Functional size traits in crowded vegetation: when bigger isn’t better

Abstract

Why do some species within crowded vegetation have greater numerical abundance than others? Functional traits are often explored as predictors of success under certain habitat conditions, and in the context of competition, traditional plant competition theory is based on a ‘size advantage’ (SA) hypothesis, where larger plants are generally considered better competitors for contested resources. However, body size distributions are right-skewed at virtually all scales — from plants within populations to species within communities, within phylogenetic lineages, and across entire regional floras. And field studies in crowded natural vegetation—both woody and herbaceous—have generally failed to find evidence for a species size advantage in predicting numerical abundance. This paradox has recently been interpreted in terms of a ‘reproductive economy’ advantage (REA) whereby smaller species, although severely suppressed in size by neighbourhood crowding, are nevertheless generally more likely (than larger species) to produce at least some offspring. Traits that might be associated with greater reproductive economy, and of interest here, include smaller values for three size metrics: minimum reproductive threshold size (MIN), seed size, and leaf size. Here, we used a combination of natural field surveys, greenhouse studies and multi-season field experiments to further investigate the relative importance of SA and REA in old-field meadow vegetation in eastern Ontario. Overall, we found strong support for a reproductive economy advantage, particularly for smaller MIN – and especially when crowding from neighbouring vegetation was the most intense and when species had a long, relatively undisturbed history of growing together. We found weak and inconsistent support for the size advantage hypothesis across all studies and in many cases, we could not detect a signal for either the SA or the REA hypotheses. This was especially true when species were introduced that had no history of growing together. We also explored how dry mass predicts other metrics of plant body size that better reflect ‘space occupancy’ (SO) in situ (i.e. canopy height, area, and volume) — and found that plants/species with smaller SO body size often had greater dry mass per unit SO body size. These allometric relationships have important implications for fecundity allocation and the REA hypothesis. The combined results of these studies suggest that both SA and REA are likely important, but dependent on the local severity of crowding between near neighbours — with REA generally being more important as crowding increases, and SA (contrary to traditional theory) more important as crowding decreases. At the whole community level then, across highly variable levels of local crowding, evidence for REA in predicting abundance is likely to be obscured by SA effects and vice versa.