Disperal in Island and Mainland Voles

Dispersal was studied in the beach vole, Microtus breweri (Baird) by monthly removal of animals from an experimental 0.8—ha grid. Microtus breweri is endemic to Muskeget Island, Massachusetts. A nearby live—trap grid provided information about residents. The same design was used on the mainland to trap the meadow vole. Microtus pennsylvanicus, to compare with the beach vole, which does not undergo the regular 3— to 4—yr cycles typical of the mainland. Both populations, however, had similar peak densities (160—180/ha). In M. breweri, dispersal is a simple process. Dispersers move short distances and are a random sample of the control population for age structure and sex ratio. In M. pennsylvanicus dispersal movements are longer. More mainland ♀ ♀ disperse in summers and more ♂ ♂ in winters. The age structure of mainland dispersers is not a random sample of the control population. Seventy percent of the variance in number of dispersers is explained by the density and rate of increase of the control population in M. breweri, but only 46% is thus explained in M. pennsylvanicus. Density is the more important of the 2 variables. Twenty—nine percent of the losses in the control population are caused by dispersal on Muskeget Island with a corresponding figure of 7% for the mainland. Virtually none of the mortality in the cyclic decline on the mainland can be accounted for by dispersal. In both species, dispersers are in reproductive condition to a greater extent than residents. Mice in the island population have larger home ranges than the mainland mice. Males have larger home ranges than ♀ ♀, and on the mainland, adults have larger home ranges that subadults. Both species have larger home ranges in summers. A model is constructed in which the differences in dispersal characteristics between the 2 species are attributable to differences in availability of a place into which the dispersers can move. I suggest that the mainland population has selective dispersal leading to cycles, whereas the island has a simple drainage of excess recruits, no selection by way of dispersal, and subsequently no cycles. The model suggest how this can evolve.