Background/Question/Methods Considerable debate occurs about the necessity of knowing organism traits, such as size, resource-use efficiency, dispersal capability, etc., for understanding the structure and dynamics of ecological guilds, or communities that share a common resource. For multiple ecological guilds from several taxa, I tested the ability of trait-independent neutral theory and a theory of size-dependent niche-based community assembly to predict three community patterns: body size distributions, species abundance distributions, and species-area curves.

Results/Conclusions Guilds varied considerably in their correspondence to predictions by the different theories, with terrestrial herbivores, detritivores, and predators corresponding most strongly to niche-based theory, and plants, oceanic zooplankton, and aerial feeding birds corresponding best with neutral theory. Comparison of the two theories suggests that decreasing dispersal capability and increasing habitat dimensionality, from two dimensions to three, may both lead to increasing dispersal limitation and a declining correspondence of patterns to trait-dependent niche-based assembly. Results suggest that the importance of trait-independent “statistical mechanics” of neutral assembly and body size-dependent resource partitioning is driven by inherent dispersal capabilities (mobility) and the dimensionality (from two- to three-dimensional) of habitats for different guilds.