Papers in the Biological Sciences


Date of this Version



Published in Proceedings XIII International Ornithological Congress, Ithaca 17-24 June 1962, ed. Charles G. Sibley (Baton Rouge, LA: American Ornithologists' Union, 1963), pp. 531-543.


The ducks, geese, and swans present a host of species-recognition and isolating-mechanism problems that are equaled by few other groups of birds. As a family, the Anatidae have provided the greatest number of interspecific hybridization records of any avian family (see Gray, 1958, and Johnsgard, 1960a) both in captivity and under natural conditions. A surprising number of these hybrids have proved to be fertile, even when obtained between what appear to be well-marked genera. This remarkable capacity for hybridization indicates that isolating mechanisms must be operating effectively if species are to retain their integrity under natural conditions. Since genetic isolation is practically absent in the Anatidae, other isolating mechanisms must, of course, have evolved to take their place. Of these, the most significant appear to be behavioral differences and various morphological (plumage and soft-part) specializations that are usuaiIy associated with these behavioral differences. A comparative behavioral study of the Anatidae was undertaken from 1959 to 1961 at The Wildfowl Trust, in England, where I was able to observe in life 125 out of the approximately 140 extant species of Anatidae, including 34 of the 41 genera accepted by the most recent authority on the family (Delacour, 1954-59).

Of the 10 tribes of Anatidae accepted by Delacour (1954-59), all but 1 (Anseranatini), which is monotypic, are characterized by the inclusion of numerous closely related, often sympatric, species. It is, of course, in these groups that behavioral and morphological elaborations associated with pair formation and copulation are most conspicuous, and it is also here where the greatest amount of information regarding the development of isolating mechanisms can be gleaned. The following discussion is subdivided into sections corresponding to the tribes and included genera of Delacour (1954-59), except where results of my studies have suggested certain modifications in his classification (Johnsgard, 1961a). As most of the observations given below are my own, I have not resorted to frequent citation to support them. Where terms refer to ritualized displays, the first letter of these terms has been set in capital letters.

Included in

Ornithology Commons