Date of this Version
Delacour & Mayr's (1945) classic revision of the Anatidae took waterfowl behaviour into account to a much larger degree than had any previous classifications of the group. However, their utilization of behaviour was primarily at the tribal and generic levels, and no real attempt was made to use behaviour for determining intrageneric relationships. Thus far only Lorenz (1941, 1951-1953) has seriously attempteq this with waterfowl, and his analysis of the relationships within the genus Anas (sensu Delacour & Mayr) has provided a remarkable insight into the evolution of this group. I have attempted to expand Lorenz's behavioural delineation of relationships within this genus to include all the living species, and have also attempted to do comparative behavioural analyses for the other genera and tribes in the family Anatidae. The publication of these studies in full detail will be done at a later date (Johnsgard, unpublished) but their taxonomic implications can be summarized here. The purpose of this paper is therefore to provide a more detailed set of behavioural definitions for the anatid groups proposed by Delacour & Mayr, and to suggest ,changes in intrageneric sequences of species, in generic relationships, and in a few cases the species composition of certain tribes of the Anatidae. Except where otherwise indicated, the species and generic nomenclature used is that of Delacour & Mayr (1945) or Delacour (1954-1959).
Acknowledgments. This work was done while the author was studying at the Wildfowl Trust, Slimbridge, on a National Science Foundation postdoctoralfellowship. I would like to express my sincere appreciation to the National Science Foundation for making this study possible and to the Wildfowl Trust personnel for their assistance in every possible way. For stimulating suggestions and unpublished observations I owe D. F. McKinney, P. Scott, and W. Von de Wall my sincere thanks, and the great source of inspiration that has been provided by the writings of K. Z. Lorenz and O. Heinroth cannot go unmentioned.
Observations. Each of the major taxonomic groups will be listed below, and defined so far as possible in behavioural terms. Whenever feasible, intrageneric and intergeneric relationships that are indicated behaviourally will be mentioned.