Post-dispersal seed predation and seed bank persistence

This study examines whether post-dispersal seed predators could be an important selective force in determining the seed bank strategies of grassland plants. It tests the hypothesis that species with persistent seed banks should sustain proportionally less predation of buried seeds than species which have transient seed banks and that this should be true irrespective of seed size. Results are drawn from a field experiment examining the relative susceptibility of surface versus buried seeds for 19 herbaceous taxa exhibiting different degrees of seed bank persistence. The data were consistent with the hypothesis that seed predators (rodents) influence the seed bank characteristics of seeds. Rodents removed proportionally more large seeds than small seeds and removed a smaller proportion of seeds with persistent rather than transient seed banks, independently of seed size. On average, burial reduced seed removal by almost 50%. The decrease in rates of seed removal following burial was marked for seeds with persistent seed banks but negligible for seeds with transient seed banks. Herbaceous plants with relatively large seeds (seed mass > 1 mg) that form persistent seed banks were either completely avoided or only consumed in small quantities by rodents. In contrast, large-seeded species with transient seed banks suffer high rates of seed predation. Models of life-history evolution predict tradeoffs between seed dormancy and seed mass since dormancy and seed size are correlated traits that both reduce risk in variable environments and thus will show patterns of negative covariation. This paper presents an alternative explanation for this trade-off based on experimental evidence of a negative relationship between seed bank persistence and predation risk.