A meta-analysis of adaptive deme formation in phytophagous insect populations.

The adaptive deme formation (ADF) hypothesis predicts that herbivorous insects become locally adapted to their host plants over time. Since its inception, approximately 17 independent studies have tested ADF, and they are divided in support and rejection of the hypothesis. This field of insect evolutionary ecology has a contentious history, and the contradictory studies obscure our understanding of the general evolutionary importance of adaptive deme formation in phytophagous insects. We conducted a meta-analysis in an attempt to clarify this issue. Meta-analysis is a statistical method for quantitatively comparing and synthesizing the results of different studies in a way that is more objective than a traditional literature review. Our analysis indicates that local adaptation is an important phenomenon in diverse insect systems. Contrary to predictions of the original hypothesis, there was no evidence that insect dispersal ability, and ostensibly gene flow, was associated with local adaptive differentiation. There was some indication that breeding (parthenogenetic, haplodiploid, diplodiploid) and feeding (exophagous, endophagous) modes may influence the evolution of locally adapted demes. Our analysis supports the theory of adaptive deme formation and provides guidance for future research directions.