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

Key words: Cynomys gunnisoni; group size; inbreeding; monogamy vs. polygyny; plant patchiness; population density; prairie dogs; social systems; territoriality.

INTRODUCTION

While vertebrate social systems were once believed to represent fixed species attributes, evidence now suggests that many species are socially flexible in response to spatially or temporally varying environments (Lott 1984, 1991, Slobodchikoff 1984, 1988, Slobodchikoff and Schulz 1988). Hypotheses for social system variability (Lott 1991) have tended to focus on either ecological or demographic factors in spite of the concerns of some investigators that a given social system will rarely be attributable to any single factor (Caro and Bateson 1986). In particular, the social strategies of interacting conspecifics have largely been ignored as a means of predicting social system variation (Lott 1991).

Many species of small rodents are known to alter both the size and structure of their social groups in response to changing ecological and demographic conditions (Lott 1991). They thus represent ideal subjects for the modeling of environmental effects on social system variation. These species are frequently capable of explosive population growth leading to pronounced habitat alterations, particularly in relation to food resource distributions. Selection for considerable social plasticity would be expected under such environmentally variable conditions.

Gunnison's prairie dogs (Cynomys gunnisoni) are highly social ground squirrels whose range is limited to grasslands of the Colorado Plateau in western North America (Hall and Kelson 1959). Dense aggregations of these animals form extensive towns, or colonies, within which social groups occupy fixed territories. These groups are known to vary annually in both size and composition (Travis 1994), and are characterized by a high frequency of amicable social interactions (Rayor 1988). Groups are typically composed of one or more adult males, one or more adult females, and young of the year (Slobodchikoff 1984, Rayor 1988, Travis and Slobodchikoff 1993, Travis 1994). Most litters are sired by multiple males, and 60% of the offspring produced per litter are the result of extraterritorial fertilizations (Travis 1994).

There are several ecological correlates of social system variation in Gunnison's prairie dogs (Slobodchikoff 1984, Travis and Slobodchikoff 1993). Slobodchikoff's (1984) model predicts an association between resource abundance (as well as its distribution) and the mating system. This habitat variability--mating system model predicts group defense of resources under conditions of abundant but patchily distributed resources. These groups are expected to consist of a core of females that, when joined by a single male for the purposes of mating, will lead to the existence of a polygynous mating system. Uniformly distributed resources, on the other hand, are predicted to preclude the formation of female groups due to the negative effects of competition, thereby leading to the existence of a monogamous mating system. Slobodchikoff (1984) tested these predictions by performing habitat alterations on a prairie dog colony characterized initially by an abundance of patchily-distributed resources. Mowing of this colony effectively established a uniformity of resources without affecting overall resource abundance. Prairie dogs responded in the predicted fashion by switching from polygyny to monogamy. Travis and Slobodchikoff (1993) provided further correlative evidence in support of the habitat variability--mating system model by demonstrating social system variation between two prairie dog colonies differing only in food resource distributions. As predicted, prairie dogs occupying the colony with a highly patchy distribution of plant food resources displayed a social structure suggestive of polygyny, while prairie dogs occupying the more homogeneous site were characterized by a predominance of monogamous social systems.

The purposes of this paper are twofold. First, we will expand the habitat variability--mating system model (Slobodchikoff 1984) to include possible associations between population demography and mating systems. Second, we will consider associations of resource distribution and population demography, not only with particular mating system types, but also with certain social strategies practiced by individuals, particularly in relation to inbreeding. The model that will emerge, which we shall call the resource distribution--population density model, will represent one of the first attempts to account for interactions among ecological and demographic factors, as well as social strategies, associated with the evolution of mating system variability.

THE MODEL

Fig. 1 shows the predicted associations between plant patchiness and population density, and the size and composition of Gunnison's prairie dog social groups. This part of the resource distribution--population density model predicts that at low plant patchiness, social groups will remain small, consisting only of single pairs of males and females, regardless of population density. The prediction is made on the basis of the original habitat variability--mating system model, in which larger groups are expected to form only as necessary for the adequate defense of resources. Under conditions of sustained resource uniformity, increasing population density will not, therefore, become associated with the formation of larger groups, but with the formation of a greater number of monogamous pairs. Under conditions of high plant patchiness, we predict that population density will correlate with group size in such a way that, at intermediate densities, groups of females will join with individual males in a system of apparent polygyny. At high densities, they will be unable to exclude additional males and groups will begin to include multiple males and females. Finally, we expect that, at low population densities coupled with high plant patchiness, the existence of groups of any kind will be precluded, due to difficulties related to resource defense in the absence of a sufficient pool of individuals from which to recruit group members.