Coexistence mechanisms of mycophagous drosophilids on multispecies fungal hosts: aggregation and resource partitioning

In nature various mechanisms that provide some refuges for inferior species to escape from competition act together to promote the species coexistence in a local community. We propose a method to evaluate simultaneously the relative importance of different mechanisms for the species coexistence, focusing on a mycophagous drosophilid community depending on a multispecies fungal host system, where two mechanisms – aggregation over patches of single fungal species and resource partitioning at the fungal species level – can act together. The data are based on the numbers of adult insects emerging from naturally occurring fungi collected from a temperate forest in northern Japan. 2. We confirm the finding of Sevenster & van Alphen (1996) that large patches are low‐density refuges despite the large absolute numbers of competitors on them, and therefore adopt new measures for intra‐ and interspecific aggregations and the species‐persistence criterion, which were developed by Sevenster (1996) to take account of variable patch sizes in calculating the aggregation measures. 3. The calculated measures are compared among three analyses for fungal resources: AR‐analysis (the original data‐set from multispecies fungal patches) in which both aggregation and resource partitioning can act together; R‐analysis (combining the data for all patches of each fungal species) in which only resource partitioning at the fungal species level can act; and A‐analysis (calculating the measures independently for each fungal species) in which only aggregation process can act as a coexistence mechanism. 4. Intraspecific aggregation over multispecies fungal patches is strengthened by both aggregation and resource partitioning processes. 5. We reconfirm the finding of Sevenster & van Alphen (1996) that interspecific associations are consistent over years, and support their claim that such consistent associations should be taken as real aspects of the community structure. 6. Measures for the persistence criterion ( T ), the ‘relative effect of competitor aggregation’, are less than unity, satisfying the necessary and sufficient condition for the species persistence, in most drosophilid species concerned; these tend to be smaller in AR‐analysis than in R‐ or A‐analysis, suggesting facilitation of the species coexistence by both aggregation and resource partitioning in the mycophagous drosophilid community studied. 7. Differences in T ‐value between AR‐ and R‐analyses and between AR‐ and A‐analyses may indicate relative contributions of aggregation and resource partitioning, respectively, to the species coexistence. We propose a method to compare various communities in a multi‐dimensional space with regard to the relative importance of different coexistence mechanisms.