Abstract: Feral pony mares, Equus caballus, at Assateague Island, Maryland, formed linear dominance hierarchies within bands. Generally, older mares dominated younger mares, and larger mares dominated smaller mares. Large mares initiated aggression more often than small mares when age was controlled for but, surprisingly, older mares initiated aggression less often than younger mares when size was controlled for. Thus, mares peak in aggressiveness fairly soon after achieving full size and then, while maintaining or improving their rank in the domainance hierarchy, progressively reduce their involvement in aggression as they grow older, Involvement in aggression per mare increased as number of mares in the group increased; this effect was independent of nearest-mare distances. Frequency of involvement in aggression did not differ between mares that had changed bands within the year and mares whose band association had continued for a year or more. Aggression was directed more frequently than expected at subordinate mares who were nursing, and also occurred more frequently than expected at water holes. The proportion of aggressive encounters during grazing closely matched the total proportion of time spent grazing. Subordinate mares with foals received aggression more often than subordinate mares without foals. The high frequency of aggression associated with foals and nursing suggests that interference with reproduction of subordiantes is an important mode of competition between mares. Such interference may be common in animals that feed on dispersed resources and live in small, cohesive groups.

Abstract: n a herd of semi-captive plains zebras interventions, which occurred within the harems, were investigated in order to answer the question why zebras interfered. These interventions are of interest because they regulate the contacts between companions and because, as corrective and preventive measures, they reveal the normative principles underlying the behaviours by which animals structure their social environment. An attempt was made to deduce 1) the internal norms of the interferer; 2) his short term aims; 3) his tactis and 4) his perception of the social environment. The analysis revealed that in the case of an affiliative interaction foals, yearlings and adult mares started to interfere if a friend had an affiliative contact with another zebra. In view of the interferer's behaviours it was concluded that their aim was to break off the ongoing interaction and that zebras tended to protect friendship bonds. Foals and yearlings further interfered if their mother was being threatened, attacked or sexually approached by a stallion. In view of the interferer's behaviours its aim was to prevent iminent interactions or to break off ongoing interactions. This suggests that these interventions were of a protective nature. The interferer's behaviours in both contexts ware very much alike. Mares tended to interfere if their foal/yearling or adult daughter was threathened or aggressed or if a mare friend was being sexually approached by a stallion. This type of intervention was of a protective nature. Stallions in a multi male harem showed a high tendency to interfere in courtship interactions. From the resemblance between interventions in courtship and in aggressive interactions it is concluded that, at leat in a number of cases, the individuals have perceived courtship behaviour by the stallion as a threat towards the mare involved.

Abstract: In a trial of cage designs for laying hens, eggs were discovered in dust baths where access was restricted by a closed door during the normal laying period (08:00-13:00 h). Observations showed that the hens in these dust bath treatments had developed methods of opening the doors in order to lay in the baths. Three different methods of opening were observed. An average time of 34.4 min was spent attempting to open the doors before access was finally achieved. This implies a strong nesting motivation in these hens. The proportion of eggs laid in the dust baths increased (with occasional fluctuations) over a 24-week period. Door opening is likely to have initially developed in one individual in each cage through a trial and error basis, and then have been learned by cage mates through imitation. The speed and efficiency of door opening was not found to increase with experience or time.

Abstract: The effect of social rank on the feeding behaviour of female red deer, Cervus elaphus L., on the Isle of Rhum, Scotland, was investigated. Hinds were less likely to approach and more likely to leave the vicinity of other individuals if these hinds were dominant to them. Movements away by subordinates were more likely to involve a break from feeding. Feeding rate, as measured by bite rate, increased with distance from dominant neighbours, but was unaffected by the distance to subordinates. It appears that aggressive interactions had little direct effect on access to food. Instead, it is suggested that feeding competition in red deer hinds is largely a passive process, operating through the avoidance of conflict by subordinates.