Operating from a research station situated in one of the last primary forest tracts of Panay Island, Philippines, the interaction between the flying fox species Ptenochirus jagori (Megachiroptera, Pteropodidae) acting as a frugivore and the forest was examined. During 27 months of field work (October 1998 to May 2001), the first data on home range of these nocturnal animals could be gathered by monitoring individuals using radiotelemetry. The location of feeding places made it possible to test functional hypotheses of seed dispersal (e.g. the escape hypothesis). Moreover, germination experiments were conducted in order to evaluate the consequences of seed treatment by the bats to seed germinability. Additional studies dealt with the spatio-temporal availability of exploited plants in conjunction with habitat use and feeding strategy. With respect to nutrition, leaf-eating (folivory) as a potential protein supplementation was of particular interest. Leaves consumed by bats as well as those of randomly selected trees were analysed for their nutrient content, proteins included. By comparing site characteristics of roosting places (e.g. woody vegetation and foliage distribution around roosting trees) insight was gained into an important aspect of P. jagori's life habits. Roost-related home ranges of five animals of both sexes varied from 8.4 to 30.9 ha during a mean intensive tracking period of 11.5 days, with no significant difference between the sexes. Within their home range, single bats concentrated their nocturnal activities on separate feeding areas ranging from 0.03 to 0.97 ha in size. A cluster analysis revealed that these core-use areas as a whole represented 3 to 14 % of the momentary home ranges. Bats used the remaining parts of their home range for commuting between roosts and feeding areas. Mean travel distances from day roosts to feeding places were 450 to 500 m for both sexes. Upon deserting a day roost bats moved only hundreds of meters to occupy a new roost. After several roost switches, some bats even returned close to formerly occupied tree cavities. Thus, overall home range size of tracked bats did not change substantially after months. Ptenochirus jagori is considered to be relatively site faithful judging from the present results. The spatio-temporal availability and the density of used plant species is likely to be the major reason for locally restricted movements. Bats always visited several foraging areas every night. Abundant fig trees of secondary forest which were fruiting several times a year constituted a large fraction of the species? diet. However, P. jagori also feeds opportunistically on (fig) fruits of primary forest, obviously once they are available. The small roost-related home ranges indicate that the bats do not invest much time and energy in searching for fruiting conspecifics of a given fig species. A core diet consisting of figs caused individuals to often visit secondary forest. The flying foxes showed no strong habitat preferences. The predictability of foraging movements with regard to sequence and direction, the recapture of bats after months at identical sites or nearby, and the short foraging flights within visited areas leads one to suppose some kind of ?trap lining? behaviour (Janzen 1971) minimizing travel distances and energy cost: individual bats may well remember food locations which they regularly patrol. In the context of minimizing travel distances, one individual most probably switched its day roost in response to spatial changes in the availability of food resources. When comparing day roost areas of P. jagori with those randomly selected by an observer, the latter contained more trees per plot. Within the former, the horizontal distribution of foliage was more variable, roost trees had a larger diameter at breast height, and the size of cavity entrances was larger. Individual bats did not specifically select fruiting trees for roosting. The results are discussed in terms of roost accessibility and temperature properties as well as of predator avoidance. Leaves chosen by bats did not differ significantly in their protein content compared to control leaves. Mean protein content of the former is similar to that of fig fruits. Folivory is therefore likely contributing to a balanced diet rather than counteracting deficiencies in protein supplementation. The effect of Ptenochirus jagori on seed germinability ranges from neutral to favourable. With their typical feeding behaviour of both ejecting and defecating large seeds of two non-fig species, bats ?free? seeds from being enclosed in fruit pulp and thus probably doomed to die. With respect to two examined fig species, Labnog (Ficus septica (Moraceae)) and Green Lonok (Ficus sp.), the germination success of seeds from ejecta, faeces and manually defleshed control seeds was similar. Ejected and defecated seeds of Green Lonok, however, germinated at a significantly higher rate compared to seeds of a second control group for which whole fruits were planted. Among treated seeds no pronounced differences were recorded in the temporal distribution of seed germination. Ptenochirus jagori does not feed directly at the mother plant, but carries the fruits to distant ?feeding roosts?. The seed shadow around individual trees of two fig species better fitted an inverse power function than a negative exponential one. The bats deposit most of the seeds at distances of less than 50 m away from fruiting trees. Through their ability to retain seeds in the gut over the day-resting period, bats play potentially the role of long-range dispersers. In the case of Green Lonok, differential ingestion of parasitized and unparasitized seeds may render dispersal more effective: significantly more unparasitized, i.e. viable, seeds occur in faeces compared to ejecta. Experiments dealing with the escape hypothesis were successful in the fig Green Lonok and in Pygeum vulgare (Rosaceae): via dispersal flying foxes increase the germination probability of seeds since significantly more seeds die under the tree crown.