Food selection by the rabbit.

1. The composition of a sward may reflect the food preferences of grazing animals. These preferences are themselves affected by certain features of the vegetation including the species present, their relative abuMance, growth stage and distribution within the sward. 2. Some of these aspects of plant-animal interaction were studied in relation to selection of food by the rabbit (Oryctolagus cuniculus Le). 3. A technique of faecal analysis, based on measurement of areas of epidermal fragments, was used to determine the diets of rabbits feeding on lowland grasslands. 4. Correction factors to allow for the effects of differential digestion of different grass species were derived by feeding captive wild rabbits known proportions of seven common grasses. 5. The relative proportions of different grasses in the faeces of wild rabbits were found to be positively correlated with their relative availabilities at eight differing grassland sites. Distinct preferences resulted in consistent disproportional selection of particular species. 6. Application of correction factors to the faecal analysis data did not alter the results qualitatively due to a positive correlation between palatability and digestibility. 7. The annual grazing pattern of a population of wild rabbits was studied over two successive years by faecal analysis. Consistent seasonal variations in species utilization occurred, including increased selection for oertain irasses just prior to flowering, followed by substantial intale of grass seeds. S. Preferences shown by captive wild rabbits in ccntralled feeding experiments paralleled the preferences of rabbits in the field. 9. Controlled choice experiments on domestic rabbits indicated that the relative frequencies of two food-types in a mixture, pre-feeding and previous experience affected selection. The behavioural responses were discussed in terms of conditioning and differential satiation. 10. Computer simulation based on the results of the present study and further experimental work was suggested as a means of further investigating the interaction between vegetation composition and rabbit grazing behaviour