Biotic filters rather than abiotic filters should influence the community structure of animals such as insectivorous bats with life histories characterized by low fecundity. low predation risk. long life expectancy, and stable populations. I investigated the relative influence of biotic filters on various parameters that define bat community structure, based on predictions from competition and coevolution hypotheses. Each of the competing hypotheses was considered from a number of different perspectives to provide multiple opportunities from which to uncover the mechanism that could result in non-random patterns of community structure. I used a battery of null model and multivariate analyses to test the deterministic or nondeterministic nature of phenotypic, trophic, and species composition patterns of insectivorous bat communities - which I called ensembles. I obtained data for these analyses by sampling the local and regional species richness of insecti vorous bats of southern Africa between 2001 and 2004. Rarefaction and species richness estimators indicated that species inventories of the fynbos, forest, and savanna ensembles, and the Cape Floristic Kingdom regional pool, were complete. As hypothesised, competition and coevolution filters strongly influenced parameters that define bat community structure. My results show that the phenotypic and trophic community structure of insectivorous bats at a local scale exhibit non-random patterns consistent with competition and coevolution hypotheses. Except for an even body size distribution, non-random patterns are not ubiquitous across ensembles. This is linked to multiple causations of competition and coevolution filters operating at a local scale. There was evidence that competition influenced body size distribution across ensembles, and echolocation and dietary patterns in ensembles with high species richness or abundance. At the same time, coevolution filters - mediated by prey defence - strongly influence dietary niche patterns. and to a lesser degree, echolocation patterns. Thus, the non-random phenotypic and trophic patterns of ensembles across southern Africa reflect competition and coevolution filters operating in tandem, or separately, at a local scale. Moreover, abiotic filters such as geographic distribution ranges of small and large-bodied species, extinction risk, and the physics offlight and sound probably also interacted at local and/or regional scales to influence the structure of coexisting bats at a local scale.