female group size, r=-0.180, n=9 (all P's > 0.05). Deer 

 numbers were 3.15 times higher at the high than the low, so 

 the available range of densities should have been sufficient 

 to show differences if they occurred. Although statistically 

 non-significant, the negative correlation coefficients may 

 indicate a tendency toward smaller group size as density 

 increased. 



General observations suggested that mature females were 

 socially oriented towards their fawn(s) first, and exhibited 

 minimal relations with other adult deer. However, as the 

 biological year progressed from fawning, females increasingly 

 associated with other females, especially during years of poor 

 fawn survival. This suggested that female group size was 

 related to fawn survival. A test of that hypothesis with 

 correlation analysis indicated a strong negative relationship 

 (Figs. 7.7 and 7.8) between the number of females per group 

 and fawn: 100 female ratio in the population during autumn 

 ( r =_0.907, P<0.01, n=14). A similar relationship was 

 indicated with respect to numbers of fawns: 100 females in the 

 population during early winter (r=-0.920, P<0.01, n=15). Data 

 for autumn 1985 were not included in the analysis; that data 

 point did not fit the linear regression line, probably because 

 most fawn mortality had occurred by mid-summer such that 

 females had a longer than usual time to reassociate prior to 

 the autumn survey. These correlations indicated that the 

 higher fawn production and survival through early winter, the 

 less likely that females were associated with other females. 

 Group size increased as fawn survival declined both within and 

 between years . 



The same data displayed in a different manner (Fig. 7.9) 

 led to additional conclusions. During early autumn, the 

 largest decrease in fawn: female ratio occurred between single 

 females and groups of 2 females. The degree of decline 

 between these as compared to the decline between other sized 

 groups indicated that the second female in the group was the 

 most likely of all additional females to not have a fawn. 

 Data from marked deer verified this and indicated that the 

 second female in the group (first one added in autumn) was 

 most likely an unproductive yearling daughter. 



For the early winter period, the change in fawn: female 

 ratio was about the same from 2 to 3 females /group as it was 

 from 1 to 2 females/group. Again, data from marked deer 

 indicated that by early winter, the third female added to a 

 group was most likely to be either a second yearling daughter 

 that had not rejoined the family group by early autumn or an 

 unproductive 2-year-old daughter. Little difference in 

 fawn: female ratios occurred among female group sizes 3, 4, 5, 

 and 6 during early winter, indicating that the fourth, fifth, 

 and sixth females added to a group were about equally likely 



191 



