DORMOUSE DISTRIBUTION - SURVEY TECHNIQUES, INSULAR ECOLOGY AND SELECTION OF SITES FOR CONSERVATION

1. A stratified sample of 238 woodlands in Herefordshire was surveyed to determine the incidence of dormice in relation to age, area and isolation of the site. Characteristically gnawed hazel-nut shells were sought as indicators of dormouse presence. 2. Dormice were more frequently found in ancient woodland than in recent woodland, supporting a previous hypothesis that they are an insular species. 3. Both site isolation and area influenced distribution, in accordance with insular ecological theory. Prevalence of woodland and boundaries around a site affected dormouse incidence. The results suggest that isolated woodlands less than about 20 ha in extent are unlikely to support viable dormouse populations. 4. Dormouse distribution was predicted for each woodland in Herefordshire, on the basis of woodland area and isolation. This included woodlands lacking hazel, where dormice would be difficult to detect by field survey. The data imply a possible metapopulation model. 5. Management policies should consider hedgerows and other arboreal links between a woodland and its neighbours, not just habitat within the site itself. Re-introductions and conservation management are most likely to be successful in larger, ancient woodland sites and most likely to be useful if these are close to, or linked with, adjacent woodland.