Background/Question/Methods Closely related species are expected to compete strongly with each other due to their similar phenotypes and ecologies. Yet, even the basic prediction that plants will perform worse when grown with close relatives (e.g., congeners) compared to with more distant relatives remains largely untested in a multi-species setting, and related species can be found co-occurring at fine spatial scales in some systems. Given that similarities in phenotype and ecology will likely influence other ecological interactions that can alter the outcomes of competition, the consequences of relatedness on species coexistence need to be considered in a broader ecological context. Using seven perennial, herbaceous Asteraceae (three Solidago (goldenrod), three Symphyotrichum (aster) species, and related Euthamia graminifolia) that frequently co-occur in old-field habitats, I experimentally examined plant performance in field mesocosm communities of equal initial densities but differing relatedness (goldenrods or asters only, vs. a confamilial mixture) and subjected to a variety of ecological contexts. Over two growing seasons, I manipulated several biotic and abiotic conditions, including vertebrate browsing (exclusion by fencing), insect herbivory (insecticide spray versus natural and supplemented insect densities), and soil moisture (water addition versus ambient).

Results/Conclusions Overall, across all species and treatments, plant performance (measured as above-ground biomass) in congeneric mesocosms did not show a consistent reduction compared to confamilial mesocosms, not supporting the prediction of more intense competition in communities composed of closer relatives. Rather, the relative abundances of the species, as well as total community productivity, were contingent on the ecological contexts and varied between the years and genera: by year 2, Symphyotrichum species benefited from growing with more distant relatives, whereas Solidago species did not, or showed the opposite outcome. In addition, while the watering treatment had little influence on community productivity, and insect herbivory caused similar biomass reduction in congeneric and confamilial communities, their main and interactive effects differed greatly among the species. Cumulative damage by phytophagous insects varied by more than ten-fold among the species, and implied associational susceptibility of preferentially attacked Solidago species in confamilial communities. Overall, my results suggest that the ecological context and idiosyncratic responses of component species may, at least initially, over-ride the net effects of relatedness on community structure, and highlight the need to include long-term interactions and ecological processes in our efforts to understand patterns of coexistence among close relatives.