STATISTICAL TESTS OF HOST-PARASITE COSPECIATION

Evolution. 1997 Apr;51(2):410-419. doi: 10.1111/j.1558-5646.1997.tb02428.x.

Abstract

A history of cospeciation (synchronous speciation) among ecologically associated, but otherwise distantly related, species is often revealed by a strong correspondence of their phylogenies. In this paper, we present several tests of cospeciation that use maximum-likelihood and Bayesian methods of phylogenetic estimation. The hypotheses tested include: (1) topological agreement of phylogenies for coevolving groups; (2) identical speciation times of associated species; and (3) identical evolutionary rates in genes of associated species. These tests are applied to examine a possible instance of host-parasite coevolution among pocket gophers and lice using mitochondrial COI DNA sequences. The observed differences between gopher and louse trees cannot be explained by sampling error and are consistent with a rate of host switching about one-third the host speciation rate. A subset of the gopher-louse data is consistent with a common history of evolution (i.e., the topologies and speciation times are identical). However, the relative rate of nucleotide substitution is two to four times higher in the lice than in the gophers.

Keywords: Birth-death process; conditional probability test; cospeciation; likelihood-ratio test; maximum likelihood; maximum-posterior probability.