Studies of histology and development of rainbow trout ovaries indicate that the degree of follicular atresia is associated with degree of starvation during the maturation period. Experimental stocks raised under varying starvation regimes showed that the greater the starvation, the higher the proportion of maturing follicles which regressed. Thus the result was a lower number of eggs brought to maturity in starved fish.Observations on natural populations indicate wide differences in egg size, with corresponding inverse differences in egg number at maturity. Egg size is apparently not effected by diet, but rather through natural selection depending on intraspecific competition. There appeared to be little correlation between egg size and fish size, but some difference did exist between age classes. Variations in egg number are therefore attributable to fish size, egg size and adequacy of diet. Starvation has also a lasting effect on the proportion of a given age-class which matures.

It has been claimed that the short-term forgetting shown by the Peterson technique is entirely due to proactive interference from prior experimental items. Two experiments investigated this by studying forgetting when prior items were avoided by testing subjects only once. Both experiments showed significant forgetting, although the degree of forgetting was less than with a multitrial procedure. On the basis of this and other results it is suggested that the Peterson technique comprises two components, a primary memory component which decays within 6 sec, and a more stable secondary memory component. Forgetting with the multitrial procedure is attributed principally to the need to use temporal retrieval cues to avoid confusion between successive items; longer retention intervals are associated with reduced temporal discriminability and hence poorer recall.

Subjects performed a memory task on two occasions, one in the morning and the other in the afternoon. The task comprised two components, one involved immediate recall of sequences of nine digits, the other involved the repeated item technique devised by Hebb (1961), in which one nine-digit sequence is surreptitiously repeated, each repetition being separated by two non-repeated sequences. Performance on the immediate memory task was better in the morning than the afternoon. The repeated item was recalled more accurately than non-repeated items, but this effect was not influenced by time of day. An explanation in terms of the relationship between arousal and memory reported by Kleinsmith and Kaplan (1963) is suggested.

Sixty narwhals (Monodon monoceros) were sampled in the vicinity of Pond Inlet, N.W.T., during the summer of 1978 and 1979. Concentrations of Pb, Cu, Cd, Hg, Se, As, and Zn were measured in liver, kidney, muscle, and blubber. All elements except As were lowest in concentration in the blubber and highest in kidney and liver; Cd and Zn were generally higher in kidney than in liver, while the converse was true for Pb, Cu, and Hg. Cadmium levels in liver and kidney had high interanimal variability but were generally higher than would be expected based on existing marine mammal Cd content data. Certain metals were correlated with animal size and sex. Mercury in kidney, muscle, and blubber and Cd in muscle appear to accumulate during growth. A number of interelement associations with extremely high probabilities were found: among them, that between Hg and Se and Cd and Zn in liver and kidney tissues. The latter association is thought to be related to the presence of metallothionein. The data presented are discussed in terms of what little is known of the biology of the narwhal and represent a baseline for a resource organism which is harvested in an area already subject to mining activity, the level of which is due to increase in the future.

Murdock has shown that the immediate free recall of a list of words is adversely affected by a concurrent card-sorting task, with degree of impairment increasing with the number of sorting alternatives. He interprets this in terms of a limited capacity STM mechanism. The present study repeats Murdock's study with the addition of an STM control, a condition in which recall was delayed for 30 sec. by a rehearsal-preventing task. Card-sorting load influenced the long-term component (measured by delayed recall), but not the short-term component (measured by subtracting delayed from immediate recall). It is suggested that the limited capacity system affects input into LTM rather than STM.It has frequently been suggested that short-term memory (STM) may be based upon a mechanism of limited informational capacity (Broadbent, 1958;Brown, 1958 ;Crossman, 1960). More recently Murdock has reported a number of experiments which he interprets as supporting the hypothesis of a limited capacity mechanism underlying STM. Two experiments used the technique of minimal paired associate learning. The fkst of these (Murdock, 1964) showed that although the serial position curve changed with practice, the overall level of retention remained constant since a decrease in primacy was balanced by an increase in recency. In the second experiment (Murdock, 1965a), he showed that so long as total presentation time was held constant, he could trade off repetition and presentation time per pair. In a third study subjects were required to sort cards during the presentation of a list of words which were subsequently recalled (Murdock, 1965 b). As the information load of the card sorting increased, the number of words recalled decreased, implying that the subject was dividing a limited amount of processing capacity between the two tasks.While these experiments are consistent, at least qualitatively, with a limited capacity mechanism, an attempt to generalize this conclusion to other STM studies runs into difficulties. Serial memory span, for example, seems to depend less on the information per item than on the number and similarity of items to be retained (Miller, 1956;Conrad & Hull, 1964). Similarly, rate of presentation does not have the massive effect which might be expected on the basis of a limited capacity mechanism either for serial recall (Aaronson, 1967) or the digit probe technique (Waugh & Norman, 1965).Re-examination of Murdock's free recall study, however, suggests a possible explanation of these discrepancies. Murdock's serial position data suggest, as he points out, that the last few items presented are apparently not affected by the sorting task, with consistently high recall in all conditions (the recency effect).Subsequent work has shown that free recall comprises two components: a stable long-term component, and a short-term component which disappears if recall is delayed for 15 sec. or more (Glanzer & Cunitz, 1966;Postman & Phillips, 1965). This short-term component is limited almost entirely to the last few items presented, 4-2