Features that determine the composition of avian communities have received extensive and enthusiastic attention, both empirically and theoretically (e.g., Cody 1974; Strong et al. 1984; Wiens, 1989a,b). Interspecific competition for limited resources is one influence widely regarded as critical, but others include species-specific responses to environmental conditions, predation, parasitism, commensal and mutualistic interactions, disturbance and chance, and historical events (Wiens, 1989b). Despite the attention they have received, the relative importance of the different features remains highly controversial.

Certainly the community of birds present in an area will be but a subset of the total population of each constituent species; thus, the total population size of each species is a potential influence on local communities. Since many bird communities in the temperate zone are reconstituted annually with migrants, the geographical location of a site within the breeding range is an important feature; clearly, an avian community outside the normal breeding range of a particular species is unlikely to include that species, irrespective of the suitability of vegetation and other habitat features. Further, suitable habitat encountered earlier in spring migration is more likely to be occupied than is habitat encountered later. And the breeding ranges of birds are known to shift from year to year, depending on climatic conditions, a phenomenon that represents a modification of the general breeding range at a finer scale of resolution. The return of migrant birds to areas from which they fledged or where they previously bred modifies the occupancy of a particular habitat (HildTn 1964); likewise, traditional use of breeding sites by birds can induce species to persist in unsuitable habitat (Wiens 1985). Thus site fidelity can weaken the relationship between habitat quality and number of birds on an area (Wiens 1985). Waterfowl have received less attention from community ecologists than have many other groups of birds despite the importance attributed to waterfowl by biologists more generally, as well as by the public. Exceptions include work by Nudds and colleagues in Canada and by P?ysS and colleagues in Finland (see Nudds 1992 and references contained therein). Further, waterfowl are important ecologically; in much of the prairie of the North American midcontinent, waterfowl are numerically among the most common bird species and are certainly dominant in terms of biomass. This chapter addresses some influences on waterfowl communities in mixed-grass prairie pothole habitat. It takes a temporal perspective, based on annual censuses of breeding ducks for 25 years on a specific study area at Woodworth, North Dakota. Any changes in the structural features of the habitat that may have occurred during this period at the census site were at most gradual. I examine how the waterfowl communities varied in response to influences that did change annually, such as climate, conditions of the wetlands on the study area, the regional populations of birds from which the communities were constituted, and the population at Woodworth during the previous year. The results of the analyses are interpreted relative to individual characteristics of 11 waterfowl species: Mallard (Anas platyrhynchos), Gadwall (Anas strepera), Green-winged Teal (Anas crecca), Blue-winged Teal (Anas discors), Northern Pintail (Anas acuta), Northern Shoveler (Anas clypeata), American Wigeon (Anas americana), Canvasback (Aythya valisineria), Redhead (Aythya americana), Lesser Scaup (Aythya affinis), and Ruddy Duck (Oxyura jamaicensis).