Biodiversity is a central concept in ecology, fundamental to how we conceive of the natural world, measure human impacts on it, and orient conservation efforts to protect it. When species are lost, their links are lost, and functions for which they are responsible can be lost as a result. There is evidence of an intuitive link between diversity and function, where more diverse systems should generally include a greater number of functions with a greater amount of redundancy built into them, but defining this link has proven difficult. This is an important challenge in ecology, however, because the function and stabilizing role of biodiversity is one of the most important arguments for its preservation. This thesis examines the role of biodiversity in ecosystem function and community stability grassland communities across Silwood Park. The experiment uses natural gradients in plant diversity, measured in a number of alternative ways.

Species richness can be an imperfect metric for investigating the link between diversity and ecosystem functioning but can be used as a proxy for more complex diversity measures in grassland communities at Silwood Park. There was no general relationship between plant species richness and plant biomass, and a negative relationship between plant richness and soil mesofaunal biomass. An extreme flooding event negatively impacted some soil mesofauna and not others whilst increasing biomass yield during the following year. Plant biomass stability was independent of species richness and soil mesofaunal biomass varied with either plant richness or plant biomass depending on the definition of stability being used. The inclusion of relative abundance and phylogenetic information allows us more predictive power when looking at the biodiversity-ecosystem function relationship for both primary and secondary production. Experiments such as DExtER are vital to understanding the role of diversity in natural communities.