Current models of chimpanzee social structure differ in the extent to which females range with the males and are loyal to a particular social group. We tested these models by analysing 18 years of observational data on the Gombe chimpanzees to investigate the relationship between female space use patterns and both female feeding competition and changes in the male-defended range boundaries. Females at Gombe typically spend most of their time in small overlapping core areas within the community range. Most core areas clustered into two neighbourhoods, north and south. Most females maintained a high degree of site fidelity, but altered their space use patterns to stay within a male-defended boundary. This overall pattern supports the male-bonded model of the chimpanzee social system, over the bisexual or male-only community models, but there are many exceptions. Some females were very peripheral and may have associated simultaneously with two communities. Others switched communities as adults. Thus, different individual females displayed a variety of space use patterns. Different space use patterns of individual females were associated with differences in reproductive success. Members of the northern neighbourhood had higher reproductive success than those of the south, and peripheral individuals either did very well or very poorly. Females that moved from one community to another as adults produced the fewest surviving offspring. These results suggest that female ranging patterns are influenced by both feeding competition and male territorial behaviour.
Bibliographical noteFunding Information:
We thank the research staff at the Gombe Stream Research Centre for data collection, especially H. Matama, I. Yahaya, H. Mkono and E. Mpongo. We are indebted to Tanzania National Parks, the Serengeti Wildlife Research Institute and the Tanzanian Council for Science and Technology for approval of this work over the decades. Thanks to J. Waterman for overseeing data entry early on, J. Schumacher-Stankey and all of the data enterers, and D. Scheel, T. Susman, T. Sperry and C. Ryan for programming on demand. Thanks also to D. Hawkins for statistical advice. This work was funded in part by NSF grants DBS-9021946, SBR-9319909, the Graduate School and the College of Biological Sciences of the University of Minnesota, the American Philosophical Society, the Carnegie Corporation, the Jane Goodall Institute, Milton Harris, the L.S.B. Leakey Foundation, the Minnesota Base Camp and the Windibrow Foundation. The research presented here was described in Animal Research Protocol No. 9310004, approved by the Institutional Animal Care and Use Committee of the University of Minnesota.