Seed predation decreases recruitment in Lupinus sericeus and Lithospermum ruderale

Seed predation is common in most plant communities, but we know surprisingly little about its effects on plant recruitment. While many studies have documented seed loss, relatively few have quantified how this seed loss affects plant recruitment and abundance. It is often assumed that seed predation has minor impacts on plant abundance, either because plant populations are safe-site rather than seed limited, or because of density dependence. To test these assumptions, I crossed rodent exclosure treatments with seed additions (at a range of seed densities) of two perennial grassland forbs: Lupinus sericeus (Fabaceae) and Lithospermum ruderale (Boraginaceae) in grasslands in western Montana. I quantified how post-dispersal seed predation by deer mice (Peromyscus maniculatus) influenced seedling recruitment and subsequent juvenile abundance of these forbs, and examined whether these species were seed-limited. Seedling recruitment was, on average, three times higher in rodent-exclosure than in rodent-exposed plots. In the absence of granivores there was strong seed limitation; increases in seed abundance led to correlated increases in seedling recruitment. In control plots, however, heavy predation reduced the relationship between seed number and seedling recruitment. Two summers after initial seed addition, the abundance of both L. sericeus and L. ruderale was still substantially greater in plots where rodents were excluded compared to control plots, and showed little evidence of density-dependence. These results indicate that post-dispersal seed predation can strongly influence the abundance of these prairie forbs; they also show that traditional estimates of seed limitation may prove misleading if seed predators are not considered.