This study aims to investigate the link between genetic functioning and diversity, and to establish a relationship, if any, between that diversity and the stability of the community. Extreme transient perturbations were performed, using repeat chloroform fumigations, to reduce the community to its lowest genetic level (0.7% of the DNA of the control population). Severe reduction in community diversity and selection for bacilli was initially observed, the community subsequently recovered (6 months) to its previous diversity but was altered in community composition, bacilli were no longer dominant. Perturbations using benzene resulted in a reduced genetic diversity (16S rDNA and DGGE) which did not recover (after 4 weeks). However, full recovery was observed for culturable cell numbers, DNA concentration and community level physiological profile (Biolog)., suggesting high levels of genetic redundancy within soil. Investigation of the relationship between diversity and stability was performed by perturbing soils with naturally differing levels of diversity (Sourhope>Boyndie) to equivalent toxicity, using copper and benzene. Large decreases in biomass were observed for both soils after benzene perturbation. Assuming successful equivalent toxicity, the genetic diversity of the naturally more diverse soil was more resistant to benzene perturbation than the less diverse soil. The broad scale function of both benzene-treated soil communities was unaffected (14C-labelled wheat shoot mineralisation). Narrow niche function was impaired for both soils (14C-labelled 2,4-DCP mineralisation). The Sourhope soil recovered this function by the end of the experiment, but Boyndie did not. The genetic diversity of both soils was resistant to copper perturbation despite biomass decreases. Initial shifts in physiological profiles of both soils was observed but rapidly returned to that of the control. In conclusion, diversity-stability relationships may exist in soil communities, despite generally being an extremely robust system.