We investigated the diet of coyotes (Canis latrans) on the Pifion Canyon Maneuver Site (PCMS) in southeastern Colorado from February 1983 to June 1986. Based on regression analysis, 71% of the variation in the volume of large prey (e.g., adult ungulates) in coyote scats could be explained by coyote group size; 77% of this variation could be explained by coyote group size and snow depth. Snow depth alone explained 28% of the variation in the volume of large prey in the scats. Coyote group size explained 19% of the variation in the volume of medium size prey (e.g., lagomorphs and ungulate fawns) and 38% of the variation in the volume of small prey (e.g., rodents). Coyote groups were largest in winter and smallest in summer. Most juveniles dispersed before the largest group sizes were observed. Group foraging and snow depth may influence prey selection by coyotes in southeastern Colorado. J. WILDL. MANAGE. 52(4):647-653 Diets of coyotes have been examined (Fichter et al. 1955, Gipson 1974, MacCracken 1981) but few studies have examined the relationship of prey selection and coyote social organization. In Jasper National Park, Alberta, during summer, when rodents were the major food, mean coyote group size was 1.4 (Bowen 1978, 1981). In winter, when ungulates were the major food, group size averaged 2.0. Mean group size in Wyoming was 1.3 coyotes in summer, when rodents were the major resource consumed; in winter, when ungulate carrion dominated the diet, coyote group size was 1.8 (Bekoff and Wells 1980). Bekoff and Wells (1980) noted that during 3 successive winters in Rocky Mountain National Park, Colorado, when no ungulate carrion was available, coyotes preyed on small mammals throughout the year. Mean group size for coyotes was 1.1 in summer and winter. Bowen (1978, 1981) and Bekoff and Wells (1980) concluded that when coyotes were feeding on large prey, group sizes were correspondingly larger, which increased foraging efficiency. They further observed that delayed dispersal of pups resulted in the formation of larger packs; however, seasonal variation in the diet still accounted for much of the variation in the number of pack members observed together. Messier and Barrette (1982) found no relationship between prey size and coyote group size in Quebec; they concluded that larger coyote groups could be explained by delayed dispersal of juveniles. The obj ctive of our study was to document the diet of coyotes in southeastern Colorado and examine the relationship between prey size, pack sociality, time of juvenile dispersal, and group