Neobatrachia

Ranoidea Node A

Ranoids are a large and very diverse group. In a sense they are a radiation equivalent to the "hyloids" (=bufonoids). (Realize however that there is little evidence for hyloid monophyly, where is there is data supporting ranoid monophyly, depending on the inclusion of Dendrobatidae.) Ranoids however are firmisternal, meaning that the left and right epicoracoid cartilages of the shoulder girdle are fused. In contrast, the hyloids are mostly arciferal, meaning that the cartilages overlap, and are not fused. Ranoids are a large and very diverse group. In a sense they are a radiation equivalent to the "hyloids" (=bufonoids). (Realize however that there is little evidence for hyloid monophyly, where is there is data supporting ranoid monophyly, depending on the inclusion of Dendrobatidae.) Ranoids however are firmisternal, meaning that the left and right epicoracoid cartilages of the shoulder girdle are fused. In contrast, the hyloids are mostly arciferal, meaning that the cartilages overlap, and are not fused.

Ranoids can be viewed as the Old World counterpart of the hylids, bufonids, and leptodactylids. "Ranidae" are analogous to "Leptodactylidae" in being a diverse, paraphyletic group of non-treefrog forms, but with their center of diversity is Africa and Southeast Asia, with a minor radiation of the genus Rana in the New World. Rhacophoridae and Hyperoliidae are the Old World ranoid treefrog types, analogous to hylids in Central and South America. Microhylids are perhaps the firmisternal equivalent to Bufonidae in that they are a cosmopolitan group with three apparent radiations, in South America, Africa, and southeast Asia.

One major issue in the content of the Ranoidea is the inclusion (or not) of the Dendrobatidae (poison-dart frogs). Here we follow Ford (1989) in treating dendrobatids as ranoids. Analysis of rRNA sequence data by Hay et al. (1995) does not support this, however. The four species of ranoids analyzed by Hay et al. (1995) formed a monophyletic group, but the dendrobatid was placed among the "hyloids."

Phylogenetic relationships.--The name Ranoidea was defined by Ford and Cannatella (1993) as the node-based name for the last common ancestor of hyperoliids, rhacophorids, ranids, dendrobatids, Hemisus, arthroleptids, and microhylids, and all of its descendants. Synapomorphies of Ranoidea include completely fused epicoracoid cartilages, and the medial end of coracoid being wider than lateral end (Ford, 1989b). A probable synapomorphy, depending on the position of Dendrobatidae within Ranoidea, is the insertion of the semitendinosus tendon dorsal to the gracilis muscle (convergent with some myobatrachines).

Traditionally, Ranoidea was diagnosed by the condition of the pectoral girdle described as firmisterny (Duellman and Trueb, 1986). This feature has been controversial because of ambiguity in the definition of the term. One of the simpler and more widely accepted definitions of firmisterny is the complete fusion of the epicoracoid cartilages (Peters, 1964). The girdle has been treated as a single complex unit, and many individual aspects of it (i. e., clavicle, coracoid, epicoracoid and procoracoid cartilages) have been not used in systematic analyses (Ford, 1989a). The validity of the firmisternal girdle as a single character cannot be assessed until the individual elements are examined for character independence (Ford, 1988, 1989b).

Heleophryne Node B

The redundant family name applied to this taxon is Heleophrynidae.

Allophryne ruthveni Node C

The redundant family name applied to this taxon is Allophrynidae.

"Myobatrachidae" Node D

Under most arrangements, Limnodynastinae and Myobatrachinae are included in the family Myobatrachidae. The arrangement here follows Ford and Cannatella (1993).

Rhinoderma Node E

The redundant family name applied to this taxon is Rhinodermatidae.

Hemisus Node F

The redundant family name applied to this taxon is Hemisotidae.

Unnamed Node G

No name has been proposed for this node. Under Linnean nomenclature the name Myobatrachidae has priority over Sooglossidae.

Unnamed Node H

Lynch (1973) and Duellman and Trueb (1986:character J) used the presence of intercalary elements to unite Hylidae, Centrolenidae and Pseudidae into a clade. Intercalary elements are also known in Hyperoliidae and Rhacophoridae, as well as mantelline ranids and phrynomerine microhylids. The tree presented by Duellman and Trueb (1986:Fig. 17-3) has an equally parsimonious solution that would require convergent evolution of a firmisternal girdle (character C1) if lack of homoplasy in the intercalary element (character J1) is favored. In a subtree Duellman and Trueb (1986:Fig. 17-4) used the presence of intercalary elements to unite mantellines, hyperoliids, and rhacophorids, but not phrynomerines. This tree also has an equally parsimonious alternative that would unite phrynomerines with the aforementioned clade, but requires homoplasy in the Type 2 larva (character O2').

Ford's (1989) study of dendrobatid relationships did not include centrolenids or pseudids. Ford and Cannatella continued the recognition of the clade consisting of Hylidae, Pseudidae, and Centrolenidae, which is diagnosed by the presence of intercalary elements, but did not name it. Hay et al. (1995) found that pseudids, hylids and centrolenids (each represented by one species) not to be a clade, although the latter two were sister-taxa, and all three were part of the Neobatrachia.

Recently, Ruvinsky and Maxson (1996; Molecular Phylogenetics and Evolution 5:533-547) analyzed relationships among Neobatrachia based on mitochondrial 12S and 16S sequences. Their neighbor-joining tree is presented below. Several features of their tree are noteworthy.

Bufonoids are a monophyletic group, and dendrobatids are nested within this clade. Ranoids are monophyletic also. Hylidae is polyphyletic, with some hylids being most closely related to Bufonidae, some to this clade plus Centrolenidae, and some to Pseudidae and Rhinodermatidae. Leptodactylidae as represented by Telmatobiinae and Leptodactylinae can be considered paraphyletic in one considers that Bufonidae, Hylidae, Rhinodermatidae, Pseudidae, and Dendrobatidae are all derived from Leptodactylidae, an arrangement that agrees with Lynch (1971).