Background/Question/Methods The link between environmental heterogeneity and opportunities for species coexistence is well established theoretically, but has received mixed experimental support. Understanding the causes of the discrepancy between theoretical and empirical results is an important goal. For example, we found that in a long-term experiment conducted in a low productivity grassland in SW Michigan, heterogeneous fertilization had no effect on species richness relative to uniformly fertilized plots. Instead, both tyes fertilized plots tended to become dominated by clonal species, which are present, but not dominant in unfertilized controls. We hypothesize that this pattern may come about if clonal plants are superior resource competitors, but are constrained to forage for resources at larger spatial scales than are non-clonal plants.

Results/Conclusions We present a model of competition for a single, spatially heterogeneous resource that shows the conditions for this behavior. In unfertilized plots, clonal plants are forced to average over patches of very low resource supply, potentially preventing their establishment, while in fertilized plots, the average supply rate at the scale experienced by the clonal dominants is high enough to allow invasion and potential exclusion of non-clonal species. In addition, the model makes a number of specific qualitative predictions that can be compared with experimental data. This works suggests that considering the differences in the spatial scales at which species interact with environmental heterogeneity may help explain the discrepancy between theoretical and empirical results.