. I find this line of argument neither clear nor persuasive. Let us grant that a statement about an individual's natural rights is derived merely from reasoning about ideas. How would a statement about the public's interest be different? White seems on occasion to attribute to Publius the view that the public good, happiness, or interest is itself "discernable empirically." This interpretation runs counter, however, to White's broader claim that Publius relied on reason alone to establish the meaning of moral terms. If the good or happiness of the public can be discerned empirically, why could we not discern the individual's good or happiness in the same way? On other occasions, White seems to say that Publius relies on experience merely to determine the means for attaining the public good or happiness, or else the degree to which it has been attained, rather than what that good or happiness is. If, for example, Publius were to say that the public is in a happy condition or that the public happiness is promoted by some law, he would intend these as empirical statements which can be tested by observation. The problem with this argument is that parallel statements about the good or happiness of individuals would seem to be testable in the same way: for example, the statement that individuals are enjoying their natural rights or that such enjoyment is promoted by some law. Even granting that The Federalist has a hybrid epistemology, White has not established that its statements about the public good have a different foundation from its statements about the good of the individual. White insists, nevertheless, on interpreting Publius's statements about natural rights and about the public good as if they were divided by an "epistemological gulf." According to White, Madison's and Hamilton's epistemology required them to elevate the protection of natural rights above the promotion of the public good, since statements about the former are always more certain than statements about the latter. White believes that Madison did in fact give priority to natural rights in most instances, but in one area—that of slavery—Madison was inconsistent. When his defense of the Constitution forced him to choose between the slaves' right to liberty and the slaveholders' right to property, Madison chose to defend slavery by appealing to the public good: "To insist that the slaves be given their liberty, he seems to have said in reply, would violate the right of the American people to pursue happiness by avoiding the dismemberment of the Union, and therefore he accepted the continuation of slavery." In this case, at least, Madison was willing to put what was "politic" above what was honest or just. White contends that Hamilton was much more willing than Madison to disregard the implications of their epistemology and to subordinate the people's natural rights to energy and efficiency in government. The substantive issue raised by White's analysis is whether Madison and Hamilton thought that the individual's natural rights could in fact be in conflict with the public good, and whether in cases of such conflict they would have given priority to the latter. Clearly such conflicts can arise if the right to life and liberty is inalienable—consider, for example, the case of the person drafted into military service or even the person punished for a crime. By and large, however, the authors of The Federalist thought that the public good requires the protection of individual rights, so that energetic government serves the interests of the individual as well as those of the community. This point is stated clearly in a long passage that White quotes from No. 1, where Hamilton observes "that the vigour of government is essential to the security of liberty; that, in the contemplation of a sound and well informed judgment, their interest can never be separated." White detects "code words" in this passage which lead him to conclude that Hamilton was not seriously interested in protecting the liberties that were central to the Revolution. This seems to be another instance, however, where White is guided by his preconceptions rather than by textual evidence. As for Madison's alleged inconsistency on the issue of slavery, it is true that he saw this institution as a violation of the natural rights of the slaves. It is also true that Madison regarded union as vital to the happiness of the American people, and that he was willing to guarantee the protection of property in slaves in order to bring the southern states into the union. Yet this was not a simple case, as White would have it, in which individual rights were being sacrificed to the public good. Madison believed—correctly, I think—that union would serve the interests even of slaves, since the eventual abolition of slavery would be more likely with union than with the states divided into northern and southern confederacies. Of course, the politics of the struggle for ratification prevented him from saying so. White's interpretation of the epistemology of The Federalist leads him to a distinctive position also on the long-standing debate as to whether or not the principles of the Constitution are consistent with those of the Declaration of Independence. After the turn of the present century, many historians came to view the Constitution as an effort by a wealthy elite to curb the demand, flowing from the Revolution, for equality and democracy. More recently, other interpreters, most notably Martin Diamond, have defended the Constitution on democratic grounds and have denied that its principles are opposed in any fundamental way to those of the Declaration. White incorporates elements of both views. On the one hand, he finds that The Federalist embodies the same epistemological and moral principles as the Declaration. White thus opposes, for example, Gordon Wood's suggestion that the Constitution's defenders gave new and different meanings to the terms of the Declaration. Nevertheless, he accepts the old claim that the Constitution's de- 6. See Diamond, Democracy and The Federalist: A Reconsideration of the Framers' Intent, 53 AM. POL. SCI. REV. 52 (1959). fenders were "aristocratic" or "elitist" in their view of politics. He thus reaches a distinctive conclusion about the relationship of The Federalist to the Declaration: They embody the same philosophical principles, but those common principles have elitist implications. This elitist potential remained latent during the Revolution, but it was activated in 1787-88 by such Federalists as Madison and Hamilton. In White's view, the Declaration and The Federalist are elitist in principle because they assume that men are not equal in their capacity to discern the truth about morals and politics. Once it is established that some are better able than others, because of their greater learning or their freedom from bias and passion, to see the self-evidence of moral truths or the soundness of experimental principles in the science of politics, these superior reasoners can elevate themselves above the rest and become what Locke had called "dictators of principles." White believes that these elitist assumptions find clear expression in Publius's teachings on politics. Political elitism "is blatantly present in The Federalist in spite of the many genuflections that are made in the direction of the people." Publius may sometimes seem to say that the people are capable of governing themselves, but White thinks that in these passages Publius simply flatters his readers and "plays to the gallery." White thus rejects Publius's own claim to be a friend of the people and a champion of republican government. In assessing this argument, we must note that White's various formulations of it partake of an inconstancy and imprecision that are not unusual in this book. In the most extreme formulations, Publius is said to have thought that the principles of morals and politics "could be perceived only by the few," or by "a special sort of person." If Publius had truly held such an exclusivist position, then the charge of political elitism might have some credibility. In other formulations, however, Publius's epistemology seems not to be elitist at all. Majority rule would hardly seem to be endangered by the view that "certain members of the community were better able . . . to discover the truths of natural law than the majority of the people were," or that the discovery of truth in morals and politics "could require an expertise that not all men possessed." White shifts back and forth between these formulations, apparently unaware of the difference between saying, on the one hand, that truth can be perceived only by a few, and, on the other, that the few surpass the many in perceiving the truth or that "not all" are able to perceive it. On one occasion, White quotes a lengthy passage from No. 63 in which Madison explains clearly how the judgments of the few and the many are to be related under the proposed Constitution. According to Madison, “the cool and deliberate sense of the community ought to in all governments, and actually will in all free governments ultimately prevail over the views of its rulers.” Nevertheless, there will arise “particular moments in public affairs” when the people will be drawn or misled into calling “for measures which they themselves will afterwards be the most ready to lament and condemn.” It will be salutary “in these critical moments” for a respectable body of citizens, such as the proposed Senate, to check the progress of the measure “until reason, justice and truth, can regain their authority over the public mind.” White takes these remarks as expressing an elitism or “parentalism,” since the people have to be protected by a respectable body of citizens from the tyranny of their own passions. Yet Madison’s words hardly support such an interpretation. Madison is very clear that the senatorial checks on the popular will are only occasional and temporary. In a properly constituted republic, the will of the people must prevail in the long run. Contrary to White’s interpretation, Madison does count on the people to understand their own rights and duties and to make sound judgments about the policies of government. White’s comments suggest that he has given very little thought to the relationship of philosophy and democratic government. If it is undemocratic to say that some are more perceptive than others about natural rights or the public interest, or that not everyone can perceive these things adequately, then it is difficult to envision a defensible epistemology that would support democracy. Must a democrat assume that all persons understand politics equally well? Must he identify the truth with what the people finally embrace in the marketplace of ideas, so that no independent standpoint remains for judging the popular views? White calls into question the democratic credentials of the American founders, but he gives no indication of what a suitable epistemology for democratic government might be. I have focused on White’s effort to extract a theory of knowledge from The Federalist, but he discusses other philosophical themes as well. In sections on “psychology” and “theory of action,” he has some sensible things to say about Publius’s account of human motives. He promises to treat “metaphysics” in one section of the work, but says nothing substantial about that topic, aside from some observations on causality. Although his thesis requires it, White does not bother to explore how Madison and Hamilton might have understood the being of nature, of God, and of man’s essence. He simply assumes that these writers must have shared the views that he attributes, with little evidence, to Locke. White devotes a section on "philosophy of history" to a refutation of Charles Beard's claim, in his earlier writings, that Madison was an economic determinist. In this context, White reviews Madison's account of the problem of faction and the "republican remedy" which the Constitution provides for this problem. White is able to show that economic motives were, for Madison, only one of the causes of factions. His account of the Madisonian solution to the problem of faction needs correction, however, on two points. First, Madison did not regard an expansion of a country's territory and population as sufficient in itself to produce the diversity required to check majority factions. The competing interests and ways of life would have to be multiplied through the encouragement of commerce. Publius thus advocated a large, commercial republic.7 Second, White sometimes fails to distinguish between a "faction" and an interest group. At one point, he writes that "Madison wanted society to be divided into lots of factions because he wanted to prevent any one faction from becoming a majority of the people." A bit later, he writes that Madison wanted the republic to be "divided into many interest groups that would check each other." White tends here and elsewhere to equate "interest groups" and "factions," even though, as he should know, a faction is an interest group that has become dangerous to others. As White himself observes at one point, a faction, for Madison, was bad by definition, since its members were actuated by an impulse adverse to the rights of other citizens or to the permanent and aggregate interests of the community. While Madison's policy was to multiply competing interests, he certainly did not wish to multiply factions, as White suggests. Although White attempts to put The Federalist into philosophical context, he says virtually nothing about the context of debate in which the struggle for ratification of the Constitution occurred. The reader is thus likely to be puzzled by White's discussion of remarks by Hamilton on previous and subsequent amendments. Anti-Federalist strategy by early 1788 was to insist on previous or conditional amendments to the Constitution, which would have to be enacted before ratification could take place. The alternative which the Federalists pushed successfully was ratification first, with the promise that amendments would be considered subsequently by the new Congress. If White is aware of these strategic issues, he has 7. See Diamond, The Federalist, in History of Political Philosophy 659 (L. Strauss & J. Cropsey eds. 1987). nothing to say about them. In his initial discussion, he says, correctly, that Hamilton favored subsequent amendments, but later he contradicts this by observing that Hamilton wanted to demonstrate "the wisdom of 'previous' as opposed to 'subsequent' amendment of the Constitution." Here is but another illustration of a carelessness that is all too typical of this book. Despite its pretensions, or perhaps because of them, White's study fails to illuminate the text of The Federalist. He chides Madison and Hamilton for their "self-elevation," for thinking of themselves as superior to those who know less than they do. Yet this is exactly the posture that White assumes in relation to The Federalist. He approaches this text not with the openness of an inquirer who hopes to learn something, but with the dogmatist's confidence that he knows, better than its authors, the philosophical assumptions on which their work is based. He imposes on The Federalist a simplistic epistemology that is unlikely to have been held either by Publius or by the philosophers on whom Publius is said to have relied. White's conclusions about Publius's moral philosophy are based on thin and often tortured evidence. He leaves the problem of Publius's "metaphysics" largely unexplored. For no good reason, he questions the genuineness of Publius's dedication to popular government. White says some worthwhile things on some topics, but a carelessness of expression mars his study throughout. The idea of exploring the philosophical assumptions of The Federalist is a good one, but unfortunately this book has little to teach us on this subject.
Laser gyro signal filtering by combining CEEMDAN and principal component analysis Rongrong Huang\textsuperscript{1}, Lei Yan\textsuperscript{2}, Jing Liu\textsuperscript{3} \textsuperscript{1,2}College of General Quality Education, Wuchang University of Technology, Wuhan, 430223, China \textsuperscript{3}Wutaizha Primary School, Wuhan, 430000, China \textsuperscript{1}Corresponding author E-mail: \firstname.lastname@example.org, \email@example.com, \firstname.lastname@example.org Received 7 April 2021; received in revised form 9 July 2021; accepted 6 August 2021 DOI https://doi.org/10.21595/jve.2021.21860 Copyright © 2021 Rongrong Huang, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract. In order to suppress the random shift error of laser gyro and improve the practical precision of inertial navigation system, an improved gyro filtering method is proposed by combining the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and principal component analysis (PCA). Firstly, the gyro signal is decomposed by CEEMDAN, and the noise energy of each intrinsic mode function (IMF) is estimated according to the distribution model of noise energy. Then, on basis of noise energy, the principal component analysis is used to remove the noise IMF to achieve the final denoising of gyro signal. In the proposed method, CEEMD can improve the mode mixing and denoising effect of gyro signal. Moreover, PCA is used to decompose each IMF. According to the noise energy, the noise of each IMF is removed adaptively to avoid the selection of noise IMF and better retain the useful information of the signal. The proposed method is completely dependent on the characteristics of gyro signal and has good adaptability and strong denoising ability. Furthermore, the filtered effect of different methods is analyzed by overlapping Allan variance. The experimental results show that the proposed method can suppress the gyro random drift more efficiently, and the effect of removing noise is better than EMD threshold method and EMD correlation coefficient method. Keywords: Gyro random drift, CEEMDAN, principal component analysis, filtering. 1. Introduction Laser gyroscope is a new generation of inertial measurement element. Compared with traditional mechanical gyroscope, it has many outstanding advantages and has been widely used in many fields [1]. The accuracy of the output signal of laser gyro will directly affect the alignment efficiency of inertial navigation system. Due to the influence of uncertain factors such as internal structure and external environment, the output signal of the gyro often contains a lot of random noise that seriously affects its accuracy. How to effectively reduce the random drift of the gyro and improve the output accuracy of the signal has always been a hot and difficult point in laser gyro [2]. There are currently two main methods to reduce gyro random drift [3]: one is to establish a time series model of the gyro random drift, and the Kalman filter and more are used for random drift compensation based on this; the other is to use wavelet transform and more to reduce the noise of the gyro output data. Since the gyro signal will be interfered by a variety of uncertain factors, the random drift has the characteristics of slow time-varying, nonlinear, and non-stationary, which makes it impossible to establish an accurate drift model. Therefore, a time series model is difficult to obtain the ideal compensation effect, and the current gyro signal processing mainly adopts noise reduction method [3]. To solve the problem of large random drift of MEMS gyroscope, scholars at home and abroad have adopted many methods to deal with it. For example, a digital low-pass filter was used to filter out high-frequency noise [4], but this filter needs to be designed according to experience and is not suitable for the aliasing of noise spectrum and signal spectrum. The wavelet denoising method was proposed and the wavelet coefficients were optimized through sparse redundancy representation [5], but the appropriate threshold value and wavelet basis function is difficult to be selected in the wavelet filtering. An advanced neural architecture search cyclic neural network (NAS-RNN) method was adopted [6], but the basic structure and modules used in the NAS algorithm are difficult to rely on manual design, and require a large amount of computation. The time series model with Kalman filtering method is the most commonly used method for MEMS gyroscope error compensation [7], but the filtering accuracy is low. The empirical mode decomposition (EMD) method does not require any prior knowledge of the signal [8], nor does it need to establish an error model. The original signal can be directly decomposed into several intrinsic mode functions (IMFs) and a margin, and then the high-frequency component containing noise is removed and the low-frequency component is reconstructed to achieve the original signal. In reference [9], the dominant component of noise was screened by calculating the variance of the autocorrelation function of each IMF. The IMF component with a small variance value can be considered as the dominant component of noise. In reference [10], the correlation coefficient method was used to screen IMF. When the local minimum appeared for the first time in the correlation coefficient graph, the IMF component before the extreme point could be determined as the noise component. In reference [11], IMF components were divided into noise IMF, aliasing IMF and signal IMF according to Pearson correlation coefficient criterion, and the aliasing component was processed with good denoising effect. However, there is no definite criterion for the selection of the IMF with noise. The traditional EMD denoising method directly removes the high-frequency component dominated by noise, which will lose some useful information and lead to the deviation of reconstructed signal. In reference [12], EMD algorithm with traditional time series modeling filtering method was used to compensate the error of gyroscope. Although good filtering effect is achieved, several high-order Kalman filters are used in the filtering process, leading to poor real-time performance. In order to overcome the influence of EMD mode mixing and noise IMF selection on denoising, a gyro signal denoising method combining CEEMDAN and principal component analysis (PCA) is proposed in this paper. This method uses CEEMD instead of EMD to achieve almost perfect signal reconstruction. It improves the mode mixing and denoising effect of gyro signal. Moreover, PCA is used to decompose each IMF, and the principal components to be retained are selected adaptively according to the noise energy, thereby removing the noise of IMF r, avoiding the screening of noise IMF, and further improving the denoising effect of gyro signal. The proposed method is completely dependent on the characteristics of gyro signal, without judging the IMF noise term and setting the threshold value. In this paper, the noise reduction experiments of laser gyro test signals are carried out, and the experimental results are analyzed by overlapping Allan variance. The results show that the proposed method can suppress the gyro random drift more effectively than the existing EMD noise reduction methods. 2. CEEMDAN and principal component analysis 2.1. CEEMDAN The CEEMDAN algorithm can effectively solve the problem of modal aliasing caused by EMD by adding adaptive white noise in each stage of decomposition. Meanwhile, it overcomes the problem of reconstruction errors caused by EEMD through adding white noise. The CEEMDAN algorithm is shown below [13]: Step 1: Find the first-order modal component. Add positive and negative pairs of Gaussian white noise $(-1)^m \varepsilon n^i(t)$ to the original signal $x(t)$ in order to form a new signal $x(t) + (-1)^m \varepsilon n^i(t)$, and $m \in \{1,2\}$. $\varepsilon$ is the amplitude. $n^i(t)$ is the white noise sequence added for the $i$ time and obeys the standard normal distribution, and $i$ is the number of auxiliary noises, $i = 1,2,\cdots,N$. EMD of the new signal is obtained: \[ x(t) + (-1)^m \varepsilon n^i(t) = IMF_1^i(t) + r_1(t). \] (1) At this time, $N$ first-order components $IMF_1^i(t)$ are obtained. The average value through $N$ $IMF_1^i(t)$ is found and the final weight of the first stage is obtained: $$\overline{IMF}_1(t) = \frac{1}{N} \sum_{i=1}^{N} IMF_1^i(t). \tag{2}$$ From Eq. (1) and Eq. (2), the first-order residual component $r_1(t)$ is obtained: $$r_1(t) = x(t) + (-1)^m \varepsilon n^i(t) - \overline{IMF}_1(t). \tag{3}$$ Step 2: Find the second-order modal component. Add positive and negative pairs of Gaussian white noise $(-1)^m \varepsilon n^i(t)$ to the remaining component $r_1(t)$ to form a new signal $r_1(t) + (-1)^m \varepsilon n^i(t)$. EMD is used to decompose again: $$r_1(t) + (-1)^m \varepsilon n^i(t) = IMF_2^j(t) + r_2(t). \tag{4}$$ Then, the second-order component $\overline{IMF}_2(t)$ is obtained by averaging $NIMF_2^j(t)$: $$\overline{IMF}_2(t) = \frac{1}{N} \sum_{i=1}^{N} IMF_2^j(t). \tag{5}$$ Finally, the second-order residual component is obtained: $$r_2(t) = r_1(t) + (-1)^m \varepsilon n^i(t) - \overline{IMF}_2(t). \tag{6}$$ Step 3: Repeat step 2 until the remaining signal cannot be decomposed. Suppose that $K$ average IMF components are obtained at the end of the algorithm, the final residual signal $R(t)$ is: $$R(t) = x(t) - \sum_{k=1}^{K} \overline{IMF}_k(t).$$ The original signal can be expressed as: $$x(t) = \sum_{k=1}^{K} \overline{IMF}_k(t) + R(t).$$ ### 2.2. Principal component analysis PCA is a typical decorrelation algorithm [14], which has been widely used in dimensionality reduction, data compression, and noise removal. Suppose $X$ is a $m \times n$ matrix, i.e.: $$X = \begin{bmatrix} X_1 \\ X_2 \\ \vdots \\ X_n \end{bmatrix} = \begin{bmatrix} x_1^1 & x_2^2 & \cdots & x_n^n \\ x_1^2 & x_2^2 & \cdots & x_n^n \\ \vdots & \vdots & \ddots & \vdots \\ x_1^m & x_2^m & \cdots & x_n^m \end{bmatrix}. \tag{7}$$ Take the mean of $X_i$ as $\bar{X}_i$, $\bar{X}_i = \frac{1}{n} \sum_{j=1}^{n} x_i^j$, and let $\bar{X} = [X_1 - \bar{X}_1, X_2 - \bar{X}_2, \cdots, X_m - \bar{X}_m]^T$, then the covariance matrix $\Omega$ of $X$ is $\Omega = \frac{1}{n} X \cdot X^T$. The purpose of PCA transformation is to find an orthogonal matrix $P$. The components in $X$ are de-correlated by the transformation $Y = PX$, and the covariance matrix of $Y$ is diagonal matrix. Because $\Omega$ is a symmetric matrix, and $\Omega$ can be expressed by singular decomposition: $$\Omega = \Phi \Lambda \Phi^T = [\phi_1, \phi_2, \cdots, \phi_m],$$ where, $\Phi = [\phi_1, \phi_2, \cdots, \phi_m]$ is an orthogonal matrix, $\Lambda = \text{diag}\{\lambda_1, \lambda_2, \cdots, \lambda_m\}$. $\lambda_1, \lambda_2, ..., \lambda_m$ represents the characteristic root of covariance matrix and satisfy $\lambda_1 \geq \lambda_2 \geq \cdots \geq \lambda_m$. $\phi_i$ represents the eigenvector corresponding to $\lambda(x)$. If $k$, and then the disjoint operation of each component in data $X$ can be realized by transforming $imf_k = \{d_1, d_2, \cdots, d_N\}$. Components in matrix $\varepsilon(imf_k)$ are independent to each other. In addition to de-correlation, another important feature of PCA is to de-noise signals by optimizing the selection of some principal components. After PCA decomposition, the noise-containing signal is mainly concentrated in the first few principal components, and the noise is distributed in each principal component to varying degrees. Therefore, as long as the first few principal components are retained for reconstruction, the noise in the signal can be significantly removed and a good de-noising effect can be achieved. That is, if let $\tilde{Y} = [Y_1^T, Y_2^T, \cdots, Y_H^T, 0, \cdots, 0]^T$, $K < m$, and then: $$\tilde{X} = P^T \tilde{Y} = \sum_{i=1}^{H} \phi_i Y_i,$$ is the result of denoising the original noise-containing signal $X$. 3. Gyro signal denoising based on PCA and CEEMDAN 3.1. Energy composition model of gyro signal IMF The laser gyro signal $x(t)$ is decomposed by CEEMDAN. The IMF at layer $k$ is $\overline{IMF}_k = \{d_1, d_2, \cdots, d_N\}$, and the energy of $\overline{IMF}_k$ $\varepsilon(\overline{IMF}_k)$ is defined as [15]: $$\varepsilon(\overline{IMF}_k) = \overline{IMF}_k \cdot \overline{IMF}_k^T = \sum_{n=1}^{N} d_n^2,$$ where, $imf_1$ represents the length of $\overline{IMF}_k$. For the sake of discussion, let $f_k = \overline{IMF}_k$, assuming that: $$f_k = g_k + V_k,$$ where, $g_k$ represents the signal component contained in $f_k$, and $V_k$ represents the noise component in $f_k$, there are: $$f_k - E(f_k) = g_k + V_k - E(g_k + V_k) = [g_k - E(g_k)] + [V_k - E(V_k)].$$ Among them, $E(.)$ indicates the expectation. According to EMD, the resolution characteristics of zero mean white noise $E(V_k) = 0$, and let: $$\hat{f} = f - E(f), \quad \hat{g} = g - E(g).$$ Then the above equation becomes: \[ \hat{f}_k = \hat{g}_k + V_k, \] (11) The energy of $ \hat{f}_k, \hat{g}_k $ and $ V $ are $ \varepsilon(\hat{f}_k), \varepsilon(\hat{g}_k) $ and $ \varepsilon(V_k) $ respectively. According to Eq. (9): \[ \varepsilon(\hat{f}_k) = \hat{f}_k \cdot \hat{f}_k^T = (\hat{g}_k + V_k)(\hat{g}_k + V_k)^T = \hat{g}_k \cdot \hat{g}_k^T + \hat{g}_k \cdot V_k^T + V_k \cdot \hat{g}_k^T + V_k \cdot V_k^T. \] Because signal $ \hat{f}_k $ has nothing to do with the noise $ V_k $, and: \[ \hat{f}_k V_k^T = [g_k - E(g_k)][V_k - E(V_k)]^T \approx 0, \quad V_k \hat{g}_k^T = [V - E(V)][\hat{g}_k - E(\hat{g}_k)]^T \approx 0. \] So: \[ \varepsilon(\hat{f}_k) \approx \hat{g}_k \hat{g}_k^T + V_k V_k^T = \varepsilon(\hat{g}_k) + \varepsilon(V_k). \] (12) That is, the energy of $ \hat{f}_k $ is mainly composed of signal energy $ \varepsilon(V_k) $ and noise energy $ \varepsilon(V_k) $. When $ \hat{f}_k $ is denoising, if the noise with energy $ \varepsilon(V_k) $ can be removed from $ \hat{f}_k $, then it can be considered that the remaining part is all signal information and no longer contains noise. The energy of $ \varepsilon(V_k) $ in $ \hat{f}_k $ is unknown. Since the signal and noise are mixed together, and it is usually impossible to find the noise energy contained in $ \hat{f}_k $. However, based on the energy model of the noise signal decomposed by EMD [17, 18], the energy of $ \varepsilon(V_k) $ in $ \overline{IMF_k} $ can be approximated. After EMD decomposition of the signal contaminated by white noise, the embedded mode function of the first layer $ \overline{IMF_1} $ is basically composed of noise and contains only a small amount of signal information. Assuming that $ \overline{IMF_1} $ is completely composed of noise, that is, $ \varepsilon(V_1) = \varepsilon(\overline{IMF_1}) $, then the energy $ \varepsilon(V_k) $ of the noise contained in the $ k $-th IMF can be approximately obtained according to the following formula: \[ \varepsilon(V_k) = \frac{\varepsilon(V_1)}{\beta} \rho^k, \quad k \geq 2, \] (13) where, $ m $ and $ r $. According to Eq. (2) and Eq. (3), the noise energy in $ \hat{f}_k = \overline{IMF_k} - E(IMF_k) $ is the same as that in $ \hat{f}_k = \overline{IMF_k} $. Therefore, the energy $ \varepsilon(V_k) $ of noise contained in $ \hat{f}_k $ can be approximately obtained through Eq. (13). Simply let $ \varepsilon(V_1) = \varepsilon(\overline{IMF_1}) $, regardless of the detailed information contained in $ \overline{IMF_1} $, it will lead to the excessive estimation of noise energy, which is not conducive to the preservation of signal details. If PCA is applied to $ \overline{IMF_1} $ to extract some detailed signal information, the noise energy value of $ \varepsilon(V_1) $ in $ \overline{IMF_1} $ can be estimated more accurately. ### 3.2. Principal component selection for IMF denoising at each layer After CEEMDAN decomposition of the gyro signal, there are different degrees of noise in each IMF layer. PCA is used to further remove the noise of each IMF layer. Still make $ f_k = \overline{IMF_k} $ and $ \hat{f}_k = \hat{f}_k - E(f_k) $, and let $ Z = (Z_1, Z_2, \cdots, Z_N)^T = (f_k)^T $ and $ X = (X_1, X_2, \cdots, X_N)^T = (\hat{f}_k)^T $. Let $ m_Z = E(Z) $ and $ m_X = E(X) $, due to $ X = Z - m_Z $, obviously $ m_X = 0 $, and the covariance matrix of $ Z $ is obtained $ C_Z = E((Z - m_Z)(Z - m_Z)^T) = E\{(X - m_X)(X - m_X)^T\} = C_X $. Therefore, using PCA to denoise $ f_k $ is equivalent to using PCA to denoise $ \hat{f}_k $. Let $ \lambda_1 \geq \lambda_2 \geq \cdots \geq \lambda_N $ be the characteristic value of $ C_X $, and $ \phi_1, \phi_2, \cdots, \phi_N $ be the corresponding eigenvectors. Let $ \Phi = [\phi_1, \phi_2, \cdots, \phi_N]^T $, i.e. $ \Phi $ is an orthogonal matrix. Define: \[ Y = (Y_1, Y_2, \cdots, Y_N)^T = \Phi(X - m_X). \] According to the decomposition characteristics of PCA, the noise in $ X $ is distributed in all the components in $ Y_i $, while the signal is mainly concentrated in the components of the first few layers. If the first $ H $ eigenvectors are selected to form a new transformation matrix \[ \Phi_H = [\phi_{1}, \phi_{2}, \cdots, \phi_{H}, 0, \cdots, 0]^T, \] then the approximate value $ \bar{X} $ of the original signal $ X $ can be obtained from $ Y_H $ by the inverse transformation: \[ \bar{X} - m_X = \Phi_H^T Y_H = \sum_{i=1}^{H} \phi_i Y_i. \] Therefore, $ \bar{X} = \sum_{i=1}^{H} \phi_i Y_i + m_X $ is the signal de-noised by $ \hat{f}_k $, and: \[ \Delta X = X - \bar{X} = \sum_{i=H+1}^{N} \phi_i Y_i, \] (14) is equivalent to the noise removed from $ \hat{f}_k $. When PCA is used to remove the noise in $ \text{imf}_k $, an appropriate number of principal components is selected for reconstruction. Usually, according to the cumulative contribution rate of the first $ H $ principal components $ r = (\sum_{i=1}^{H} \lambda_i / \sum_{i=1}^{N} \lambda_i) $ to determine the principal components retained. However, the selection of principal component is not simple: if the cumulative contribution rate $ r $ is too high, there will be a lot of residual noise, resulting in incompletely removal of noise; If the cumulative contribution rate $ r $ is too small, more signal details will be lost. Moreover, the intensity of noise contained in each IMF layer is different. Therefore, in the denoising process of $ \overline{\text{IMF}}_k $, the cumulative contribution rate $ r $ cannot be set to a fixed value. Based on the distribution characteristics of noise energy in $ \overline{\text{IMF}}_k $, this paper adaptively determines the value of the cumulative contribution rate $ r $ in denoising. ### 3.2.1. Principal component selection of $ \overline{\text{IMF}}_1 $ From CEEMDAN’s decomposition characteristics of noise signals [16], it can be seen that IMF is basically composed of noise and only contains a small amount of signal details. After PCA decomposition, the signal is basically only concentrated in the first principal component. Therefore, when using PCA to de-noise $ \hat{f}_1 $, only the first principal component is retained for reconstruction. That is, make $ X = \hat{f}_1 $, and $ X $ de-noised by PCA is the signal $ \bar{X} = (\phi_1, 0, 0, \cdots, 0)Y = \phi_1 Y_1 $. At this point, the noise removed from $ X $ is: \[ \Delta X = X - \bar{X} = \sum_{i=2}^{N} \phi_i Y_i. \] (15) Therefore, the energy of noise $ \varepsilon(V_1)\varepsilon(\Delta X) $ contained in $ \hat{f}_1 $ can be obtained. According to the noise energy distribution in the EMD model Eq. (13) and $ \varepsilon(V_1) $, $ \hat{f}_k $ value contained in the noise energy is calculated. ### 3.2.2. Principal component selection of $ \overline{\text{IMF}}_k $ ($ k \geq 2 $) The following discusses the principal component selection method of $ \hat{f}_k = \overline{\text{IMF}}_k - E(\overline{\text{IMF}}_k) $ at $ k \geq 2 $. From Eq. (12), the energy of $ \hat{f}_k $ is mainly composed of ideal signal energy and noise energy. If appropriate $ H $ is selected during the PCA reconstruction, the energy of noise $ \Delta X $ deleted in Eq. (14) is the same as that of $V_k$ contained in $\hat{f}_k$. That is, $H$ is selected, let: $$\varepsilon(\Delta X) = \varepsilon(V_k).$$ It is considered that the noise has been completely removed, and the remaining principal component is the ideal signal $\hat{g}_k$ without noise. In order to facilitate the calculation and reduce the error, the above Formula is modified to: $$\frac{\varepsilon(\Delta X)}{\varepsilon(X)} = \frac{\varepsilon(V_k)}{\varepsilon(X)}. \tag{16}$$ It can be seen from Eq. (14) that the energy of the deleted noise $\Delta X$ is: $$\varepsilon(\Delta X) = (\Delta X)^T \Delta X = \sum_{i=H+1}^{N} \sum_{j=H+1}^{N} Y_i^T \phi_i^T \phi_j Y_j = \sum_{i=H+1}^{N} Y_i^2 = \sum_{i=H+1}^{N} Y_i Y_i^T.$$ Since $Y_i = \phi_i^T (X - m_X)$, and: $$\varepsilon(\Delta X) = \sum_{i=N}^{N} \phi_i^T (X - m_X)(X - m_X)^T \phi_i = N \sum_{i=N}^{N} \phi_i^T \left[ \frac{1}{N} (X - m_X)(X - m_X)^T \right] \phi_i$$ $$= N \sum_{i=N}^{N} \phi_i^T E[(X - m_X)(X - m_X)^T] \phi_i = N \sum_{i=N}^{N} \phi_i^T C_X \phi_i = N \sum_{i=N}^{N} \phi_i^T \lambda_i \phi_i$$ $$= N \sum_{i=N}^{N} \lambda_i.$$ Due to $X = \sum_{i=1}^{N} \phi_i Y_i + m_X$ and $m_X = 0$, and the energy of signal $X$ is: $$\varepsilon(X) = X^T X = \left( \sum_{i=1}^{N} \phi_i Y_i \right) \left( \sum_{j=1}^{N} \phi_j Y_j \right) = \sum_{i=1}^{N} \sum_{j=1}^{N} Y_i^T \phi_i^T \phi_j Y_j = \sum_{i=1}^{N} Y_i Y_i^T$$ $$= N \sum_{i=1}^{N} \phi_i^T \frac{1}{N} (X - m_X)(X - m_X)^T \phi_i = N \sum_{i=1}^{N} \phi_i^T E[(X - m_X)(X - m_X)^T] \phi_i$$ $$= N \sum_{i=1}^{N} \phi_i^T C_X \phi_i = N \sum_{i=1}^{N} \lambda_i.$$ When using PCA to de-noise $\hat{f}_k$, if the first $H$ principal components are selected for reconstruction, the ratio of the energy of the deleted noise $\Delta X$ to that of the original signal $X$ is $$\frac{\varepsilon(\Delta X)}{\varepsilon(X)} = \frac{\sum_{i=H+1}^{N} \lambda_i}{\sum_{i=1}^{N} \lambda_i}.$$ It can be seen from Eq. (16) that in order to make the energy of deleted noise equal to that of the noise contained in $\tilde{f} = f - \hat{f}$, an appropriate $H$ should be selected as that $$\frac{\sum_{i=H+1}^{N} \lambda_i}{\sum_{i=1}^{N} \lambda_i} = \frac{\varepsilon(V_k)}{\varepsilon(X)}.$$ It can be considered that all noises in $\tilde{f} = f - \hat{f}$ has been removed, and the remaining part is all composed of signal information without noise. $\tilde{f}(n)$ can be calculated by Eq. (7) and noise energy model Eq. (13), and $\varepsilon(X)$ can be calculated directly. When choose $H$, it is hard to be sure that $\frac{\sum_{i=H+1}^{N} \lambda_i}{\sum_{i=1}^{N} \lambda_i} = \frac{\varepsilon(V_k)}{\varepsilon(X)}$ is true. In this paper, the value of $H$ is selected according to the following method: if there is $\{\phi_i\}$, Eq. (17) is established, and let $r_N$: $$M = m_0 \leq \frac{\varepsilon(V_k)}{\varepsilon(X)} \leq \text{imf}_k, \quad (k = 1,2 \cdots, N). \tag{17}$$ ### 3.3. De-noising steps of gyro signal based on CEEMDAN and PCA The laser gyro signal denoising algorithm based on CEEMDAN and PCA proposed in this paper are as follows: Conduct CEEMDAN on the gyro signal $x(t)$, and set its intrinsic modal function as $\overline{\text{IMF}}_k$ ($k = 1,2 \cdots, K$), and the remainder as $R(t)$; Let $\varepsilon(V_1) = \varepsilon(\overline{\text{IMF}}_1)$, and calculate the energy $\varepsilon(W_k)$ ($k \geq 2$) of the noise contained in $\overline{\text{IMF}}_k$ according to Eq. (13). Let $\hat{f}_k = \overline{\text{IMF}}_k - E(\overline{\text{IMF}}_k)$, $k = 1,2,\cdots,K$. PCA decomposition is performed on $\hat{f}_k$. Through Eq. (17), the principal component $H$ retained in the denoising of feature space is calculated. According to $\tilde{X} = \sum_{i=1}^{H} \phi_i Y_i + m_X$, the denoising result is found in the feature space, and the original image $\hat{f}_k'$ is found through iteration. The de-noising result of $\overline{\text{IMF}}_k$ is $\overline{\text{IMF}}'_k = \hat{f}_k' + E(\overline{\text{IMF}}_k)$. Accumulate and reconstruct $\overline{\text{IMF}}'_k$ ($k = 1,2,\cdots,K$) to get the de-noised gyro signal $x'(t) = \sum_{k=1}^{N} \overline{\text{IMF}}'_k(t) + R(t)$. ### 4. Experimental analysis #### 4.1. Experiment 1 The experimental data in this article comes from the drift test of a laser gyro under a static base at normal temperature (20 °C) (the nominal gyro drift is 1 °/h). The sampling interval is 1 s, and the output signal of 2000 epochs on the x-axis is taken for experimental analysis (similar to the y and z-axes). The experimental data is shown in Fig. 1. In order to compare the denoising effect, the gyro signal is de-noised by the EMD threshold method [12], the EMD correlation coefficient method [9] and the proposed method in this paper. In the denoising of EMD threshold method and EMD correlation coefficient method, the number of decomposition layers is 9. After the experimental signal is decomposed by CEEMDAN under the proposed method, 10 IMF components and one remainder are obtained. #### 4.1.1. Direct comparison of denoising gyro signals The laser gyro drift signal is de-noised by the EMD threshold method, the EMD correlation coefficient method and the method, and the results are shown in Fig. 2, Fig. 3, and Fig. 4 respectively. Compared with the signals de-noised by the EMD threshold method and the EMD correlation coefficient method, the signal de-noised by the proposed method is smoother. Calculate the mean value and variance of the original signal and the denoising signal respectively, and the results are shown in Table 1. It can be seen that the mean values of the three methods after denoising are basically the same. The variance of the proposed method after denoising is small, which shows that the random noise in the gyro drift data is better eliminated. \| \| Original signal \| EMD threshold method \| EMD correlation coefficient method \| Proposed method \| \|----------------------\|-----------------\|----------------------\|------------------------------------\|-----------------\| \| Mean (pulse/s) \| 2.4733 \| 2.3384 \| 2.2429 \| 2.0491 \| \| Variance (pulse/s)^2 \| 104.8087 \| 3.3047 \| 2.7739 \| 1.5620 \| 4.1.2. Comparison of overlapping Allan variance of denoising gyro signal In order to further analyze the denoising effect of the method in this paper on the gyro signal, the overlapped Allan variance is used to compare and analyze the denoising results of the three methods [17]. Overlapping Allan variance is an improvement to ordinary Allan variance. It has a larger confidence interval than ordinary Allan variance analysis and is suitable for error analysis of non-stationary random gyro signals. Let the overlapped Allan standard deviation of the signal be $\sigma(\tau) \tau = n\tau_0$, and $\tau = n\tau_0$ be the sampling interval. The double logarithmic curve of $\sigma(\tau) \sim \tau$ can describe the different random error components of the gyro signal. The original signal is obtained by the overlapping Allan analysis method, and the coefficients of the five source errors of the signal after the three methods are de-noised, including: quantization noise ($Q$), angle random walk ($N$), bias instability ($B$), rate random walk ($K$) and rate ramp ($R$). The results are shown in Table 2. It can be seen from Table 2 that after the filtering processing of the three noise reduction algorithms, the error coefficients is reduced, indicating that the three methods have a certain denoising effect on the gyro signal. Compared with the EMD threshold method and the EMD correlation coefficient method, the various indicators of the signal after noise reduction in this method are low. It indicates that the gyro signal after the EMD threshold method and the EMD correlation coefficient method still contains a certain degree of error components. The method in this paper further weakens the various error components of the gyro signal. It can also be seen from Table 2 that quantization noise is the main factor causing the random errors in the gyro signal. After noise reduction by EMD threshold method, EMD correlation coefficient method and this proposed method, the quantization noise error coefficient of the signal is reduced to the original 13.82%, 11.14% and 6.46%, respectively. Table 2. Comparison of noise error coefficient before and after denoising \| Algorithm \| $Q$ (µrad) \| $N$ / (°/h$^{1/2}$) \| $B$ (°/h) \| $K$ / (°/h$^{3/2}$) \| $R$ (°/h$^2$) \| $Q$ noise reduction front-to-rear ratio (%) \| \|----------------------------\|------------\|---------------------\|-----------\|---------------------\|---------------\|---------------------------------------------\| \| Original signal \| 6.3632 \| 0.1372 \| 0.4879 \| 2.3392 \| 6.2236e-04 \| 100 \| \| EMD threshold method \| 0.8792 \| 0.0287 \| 0.2190 \| 0.4068 \| 1.2144e-04 \| 13.82 \| \| EMD correlation coefficient method \| 0.7091 \| 0.0194 \| 0.1613 \| 0.3089 \| 4.2933e-05 \| 11.14 \| \| Proposed method \| 0.4113 \| 0.0086 \| 0.0924 \| 0.1322 \| 1.2977e-05 \| 6.46 \| ### 4.2. Experiment 2 In this experiment, the low cost MPU-6050 gyroscope is studied. At room temperature, the calibrated MEMS inertial sensor is fixed on the horizontal test bench through a clamp, and the sampling frequency is set to 100 Hz. After power-on, the data is collected continuously for 30 s after preheating for 0.5 h. The Y axis sampling data of the gyroscope is used as the original output data of the gyroscope. After calculation, the original signal of gyroscope random drift error is shown in Fig. 5. The de-noised results of EMD threshold method, EMD correlation coefficient method and the proposed method are shown in Fig. 5. To further evaluate the effect of various denoising methods, root mean square error (MSE) and signal-to-noise ratio (SNR) are used to measure the denoising effect: $$MSE = \sqrt{\sum_{i=1}^{N}[f(i) - f'(i)]^2},$$ \hspace{1cm} (18) $$SNR = 10 \times \lg \left( \frac{\sum_{i=1}^{N} f^2(i)}{\sum_{i=1}^{N}[f(i) - f'(i)]^2} \right),$$ \hspace{1cm} (19) where, $f(i)$ is the original signal, and $f'(i)$ is the de-noised signal. The larger the SNR is, the smaller the root mean square error is, and the better the denoising effect will be. MSE and SNR are calculated respectively as shown in Table 3. The EMD Threshold Method denoising method directly removes the coefficients that are smaller than the threshold value in each IMF layer, resulting in partial information missing. The EMD correlation coefficient method directly removes the modal units with small correlation coefficients, which also leads to the loss of gyro signal. While using the proposed method to de-noise, the signal waveform is well preserved. Compared with the EMD threshold method, the MSE of the proposed method is reduced by 12% and the SNR is improved by 13%. Compared with the EMD correlation coefficient method, the MSE of the proposed method is reduced by 12% and the SNR is increased by 13%. Table 3. Evaluation of gyroscope signal denoising effect \| Method \| MSE \| SNR \| \|-------------------------\|-------\|----------\| \| Original signal \| 0.1065\| 15.3005 \| \| EMD threshold method \| 0.0914\| 22.3445 \| \| EMD correlation coefficient method \| 0.0943\| 23.7260 \| \| The proposed method \| 0.0827\| 25.3522 \| Experiment 1 and experiment 2 are carried out by Windows10 system, 2.90G Hz i5 processor of CPU, 16.0GB memory, and the software uses the Matlab2019A. In the experiment, the number of decomposition layers of CEEMDAN is set as 9. In experiment 1, the average running time of the three denoising methods is about 0.60227 s, 0.62782 s and 1.06885 s, respectively. In experiment 2, the average running time of them is about 0.38944 s, 0.40896 s and 0.71315 s, respectively. ![Graphs](image) Fig. 5. Denoised results of three methods in Experiment 2 Through comparative analysis, it can be seen that in the proposed method, due to the need of multiple iterative decomposition for CEEMDAN and the use of PCA to remove the noise in each IMF, the running time is increased to a certain extent compared with the EMD Threshold Method and the EMD Correlation Coefficient Method. In the Matlab2019A environment, the average time of the implementation phase of the proposed method is about 0.891 s. It can be seen that the proposed method can quickly obtain the denoising results of gyro signals in practical applications. If the algorithm is written into the hardware by C or other compiled languages, the running speed will be further improved. Therefore, the proposed method in this paper can meet the requirements in terms of real-time performance under certain conditions. 5. Conclusion 1) In order to overcome the shortcomings of the existing EMD gyro signal denoising algorithm, this paper combines CEEMDAN and PCA to propose an improved gyro signal denoising method. 2) According to the noise energy contained in IMF in each layer of gyro signal, the principal components of IMF that should be retained after PCA decomposition is adaptively selected to realize the denoising of gyro signal. 3) In the denoising process, the proposed method can adaptively calculate the model parameters according to the characteristics of gyro signal. Therefore, the denoising process is only related to the characteristics of gyro signal and does not require cumbersome parameters and threshold adjustments. 4) The noise reduction of the measured gyro random drift signal is carried out, and the noise reduction effects of different denoising methods are compared and analyzed by using direct comparison method and overlapping Allan variance method. 5) The experimental results show that compared with EMD threshold denoising method and EMD correlation coefficient denoising method, the noise reduction effect of the proposed method improves to a certain extent. It can remove noise, reduce the error components of gyro signal, and improve the accuracy of the inertial guidance solution more effectively. Moreover, the proposed method in this paper has better stability and adaptability in the processing of gyro signals. References [1] Y. V. Filatov, P. A. Pavlov, A. A. Velikoseltsev, and K. U. Schreiber, “Precision angle measurement systems on the basis of ring laser gyro,” Sensors, Vol. 20, No. 23, p. 6930, Dec. 2020, https://doi.org/10.3390/s20236930 [2] M. Wang, X. Dong, C. Qin, and J. Liu, “Adaptive H∞ Kalman filter based random drift modeling and compensation method for ring laser gyroscope,” Measurement, Vol. 167, p. 108170, Jan. 2021, https://doi.org/10.1016/j.measurement.2020.108170 [3] K. Liu et al., “Noise analysis of a passive resonant laser gyroscope,” Sensors, Vol. 20, No. 18, p. 5369, Sep. 2020, https://doi.org/10.3390/s20185369 [4] F. S. Tian, “Research on navigation solution of multi-rotor UAV based on multiple sensors,” Harbin Institute of Technology, 2017. [5] J. Song, Z. Shi, B. Du, L. Han, H. Wang, and Z. Wang, “MEMS gyroscope wavelet de-noising method based on redundancy and sparse representation,” Microelectronic Engineering, Vol. 217, p. 111112, Sep. 2019, https://doi.org/10.1016/j.mee.2019.111112 [6] Z. Zhu, Y. Bo, and C. Jiang, “A MEMS gyroscope noise suppressing method using neural architecture search neural network,” Mathematical Problems in Engineering, Vol. 2019, pp. 1–9, Nov. 2019, https://doi.org/10.1155/2019/5491243 [7] Liu X. B., Chen G. W., Wang D., and Wang D. F., “Analysis and compensation of drift and noise in MEMS Gyroscope,” Chinese Journal of Sensors and Actuators, Vol. 31, No. 3, p. 368, 2018. [8] N. E. Huang et al., “The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis,” Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, Vol. 454, No. 1971, pp. 903–995, Mar. 1998, https://doi.org/10.1098/rspa.1998.0193 [9] Liu M., Liu J. H., Chen J. H., and Peng F. C., “MEMS Gyroscope noise reduction method based on improved EMD,” Chinese Journal of Sensors and Actuators, Vol. 33, No. 5, pp. 705–710, 2020. [10] He J. N., Zhong Y., and Li X. F., “Multi-scale prediction of MEMS gyroscope random drift based on EMD-SVR,” Journal of Measurement Science and Instrumentation, Vol. 11, No. 3, pp. 290–296, 2020. [11] Liu Wenta, Liu Jiaya, and Shen Qiang, “Integrated modeling and filtering of fiber optic gyroscope’s random errors,” Opto-Electronic Engineering, Vol. 45, No. 10, p. 180082, 2018. [12] Yang J. H. et al., “A modeling method for random errors of micromechanical gyroscope based on the improved EMD,” Chinese Journal of Scientific Instrument, Vol. 40, No. 12, pp. 196–204, 2019. [13] M. E. Torres, M. A. Colominas, G. Schlotthauer, and P. Flandrin, “A complete ensemble empirical mode decomposition with adaptive noise,” in ICASSP 2011 – 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4144–4147, May 2011, https://doi.org/10.1109/icassp.2011.5947265 [14] Lei Zhang, R. Lukac, Xiaolin Wu, and D. Zhang, “PCA-based spatially adaptive denoising of CFA images for single-sensor digital cameras,” IEEE Transactions on Image Processing, Vol. 18, No. 4, pp. 797–812, Apr. 2009, https://doi.org/10.1109/tip.2008.20011384 [15] P. Flandrin, G. Rilling, and P. Goncalves, “Empirical mode decomposition as a filter bank,” IEEE Signal Processing Letters, Vol. 11, No. 2, pp. 112–114, Feb. 2004, https://doi.org/10.1109/lsp.2003.821662 [16] X.-D. Niu, L.-R. Lu, J. Wang, X.-C. Han, X. Li, and L.-M. Wang, “An improved empirical mode decomposition based on local integral mean and its application in signal processing,” Mathematical Problems in Engineering, Vol. 2021, pp. 1–30, Feb. 2021, https://doi.org/10.1155/2021/8891217 [17] K. Maciuk, J. Kudrys, M. Bagherbandi, and I. V. Bezmenov, “A new method for quantitative and qualitative representation of the noises type in Allan (and related) variances,” Earth, Planets and Space, Vol. 72, No. 1, p. 186, Dec. 2020, https://doi.org/10.1186/s40623-020-01328-6 Huang Rongrong received master’s degree from Chongqing Normal University, in 2013. Now she works at Wuchang University of Technology. Her current research interests include system theory. Yan Lei received master’s degree from South-Central University for Nationalities, in 2017. Now she works at Wuchang University of Technology. Her current research interests include nonlinear dynamic system. Liu Jing received bachelor’s degree in China University of Geosciences in 2016. Her major is mathematics and applied mathematics. Now she works at Wutaizha primary school as a teacher in Wuhan, China.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.