Effects of forest fragmentation on the winter body condition and population parameters of an habitat generalist, the wood mouse Apodemus sylvaticus: a test of

Three main causal hypotheses have been proposed to explain the inverse relationships between habitat patch size and density of gen- eralist mouse species in fragmented habitats: 1) enhanced habitat conditions as habitat patch size decreases; 2) inhibited emigration of excess indi- viduals in small and isolated habitat patches; and 3). reduced territoriality in small patches because they are occupied temporarily by non- reproductive individuals. From the mechanism underlying each hypothesis, we derived predictions on the effects of fragment size on the body condition of individuals (measured both as absolute body size and as body mass relative to body size) and some demographic parameters of mouse populations redated to reproductive output (sex-ratio and proportions of sexually active and recently-born individuals), and we tested such predic- tions with data from wood mice Apodemus sylvaticus wintering in three Spanish forest archipelagos in which the inverse relationship between for- est patch size and mouse abundance had been previously proven. No differences in average body size or in average body mass relative to body size were detected among fragments. Mouse populations wintering in small fragments showed more male-biased sex-ratios, a larger proportion of sex- ually active adults and fewer juveniles as compared to mouse populations wintering in large fragments nearby. Results clearly rejected the third hypothesis and did not support the second one. It thus seemed that habitat conditions for mice improved as forest fragment size decreased, although the expected positive effects on individuals could have been prevented by relaxed territoriality and increased food resource depletion by denser mouse populations. Bearing in mind the negative effects of dense wood mice populations on the distribution, abundance and population dynamics of forest species, this apparent enhancement of habitat conditions for mice in small forest fragments could have far-reaching conse- quences for the long-term persistence of such fragments.