A HISTORY OF HOST ASSOCIATIONS AND EVOLUTIONARY DIVERSIFICATION FOR OPHRAELLA (COLEOPTERA: CHRYSOMELIDAE): NEW EVIDENCE FROM MITOCHONDRIAL DNA.

Abstract

A rapidly growing supply of phylogenetic trees has fueled a burst of insights into the ways that phylogenetic data might inform our study of evolution (e.g., Lauder 1990; Wanntorp et al. 1990; Brooks and McLennan 1991; Harvey and Pagel 1991; Swofford and Maddison 1992). One area in which these approaches have proven particularly useful is the study of coevolution (Mitter and Brooks 1983; Brooks 1988; Miller 1991; Mitter et al. 1991; Brooks and McLennan 1993), the cladogenetic and selective influences that ecologically coupled taxa exert on each other through evolutionary time. Those who study the coevolution of herbivorous insects and their host plants have repeatedly noted both the relative host specificity of most herbivore species and the taxonomic conservatism that often characterizes the host affiliations of insect genera and families (Dethier 1954; Ehrlich and Raven 1964; Ward and Spalding 1993). Certain explanations for these patterns can be tested phylogenetically. For example, if herbivore lineages remain strictly associated with their host species over sufficient periods of evolutionary time, vicariance events that isolate populations of the host will also isolate populations of the herbivore. To the extent that these events lead to the formation of new species, this combination of host loyalty and parallel cladogenesis (Benson et al. 1975; Mitter and Brooks 1983; Spencer 1988) can account for both host specificity and host conservatism. If this allopatric cospeciation scenario (Brooks 1979) holds, the phylogenies of host plant taxa and their insect parasites will be at least roughly concordant. Topological contradictions between these trees, however, suggest that herbivore lineages have switched their affinities from one plant group to another subsequent to the diversification of these plants (Jermy 1984; Brooks 1988). To date, most of the handful of pertinent studies (reviewed in Mitter and Farrell 1991) have found evidence for such host shifts, but a few notable exceptions are strongly consistent with the cospeciation model (Farrell and Mitter 1990; B. Farrell in prep.). But if host shifts commonly occur, why do they involve the particular (and often related) plant species that they do, given that selection pressures favoring the use of locally common nonhosts must be omnipresent (Futuyma 1983)? Such conservatism suggests that the spectrum of likely host shifts may be limited by constraints on adaptively relevant genetic variation, perhaps because of the host associations of an her-