tended to remain solitary as long as their fawns survived, 

 although regrouping with yearling daughters was common after 

 the fawn-rearing period. When both the matriarch and her 

 daughters of fawn-bearing age recruited fawns, they often 

 remained apart; if they regrouped, it was usually during late 

 winter or spring. Fawns, the deer most vulnerable to 

 predation, occurred in the smallest social groups; larger 

 groups contained mostly adult deer. Because solitary females 

 and their fawn(s) primarily used relatively denser cover 

 (Riley and Dood 1984), the predator avoidance hypothesis that 

 large ungulates are most likely to avoid detection by 

 predators in dense cover when alone or in small groups seemed 

 to hold. During the first 45 days of life, however, females 

 and fawns made considerable use of open habitat (Riley and 

 Dood 1984) and did not group with other deer. 



The only time that deer made extensive use of open 

 habitats was during spring when feeding on ubiquitous new 

 growth that appeared first in open habitats. For the limited 

 time period that larger groups fed together on open areas, 

 forage competition was not a problem. The most efficient way 

 to use the high density of available food and detect predators 

 was probably to feed in large groups. These groups were only 

 temporary feeding aggregations and after feeding, the deer 

 separated into smaller family groups when they moved back to 

 denser cover to bed. 



We concur with Hirth (1977) that both predator avoidance 

 and optimum feeding efficiency considerations may interact 

 with habitat type to help determine group size. However, our 

 data indicate that, in areas where a choice of habitat type is 

 available, females with fawns prefer solitude and may preempt 

 denser cover. The most observable groups are the larger 

 groups that often use more open habitat, however, these larger 

 groups did not necessarily prefer or always use open habitat. 

 On our area, larger group size was more related to poor fawn 

 survival than to the type of habitat the deer were using. 

 Thus, in late summer and early autumn 1985, deer were in 

 larger groups than usual even though they were not using more 

 open habitat than normal. It is possible that the larger 

 group sizes Hirth (1977) observed on the Welder Refuge as 

 compared to the George Reserve were at least partially related 

 to the poorer fawn survival on the Welder Refuge as compared 

 to the George Reserve. 



We agree with Hirth (1977) that density of deer did not 

 affect group size. On our area, the largest groups were 

 observed during years of poor fawn survival that included 

 years of both high and low deer density. However, we do not 

 agree that mule deer are inherently more gregarious than 

 white-tailed deer (DeVos et al . 1967). Very small family 

 groups appeared to be the preferred social group, especially 



209 



