THE DIET OF THE BARN OWL TYTO-ALBA IN COUNTY CORK IN RELATION TO LAND-USE

Mammalian prey of barn owls Tyro alba in various district electoral di visions in County Cork was examined against areas under cereals, sugar -beet, potatoes, other crops, hay, pasture, silage, rough grazing, woo dland and the residual area-including farms of less than 5ha, commerci al conifer stands and unimproved land-in those divisions. House mice M us domesticus, the most variable constituent in the prey (<1% to 75%) and occurring mainly around farmsteads, were strongly associated with other crops, the largest component of which is fodder beet. The latter is correlated with cereal-based fodder concentrates stored on farm, w hich, we argue, largely determine the population of house mice. Other consistent positive associations were between field mice Apodemus sylv aticus and residual area; pygmy shrew Sorer minutus and woodland; and common rat Rattus norvegicus and potatoes. Using polychotomous stepwis e logistic regression, mathematical models were constructed with land categories as independent variables and numbers of the different prey species as dependent variables. Such models can predict prey from data on land use. The most important explanatory variable was other crops, particularly for house mice. Pasture, which provides minimal cover an d food for small mammals, was not an explanatory variable. Refinements in the collection of data, to improve models, are suggested.