Summary

On oceanic islands, some large diurnal megachiropteran bat species (flying foxes; Pteropus spp.) frequently use thermal or slope soaring during foraging flights to save energy. We compared the flight morphology and gliding/soaring performance of soaring versus non-soaring Pteropus species, one pair on American Samoa and one pair on the Comoro Islands, and two other soaring/flap-gliding species and one non-soaring species. We predicted that the soaring species should have a lower body mass, longer wings and, hence, lower wing loadings than those species that use mainly flapping flight. This would give a lower sinking speed during gliding, a higher glide ratio, and enable the bats to make tighter turns with lower sinking speeds than in the non-soaring species. We theoretically calculated the gliding and circling performances of both the soaring and non-soaring species. Our results show that there are tendencies towards longer wings and lower wing loadings in relation to body size in the gliding/soaring flying foxes than in the non-soaring ones. In the species-pair comparison of the soaring and non-soaring species on American Samoa and the Comoro Islands, the soarers on both islands turn out to have lower wing loadings than their non-soaring partners in spite of opposite size differences among the pairs. These characteristics are in accordance with our hypothesis on morphological adaptations. Most differences are, however, only significant at a level of P<0.1, which may be due to the small sample size, but overlap also occurs. Therefore, we must conclude that wing morphology does not seem to be a limiting factor preventing the non-soarers from soaring. Instead, diurnality in the soaring species seems to be the ultimate determinant of soaring behaviour. The morphological differences cause visible differences in soaring and gliding performance. The glider/soarers turn out to have lower minimum sinking speeds, lower best glide speeds and smaller turning radii than the non-soarers. When the wing measurements and soaring performance are normalized to a body mass of 0.5 kg for all species, the minimum sinking speed becomes significantly lower (P<0.05) in the three soaring and the one flap-gliding species (0.63 m s(−)(1)) than in the three non-soaring species (0.69 m s(−)(1)). Interestingly, the zones in the diagrams for the glide polars and circling envelopes of these similar-sized bats become displaced for the glider/soarers versus the non-soarers. The glide polars overlap slightly only at the gliding speeds appropriate for these bats, whereas the circling envelopes do not overlap at the appropriate bank angles and turning radii. This points towards adaptations for better gliding/soaring performance in the soaring and gliding species.