LAUSR

We tested whether the recently proposed two-part measure of degree of invasion (DI) of a community relating exotic proportion of cover to exotic proportion of richness can characterize patterns of plant invasions at multiple savannah sites in Southern Africa. Regression analysis was performed on transformed data to assess how this two-part measure of DI compares to other metrics of community invasibility. The results indicate that at the plot level, the absolute cover of exotics was not significantly related to native cover for three sites out of four assessed (R2 ≤ 0.17; P > 0.05). Also, at all four sites, no significant relationships were detected between native and exotic plant richness at both the 1-m2 and 400-m2 plot scales. By contrast, significant (P < 0.05) positive linear relationships were observed between exotic proportion of richness and exotic proportion of cover at all sites (R2 was as high as 0.67 and 0.97 for two sites). Our results indicate that the new two-part measure of DI is able to characterize patterns of plant invasions across plant communities in African savannahs.