FORAGING ECOLOGY OF A NEOTROPICAL BAT, CAROLLIA PERSPICILLATA

The foraging ecology of a Neotropical, frugivorous bat, Carollia perspicillata (Phyllostomidae), was studied in a seasonal dry forest in Costa Rica. Movement patterns were examined by capturing bats in mist nets at seven foraging sites. The fruit resources at these sites were mapped and censused regularly. Fruit preference tests were performed with free-flying bats in a flight cage. Activity at the day roost and at feeding sites was studied using a night vision scope.;The movement patterns of C. perspicillata were those of a refuging species that dispersed nightly from a central place in a manner that diminished the over-exploitation of resources by conspecifics near the roost. Resource utilization overlap with conspecifics from other roosts was decreased by low foraging site overlap. Overlap with other frugivorous bat species was reduced by low diet overlap.;Resource utilization was influenced by factors relating to resource abundance, competitive interactions, social interactions, foraging experience, predation risk and reproductive efforts. The influence that a factor had on the foraging behavior of a bat was dependent on its age, sex, and reproductive condition. The bats responded to both short- and long-term changes in parameters that influence foraging. Reproduction was timed to coincide with peaks in fruit abundance.;Food selectivity was determined by factors related to the predictability and locatability of resources. Bats preferred fruits from shrubs found in large patches that produced small amounts of ripe fruits nightly over a period of several weeks.;Behaviors such as the selection of fruits which are easily removed from plants, the use of feeding roosts away from resources, and the use of the day roost as a feeding roost during moonlit nights, suggest that the bats attempt to minimize the times of greatest prodation risk.;The foraging behavior of C. perspicillata was influenced by the dynamic interaction of factors that included both energetic (resource abundance) and non-energetic (predation risk) considerations.