A Molecular Test of Bat Relationships: Monophyly or Diphyly?

Abstract

?Two conflicting hypotheses concern the origin of flying mammals. The traditional hypothesis states that the two major groups of bats, the microchiropterans and the megachiropterans, are sister groups that constitute the taxon Chiroptera. In contrast, the diphyly hypothesis suggests that megachiropterans are more closely related to primates than to microchiropterans. Different suites of morphological characters provide support for each of these hypotheses, and previous molecular studies have not provided a clear resolution of the problem. We analyzed a region of the mitochondrial 12S ribosomal RNA gene from ll species of mammals, including 2 species of megachiropterans, 2 species of microchiropterans, a primate, a colugo (Dermoptera), a tree shrew (Scandentia), and 4 outgroups, to test the diphyly hypothesis. A phylogenetic analysis of 257 base pairs resulted in two shortest unrooted trees that significantly support the monophyly of the bats and also suggest that the colugo is more closely related to primates than to the bats: (((Primates, Dermoptera) Scandentia) (Microchiroptera, Megachiroptera)). The topology sup? porting the diphyly hypothesis is 10 steps longer than the most-parsimonious unrooted tree. Although the traditional hypothesis is supported with respect to bat monophyly, the rDNA data support the nontraditional grouping of colugo and primates (a hypothesis also supported by neurological data). [Bat monophyly, phylogenetic confidence, Chiroptera, Archonta.] All mammals capable of true flight tra? ditionally have been recognized as a monophyletic group, Chiroptera. The two major lineages of bats have been united based on several unique musculoskeletal specializations associated with the pres? ence of wings (Baker et al., 1991b). Re? cently, doubt has been raised over the support for a bat dade. Are the large fruiteating bats of the Old World (Megachiroptera) the closest living relatives of the smaller predominantly insectivorous cos? mopolitan bats known as microchiropterans? The monophyly of Chiroptera was questioned first by Smith (1977) and Smith and Madkour (1980). They concluded that derived features of the penis, neurosensory system (size of neocortex), and limb joints suggest a diphyletic origin for bats, or that mammals evolved flight twice. The diphyly hypothesis also has been support? ed by Pettigrew (1986) based upon several features related to the patterns of connec? tion between the retina and midbrain (su? perior colliculus) that are shared in pri? mates and megachiropterans. Predictably, these findings have sparked a controversy 1 Present address; Department of Biology, Texas Wesleyan University, Fort Worth, Texas 76105, USA. over bat relationships. The evidence sup? porting opposing sides of the argument was recently summarized by Pettigrew (1991a, 1991b), Baker et al. (1991b), and Simmons et al. (1991). The disagreement among existing mor? phological studies (Simpson, 1945; Smith, 1977; Smith and Madkour, 1980; Novacek, 1982; Pettigrew, 1986; Shoshani, 1986; Wible and Novacek, 1988; Pettigrew et al., 1989; Thewissen and Babcock, 1991) makes this particular problem an excellent can? didate for molecular analysis. Which suite of characters are truly the result of com? mon ancestry and which are convergent? The differences between the megabats and microbats are widely known (Pettigrew et al., 1989), but synapomorphies for Chirop? tera have been much more elusive. The bats are generally regarded as part of the superorder Archonta (Novacek, 1989), along with Primates, Dermoptera (colugos or flying lemurs), and Scandentia (tree shrews). Bats and colugos traditionally have been recognized as sister taxa (Fig. la), whereas Pettigrew et al. (1989) hypothe? sized a sister-group relationship between one of the groups of bats (megachiropterans) and primates (Fig. lb). There is very little consensus regarding the relation-