Social behaviour and density regulation in house mice living in large enclosures

Abstract

One of the important issues in the population biology of house mice (Mus musculus L.) is the extent to which social grouping influences gene flow and demographic processes. This species is territorial, at least in moderately large captive colonies (Eibl-Eibesfeldt 1950; Crowcroft 1955, 1973; Anderson 1961; Crowcroft & Rowe 1963; Anderson & Hill 1965; Reimer & Petras 1967). Evidence also suggests that territories are not necessarily individual but held by social groups (Eibl-Eibesfeldt 1950; Young, Strecker & Emlen 1950; Reimer & Petras 1967; Selander 1970; Busser, Zweep & van Oortmerssen 1974). It is still not known, however, whether this arrangement produces a static genetic structure, or it the combination of group cohesion with active dispersal leads to a constantly changing genetic situation. DeFries & McClearn (1972), for example, state that 'stable populations of house mice are comprised of small breeding units, or demes, with little genetic interchange among them'. Moreover, Ehrlich & Raven (1969) use the evidence for micro-geographical differentiation in this species to argue that species are not primarily held together by gene flow as is often assumed. On the other hand, the house mouse is well known as a very successful colonizing or 'weed' species. Social behaviour may also influence birth, death and emigration rates. It is difficult to study social behaviour and geographical structuring simultaneously; the one requires intensive observation on small captive groups (see Archer (1970) for review), whilst the other requires study of feral populations. The study reported here has been an attempt to narrow this gap between laboratory and field studies. My plan was to construct enclosures large enough for the development of spatial substructure, yet small enough to obtain direct evidence of social behaviour. The experimental design differed little from some commensal populations of Mus which live with abundant food and shelter in the midst of large areas of inhospitable habitat. The main difference was that normal dispersal was prevented, but, of course shortrange movements remained possible, and escape behaviour could be monitored. Moreover, enclosures are valuable for studying frustrated dispersal and hence the normal role of dispersal behaviour (Lidicker 1975). Supplies of food, water and shelter were provided in excess of demand, although food and water were dispersed only at fixed locations. I intended to study social and demographic behaviour in the absence of constraints by these basic resources. This arrangement stressed the cohesiveness of social units and provided a test of the potential responses of this species to extreme conditions.