View Anura Tree

The node-based name Anura was defined by Ford and Cannatella (1993) as the last common ancestor of living frogs and all its descendants. According to this definition, †Triadobatrachus is not part of Anura, following Trueb and Cloutier (1991). The late Jurassic fossil †Notobatrachus degiustoi has been considered as having uncertain placement with respect to Anura--possibly the sister-group, possibly not (Cannatella, 1985); or related to Leiopelma (Estes and Reig, 1973). Baez and Basso (1996) placed it as the sister-group of Anura.

†Vieraella herbstii is another relatively complete early Jurassic fossil, but is less well-preserved than †Notobatrachus (Estes and Reig, 1973). The presence of nine presacral vertebrae places it among the basal salientians, but other characters are not sufficiently preserved to permit definitive placement in Anura. A third well-preserved Jurassic fossil taxon, †Eodiscoglossus santonje, has eight presacral vertebrae (Estes and Reig, 1973), and thus is clearly within Anura. †Prosalirus bitis is from the Early Jurassic (Pliensbachian) of Arizona; the putative limbed caecilian †Eocaecilia is from the same formation. It is reputed to be the earliest known "frog" in the sense that it has froglike features such as a urostyle and elongate, anteriorly directed ilia. For more details see the leaf page on †Prosalirus.
The monophyly of Anura has rarely been questioned. Griffiths (1963:279) considered Anura to be diphyletic; i.e., Ascaphus and Leiopelma comprised one lineage, and all other frogs a second, "...stemming independently from either different levels of a single proanuran organization or different proanuran stocks." Rocek (1981) considered †Triadobatrachus to be within Anura, and placed Pelobatidae (including only Pelobates and some related fossils), and somewhat tentatively, Pipidae, †Palaeobatrachidae, and Rhinophrynidae, in Archaeosalientia. †Triadobatrachus and remaining anurans were included in Neosalientia. The character distinguishing the two groups is a median dermal bone, the interparietal (Reinbach, 1939), which Rocek homologized with the median extrascapular of the osteolepiform fish †Eusthenopteron. The interparietal is found in Pelobates and some fossil relatives, but not in any pipoid frogs. Milner (1988) provided a cogent review of the interpretation of this dermal element.

In a study of tetrapod phylogeny, Hedges et al. (1990) analyzed 123 phylogenetically informative sites from 18S ribosomal RNA of 21 tetrapods, including 4 salamanders, 4 caecilians, and 1 species each from Bufonidae, Discoglossidae, Hylidae, Leptodactylidae, Microhylidae, Pelobatidae, and Sooglossidae. Bootstrap analyses using both maximum-parsimony and neighbor-joining algorithms did not support the monophyly of Anura, Caudata, or Gymnophiona, but a monophyletic Amphibia was supported with a bootstrap value of 100%. Hedges and Maxson (1993) and Hillis et al. (1993) presented analyses of anuran relationships based on DNA sequence data from the mitochondrial and nuclear ribosomal genes, respectively.

Hedges et al. (1990) also analyzed 35 variable sites from 28S rRNA of four species of frogs, from Discoglossidae, Hylidae, Pelobatidae, and Pipidae. Bootstrap analysis of maximum parsimony trees yielded a monophyletic Anura, but with no resolution among the four taxa. The neighbor-joining analysis indicated a sister-group relationship between the discoglossid and pipid, and between the bufonid and hylid, but the bootstrap value for both of these nodes was less than 50%. Neither analysis yielded a monophyletic Amphibia.

Hay et al. (1995) is the most comprehensive molecular systematics treatment of the relationships among the families of amphibians. Their neighbor-joining tree yielded a monophlyletic Anura, Caudata, and Gymnophiona.

"Archaeobatrachia"

Archaeobatrachians have generally included discoglossoids, pipoids and pelobatoids (Duellman, 1975; Reig, 1958). As discussed on other pages, the synapomorphies of Bombinanura and Pipanura collectively demonstrate that "Archaeobatrachia" is paraphyletic. The informal term archaeobatrachian is a convenient term for anurans that are not part of Neobatrachia.

Bombinanura

Bombinanura, a node-based name, was defined by Ford and Cannatella (1993) to be the most recent common ancestor of living Bombinatoridae and Discoglossanura, and all its descendants. Synapomorphies of Bombinanura include fusion of the halves of the sphenethmoid, eight presacral vertebrae, absence of the m. epipubicus, and absence of the caudalipuboischiotibialis muscle (Cannatella, 1985). Subclades of Bombinanura include Bombinatoridae and Discoglossanura.

Discoglossidae and Discoglossoidea

Most analyses of molecular sequence data have placed Alytes, Discoglossus, Bombina, and Barbourula in a clade, to which the name Discoglossidae has been applied. In contrast, Ford and Cannatella (1993) listed derived morphological features that would require paraphyly of this group , and they recognized both Discoglossidae (for Alytes and Discoglossus) and Bombinatoridae (for Bombina and Barbourula). A synthesis of the two datasets has not been attempted. There are apparently no published synapomorphies for "Discoglossidae," and the dissimilarity of Alytes and Discoglossus, on one hand, and Bombina on the other, has often been noted (e.g., Lanza et al., 1976), suggesting a deep divergence between the two lineages. Griffiths (1963) stated that the diagnostic feature of Discoglossidae (sensu lato) was a triradiate sternum. However, this feature is also present in Leiopelma. The similarity can be interpreted either as shared plesiomorphy between Leiopelma and Discoglossidae, or a derived feature that unites Leiopelma more closely to Discoglossidae than to Ascaphus.

Pipanura

The node-based name Pipanura was proposed by Ford and Cannatella (1993) for the most recent common ancestor of living Mesobatrachia + Neobatrachia, and all its descendants. The subordinal name Ranoidei was coined for this clade by Sokol (1977), but that name was re-assigned to a less inclusive taxon by Dubois (1983, 1984). Sokol's (1977) use of this name was unfortunate because the informal name, ranoid, is homonymous with the widely used name Ranoidea.

Synapomorphies include a sinistral spiracle in the larvae (a characteristic feature of Orton's Type 4 tadpole), absence of free ribs in adults, torsion in the carpal elements, the presence of vocal sacs, and fusion of the trigeminal and facial ganglia (Cannatella, 1985; Sokol, 1975).

Acosmanura

Pelobatoidea

The name was applied to the node that is the common ancestor of living Megophryidae, Pelobatidae, and Pelodytes. Synapomorphies of Pelobatoidea include the presence of a palatine process of the maxilla and ossification of the sternum into a bony style (Cannatella, 1985). Duellman and Trueb (1986) listed the presence of a dorsal gap in the cricoid ring as a synapomorphy for this clade. However, there is no gap in Scaphiopus, Spea, Pelobates (except for the smallest species, P. fuscus), or several megophryids; at best, the presence of a dorsal gap would be an ambiguous synapomorphy. Relationships among the living pelobatoids are an unresolved trichotomy of Megophryidae, Pelobatidae, and Pelodytes.

Pipoidea

Pipoidea was defined by Cannatella and Ford (1993) to be the most recent common ancestor of living Pipidae + Rhinophrynidae, and all its descendants. Pipoidea is diagnosed by several distinctive synapomorphies, including the absence of mentomeckelian bones (see comments below on †Palaeobatrachidae), absence of lateral alae of the parasphenoid, fusion of the frontoparietals into an azygous element, greatly enlarged otic capsules, and a tadpole with paired spiracles, and lacking beaks and denticles (Orton Type 1 tadpole). Clades of Pipoidea include Pipidae, †Palaeobatrachidae, some unplaced fossil "pipids," and Rhinophrynidae.

Pipimorpha

The new stem-based name Pipimorpha was defined by Ford and Cannatella (1993) to be those taxa that are more closely related to living Pipidae than to living Rhinophrynus. Pending a more detailed assessment of the relationships of †Palaeobatrachidae, Pipidae, and the fossil "pipids," we consider †Palaeobatrachidae, †Thoraciliacus, †Cordicephalus, †Saltenia, †Shomronella, and †Eoxenopoides to be part of Pipimorpha.

"Mesobatrachia"

Ford and Cannatella (1993) applied the name Mesobatrachia to the node that is the most recent common ancestor of the living Pelobatoidea and Pipoidea. This relationship was based on synapomorphies discussed by Cannatella (1985). Mesobatrachia as proposed by Laurent (1979) was a paraphyletic group. Cannatella (1985) first applied the name to a clade. Synapomorphies reported for Mesobatrachia included closure of the frontoparietal fontanelle by juxtaposition of the frontoparietal bones, partial closure of the hyoglossal sinus by the ceratohyals, absence of the taenia tecti medialis, and absence of the taenia tecti transversum (Cannatella, 1985; Sokol, 1981).Most other previous taxonomies, and recent molecular phylogenetic analyses, have placed the Pelobatoidea as the sister-group to Neobatrachia, rather than to Pipoidea. Thus, the Mesobatrachia as formerly proposed is not monophyletic, and therefore no longer recognized.