A Reevaluation of the Evidence Supporting an Unorthodox Hypothesis on the Origin of Extant Amphibians
Total Page:16
File Type:pdf, Size:1020Kb
Contributions to Zoology, 77 (3) 149-199 (2008) A reevaluation of the evidence supporting an unorthodox hypothesis on the origin of extant amphibians David Marjanovic´, Michel Laurin UMR 7179, Équipe ‘Squelette des Vertébrés’, CNRS/Université Paris 6, 4 place Jussieu, case 19, 75005 Paris, France, [email protected] Key words: Albanerpetontidae, Brachydectes, coding, continuous characters, data matrix, Gerobatrachus, Gym- nophioniformes, Gymnophionomorpha, Lissamphibia, Lysorophia, morphology, ontogeny, paleontology, phylogeny, scoring, stepmatrix gap-weighting Abstract Contents The origin of frogs, salamanders and caecilians is controver- Introduction ............................................................................ 149 sial. McGowan published an original hypothesis on lissam- Nomenclatural remarks ......................................................... 152 phibian origins in 2002 (McGowan, 2002, Zoological Journal Phylogenetic nomenclature .............................................. 152 of the Linnean Society, 135: 1-32), stating that Gymnophiona Rank-based nomenclature ............................................... 154 was nested inside the ‘microsaurian’ lepospondyls, this clade Abbreviations ........................................................................... 154 was the sister-group of a caudate-salientian-albanerpetontid Methods .................................................................................... 155 clade, and both were nested inside the dissorophoid temno- Addition of Brachydectes and homology of its spondyls. We have investigated McGowan’s data matrix and dermal skull bones ............................................................. 155 disagree with the scoring of 35% of the cells. All taxa and all Ontogeny and phylogenetic position of Doleserpeton . 157 but two characters are affected. In some cases, we have a dif- Ontogeny and phylogenetic position of Brachydectes .. 157 ferent interpretation about correspondence between mor- Addition of Gerobatrachus and its ontogeny and phology and character states, or we delimit states differently phylogenetic position ....................................................... 158 (or use information that was unknown in 2002). In others, we Phylogenetic analysis ........................................................ 159 report probable typographic errors. When these cells and Rooting the tree.................................................................. 159 characters are revised, the most parsimonious trees – now Interpretation of the OTUs ............................................. 160 longer by almost 64% – support one of the three commonly Revision of the matrix ...................................................... 163 advocated hypotheses, namely a monophyletic Lissamphibia Results ....................................................................................... 163 nested, together with its sister-group Albanerpetontidae, Discussion ................................................................................. 166 within the temnospondyls (next to Doleserpeton) – even Implications of the size of the matrix ........................... 166 though we did not add any characters or taxa to the very Implications of the quality of the matrix ..................... 167 small data matrix. This exemplifies the impact of errors in Analyses without Gerobatrachus ..................................... 168 data matrices on the results of phylogenetic analyses. Adding Interpretation of Gerobatrachus and effects the lysorophian Brachydectes, however, results in the Lissam- of its addition to our analysis ......................................... 168 phibia-Albanerpetontidae clade becoming the sister-group of ‘Microsaur’ phylogeny ...................................................... 169 Brachydectes and settling within the lepospondyls rather than The phylogenetic position of Albanerpetontidae ....... 169 the temnospondyls, thus supporting another of the previously Acknowledgments ................................................................... 169 published three hypotheses. This latter finding does not References ................................................................................. 170 change if the recently described Gerobatrachus is also added. Appendix 1................................................................................ 177 Finally, when Doleserpeton is interpreted as morphologically Appendix 2................................................................................ 198 immature (which means scoring three characters as unknown Appendix 3................................................................................ 199 instead of known), Lissamphibia and Albanerpetontidae are again nested within the ‘microsaurian’ lepospondyls, even though Brachydectes is not included in this analysis. This, too, does not change if Gerobatrachus is added and likewise treat- Introduction ed as morphologically immature. Bootstrap supports are rather low under all assumptions. Such lability was to be ex- The origin of lissamphibians remains highly conten- pected from the small size of the data matrix. tious, despite decades of intensive research (Vallin 150 D. Marjanovic´ and M. Laurin – Reevaluation of extant amphibian origins and Laurin, 2004; Pawley, 2006: appendix 16; An- other than to Amniota, as do all morphological derson, 2007; Ruta and Coates, 2007; Marjanovic´ analyses based on extant taxa, but paleontological and Laurin, 2007; Anderson et al., 2008a). Our hy- analyses suggest several incompatible hypotheses potheses about the relationships between the rele- about lissamphibian origins. vant extinct taxa (from the late Paleozoic and early Currently, three main hypotheses (all with vari- Mesozoic) have been fairly stable in phylogenetic ants) on lissamphibian origins are discussed based analyses over the last two decades (the various on morphological data. The first hypothesis (Fig. groups of ‘lepospondyls’ are the obvious exception), 1a-d) advocates a single origin within dissorophoid while widely divergent opinions persist on where to temnospondyls (Bolt, 1969 [with reservations], insert the anurans, urodeles and gymnophionans 1977; Milner, 1988, 1993; Panchen and Smithson, into this tree. Part of the problem is a stratigraphic 1988; Trueb and Cloutier, 1991; Lombard and Sum- gap between the oldest known representatives of ida, 1992; Ahlberg and Milner, 1994; Ruta et al., the lissamphibians on the one hand and the first ap- 2003; Ruta and Coates, 2007; Jenkins and Walsh in pearance of most or all of their proposed sister- Jen kins et al., 2007). The second hypothesis (Fig. groups on the other, as noted by Schoch and Milner 1e) advocates a single origin within lepospondyls (2004). (Laurin and Reisz, 1997, 1999; Laurin, 1998a, b; Early phylogenetic analyses of paleontological Vallin and Laurin, 2004; see also Pawley, 2006: 239 data divided most early limbed vertebrates into an and appendix 16). In the third hypothesis (Fig. 1f, amphibian clade composed of temnospondyls and g), the extant amphibians are deemed to have origi- lepospondyls, and a reptiliomorph clade composed nated from at least two, but usually three groups of of embolomeres, seymouriamorphs, diadecto- Paleozoic stegocephalians (Carroll and Currie, morphs and amniotes (Gauthier et al., 1988; Trueb 1975; Carroll and Holmes, 1980; Schoch and Car- and Cloutier, 1991; Lombard and Sumida, 1992; roll, 2003; Carroll et al., 2004; Carroll, 2007; Carroll Ahlberg and Milner, 1994; Fig. 1a, b). However, in Jenkins et al., 2007; Anderson, 2007; Anderson et nearly all recent paleontological studies indicate al., 2008a), once or twice within lepospondyls (gym- that the lepospondyls are closer to the amniotes nophionans and sometimes urodeles), and once or than the temnospondyls and the seymouriamorphs twice within dissorophoid temnospondyls (anurans are, and this topology is now nearly consensual and sometimes urodeles). Unlike the two monophy- (Carroll, 1995, 2007: fig. 77; Laurin and Reisz, 1997, ly hypotheses, it is not compatible with the results 1999; Laurin, 1998a, b; Anderson, 2001, 2007; Ruta of the molecular analyses cited above, and has so et al., 2003; Vallin and Laurin, 2004; Pawley, 2006; far only been supported by two very similar pub- Ruta and Coates, 2007; Anderson et al., 2008a; see lished data matrices (Anderson, 2007; Anderson et Fig. 1c-e, g-h) – only the position of the extant am- al., 2008a) (Fig. 1g). phibians within this tree is not. This latter question The phylogenetic analysis by McGowan (2002) has not yet been resolved, even though it has been supports a fourth hypothesis (Fig. 2): the extant assessed using anatomical data on extant (Carroll amphibians are diphyletic, the (‘microsaurian’) lep- and Currie, 1975; Carroll and Holmes, 1980) and ospondyls are stem gymnophionans, and the lepo- extinct (Milner, 1988, 1993; Laurin and Reisz, 1997, spondyls including the extant amphibians are nested 1999; Laurin, 1998a, b; Anderson, 2001, 2007; Ruta within the (dissorophoid) temnospondyls, contra- et al., 2003; Carroll et al., 2004; Schoch and Milner, dicting the consensus that Temnospondyli and 2004; Vallin and Laurin, 2004; Pawley, 2006; Ruta Lepo spondyli form mutually exclusive clades (Car- and Coates, 2007; Anderson et al., 2008a) forms, as roll, 1995; Vallin and Laurin, 2004; Pawley, 2006; well as developmentary (Schoch and Carroll, 2003; Ruta and Coates, 2007; Anderson, 2007; Anderson Schoch, 2006; Anderson, 2007)