DISPERSAL AND DESTRUCTION IN A MULTIPLE HABITAT SYSTEM - AN EXPERIMENTAL APPROACH USING PROTIST COMMUNITIES

Many habitats occur as discrete fragments, and the populations in such communities persist at a regional level, despite periodic local extin ction. One component of local extinction is disturbance, or in the ext reme case destruction, of individual patches. The impact of such distu rbances on the ability of the entire system (the metacommunity) to ret ain species is clearly a significant question in many real systems. Th e problem has been the subject of a number of recent theoretical inves tigations, but this study addresses the question using an experimental 'model' system of protist communities, where patch destruction rates and between-patch dispersal rates can be manipulated. Replicate blocks of eight microcosms (patches) were inoculated with 13 species of prot ists and then subjected to 3 levels of destruction and 2 levels of bet ween-patch dispersal in a factorial design. Communities were sampled a fter 6 and 16 weeks. The results support the simple prediction that me tacommunities subject to higher rates of patch destruction suffer grea ter species losses, and that increasing dispersal rates between patche s moderates, but does not entirely counteract, that effect. Differing dispersal rates alone had little effect on either community or metacom munity diversity. There was no evidence that disturbance enhanced dive rsity, either within or across patches, but there was some suggestion that there were consistent differences in community structure in respo nse to combinations of destruction/dispersal treatments. The possible mechanisms underlying the results, including the effect of low local a bundance on colonization rate, are discussed.