Ecologic Overlap and Interspecific Strife in Breeding Population of Anna and Allen Hummingbirds

Abstract

Analyses of interspecific competition in natural populations are fundamental to modern theory in ecology and evolution. Yet, at least among higher vertebrates, few examples with supporting details are on record, and current theories require corroboration in the field (Errington 1946; Crombie 1947; Richards 1948; Lack 1949). This will come slowly, chiefly because the proper recording and interpretation of competitive incidents in the field require a long and intimate field acquaintance with a given species and because the details of such incidents are often elusive or subtle. Also, the true nature of adaptive differences between closely similar species may escape detection unless these are observed in competition (Lack 1944; Crombie 1947). The object of this paper is to examine ecological differences between two sympatric species of hummingbirds which overlap in both geographic and habitat distribution, providing an example of what, in Schmalhausen's (1949) classification of competitive situations, is termed active intergroup competition. The Anna hummingbird (Calypte anna) and the Allen hummingbird (Selasphorus sasin, formerly S. alleni) are almost restrictedly Californian in breeding distribution, and their breeding ranges overlap along the coast from the San Francisco Bay region to Ventura County (Grinnell and Miller 1944). The evidence to be presented here was obtained in the course of a five-year census, in the period from 1943 to 1947, of the breeding population of hummingbirds in Woolsey Canyon, which drains into Strawberry Creek on the west side of the Berkeley Hills. The study area, now a part of the University of California campus, totals about 125 acres and, until 1948, supported a complex of woodland and chaparral vegetation surrounded by open grassy slopes such as typically fills the canyons in the central Coast Ranges (see Fig. 1). Studies of territoriality in hummingbirds reported earlier (Pitelka 1942) have been continued, and while data presented in this paper concern specifically the populations of Woolsey Canyon, much original information on behavior to be reported elsewhere has been used in interpreting the data on these populations.