DO SMALL-SEEDED SPECIES HAVE HIGHER SURVIVAL THROUGH SEED PREDATION THAN LARGE-SEEDED SPECIES?

Abstract

Seed ecologists have often stated that they expect larger-seeded species to have lower survivorship through postdispersal seed predation than smaller-seeded species. Similar predictions can be made for the relationship between survivorship through predis- persal seed predation and seed mass. In order to test these predictions, we gathered data regarding survivorship through 24 hours of exposure to postdispersal seed predators for 81 Australian species, and survivorship through predispersal seed predation for 170 Aus- tralian species. These species came from an arid environment, a subalpine environment, and a temperate coastal environment. We also gathered data from the published literature (global) on survivorship through postdispersal seed predation for 280 species and survi- vorship through predispersal seed predation for 174 species. We found a weak positive correlation between seed mass and the percentage of seeds remaining after 24 hours of exposure to postdispersal seed predators at two of three field sites in Australia, and no significant relationship across 280 species from the global lit- erature, or at the remaining field site. There was no significant relationship between seed mass and survivorship through predispersal seed predation either cross-species or across phylogenetic divergences in any of the vegetation types, or in the compilation of data from the literature. Postdispersal seed removal was responsible for a greater percentage of seed loss in our field studies than was predispersal seed predation. On average, 83% of diaspores remained after 24 hours of exposure to postdispersal seed removers, whereas 87% of seeds survived all predispersal seed predation that occurred between seed formation and seed maturity. Mean seed survival was higher in the field studies than in the literature compi- lations, and species showing 100% survival were heavily underrepresented in the literature. These differences may be due to biases in species selection or publication bias. Seed defensive tissue mass increased isometrically with seed mass, but there was no significant relationship between the amount of defensive tissue per gram of seed reserve mass and survivorship through postdispersal seed predation.