Abstract: This paper reports quantitative leadership differences in semi-captive bar-headed geese (Anser indicus) at different times of the year, and in different types of groups. Leading is defined here as causing the departure or determining the direction of movement of the whole group. No permanent and exclusive leader of a pair or family group was found, rather relative leading frequencies of male, female and young showed a definite shifting pattern. Females led more often than their mates prior to breeding, and on nest pauses during the incubation period, but less often in summer, autumn and early winter. In families there was no difference between the frequencies of male and female leading. Family females led relatively more often than those of pairs without offspring. This difference was related to the presence, not the number, of young. Goslings led the family about as often as the parents during the rearing period in early summer, less often in autumn, winter and next spring. Such differences and changes are to be expected where competence in particular tasks and dependence on partners vary between group members, and where different situations require different abilities. For the geese, the results can be related to the different options of group members and to the different benefits they derive from leaving (or 'staying put') or following (or waiting for the others) in different situations.

Abstract: Ten feral mares free-roaming in Maryland, USA, were inoculated with porcine zonae pellucidae (PZP) protein before the breeding season for three consecutive years (1988-90). Ovarian function was monitored for 51 days during the peak of the breeding season after the third annual PZP inoculation, in seven of these mares and in four untreated control mares, by means of urinary oestrone conjugates and nonspecific progesterone metabolites. None of the ten inoculated mares became pregnant in 1990, compared with 55% of 20 control mares, which included two of the four monitored for ovarian function. Three of the untreated mares demonstrated apparent normal ovarian activity, characterized by preovulatory oestrogen peaks, concurrent progesterone nadirs at ovulation, breeding activity, and luteal-phase progesterone increases after ovulation. Two of the seven monitored PZP-treated mares demonstrated ovulatory cycles that did not result in conception. One was pregnant as a result of conception in 1989 and demonstrated a normal, late-gestation, endocrine profile. The remaining four PZP-treated mares revealed no evidence of ovulation, and urinary oestrogen concentrations were significantly depressed. The experiments indicated that (i) a third consecutive annual PZP booster inoculation is greater than 90% effective in preventing pregnancies in mares and (ii) three consecutive years of PZP treatment may interfere with normal ovarian function as shown by markedly depressed oestrogen secretion.

Abstract: 1. The role of lion Panthera leo predation in the dynamics of blue wildebeest Connochaetes taurinus and zebra Equus burchelli populations was investigated through simulation models. The data used in the models were from intensive observations over 4 years in the south-east of the Kruger National Park. 2. Population estimates of wildebeest and zebra were made from aerial surveys, sex and age ratios from ground counts. Lion numbers were determined from observations of marked and radio-collared animals. Predation was studied by following lions for continuous periods of up to 336 h. 3. Two models were constructed. Model 1 ascertained the number of killing lions (adult females) that could be supported by each prey population while remaining stable. A single model was constructed for the sedentary wildebeest population. A summer and winter model was constructed for the semi-migratory zebra population. The sensitivity of the parameters in the model was tested by changing their value by 10%. In model 2, the kill age structure for each species was changed to determine the number of killing lions the altered prey selection parameters could support. 4. There was no difference in the vulnerability of either species to predation. Zebra foals (<1 year) were killed more frequently than expected. No selection for sex or by season could be found for either species. 5. Model 1 predicted that the wildebeest population stabilizes with 7.7 killing lions, close to the number in the study area. The winter zebra population stabilizes with 6.8 killing lions and the summer zebra population with 19.4. Manipulation of kill rate followed by adult fecundity rate had the greatest effect on population size of both species. In model 2, wildebeest predation was made selective towards calves and zebra predation was made non-selective for sex and age. With these parameters the wildebeest population stabilizes with 10.7 killing lions and the zebra population with 5.4 in winter and 15.1 in summer. 6. The models suggest that lion predation affected wildebeest more severely than zebra during the study. This was through the way in which lions selected their prey, and because of the sedentary behaviour of the wildebeest, as opposed to the semi-migratory behaviour of the zebra.

Abstract: In nature foragers must exploit resources that vary randomly in both the energy acquired per item (reward) and the time required to pursue, capture and process an item (delay). Furthermore, rewards and delays associated with particular resources may often covary significantly. An analytical model asks how variance-covariance levels for rewards and delays could influence choice of resources when lack of information or cognitive limitation implies that a consumer attempts to maximize its short-term rate of energy gain. Both greater expected reward and reduced expected delay clearly should enhance preference for a resource. The model predicts that increased delay variance and reduced reward-delay covariance should increase a forager's preference for a resource. A forager should be risk-averse towards reward variance when the reward-delay covariance is positive, but should become risk-prone towards reward variance when the reward-delay covariance is negative.