An Experimental Analysis of the Food Location Behavior of Frugivorous Bats

Abstract

Models of optimal foraging often separate the time and energy expended in getting to a feeding area (commuting costs) from the time and energy expended in searching for food (search costs) in animals that are "pure searchers." We predict that under certain circumstances (e.g., a relatively uniform resource distribution) searching and commuting behavior will be synchronous, whereas under other circumstances (e.g., a relatively patchy resource distribution) these two behaviors will be temporally separated. We have tested these predictions by studying the foraging and food location behavior of several species of Costa Rican frugivorous bats using mist—netting programs, radiotelemetry, and fruit relocation experiments. Concurrent observations were made on the phenology and distribution patterns of the six fruit species used in the relocation experiments. The mist—netting program allowed us to determine the food habits of the common frugivorous bats in our study area and to identify the major chiropteran dispersal agents of the six fruit species. The radio—tracking program provided detailed information on the foraging behavior of one of the most important seed dispersal agents, Carollia perspicillata, whose individuals commute up to 2.7 km from a central day roost before beginning to feed. Results of the fruit relocation experiments, in which ripe fruits of the six species were individually placed on "fruits poles" located either close to or at a considerable distance (>0.5 km) away from conspecific plants in areas known to be used by bats, indicated the following: (1) fruits of two species of Piper, which are highly preferred by Carollia and which occur in low nightly densities for extended periods of time, has as high (°.90) a probability of being found away from conspecific plants as when they were near conspecifics; (2) fruits of two species of Ficus and Muntingia calabura, which are very patchily distributed in time and/or space, has a significantly higher probability (.30—.50 vs. .02—.12) of being found near rather than away from conspecifics; and (3) when placed on poles with Piper fruits, fruits of Ficus ovalis, Chlorophora tinctoria, and Muntingia calbura had a significantly higher probability (.40—.60 vs. .08—.13) of being found than when placed in similar areas without Piper fruits. These results indicate that certain bats (e.g., Carollia perspicillata and Glossophaga soricina) are constantly "on the alert" for ripe Piper fruits while commuting, whereas bats that eat Ficus, Chlorophora, or Muntingia fruits (e.g., G. soricina or Arbibeus jamaicensis) are not constantly "on the alert" for these fruits while commuting. Bats feeding on Piper probably search and commute simultaneously, whereas those feeding on Ficus or Muntingia separate commuting and searching behavior. The food location behavior of frugivorous bats appears to be highly responsive to differences in the spatiotemporal distribution patterns of their food resources.