Rafael O. de Sá, Taran Grant, Arley Camargo, W. Ronald Heyer, Maria L. Ponssa, Edward Stanley
South American Journal of Herpetology 9 (s1), (1 September 2014) https://doi.org/10.2994/SAJH-D-13-00022.1
KEYWORDS: behavior, distribution, life history, morphology, taxonomy, vocalization
A phylogeny of the species-rich clade of the Neotropical frog genus Leptodactylus sensu stricto is presented on the basis of a total evidence analysis of molecular (mitochondrial and nuclear markers) and non-molecular (adult and larval morphological and behavioral characters) sampled from > 80% of the 75 currently recognized species. Our results support the monophyly of Leptodactylus sensu stricto, with Hydrolaetare placed as its sister group. The reciprocal monophyly of Hydrolaetare and Leptodactylus sensu stricto does not require that we consider Hydrolaetare as either a subgenus or synonym of Leptodactylus sensu lato. We recognize Leptodactylus sensu stricto, Hydrolaetare, Adenomera, and Lithodytes as valid monophyletic genera. Our results generally support the traditionally recognized Leptodactylus species groups, with exceptions involving only a few species that are easily accommodated without proposing new groups or significantly altering contents. The four groups form a pectinate tree, with the Leptodactylus fuscus group diverging first, followed by the L. pentadactylus group, which is sister to the L. latrans and L. melanonotus groups. To evaluate the impact of non-molecular evidence on our results, we compared our total evidence results with results obtained from analyses using only molecular data. Although non-molecular evidence comprised only 3.5% of the total evidence matrix, it had a strong impact on our total evidence results. Only one species group was monophyletic in the molecular-only analysis, and support differed in 86% of the 54 Leptodactylus clades that are shared by the results of the two analyses. Even though no non-molecular evidence was included for Hydrolaetare, exclusion of that data partition resulted in that genus being nested within Leptodactylus, demonstrating that the inclusion of a small amount of non-molecular evidence for a subset of species can alter not only the placement of those species, but also species that were not scored for those data. The evolution of several natural history and reproductive traits is considered in the light of our phylogenic framework. Invasion of rocky outcrops, larval oophagy, and use of underground reproductive chambers are restricted to species of the Leptodactylus fuscus and L. pentadactylus groups. In contrast, larval schooling, larval attendance, and more complex parental care are restricted to the L. latrans and L. melanonotus groups. Construction of foam nests is plesiomorphic in Leptodactylus but their placement varies extensively (e.g., underground chambers, surface of waterbodies, natural or excavated basins). Information on species synonymy, etymology, adult and larval morphology, advertisement call, and geographic distribution is summarized in species accounts for the 30 species of the Leptodactylus fuscus group, 17 species of the L. pentadactylus group, eight species of the L. latrans group, and 17 species of the L. melanonotus group, as well as the three species that are currently unassigned to any species group.