SELECTION FOR COMPETITIVE ABILITY: NEGATIVE RESULTS IN DROSOPHILA.

Abstract

The occurrence of competition between populations may result in selection pressures favoring increases in competitive ability. Competition is used in the following sense: competition occurs when two or more species or genotypes experience depressed fitness (growth rate or carrying capacity) attributable to their mutual presence in an area. This is a modification of a definition given by Emlen (1973, p. 307). Competitive ability is used to mean the sum of all those characteristics of an individual relevant to competition. It is likely in most species that the components of competitive ability are polygenic traits, and as such, affected by epistasis, geneenvironment interactions, and correlations with other traits. Consequently, the evolutionary effects of the occurrence of competition are not straightforward. In particular, the question, Does competitive ability change rapidly and predictably in response to selection pressures arising from the occurrence of competition? has no obvious answer. This is a matter of some interest, as failure to improve rapidly could result in exclusion by superior competitors and, if this were the case, preadaptation would be important to the structure of natural communities and the success of invasions by new species. In efforts to examine the evolution of competitive ability, laboratory investigations of the effect of intraspecific competition have been done for some short-lived insect species. The results of these studies were dissimilar. In some populations significant improvements were found within a small number of generations (Seaton and Antonovics, 1967; Bryant and Turner, 1972; Sulzbach and Emlen, 1979), while in others no change was detected (Sokal et al., 1970; Dawson, 1972; Sulzbach and Emlen, 1979). Cases of rapid evolution may be dependent upon giving the larvae or ovipositing females of one of the competing populations a temporal headstart in mixtures (Bryant and Turner, 1972). However, such initial asymmetries have been found to be insufficient in themselves to cause rapid increases in competitive ability, and it was proposed that rapid changes may be exceptional and highly dependent on the particular populations used (Sulzbach and Emlen, 1979). Most studies of the evolution of intraspecific competitive ability have employed samples from single laboratory populations, and the period of selection under constant conditions was short, generally less than ten generations. The present experiments employ hybrid populations and a selection protocol maintained for a longer period.