Origin and Phylogenetic Relationships of Living Amphibians

Points of View Syst. Biol. 51(2):364–369, 2002 Tetrapod Phylogeny, Amphibian Origins, and the Denition of the Name Tetrapoda MICHEL LAURIN Equip´ e “Formations squelettiques” UMR CNRS 8570, Case 7077, Universite´ Paris 7, 75005 Paris, France; E-mail: [email protected] The most detailed, computer-generated Ahlberg and Milner, 1994) or did not include phylogeny of early amphibians has been amphibians (Gauthier et al., 1988), thus published recently (Anderson, 2001). The precluding identication of the dichotomy study was based on an analysis of 182 os- between amphibians and reptiliomorphs teological characters and 49 taxa, including (the clade that includes amniotes and the 41 “lepospondyls,” 7 other Paleozoic taxa extinct taxa that are more closely related representing other major groups (seymouri- to amniotes than to lissamphibians). The amorphs, embolomeres, temnospondyls, latest phylogeny (Anderson, 2001) suggests Downloaded By: [University of Tennessee] At: 00:39 18 January 2008 etc.), and a single lissamphibian (the oldest that despite initial skepticism (Coates et al., known apodan, the Jurassic taxon Eocae- 2000), the new pattern of stegocephalian cilia). The publication of this phylogeny is relationships appears to be well supported welcome, but several points raised in that and relatively stable, at least as it pertains to paper deserve to be discussed further. Espe- the relationships between major taxa such cially problematic are the views Anderson as Temnospondyli, Embolomeri, Seymouri- expressed about the polyphyletic origin of amorpha, Amniota, and the paraphyletic extant amphibians and about the application group informally called lepospondyls of phylogenetic nomenclature. (Figs. 1a, 1b). TETRAPOD PHYLOGENY Anderson (2001) found that the lep- Origin of Extant Amphibians ospondyls are more closely related to This brings us to the more difcult ques- amniotes than to seymouriamorphs and em- tion of amphibian origins. Many authors bolomeres. This result had been previously have considered that extant amphibians (i.e., reported by the other authors who had done anurans, urodeles, and apodans) form a a computer-assisted phylogenetic analysis of clade (Lissamphibia) that excludes all known the main early stegocephalian taxa (Carroll, Paleozoic taxa. The most common opin- 1995; Laurin and Reisz, 1997, 1999; Laurin, ion (e.g., Gauthier et al., 1988:105; Trueb 1998; Ahlberg and Clack, 1998; Paton et al., and Cloutier, 1991; Ahlberg and Milner, 1999). These results (Figs. 1a, 1b) are clearly 1994) is that this lissamphibian clade is incompatible with the more traditional part of Temnospondyli (Figs. 1c, 1d). How- stegocephalian phylogeny (Figs. 1c, 1d) ex- ever, this opinion has never been substan- emplied by Gauthier et al. (1988), Panchen tiated by a published data matrix that in- and Smithson (1988), Lombard and Sumida cluded lissamphibians, temnospondyls, and (1992), and Ahlberg and Milner (1994). other potential relatives of lissamphibians However, those older studies either were (such as lepospondyls). For instance, the not based on a data matrix (Panchen and computer-assisted phylogenetic analysis of Smithson, 1988; Lombard and Sumida, 1992; Trueb and Cloutier (1991) included only 364 2002 POINTS OF VIEW 365 Downloaded By: [University of Tennessee] At: 00:39 18 January 2008 FIGURE 1. Stegocephalian phylogeny. (a) Recent phylogeny, as found in Laurin and Reisz (1997, 1999) and Laurin (1998). Similar variants have been published by Carroll (1995), Ahlberg and Clack (1998), and Paton et al. (1999). (b) Phylogeny of Anderson (2001). (c) Classical stegocephalian phylogeny, as exemplied by Gauthier et al. (1988), Panchen and Smithson (1988), Lombard and Sumida (1992), and Ahlberg and Milner (1994). (d) Phylogeny within temnospondyls, as assessed by Trueb and Cloutier (1991). (e) Polyphyletic origin of extant amphibians, adapted from the sources cited in (c) and Carroll (1988). All phylogenies have been simplied by eliminating some terminal taxa or regrouping them into larger clades. 366 SYSTEMATIC BIOLOGY VOL. 51 temnospondyls and lissamphibians (Fig. 1d). tentatively suggests an origin of caecilians A second, recently proposed hypothesis [apodans] separate from frogs [anurans] (as- (Fig. 1a) is that Lissamphibia is nested within suming a temnospondyl origin), has received lepospondyls (Laurin and Reisz, 1997, 1999; support from molecular (Feller and Hedges, Laurin, 1998). The third hypothesis that 1998) and developmental (Carroll et al., 1999) needs to be considered here is that extant am- studies” (p. 182). However, the molecular phibians do not form a monophyletic group phylogeny presented by Feller and Hedges unless at least some Paleozoic taxa are also (1998), far from supporting the suggestion by included. This hypothesis has not usually Anderson (2001), is incompatible with it. In- been formulated in a cladistic context (ex- deed, if the phylogeny of Anderson (2001) cept by the variant presented in Milner, 1993), is accepted, and if anurans truly are tem- but a free translation into a cladistic form is nospondyls, then apodans must necessarily shown here (Fig. 1e). Indeed, this approach be more closely related to amniotes than to has been championed by Carroll and some anurans because apodans are lepospondyls, of his former students (Carroll and Currie, and lepospondyls are more closely related 1975; Carroll and Holmes, 1980; Carroll, to amniotes than to temnospondyls (Fig. 2a). 1988). According to that hypothesis, apodans In contrast, Feller and Hedges (1998) found and urodeles are derived from lepospon- a monophyletic Lissamphibia (i.e., apodans dyls, whereas anurans are derived from are more closely related to anurans than to temnospondyls. amniotes). Most, if not all other molecular Anderson (2001:182) also subscribes to a phylogenies (Hedges et al., 1990; Hedges polyphyletic origin of extant amphibians, and Maxson, 1993; Hay et al., 1995; Zardoya but his results do not warrant this opin- and Meyer, 2001) show a similar pattern ion. Indeed, in his phylogeny, the early apo- (Fig. 2b). Because Feller and Hedges (1998) dan Eocaecilia (the only potential lissam- did not have an outgroup, they rooted their Downloaded By: [University of Tennessee] At: 00:39 18 January 2008 phibian included) is deeply nested within tree between lissamphibians and amniotes. lepospondyls, a result compatible with both Thus, rerooting the tree could have ren- the second and third hypotheses above dered it compatible with the suggestion by (in which modern amphibians are mono- Anderson (2001) and Carroll (1988) that apo- phyletic and polyphyletic, respectively) re- dans and anurans are lepospondyls and tem- garding amphibian origins (e.g., Carroll and nospondyls, respectively. However, Hedges Currie, 1975; Carroll and Holmes, 1980; et al. (1990) and Hedges and Maxson (1993) Laurin and Reisz, 1997, 1999). Anderson then did have an outgroup, and their phyloge- concludes, “The present hypothesis, which nies, if correct, show a monophyletic and FIGURE 2. Relationships between extant amphibians and amniotes. (a) Phylogeny postulated by Anderson (2001). (b) Phylogeny proposed in various molecular studies (Hedges et al., 1990; Hedges and Maxson, 1993; Hay et al., 1995; Feller and Hedges, 1998; Zardoya and Meyer, 2001). 2002 POINTS OF VIEW 367 strongly supported Lissamphibia (Fig. 2b). or Neotetrapoda proposed by Gaffney (1979) Thus, the molecular phylogenies are incom- constitutes a valid phylogenetic denition. patible with the diphyletic or triphyletic Indeed, de Queiroz and Gauthier (1990, 1992, origin of extant amphibians suggested by 1994) did not recognize a name on a clado- Anderson (2001) and Carroll (1988). gram (as found in Gaffney, 1979) or other The molecular data cannot test di- implicit references to a clade as a valid rectly whether extant amphibians are tem- phylogenetic denition. Several passages in nospondyls, lepospondyls, or both, but do those articles tackle this issue. For instance, clearly suggest that they cannot be descen- de Queiroz and Gauthier (1990:309) stated, dants of both of these Paleozoic groups un- ”Thus, an evolutionary ostensive denition, less temnospondyls and lepospondyls form a hereafter referred to as a phylogenetic def- clade that excludes amniotes. Therefore, the inition, consists of pointing to a clade, that nding that lepospondyls are more closely is, to an ancestor and its descendants.” The related to amniotes than to temnospondyls authors then explain that this can be accom- seems to rule out a polyphyletic origin of ex- plished in three ways, corresponding to the tant amphibians. node-based, stem-based, and apomorphy- based denitions. Gaffney (1979) did not explicitly dene the names Tetrapoda or PHYLOGENETIC NOMENCLATURE Neotetrapoda. Anderson (2001:171) states: “: : : I do not Third, Anderson’s (2001:171) statement adhere to the principle of priority of explicit that “Gaffney’s Neotetrapoda is clearly a phylogenetic denition advocated by Laurin crown taxon and thus takes priority over when applied to historically laden names. In- a crown-based denition of Tetrapoda” is stead I follow de Queiroz and Gauthier (1990, problematic because it is far from obvious 1992, 1994), who advocate following the rst (or certain) that Gaffney’s concept of Neote- Downloaded By: [University of Tennessee] At: 00:39 18 January 2008 use of a name with a specic clade, which trapoda was crown-based. Anderson’s (2001) in this case (Tetrapoda) would be Gaffney opinion may stem from the observation that (1979) : :

Load more