Genetic and Demographic Differences between Dispersers and Residents in Cycling and Noncycling Vole Populations

Abstract

Dispersal was studied for one year in Massachusetts in a cycling (main- land) and a noncycling (island) population of Microtus. On resident and removal trap- ping grids demographic data and genetic data at the transferrin locus were collected. During the period of study the population of Microtus pennsylvanicus in Manomet was in the low and early increase phase of a multiyear population cycle. The M. breweri population on Muskeget Island maintained a seasonal cycle. The present study contrasted both the genetic and demographic attributes of dis- persing and resident voles in Manomet and on Muskeget to determine the nature of the changes in dispersal and whether dispersing voles are a random sample of the control population. Dispersal in both areas was related to density. One of four measures of dispersal, the number of immigrants, was correlated with population density on the resident control grid in Manomet. Two measures of dispersal, the recovery ratio and number of immigrants, were correlated with density on Muskeget and the relative recruitment index was correlated with both density and growth rate. Voles on Muskeget colonized the removal grid faster than voles on the mainland. Twice as many dispersers per new recruit were found on Muskeget. In both areas dispersers differed from resi- dents in sex ratio, male weight, and male transferrin allele and genotype frequency. An excess of male dispersers was observed both in Manomet and on Muskeget although the heterogeneity between residents and dispersers was confined to the adult age group in Manomet. Heavier males dispersed during the trapping period in both areas. An excess of the Tfc allele was found in dispersing island and mainland voles. This was manifested as an increase in TfCl TfC genotypes dispersing in Manomet and an increase in dispersing TfCITfJ genotypes on Muskeget. There were two striking contrasts between dispersers at Muskeget and Manomet. First, on Muskeget Island female emigrants were a random sample of the control pop- ulation with respect to every attribute measured including age structure, weight, and allele and genotype frequency. In contrast, an excess of Tfc homozygotes was found among dispersing female voles in Manomet and mean body weight of emigrants was less than that of residents. Second, although there were genetic differences between dispersing and resident male voles in both areas, they were confined to different age groups. On Muskeget the adult age class showed heterogeneity between the control and removal grids, whereas in Manomet it was the subadult class. If, as suggested by Tamarin (1977a, 1978), the only difference between Muskeget and Manomet is the lack of an adequate dispersal sink on Muskeget, then dispersing animals that differ from residents in Manomet and those that are similar to residents on Muskeget Island should represent animals predisposed to disperse. The present results suggest that female and subadult male voles play a key role in regulating population cycles. Microtine rodents characteristically undergo 3- to 4-year cycles in population den- sity. Several theories have attempted to explain the regulation of these cycles through food, predation, stress, and genetically controlled behavior. A thorough review is provided by Krebs and Myers (1974). The present study investigates the influence of one of these mechanisms, selective dispersal, on the dynamics of two vole populations. The importance of dispersal as a mechanism regulating population density was first shown experimentally by Krebs et al. (1969). Enclosed voles demonstrated aberrant demography in which numbers built up to very high density, with resultant habitat destruction causing a sharp decline in numbers. Chitty (1967, 1970) proposed a ge- netic-behavior hypothesis in which genotypes are differentially tolerant to conditions