[997 Asiatic Herpetological Research Vol. 7. pp. 68-79 Taxonomic Review of the Megophryid Frogs (Anura: Pelobatoidea) Amy Lathrop Centre for Biodiversity and Conservation Biology of the Royal Ontario Museum, 100 Queen's Park, Toronto. Ontario, Canada M5S 2C6 and Department of Zoology, University of Toronto Abstract.-The family Pelobatidae has recently undergone notable reorganization. Once composed of two subfamilies, Pelobatinae and Megophryinae. it now consists of only the former. The family Megophryidae has recently been elevated to represent Asian members once in the family Pelobatidae; megophryids are diagnosed by three synapomorphies. A phylogenetic analysis of the Archeaobatrachia by Cannatella (1985) included three species from two genera within Megophryidae. Paraphyly in the genus Megophrys was supported by one character, the articulation of the urosacrum. A re-analysis of Cannatella's (1985) data indicated that alternative patterns of the relationships among megophryids were plausible and that the homology of the urosacral articulation should be examined in more detail. Key words: Amphibia, Anura, Pelobatoidea. Megophryidae, taxonomy Introduction The classification of anurans has been a matter of much recent debate. Several different opinions exist about their relationships (see Ford and Cannatella. 1993 for a review of the recent work on anuran rela-tionships), and no fewer than three different taxono-mies are currently in use (Duellman and Trueb, 1986; Dubois 1985, 1986; Ford and Cannatella, 1993). Pre-vious classifications (prior to Duellman. 1975) have been based on a single complex of characters. As a result, when alternative character schemes were derived, earlier classifications were found to be incon-gruent. The source of the incongruities have been the choice of characters used to construct each classifica-tion (Noble [1922, 1931] vertebral centrum, pectoral girdle, thigh musculature, and dentition; Griffiths [1963] hyolaryngeal structures, skull and limb mor-phology, and reproduction and development, vertebral column; Hecht [1963] and Starrett [1973] tadpole types). Each classification directly reflected the char-acters used in the particular study. For example, the classification proposed by Hecht (1963) in which Orton*s (1957) larval characters were used to con-struct a classification that differed markedly from one developed the same year based on adult morphology (Griffiths, 1963). The history of these characters and their role in the classification of anurans is reviewed in the works of Lynch (1973) and Duellman (1975). Noble (1922), reflected on the problems involved with constructing a classification based on either a single character or a complex of characters. However, the trend continued until Tihen (1965) and Inger (1967) provided classifications based on multiple characters. Subsequently, others have adopted a simi-lar approach to the classification of anurans (Kluge andFarris, 1969; Lynch, 1973; Duellman, 1975; Lau-rent. 1979; Dubois. 1986). More recently. Ford ( 1989, 1993) undertook an analysis of the Neobatrachia, in an attempt to identify the phylogenetic position of Dendrobatidae among all anurans. Similarly, Canna-tella (1985) generated a phylogeny for the Archaeo-batrachia. Both of these thorough phylogenetic analyses culminated in a phylogeny for all major groups of frogs (Ford and Cannatella. 1993). They have applied contemporary techniques to generate a taxonomy reflecting natural groupings (de Queiroz and Gauthier. 1992). i.e., evolutionary history. Anuran classification has progressed since Reig (1958) divided the Anura into four suborders: Amphi-coela, Aglossa, Archaeobatrachia. and Neobatrachia. Based on the analysis of Ford and Cannatella ( 1993), the Amphicoela and Archaeobatrachia are artificial groupings. However. Neobatrachia and Aglossa reflect natural groupings (the latter now called Pipoidea). They confirmed Lynch *s (1973) observa-tion that members of Pelobatidae represent a transi-tional group between the basal anurans and
BioStor
