Is habitat linearity important for small mammal communities on farmland?

1. Farmland is readily divisible into linear habitats such as hedges, and non-linear habitats such as fields and woodlots. In agricultural landscapes, conservationists have generally focused on enhancing linear habitats, but there are few data from which to judge whether or not this is a good strategy for biodiversity. 2. We investigated whether the linear or non-linear character of habitat patches, mediated by edge effects, has an impact on the abundance, diversity and richness of the small mammal communities that live within and between them. 3. In particular, we hypothesized, first, that edge effects cause narrow linear habitats to be avoided by specialists such as the bank vole Clethrionomys glareolus , but not by generalists such as the wood mouse Apodemus sylvaticus ; secondly, that edge effects lead to specialists being present in atypical habitat, through excursions at the interface between two habitats. 4. We tested these hypotheses by live-trapping small mammals in grids in the centres of three non-linear farmland habitats (woodlots, set-aside and crop fields) and in field boundaries consisting of a series of adjoining linear habitats. We compared mammal communities in the non-linear habitats with the field boundary as a whole, and with the individual linear habitat elements within the boundary. 5. There was no evidence that specialists avoided linear habitats. Indeed, the field boundary was the most species-rich habitat surveyed, and bank voles were more abundant in linear hedgerow than in non-linear woodland. Bank voles were present in linear set-aside and in the crop edge, but never in non-linear blocks of set-aside or crop, implying that they diffused out of the hedgerow into the adjacent habitats. There was no evidence of an effect of habitat linearity on field voles Microtus agrestis , wood mice or common shrews Sorex araneus . 6. Our results suggest that on uncropped land such as set-aside, the linear or non-linear character of habitats will make little difference to small mammal abundance and diversity. We advocate similar assessments for other taxa so that the effects of farm management and habitat configuration on biodiversity can be understood more fully.