Estimating dispersibility of Acer,Fraxinus and Tilia in fragmented landscapes from patterns of seedling establishment

Relative dispersibility of Tilia americana L., Acer saccharum Marsh. and Fraxinus pennsylvanica Marsh, was inferred from the ratio among species-specific regression coefficients () computed from seedling density-distance plots. Density counts were made in spatially-uniform old fields adjacent to single seed sources or monotypic fencerows. Resultant seedling shadow curves approximate the negative exponential form expected for many seeds (log y=a–X). This basic curve shape fit species of differing dispersibility, dispersal under a range of wind directions and one-year-old or all-aged cohorts. The ratios of were 1:2.6:3.2 for Tilia, Acer and Fraxinus, respectively, in order of increasing dispersibility. Vegetation patches isolated from seed sources by several hundred meters or more should have extremely low input of seeds, especially Tilia and Acer.The finding that Fraxinus disperses farther than Acer was unexpected, since the samaras of the former have faster terminal velocities. The relationship can be explained by better performance of Fraxinus samaras in the stronger winds experienced by trees in open landscapes, poorer formation of the samara abscission layer, and release of samaras following leaf abscission and during the winter when winds are the strongest. Both the samara plan and dispersal phenology need to be considered in estimating relative dispersibility among species.