Competitive overgrowth in interactions among sessile colonial invertebrates: A comparison of stochastic and phenotypic variation

Abstract

The outcome of interspecific competition has been shown to be variable in a variety of field and laboratory studies. This variability has been attributed to both stochastic and deterministic processes and, furthermore, has been used as evidence for deviation from at least three hypothetical competitive models. Here, I evaluate two sources of variation in the competitive interaction between two species in terms of their relative impact on the abundance of one competitor (a population response) and on species richness (a community response). Several levels of stochastic variation in the outcome of encounters between the highest and lowest ranked species in a multispecies assemblage provided a null, one-phenotype model for comparison against a two-phenotype model. Two competitive phenotypes of the lowest ranked species differed in terms of their ability to overgrow or be overgrown by the highest ranked species. Even low levels of phenotypic variation resulted in the persistence of the lowest ranked species and enhanced species richness. Comparable low levels of stochastic variation did not. Even competitive equivalency (Kay and Keough, 1981; Russ, 1982) in the stochastic model did not result in enhanced species richness. The question as to whether competitive intransitivity results from stochastic, environmental uncertainty or intrinsic determinism within real sessile assemblages remains unresolved at this time. The results of these simulations suggest that natural selection is more likely to favor competitive phenotypes which contribute to the formation of intransitive networks in assemblages with few rather than many species. Furthermore, the successful discrimination between phenotypic and stochastic variation will be much more difficult to attain in species-rich assemblages.