Reproductive Ecology of Fruit Bats and the Seasonality of Fruit Production in a Costa Rican Cloud Forest

In a premontane cloud forest in Costa Rica, fruit biomass of bat-visited plants showed two seasonal peaks per year. The first peak occurred during the dry/wet seasonal transition (April-May), and the second fruiting peak occurred in the late wet period (September-October). Lactation in two common fruit-eating bats, Artibeus toltecus and Sturnira ludovici, was also bimodal and coincided with seasonal peaks in fruit abundance and nutrient availability. Both bat species fed heavily on the fruits of abundant second-growth shrubs, although during the late wet season A. toltecus consumed mainly the fruits of mature forest tree species. Fruits eaten by premontane bats are high in water content, soluble carbohydates, and protein, but low in lipids. Fruits consumed by bats during lactation are extremely high in soluble carbohydrates but lower in protein on a per-gram basis than other fruits available at this interval. In Monteverde, bats did not eat insects during lactation periods as a means of increasing nutrient uptake. This suggests either that protein availability during fruiting peaks is sufficient or that lactating bats might rely on protein reserves established prior to parturition. TROPICAL FORESTS SUPPORT LARGE ASSEMBLAGES of vertebrate frugivores (Orians 1969, Fleming 1973) because many tropical plants produce fleshy, edible fruits (Frankie et al. 1974, Croat 1975, Howe & Smallwood 1982). Yearround fruit availability allows groups like fruit-eating bats (Phyllostomidae) to specialize on this food source (Gardner 1977, Fleming 1982, Heithaus 1982). Within this plant-frugivore system, however, reproductive patterns are distinctly seasonal. Peaks in fruit abundance generally coincide with or are triggered by increasing rainfall (Smythe 1970, Daubenmire 1972, Frankie et al. 1974, Opler et al. 1976, Hilty 1980, Foster 1982a, b), and it is often assumed that reproductive peaks for frugivorous bats are ultimately determined by food availability (Lim 1970, Fleming et al. 1972, Humphrey & Bonaccorso 1978, Bonaccorso 1979, Racey 1982, Thomas and Marshall 1984). Selection should favor the timing of peak reproductive stress to coincide with enhanced food or nutrient availability because reproduction is energetically costly, especially in mammals where lactation can increase female energy demands by 66-133 percent (Migula 1967, Randolph et al. 1977, Millar 1978). This study was undertaken to examine the timing of reproduction in two species of fruit bats, S. ludovici and A. toltecus (Phyllostomidae), with respect to fruit availability. Specifically, I asked if lactation schedules in these two abundant bats coincided with seasonal peaks in fruit abundance. No study to date has attempted to measure fruit abundance directly while simultaneously monitoring reproductive activity in bats. Instead, most relied on casual observations of fruit abundance or on data collected by other workers from different locations or from different years (Racey 1982). Only one study has measured insect abundance, rainfall, and lactation schedules for tropical insectivorous bats (McWilliam 1982), revealing close synchrony between peak insect abundance and nursing periods. However, annual variation in rainfall and fruit availability can be high in some tropical forests (Rand & Rand 1982, Foster 1982a, Terborgh 1983); this finding suggests the hypothesis that selection may lead to synchrony between lactation and long-term modal peaks in fruit availability rather than coincidence with a given annual peak. During the breeding season, many tropical fruit-eating birds consume a higher percentage of insects and small vertebrates to increase protein intake (Morton 1973). Several food habit studies of fruit-eating phyllostomids show that insects appear regularly in the diet (Fleming et al. 1972, Ayala & D'Allesandro 1973, Howell & Burch 1974, Gardner 1977), although the extent of insectivory during lactation remains unclear. In contrast, Heithaus (1982) reported at least three species of phyllostomids that could be classified as obligate frugivores. Several authors have speculated that tropical fruits must be more nutritious than temperate fruits because tropical frugivores are more dependent on fruits than are temperate species (McKey 1975, Howe & Estabrook 1977, Herrera 1981). Most recently, Thomas (1984) argued that energy budgets and foraging strategies of paleotropical fruit-eating bats differ substantially from Neotropical phyllostomids because the former eat fruit exclusively whereas the latter supplement a fruit diet to a large degree with in-