Population biology and estimates of abundance of fruit bats (Pteropodidae) in Philippine submontane rainforest

Thirteen species of megachiropteran bats were captured in primary forest at 830-1000 m on Negros Island in the central Phillippines; 1229 individuals were marked and released. Individuals of six species were recaptured, for an overall recapture rate of 13%. For these six species, mean movement distance ranged from 0.16 to 0.75 km. For the two species with larger sample sizes, HaplonycterisJischeri Lawrence and Ptenochirus jagori Peters, adult males moved only one-half as far as females or subadult males. For three species, Cynopterus brachyotis Muller, H.fischeri and P. jagori, minimum longevities in the wild of four or five years were documented. For the same species, estimates ofcombined subadult and adult annual survival were 60 to 80%. Density estimates for six species ranged from 0.2 to 3.7 individuals per hectare, with a cumulative density of about 10 bats per hectare. Considerable differences in proportional abundance were found between different subsamples of netted bats, but most of this variation could be accounted for by variation between habitats. Four species (Cynopterus brachyotis, Eonycteris spelaea Dobson, Macroglossus minimus Geoffroy, and Rousettus arnplexicaudatus Geoffroy) that were common in agricultural habitats distant from forest were strongly associated with clearings and rare in forest; three species (Harpyionycteris whiteheadi Thomas, Nyctimene rabori Heaney and Peterson, and Ptenochirusjagori) that were as common in clearings as in forest, or slightly more common in clearings, are absent from sites more than a kilometre from forest or forest patches, and appear to use clearings principally as flyways; and one species (Haplonycterisfischerr3 that was much more common in forest than in the adjacent clearings is elsewhere present only in or very near forest or in forest patches. Estimates of density of small subcanopy fruit bats that are derived from recapture data were found to be correlated with rank-order abundance in moderate (c. 100 fruit bats) net-samples in forest, and directly proportional to abundance of fruit bats in large (c. 300500) samples in forest. This indicates that large netting samples can provide meaningful estimates of relative abundance for these species.