Abstract
In a recent report, Boles et al. (2004) compared the effect of an environmental stress on high-diversity (wild-type) and low-diversity (recA mutant) biofilm communities of the bacterium Pseudomonas aeruginosa. They found increased resistance to an environmental perturbation in the diversified community, which they interpreted as support for the insurance hypothesis' an ecological model that predicts that a more diverse community will be better able to resist an external stress. The report is important in that it represents one of very few attempts to examine biofilm communities in the light of ecological theory. Nevertheless we question the authors' conclusion for two reasons, one general and one specific. We also question the authors' interpretation that diversification in a biofilm is somehow a programmed response to that lifestyle.
In the ecological literature insurance effects are defined as ... any long-term effects of biodiversity that contribute to maintain or enhance ecosystem function in the face of environmental fluctuation (Yachi & Loreau, 1999). The mechanism underlying these effects is functional redundancy between community members: a more diverse community is more likely to have members that respond differently to an environmental stress. If there is some amount of redundancy between these members then overall ecosystem functioning can be maintained even if some members fail. This intuitive idea has been around for many years, but was first formally stated in 1999 by Yachi & Loreau. They modelled communities of differing levels of diversity and examined how this diversity affected community productivity in the face of varying degrees of fluctuating environmental stress. They found that increasing community diversity could increase average community function; however, this outcome depended on their being an asynchronous response of the members of the community such that members responded differently to different environmental stresses. For example, member A is more successful than member B in environment 1, but member B is more successful than member A in environment 2 (i.e. a trade-off in response between environments). A key finding of their study was that this trade-off is a necessary aspect of the insurance hypothesis. If member A was more successful in both environments 1 and 2, there would be no difference in community productivity between a monoculture of A and a mixture of A and B.