GENE FLOW IN HOUSE MICE: INTRODUCTION OF A NEW ALLELE INTO FREE‐LIVING POPULATIONS

Abstract

It is a widespread view that house mice, Mus musculus, live in behaviorally isolated tribes or demes often simply referred to as social groups. However, several problems with the basis for this view may be identified: (1) no trapping system can show that mice restrict their breeding to a particular tribe; (2) low vagility is not equivalent to an exclusive home range: an occasional copulation between groups is very hard to disprove; and (3) sampling by removal disrupts existing social groups by selectively removing dominant mice which causes immigration into the area where removal occurred (Adamczyk and Walkowa, 197 1; Baker, 1980, provides extensive references that are omitted throughout this report). Studies of the genetic and population structure in house mice that are reported here grew out of the need to resolve a paradox. Whereas the observed segregation ratio of the lethal alleles (tlu) at the T locus predicted high frequencies of tl' in natural populations, field data revealed low frequencies (Bruck, 1957; Bennett, 1975). This difference between observed and expected gene frequencies was explained by assuming that house mice lived in semiisolated, small, tribal groups, in which genetic drift would play a major role in determining gene frequencies bringing the expected gene frequencies into concordance with observations (Lewontin and Dunn, 1960). Levin et al. (1969) simulated