Interaction networks among species may develop over time in plant communities, making them more productive and stable. Species complementarity increases over time as shown in several biodiversity–ecosystem functioning experiments. This increase has been interpreted as potentially a consequence of complementary resource use or pathogen load among species. However, there may be selection for reduced competition via increased niche differentiation, as indicated by greater trait divergence, or for facilitation, such that plant evolutionary legacy may drive positive biodiversity effects. Plant–soil feedbacks have been proposed to contribute to increasing complementarity effects, but these may be affected by co-evolution; for example selection in monocultures may be for plant individuals with increased defense against pathogens commonly accumulating in monocultures. We ask whether increasing complementarity effects are driven by soil legacy, the co-evolution of soil biota, interacting with plant legacy. In 2010, after eight years of co-selection, we collected plants of twelve grassland species and soil inocula from experimental grassland communities of varying species richness and composition established in 2002 in Jena, Germany. After propagating the collected material, we constructed experimental monocultures and mixture communities with monoculture legacy and mixture legacy in a glasshouse experiment and common garden experiment to test these hypotheses.
Results/Conclusions
The diversity of the local plant species community selected for mixture and monoculture plant types over the eight years of the biodiversity experiment. Complementarity effects increased for plants selected in mixtures and such effects increased in the presence of a co-selected soil community. In contrast, plants with monoculture legacy performed best planted in monocultures, and appeared to have evolved increased defense against pathogens. One of the underlying mechanisms explaining increased performance of mixture communities with mixture legacy may therefore be increased trait differences between species, indicative of selection for increased niche differentiation in mixed-species plant communities. Increased facilitative plant interactions only evolved in mixed-species communities whereas reduced net competition occurred in monoculture and mixture experimental communities with monoculture and mixture plant legacies, respectively. In conclusion, increasing productivity effects in biodiversity experiments may be due to the evolution of increased complementary through increased niche differentiation and increased facilitation. These results offer novel insights into the increase in biodiversity effects over time in experimental grasslands, and suggest that species diversity may influence selection processes within plant communities. Plants within monoculture and mixture communities may be sent on different evolutionary trajectories over time through co-adaptation to selection processes occurring within such communities.