JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecological Monographs.Abstract. I studied the proximal factors influencing dispersal from the natal site in two free-living populations of Belding's ground squirrels (Spermophilus beldingi) in the Sierra Nevada of California. I tested 10 hypotheses, each suggesting a different social, ecological, or ontogenetic factor as a proximal cause of natal dispersal in this species. Using discriminant analysis I also examined effects on dispersal of several independent variables concurrently. Data from marking and live-trapping studies over three field seasons (1979)(1980)(1981) failed to support hypotheses suggesting resource shortage, ectoparasite load, social facilitation, conspecific aggression, or avoidance by conspecifics as proximal causes of natal dispersal in S. beldingi. Furthermore, dispersal is apparently not caused by changes in juveniles' response thresholds to conspecific aggression or by juveniles' attempts to avoid members of their family units, nearest neighbors, or other members of their local populations.My data supported an "ontogenetic switch" hypothesis that suggests natal dispersal by male S. beldingi is triggered by attainment of a particular body mass or body composition, or some combination of these two variables. Other data indicated that this ontogenetic switch initiates not only dispersal behavior per se, but also a larger syndrome of behaviors that appear to be functionally related to dispersal. Specifically, this syndrome involves changes in exploration, responses to frightening stimuli, and locomotor behavior exhibited by dispersers. I discuss possible relationships among variables operating at different proximal levels (sensu Tinbergen) to cause natal dispersal in S. beldingi and other mammals. I conclude that a complex suite of variables probably operates at genetic, physiological, and socioecological levels to proximally cause natal dispersal in each species. The ultimate (evolutionary) hypotheses most consistent with my data from S. beldingi are that dispersal: (1) reduces nuclear family incest, (2) optimizes inbreeding, and (3) improves access to mates.