Non-native plants may have persistent post-invasion impacts on the communities they formerly dominated. As an example, native plants may not always return to previously infested sites, leaving an opening for secondary invasions of other exotics. Areas infested with leafy spurge (Euphorbia esula) in the northern Great Plains have been controlled with the release of flea beetles (Aphthona spp.), which has been widely successful, but native plants do not always return to these sites following the decline of leafy spurge. Our objective is to determine if the native soil seed bank has been reduced in areas recently infested with leafy spurge, and if there is a correlation between the seed bank and the aboveground plant community. We conducted seed bank assays on permanent study plots in mixed-grass prairie in Theodore Roosevelt National Park, ND (USA) that (1) had never been infested with leafy spurge and were predominately native, (2) were previously infested but returned to native dominance, and (3) were previously infested but subsequently dominated by other non-natives. We used the seedling germination method for the assay, quantifying seed densities on an area basis, and conducted plant cover surveys on the same study plots where soil seed bank samples were collected.
Results/Conclusions

Overall, there were few differences in seed density of native plants between areas that were never invaded and previously invaded areas. There were fewer native perennial cool-season grass seeds in areas now dominated by other invasives, which may be explained by higher densities of non-native perennial cool-season grasses on previously invaded sites. There was less cover of native forbs and native perennial warm-season grasses in areas currently dominated by non-natives than for the other two habitat types. The non-native seed bank composition was therefore largely reflective of the current plant cover, but the same cannot be said for native plants, especially for currently invaded areas. The similarity of the germinable seed banks of native plants between uninvaded and previously invaded areas, especially those areas currently dominated by other invasives, indicates that the native seed bank has not been reduced following invasion of leafy spurge. These combined results suggest that changes in soil processes and interactions among leafy spurge, herbivores, and soil biota, rather than altered seed banks, are responsible for recruitment limitation of native plants.