Molecular Systematics of Map Turtles (Graptemys): A Comparison of Mitochondrial Restriction Site Versus Sequence Data

?Analyses of mitochondrial DNA (mtDNA) restriction sites and sequences were used to estimate phylogenetic relationships in the emydid turtle genus Graptemys. We compared resolution and concordance among four mtDNA data sets: (1) restriction sites, (2) cytochrome b sequences (380 base positions [bp]), (3) control region sequences (344 bp), and (4) combined (cytochrome b and control region) sequences. Five of the 12 currently recognized species could not be distinguished on the basis of restriction sites. Cytochrome b sequences provided even less phylogenetic information, whereas control region sequences identified all taxa. Bootstrap consensus trees for the restriction site and control region sequence data resolved three mono? phyletic clades: a pulchra group, a pseudogeographica group, and the basal G. geographica. Of the four data sets, the control region offered the greatest resolution with respect to species identi? fication and phylogenetic information. However, a character congruence approach, combining both sequence and restriction site characters, significantly bolstered bootstrap support for the major clades and enhanced phylogenetic resolution within them. Sequence divergence estimates (restriction sites: 0.0-3.2%; cytochrome b: 0.0-1.5%) for Graptemys are among the lowest values reported for a generic-level comparison involving vertebrates. These findings are discussed in light of recent drainage isolation events along the Gulf of Mexico and a proposed slow rate of mtDNA evolution in turtles. [Mitochondrial DNA; restriction sites; DNA sequences; cytochrome b; control region; molecular systematics; Graptemys.] Few molecular approaches in evolution? ary biology have matched the prominence of mitochondrial DNA (mtDNA) analysis. A mainstay gene system in population ge? netics for over a decade, animal mtDNA is gaining increased recognition for its ver? satility in systematics (Moritz et al., 1987; Harrison, 1989; Avise, 1994). The recent surge of systematic applications counters an earlier perception that mtDNA's rapid rate of evolution would limit its role in phylogenetic inference. This initial reser? vation, based largely on comparative re? striction site assays, has abated with the advent of DNA amplification and refined sequencing techniques. Through access to universal primers and knowledge of re? gional heterogeneity in base substitution 3 Present address: Department of Biological Sci? ences, University of Alabama, Tuscaloosa, Alabama 35487, USA. rates throughout the genome (Kocher et al., 1989), researchers have used mtDNA to resolve systematic relationships at sev? eral hierarchical levels. In addition to variable substitution rates among mitochondrial genes, recent studies have also disclosed substantive rate het? erogeneity for homologous DNA sequenc? es across taxa. In vertebrates, for example, sequence data suggest that mtDNA evo? lution is slower in fishes than in mammals (Kocher et al., 1989; Thomas and Beckenbach, 1989; Martin et al., 1992). Other lines of evidence point to significant rate re? ductions for mtDNA evolution in a second major vertebrate group, the turtles (Avise et al., 1992). These findings undermine as? sumptions of a universal clock for animal mtDNA (Wilson et al., 1985) and may pose complications for certain phylogenetic comparisons. Nonetheless, there are con? structive attributes associated with both