Variation in body size, sexual dimorphism and age-specific survival in stoats, Mustela erminea (Mammalia: Carnivora), with fluctuating food supplies

Most hypotheses attempting to explain the evolution of pronounced sexual dimorphism in body size in the three species of weasels (Mustela erminea, M. frenata, M. nivalis) assume that sexual dimorphism is a long-term adaptation, associated with the different reproductive strategies of the two sexes. We here examine an auxiliary hypothesis which predicts that the degree of sexual dimorphism may also vary over the short-term because when food is temporarily abundant, sexual selection should favour a greater growth rate of males than of females. This hypothesis concerns a phenotypic response which could introduce temporarily increased variation into an existing genotypic trait. We document the present size and sexual dimorphism of stoats introduced last century to New Zealand from Britain in relation to between-year variation in food supply in a single habitat (forests of southern beech, Nothofagus. sp.). Southern beech trees produce heal ll crops of flowers and seed at 3-5 year intervals, which are associated with very variable supplies of important prey of stoats, including several species of seed-eating birds, litter-feeding insects, and feral house mice (Mus musculus). Alternative prey are scarce. Regressions of condylobasal length and head-body length on mouse population indices were significant in both sexes. Mean condylobasal length was larger in both male and female stoats born after a heavy seedfall compared with those born in non-seedfall years. However, the largest males born in years of heavy seedfall were removed by selective mortality before dhe age of 3 years, so the condylobasal lengths for old (greater than or equal to 3.0 yr) males converged on a common mean regardless of food supply in their birth year. Sexual dimorphism did not vary with food supplies (as reflected in seedfall records ol mouse population indices) al any age. First-year survivorship, at least from the age of independence, was significantly negatively correlated with density of stoats in the summer of their birth year. (C) 1997 The Linnean Society of London.