EVIDENCE FOR ECOLOGICAL CHARACTER DISPLACEMENT IN WESTERN AMERICAN CATOSTOMID FISHES

Much of recent ecological theory rests on the assumption that the primary factor responsible for the organization of natural communities, particularly of animals, is interspecific competition (e.g., MacArthur, 1972; Schoener, 1975). The validity of this assumption depends almost exclusively on inference drawn from observational evidence of two types. Morphological changes in the form of ecological character displacement (Brown and Wilson, 1959; Grant, 1972) and niche shifts correlated with geographic variation in the number or density of closely related species are generally considered the strongest available evidence for the importance of interspecific competition (MacArthur, 1972; Cody, 1974; Schoener, 1974, 1975). Grant (1972, 1975) defines ecological character displacement to be "the process by which a morphological character state of a species changes under natural selection arising from the presence, in the same environment, of one or more species similar to it ecologically . . . ." By this definition Grant would include both character convergence and divergence but would exclude cases of changes in character state due to the related process of character release. It seems clear that the interspecific interaction implicit in Grant's (1972, 1975) definition of ecological character displacement is competition. Furthermore, this causal relationship must be demonstrated if ecological character displacement is to be used as observational evidence for the occurrence of interspecific competition. Therefore, we define competitive character displacement to be a change of character state in a population that is due to natural selection arising as a result of competition with individuals of another population in the same environment. The characters may be morphological, ecological, behavioral, or physiological and the change may be either divergent or convergent. By divergent competitive character displacement we mean that a character state of a population in ecological sympatry (=syntopy, Rivas, 1964) exhibits greater difference than is the case when allopatric populations are contrasted, and that this effect can be attributed to sympatry. If the contrast of allopatric and sympatric populations reveals that the character states of the two populations are more similar in areas of sympatry than in areas of allopatry and if this effect can be attributed to competition, then the situation is one of convergent character displacement. Ecological theory predicts those situations in which convergent or divergent displacement would be expected (MacArthur and Levins, 1964, 1967; MacArthur and Wilson, 1967; Schoener, 1969a, 1969b, 1970; MacArthur, 1972). In all of these models the selection responsible for the character state change or niche shift arises as a result of interspecific competition. Grant (1972) in his critical review of the evidence for ecological character displacement, concluded that "the evidence for the ecological aspect of morphological character displacement is weak" and that "the detection of character displacement in its ecological aspect, particularly in mainland regions, will require a level of detail that has so far not been produced in a single study. " The weakness of the available evidence for competitive character displacement lies in the difficulty of establishing unambiguously that the character state of