Frugivorous Bats, Seed Shadows, and the Structure of Tropical Forests

Abstract

In this paper we present data on the seed shadows produced by frugivorous bats in a Costa Rican dry tropical forest and document the density and dispersion patterns of their food species. At Santa Rosa National Park, frugivorous bats are mobile foragers and have relatively broad diets. They create mixed-species seed shadows with most seeds being deposited around fruiting trees, under night roosts, or in their day roosts. Saplings and trees of small-seeded fruit species have clumped dispersion patterns, and mixed-species clumps involving associations among Cecropia peltata, Muntingia calabura, and Chlorophora tinctoria occur in disturbed forest sites. We discuss the general factors that influence plant distribution patterns and conclude that seedling establishment probabilities are concordant with seed-deposition probabilities, at least in smallseeded, bat-dispersed colonizing plant species. Singleand mixed-species clumps are the products of interactions between seed shadows created by frugivorous vertebrates and the non-random distribution of seed germination and seedling establishment sites. MANY TROPICAL PLANTS produce fleshy fruits and rely upon animals to disperse their seeds (Jones 1956, Frankie et al. 1974, Croat 1975, Opler et al. 1980). Tropical bird and mammal faunas contain numerous species that consume these fruits and in doing so provide dispersal services (Orians 1969, Karr 1971, Fleming 1973, Snow 1976). Several authors, including Smythe (1970), Snow (1971, 1976), Janzen (1971, 1978a), Morton (1973), McKey (1975), Howe (1977), and Howe and Estabrook (1977), have discussed the selection pressures that might contribute to phenological and energetic patterns of fruiting among plants and the foraging patterns of frugivorous animals. Except for Janzen (1970), there has been little discussion of the kinds of seed shadows produced by frugivores and the effects of these seed shadows on the density and dispersion patterns of animal-dispersed plants. Our objective in this paper is to present data on the seed shadows produced by frugivorous bats in a dry tropical forest and to examine the effects of these seed shadows on the distribution of bat-dispersed plants. The shape of a seed shadow includes two parameters: the best-fit function of seed density vs. distance from a seed source, and the heterogeneity around this function. The density-distance function is usually a leptokurtic distribution with a peak at the plant (Harper 1977, Levin 1979). This effect is found for seed shadows produced by all dislPresent address: Department of Biology, Kenyon College, Gambier, Ohio 43022, U.S.A. persal systems, whether or not they involve animals. The dispersal system, however, helps to determine the degree of kurtosis (Hubbell 1979), or the probability that seeds will be deposited at very long distances from the plant. Animals can influence this function through their foraging and defecatory behavior. These behaviors include the time spent near the fruiting plant, travel speeds, distances moved, and the time that seeds are retained. Heterogeneity is imposed upon this basic pattern because seeds may be found in much higher or lower densities than predicted by a random deviation from the generally leptokurtic distribution. This disturbance is caused by the habit of many frugivores of consuming and digesting their food in one or a few special places. These places include nest sites, calling perches or leks of birds (D. Snow 1962a,b,c, 1976; B. Snow 1970), and day roosts and night roosts of phyllostomatid bats (Vasquez-Yanes et al. 1975, Janzen et al. 1976, Heithaus and Fleming 1978, and Morrison 1978). We also introduce the concept of of a seed shadow as an additional descriptive parameter. We have observed mixed-species seed shadows for several tropical trees (Heithaus and Fleming, pers. obs.) that contain seeds from the parent species and from other species (foreign seeds). We have found that frequency vs. distance distributions for foreign species are also leptokurtic with respect to the parent plant being considered and do not represent the tail of a nearby foreign species' seed shadow. Mixed-species seed shadows are produced by frugiREPRODUCTIVE BOTANY 45-53 1981 45 This content downloaded from 207.46.13.11 on Mon, 17 Oct 2016 05:39:10 UTC All use subject to http://about.jstor.org/terms vores with broad diets. The bat Carollia perspicillata (Phyllostomatidae), for example, is one species that eats several types of fruits in one night and might produce high-diversity seed shadows. MATERIALS AND METHODS Data for this report were gathered at Parque Nacional Santa Rosa, Guanacaste Province, Costa Rica. A general description of our study area, which lies in the Premontane Moist Forest zone of Holdridge (1967), is given in Fleming et al. (1977) and Heithaus and Fleming (1978). Heithaus et al. (1975) includes a general discussion of bat-plant interactions in Guanacaste Province, Costa Rica. We captured bats in mist nets to characterize the diversity of seeds carried by six species of phyllostomatid bats and used radiotelemetry to document the foraging movements of one of these species, Caorollia perspicillata, as reported in detail in Fleming et al. (1977) and Heithaus and Fleming (1978). Data on the distribution of seed loads (=clumps of three or more seeds) deposited by bats under and around several fruiting and non-fruiting trees were obtained by counting the number of seed loads in four 2 m wide by 30 m long transects in each of the four cardinal directions. Ground litter and vegetation up to 2 m above the ground were carefully searched for seed loads in these transects. Bat seed loads can be distinguished from those of birds by their abundance and lack of urate material (Janzen 1978c). Complete results of the transect studies will be presented else-