Functional composition controls invasion success in a California serpentine grassland

P>1. Recent debates about the role of biotic resistance in controlling invasion success have focused on effects of species richness. However, functional composition could be a stronger control: species already in the community with similar functional traits to those of the invaders should have the greatest competitive effect on invaders. Still, experiments assessing effects of functional similarity have found contradictory results. 2. We used experimental communities in a serpentine grassland in California, USA, to assess the extent to which functional composition and functional diversity influenced success of two different types of invading plants: early season annuals (E) and late-season annuals (L) that have been previously shown to differ in patterns of resource acquisition. 3. We seeded known quantities of seed of six different species (three in each functional group) into experimental plots containing established communities differing in functional composition and functional diversity. The experimental communities contained different combinations of E, L, perennial bunchgrass (P) and nitrogen-fixer (N) functional groups, with functional diversity ranging from 0 to 4 groups. Each invading species was seeded into a separate quadrat within each plot to minimize competitive effects of invaders on each other. We measured both seedling and adult success of invaders for two full growing seasons to further understand mechanisms underlying biotic resistance. 4. More functionally diverse communities were less invaded overall, as measured by the average success of individual invaders. However, assessment of invaders by functional groups was more informative: Es in the extant community suppressed E invaders the most, and Ls in the extant community suppressed L invaders the most. 5. We observed a variety of interactions among extant functional groups in reducing invader success, including synergism, complementarity and 'basement' effects, where two or more groups negatively affected invaders, but combinations of groups were no more suppressive than single groups. The extant community influenced invaders more strongly through suppression of adult plant growth than through effects on seedling establishment. 6. Synthesis. Contrary to predictions from neutral theory, these results indicate that niche overlap was an important component of biotic resistance in these experimental plant communities and summed up to significant effects of species richness.