Patterns and mechanisms of conspecific and heterospecific interactions in a dry perennial grassland

P>1. Models of local stable coexistence require negative feedbacks, i.e. intraspecific interactions must be more negative than interspecific interactions. However, most competition experiments, often done in the glasshouse, have found evidence for competitive hierarchies. Measurement of interactions under realistic field conditions is necessary to assess their contribution to community dynamics, and explicit measurement of intermediaries thought to be important in interactions may allow studies to account for any variation in experimental results. 2. In this study, we compare conspecific and heterospecific interactions in a field experiment in a dry sand prairie in Michigan. We study the four dominant species at two different stages, germination and adult growth. Using seed addition and adult transplant experiments, we ask whether plants perform best in natural field monocultures of conspecifics, heterospecifics or no neighbour plots. We also measure abiotic environmental characteristics associated with each neighbourhood type and test whether performance can be explained by environmental effects. We hypothesize that plants will create competitive hierarchies because our experimental design is similar to classic competition experiments. 3. Neighbour species created consistent hierarchies in their effects on germination of all four target species, which is likely due to light limitation. However, interestingly, adult plant biomass for two of the three species (one species did not survive) was lower in conspecific monocultures compared to heterospecific or no neighbour plots, thus producing negative feedbacks. For two species, the effects of neighbours on adult growth are likely due to reduction of light and soil nitrate; for the third, however, resources could not explain the pattern that conspecific interactions were more negative than heterospecific. 4. Synthesis. These results suggest that patterns in the relative strength of conspecific and heterospecific competition depend on life-history stage. Moreover, resource uptake could explain some, but not all, of the interactions among species, suggesting that other factors such as microbial communities or other forms of niche partitioning may play a role and that field experiments are necessary to gauge their relative importance.