Abstract
Species with poor dispersal ability initially should be slower to colonize larger or rounder gaps than smaller or less circular gaps. Conversely, dispersive and seed-banking species should be less sensitive than poor dispersers to gap size and shape, and less confined near a gap's edges. I tested these ideas with a 3-year experiment in which I monitored the revegetation of gaps in grassland vegetation. Initially, species reproducing largely by clonal growth (bulbs and perennial graminoids) were the most sensitive to gap size and (to a lesser extent) shape, reaching their greatest abundances in small and/or rectangular openings. Species relying on seed dispersal (annual grasses) also tended to do best in smaller plots, but were less concentrated near plots' edges. Species relying on seed dormancy (dicots and Juncus bufonius) were least sensitive to plot size, shape, and distance from an edge. In subsequent years, these patterns often were obscured or reversed, reflecting continuing seed immigration and environmental and competitive conditions within gaps. These results indicate that species respond to gap size and shape in ways consistent with their reproductive biologies, and suggest that the importance of gaps' dimensions may vary among plant communities.