Phylogenetic relationship among cockroach families inferred from mitochondrial12S rRNA gene sequence

Abstract

A number of phylogenies exist for cockroaches that differ in the postulated relationships among families and genera. The relationship of the wood‐feeding genus, Cryptocercus, to other cockroach families and to termites, has generated considerable debate. Grandcolas (1994), based on morphological analysis, synonymized the family Cryptocercidae with Polyphagidae and placed the genus Cryptocercus in the subfamily Polyphaginae. To determine if an independent set of characters supports the placement of Cryptocercus in Polyphaginae, a phylogenetic analysis of relationships among representative genera of the five cockroach families was undertaken. DNA sequence of a ‐430 base pair portion of the mitochondrial small ribosomal subunit gene from representatives of Blattidae, Blattellidae, Blaberidae and Cryptocercus, previously published by Kambhampati (1995) and Kambhampati et al. (1996), and the homologous sequence from representatives of Polyphagidae were used in the analysis. A total of twenty cockroach taxa and three termite genera were included in the study. Because a recent study showed that Cryptocercus punctulatus consists of a species complex, DNA sequence from four individuals collected in different parts of the U.S.A. was included in the study. The trees estimated from parsimony and neighbour‐joining analyses indicated that Cryptocercus is a monophyletic clade which is most closely related to members of Blattidae. Polyphagidae is indicated as a sister group to the Blattidae + Cryptocercus complex, suggesting that Polyphagidae may belong to the superfamily Blattoidea rather than to Blaberoidea as proposed by McKittrick (1964). Blaberidae and Blattellidae were sister groups as previously proposed. Based on the present analysis, I propose that the genus Cryptocercus be retained in the family Cryptocercidae. Cockroaches