Recent studies suggest that plant-soil feedbacks (PSFs) provide mechanisms for plant diversity, succession, and invasion.  We test these hypotheses and whether different plant types, ecosystems, and experimental approaches realize different PSFs, using three meta-analytical models: a mixed model, and two different Bayes’ models, one that corrected for sampling dependence only and one that corrected for both sampling and hierarchical dependence (delta-splitting model). 

Results/Conclusions

The predominance of negative effect sizes (70%) supports the hypothesis that PSFs are a mechanism that maintains plant diversity.  Early-successional species appeared to demonstrate the most negative PSFs, but the delta-splitting model indicated that this result was based on results from a few, large studies.  PSFs were associated with non-native success and failure.  Invasive species realized the least negative PSFs, while non-natives that were not weedy or invasive realized the most negative PSFs.  Grasses and grasslands, which dominated the literature, realized the most negative PSFs, while PSFs for trees and forests were not different than zero.  Our findings provide broad support for the role of PSFs in plant community assembly, especially in grasslands, but also underscore the need for expanded testing under field conditions.