Chimpanzees are a naturally highly sociable species. Therefore, it is desirable to keep captive individuals in a social group. Although the dyadic introduction method has been commonly used to form social groups, it has not always been successful. In the present study, we tested a new method to create three all-male groups, two of which consisted of .ve individuals, and one of nine individuals. This method requires three stages. First, male dyads were put into neighboring cages in which the individuals could see each other. Tentative social rank was determined by observing the males' behaviors in these dyadic encounters. Next, they were moved to an unfamiliar environment and housed individually. Several days later, they were introduced one by one to an outside enclosure. Three allmale groups have been maintained so far, suggesting that group formation was successful. The occurrence of severe injury was considerably lower than that seen commonly in multi-male multi-female groups. Moreover, signi.cant effects of the group formation on male behavior (e.g., increased number of active social behaviors and decreased abnormal behaviors) were observed. These results demonstrate the effectiveness of our new method of forming all-male chimpanzee groups.
