Rethinking "mutualism" in diverse host-symbiont communities.

While examples of bacteria benefiting eukaryotes are increasingly documented, studies examining effects of eukaryote hosts on microbial fitness are rare. Beneficial bacteria are often called "mutualistic" even if mutual reciprocity of benefits has not been demonstrated and despite the plausibility of other explanations for these microbes' beneficial effects on host fitness. Furthermore, beneficial bacteria often occur in diverse communities, making mutualism both empirically and conceptually difficult to demonstrate. We suggest reserving the terms "mutualism" and "parasitism" for pairwise interactions where the relationship is largely independent of other species and can be verified by measuring the fitness effect experienced by both partners. In hosts with diverse microbial communities, we propose re-formulating some of the essential questions of symbiosis research - e.g. concerning specificity, transmission mode, and common evolutionary fates - as questions of community ecology and ecosystem function, allowing important biological interactions to be investigated without making assumptions about reciprocity. Understanding the fitness of host-associated bacteria is a crucial component of investigations into the role of microbes in eukaryote evolution.