The effect of environmental heterogeneity on species richness is frequently discussed in ecology. However, the empirical evidence has been contradictory as to the direction of the effect. Although some authors have considered that this might be a methodological problem, we argue that for ants, ecological interactions within the community, as interspecific competition is more important. We analyzed the plausibility of models in explaining the ant richness distribution patterns in a semi-desert

more »

... vironment. We used three predicting variables in the construction of the models to explain ant richness distribution: heterogeneity based on the amount of structures regardless of their type, heterogeneity based on the diversity of structures, and the abundance of individuals of the dominant species. We used ANOVA to chose the best model and corroborated the prediction that in this system abundance of dominant species is the best predictor of ant species richness. Neither of the heterogeneity conceptions contributed much to explain richness distribution. However, in a second analysis, we concluded that heterogeneity could affect the abundance of the dominant species. We conclude that competitive dominance is a better predictor of species richness distribution patterns than structural heterogeneity. However, the structural heterogeneity affects the distribution of dominant individuals. We suggest that some unexplained patterns observed about the relationship between heterogeneity and richness could be due to an indirect effect.