Post-dispersal seed predation can constrain native and non-native plant recruitment and consequently influence community composition. In Eastern Washington, naturalized and invasive plants dominate the meadow steppe, yet are much less common in the understory of adjacent Ponderosa Pine forests. We asked whether post-dispersal seed predators play a role in this discrepancy in the abundance of non-native plants and consequently act as agents of biotic resistance. We predicted that 1) seed predation would limit establishment of invasive plants in Ponderosa Pine forests but have little effect on invasive plants in the adjacent steppe, and 2) invasive plant establishment would be inversely proportional to seed predator abundance. Small mammal censuses were conducted during summer 2010, using capture-mark-recapture in four each of steppe and forest sites. To evaluate seedling establishment, we constructed seed addition plots within hardware cloth exclosures that either permitted (open cages) or excluded (closed cages) small mammal seed predators. Vegetation was removed inside all cages to minimize competition on plant establishment. Plots were sown with native bluebunch wheatgrass (Pseudoroegneria spicatum), naturalized cereal rye (Secale cereale), or invasive cheatgrass (Bromus tectorum) seeds (100 seeds/plot). Seedlings counts were made the following summer (2011) in all plots.
Results/Conclusions
Seedling establishment in exclosures (closed cages) did not differ between habitats (F1,82 = 0.26, p = 0.609), indicating that the establishment of test species is not limited by differences in abiotic conditions between habitats. In steppe and forest, native and naturalized species in exclosures experienced similar, significant release from predation (open vs. closed cages, all p-values < 0.01). Establishment of cheatgrass in the steppe was significantly lower in plots accessible by predators, compared to establishment in exclosures (open vs. closed cages, p = 0.002), and followed a similar, yet non-significant trend in the forest (open vs. closed cages, p = 0.123). Lower establishment of cheatgrass in the steppe correlated with high abundance of seed predators (R2 = 0.19), but high seed predator abundance was positively correlated with higher invasive plant establishment in the forest (R2 = 0.10). Our results indicate that post-dispersal seed predation provides significant biotic resistance by decreasing non-native plant establishment in both the forest and steppe. Seed predation fails, however, to solely explain the difference in invasive plant establishment observed between these habitats. Comprehensive evaluation of biotic resistance to immigrant plant populations may require integrating assessments of seed predation with other biotic factors (e.g. competition, parasitism).