JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. British Ecological Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of Animal Ecology.
INTRODUCTIONA number of detailed studies of the feeding ecology of tropical hummingbirds have appeared in recent years (e.g. Wolf 1970; Young 1971), and field work is in active progress, especially in Central America. So far, however, there seems to have been no attempt to survey the feeding habits of all the hummingbird species in an area of rich tropical forest.Wagner (1946), discussing the feeding habits of Mexican hummingbirds, concluded that in the more severely seasonal environments (such as the Mexican highlands, where he made most of his observations) hummingbirds rely to a very large extent on insect food, and that it is only in the more equable tropical regions that they depend mainly on nectar. Consonant with this, he found no close correlation between the size and shape of the beaks of the Mexican species which he studied and the flowers at which they fed, and he postulated that close matching of hummingbird beaks to flower form is to be expected only in the humid tropics. In a more recent discussion Grant & Grant (1968) were able to add little so far as tropical species were concerned, beyond citing Skutch's observation of an apparently adaptive relationship in Costa Rica between Passiflora vitifolia* and the hummingbird Phaethornis superciliosus,* the only locally-occurring species with a beak long enough to reach its nectar. It is still mainly a plausible assumption that the various sizes and shapes of hummingbirds' beaks are adapted to taking nectar from flowers of similar size and shape.While we were resident in the Arima Valley in the Northern Range of Trinidad, at latitude 100 40' N, we kept records of the feeding behaviour of all the hummingbirds seen in the valley. Records were kept systematically from February 1959 to September 1961, with the main concentration of effort during the last 14 months of this period. Observations were made intermittently, often between spells of observing other kinds of birds. They were not recorded systematically until we had been living in the valley for eight months (B.K.S.) and two years (D.W.S.). Hence it is felt that the disadvantage of their being somewhat haphazard is compensated by the fact that we were thoroughly familiar with the valley and its vegetation, and so were unlikely to have missed any of the important food plants of the main hummingbird species. The Arima Valley, which extends for several miles from the lowlands to a pass at 1800 ft (550 m) at the watershed of the range, consists of primary forest, patches of secondary forest, small holdings, patches of citrus and other fruits and larg...