Ecological and demographic effects on intraspecific variation in the social system of prairie dogs

Ecology, Sept, 1995 by Steven E. Travis, C.N. Slobodchikoff, Paul Keim

Inbreeding levels, as determined by DNA fingerprints (Fig. 5), varied between colonies in the direction that we predicted based on different population densities. Specifically, mated pairs of adults were found to be more genetically similar than expected under conditions of random mating at the low density colony (PL), thus providing clear evidence of inbreeding. We calculated similarity coefficients from n = 861 pairwise comparisons of adults at PL. The frequency distribution derived from these values is shown in Fig. 6. The overall mean coefficient of similarity at PL was 0.61 (arcsine-transformed mean = 51.33), while that of known breeding pairs (n = 9 pairwise comparisons) was significantly above this value at 0.66 (arcsine-transformed mean = 54.12), with a variance of 4.58 calculated from arcsine-transformed data (t = 3.93, df = 8, P [is less than] 0.002). At the high density colony (AH), mated individuals were less similar genetically than expected; in other words, outbreeding was the rule at AH. We calculated similarity coefficients from 820 pairwise comparisons of adults at AH (Fig. 6). The population mean at AH was 0.55 (arcsine-transformed mean = 48.06), while the sample mean, calculated from all known breeding pairs (n = 15 pairwise comparisons) was 0.51 (arcsine-transformed mean = 45.79), with a variance of 6.92 calculated from arcsine-transformed data. This difference was statistically significant (t = -3.34, df = 14, P [is less than] 0.002).

[Figures 5-6 ILLUSTRATION OMITTED]

DISCUSSION

Our results suggest that the size and structure of Gunnison's prairie dog social groups undergo highly predictable changes with increases in population density, assuming that the effects of plant patchiness are held constant. These findings are in agreement with the predictions of our model. Specifically, at low to intermediate levels of plant patchiness such as those represented by PL, group size increases directly with population density, owing to the territorial recruitment of single males and females, followed by multiple individuals of one or the other sex, and finally to multiple members of both sexes. A similar situation exists under conditions of high plant patchiness, as exemplified at AH, although monogamous social systems were rare even at the lowest densities observed. With increasing population density, the switch from a social system suggestive of monogamy to one of polygyny is not surprising, in light of numerous studies which have documented its occurrence. Avian examples include Long-billed Marsh Wrens (Telmatodytes palustris; Verner 1964), Kirtland's Warblers (Dendroica kirtlandii; Radabaugh 1972), White-crowned Sparrows (Zonotrichia leucophrys; Petrinovich and Patterson 1978), and Pied Flycatchers (Ficedula hypoleuca; Virolainen 1984). Mammalian examples include Mentawai snub-nosed langurs (Simias concolor; Watanabe 1981) and beavers (Castor canadensis; Busher et al. 1983). Increases in group size with habitat saturation are also known from Acorn Woodpeckers (Melanerpes formicivorus; Koenig et al. 1984), coyotes (Cants latrans; Andelt 1985), and jackals (Cants mesomelas; Moehlman 1986), although due, in these instances, to retention of adult young within their natal home ranges.