Molecular phylogenies can be used to test hypotheses of cospeciation between hosts and parasites by comparing both cladistic relationships and branch lengths. Molecular data can also help discriminate between competing reconstructions of the history of the host-parasite association. Methods for comparing sequence divergence in hosts and parasites are described and applied to data for pocket gophers and their chewing lice. The hypothesis of cospeciation between these two clades is strongly

more »

... ted. The lengths of homologous branches in the gopher and louse phylogenies are positively correlated, but little support is found for the hypothesis that lice are evolving an order of magnitude faster than are their hosts. A central question in the study of cospeciation is the extent to which host and parasite cladogenesis are correlated (Brooks and McLennan, 1991) . The principal test for this correlation is congruence between host and parasite phylogenies, incongruence being taken as evidence of host switching. However, equating congruence with cospeciation and incongruence with host switching is an oversimplification as incongruence may also result from the presence of multiple lineages of parasite on the same host, coupled with differential survival of those parasites (Page, 1993b; Page et al., in press). Previous work using Hafner and Nadler's (1988) allozyme data for eight pocket gophers and their parasitic chewing lice suggested that information on timing of speciation events in the two clades could be used to distinguish between these two causes of incongruence (Page, 1990b) . The availability of mitochondrial DNA sequence data for 15 gophers and their lice (Hafner et al., 1994) permits reinvestigation of this question. These data also allow further exploration of the use of host-parasite systems in the study of comparative rates of molecular evolution. A general framework underlying such comparisons has been developed by Hafner and