Portal de Periódicos da CAPES

A New Species of the Eleutherodactylus myersi (Anura: Leptodactylidae) Assembly from Ecuador

1992; The Society for the Study of Amphibians and Reptiles; Volume: 26; Issue: 2 Linguagem: Inglês

10.2307/1564862

1937-2418

John J. Wiens, Luis A. Coloma,

Lepidoptera: Biology and Taxonomy

-A new species of small, terrestrial Eleutherodactylus is described from a restricted area of interandean montane forest in central Ecuador. This species is most similar to E. orestes from southern Ecuador, but is distinguished by differences in coloration, osteology, morphometrics, and allozymes. The new species is a member of the myersi assembly of the unistrigatus species group. A phylogenetic analysis of six species of the assembly using allozymic and morphological data suggests the following relationships: ((E. pyrrhomerus + E. leon-i + ((E. ocreatus + E. trepidotus) + (new species + E. orestes))). Eleutherodactylus anae is a junior subjective synonym of E. curtipes and is not a member of the myersi assembly. RESUMEN. Se describe una nueva especie de Eleutherodactylus pequeio de habitos terrestres colectado en un refugio de bosque montano interandino en Ecuador central. Esta especie se parece mas a E. orestes del sur de Ecuador, pero se le distingue por diferencias de coloracion, osteologia, morfometria, y aloenzimas. Esta especie nueva es miembro del subgrupo (assembly) myersi en el grupo de especies unistrigatus. Un analisis filogenetico de seis especies del subgrupo usando datos de aloenzimas y morfologicos indica las relaciones siguientes: ((E. pyrrhomerus + E. leoni) + ((E. ocreatus + E. trepidotus) + (especie nueva + E. orestes))). Eleutherodactylus anae es sinonimo de E. curtipes y no es miembro del subgrupo myersi. The leptodactylid frog genus Eleutherodactylus contains more than ten percent of all anuran species (Frost, 1985), with more than 430 species described (Lynch and Burrowes, 1990). These species are currently distributed among several species groups. The largest of these is the unistrigatus group, with over 100 species found mostly in South America. Within the unistrigatus group are a number of smaller, informal taxa recognized as assemblies by Lynch and Duellman (1980). Lynch (1981) erected the myersi assembly for E. ginesi, E. myersi, E. nicefori, E. ocreatus, E. orestes, E. trepidotus, and E. vidua. Subsequently, Lynch (1984) placed E. myersi, E. ocreatus, and E. trepidotus into the pyrrhomerus assembly (which contained E. gladiator, E. leoni, E. pyrrhomerus, and E. repens), but considered the pyrrhomerus assembly a monophyletic subgroup within the more inclusive myersi assembly. With the recent addition of E. anae Rivero 1986, the myersi assembly consists of a total of 12 species from the high Andes of Colombia, Venezuela, and Ecuador. The Bosque Protector Cashca Totoras in westcentral Ecuador contains one of the last remaining stands of interandean montane forest Present addresses: 2 Department of Zoology, The University of Texas, Austin, Texas 78712-1064, USA; 3 Museo de Zoologia, Departamento de Ciencias Biol6gicas, Pontifica Universidad Catolica del Ecuador, Av. 12 de Octubre, Apartado 17-01-2184, Quito, Ecuador. in the country. Over the past several years, field parties from the Universidad Cat6lica of Quito and University of Kansas have collected representatives of a small, terrestrial Eleutherodactylus at Cashca Totoras, apparently belonging to the myersi assembly. The purposes of the present paper are to describe this species and to conduct a preliminary phylogenetic analysis of the myersi assembly. MATERIALS AND METHODS Specimens examined, including skeletal preparations and specimens examined for allozymic comparisons, are listed in Appendix 1. Museum abbreviations are: KU (The University of Kansas, Museum of Natural History); QCAZ (Museo de Zoologia, Pontifica Universidad Cat6lica del Ecuador, Quito); ULABG (Universidad de los Andes, Laboratorio de Biogeografia, Merida, Venezuela); and UPR-M (University of Puerto Rico, Mayaguez). Terminology for external features and format of the diagnosis and description follow Lynch and Duellman (1980). Measurements were taken to the nearest 0.1 mm using dial-tipped calipers. Abbreviations used in the text are as follows: E-N (eye to nostril distance); HW (greatest width of head); IOD (interorbital distance); and SVL (snout-vent length). Means reported for measurements include one standard error (+). Descriptive statistics were computed using the NCSS statistical package, and multivariate analyses were performed using the SAS statistical package on a mainframe computer. Cleared-and-stained This content downloaded from 207.46.13.125 on Sun, 17 Jul 2016 06:13:21 UTC All use subject to http://about.jstor.org/terms NEW ECUADORIAN ELEUTHERODACTYLUS skeletons were prepared following a modified version of the technique of Dingerkus and Uhler (1977). In order to assess the relationships of the new species, we performed a preliminary phylogenetic analysis of the myersi assembly using allozymic and morphological data. Liver and skeletal muscle were removed from animals freshly killed in the field (using a 15% solution of benzocaine) and were frozen immediately in liquid nitrogen for transport to the laboratory. Tissues were stored at -70 C and were used within two years of collection. Liver and skeletal muscle were homogenized separately with a teflon homogenizer in a 1:1 (v:v) mixture of tissue and distilled water. Homogenates were centrifuged at 13,445 g for 10 min at 5 C. Tissue samples were run at 5 C on horizontal starch gels composed of 12% hydrolyzed potato starch. Presumptive gene loci were visualized by histochemical staining methods (Harris and Hopkinson, 1976; Selander et al., 1971; Siciliano and Shaw, 1976). Enzyme nomenclature follows the recommendations of the International Union of Biochemistry Nomenclature Committee (1984). Loci were numbered from anode to cathode, and alleles were labeled a, b, c, etc., in order of increasing anodal mobility. Buffer systems, tissue sources, loci, and IUBNC numbers are as follows: Lithium hydroxide, muscle (Cbp-calcium binding protein, nonspecific); Poulik, muscle (Ldh-2, 1.1.1.27; Me, 1.1.1.40; Mpi, 5.3.1.8); Tris-borate-EDTA-NAD pH 9.1, liver (Ga3pdh, 1.2.1.12; G6pdh, 1.1.1.49; Ldh-1, 1.1.1.27; Mdh-1, 1.1.1.37; Mdh-2, 1.1.1.37; Sdh, 1.1.1.14); Tris-citrate-NADP pH 7.0, liver (Ald, 4.1.2.7; Icdh-1, 1.1.1.42; Icdh-2, 1.1.1.42); Tris-citrate-NADP pH 8.0, muscle (Ck, 2.7.3.2; Gpi, 5.3.1.9; Pgm, 5.4.2.2). Parsimony analysis was performed using David Swofford's PAUP (version 3.0n) and FREQPARS (which analyzes data in the form of frequencies). For the PAUP analysis, the Exhaustive Search option was used to guarantee finding the shortest tree, and character state optimizations for each stem were checked using the ACCTRAN and DELTRAN (Swofford and Maddison, 1987) optimization routines and by hand. The electrophoretic data were coded for the qualitative (PAUP) analysis by considering the locus as the character and the allelic array (combination of alleles present) as the character state (Table 1). Intraspecific polymorphisms were differentially weighted using step matrices (following P. Mabee and J. Humphries, pers. comm.). The appearance of a derived allele as a polymorphism was given an a priori weight of 0.5 as was the fixation of that allele (or apparent loss of the plesiomorphic allele). For example, the minimum hypothesized weight between the allelic arrays aa and ab would be 0.5 step, between aa and bb 1.0, between aa and cd 1.5, and between ab and cd 2.0 steps. Six morphological characters were included in the same data matrix (Table 1). Lynch (1984) used 12 morphological characters to generate a phylogenetic hypothesis for the myersi assembly (including the pyrrhomerus assembly). We used (or modified) four of these characters (17, 19, 21, 22); the remaining characters discussed by Lynch (1984) either appear to vary continuously within and between species (separation of nasals, shape of canthus rostralis, expansion of digits, presence of pads on fingers), or were not applicable at the level of our analysis (length of fingers, iris color, skin texture, size of metatarsal tubercles). For the purposes of outgroup comparison, we examined representatives of three other assemblies of high Andean Eleutherodactylus of the unistrigatus group: E. buckleyi and E. curtipes (curtipes assembly), E. devillei and E. vertebralis (devillei assembly), and E. unistrigatus (unistrigatus assembly). Among these species, the members of the curtipes assembly share the presumably derived features of a lateral flange on the frontoparietals and exostosis of the nasals and frontoparietals, and the members of the devillei assembly examined are united by the loss of vocal slits and the presence of an inner tarsal fold (Lynch, 1983). The members of the curtipes and myersi assemblies have very narrow digital pads, presumably a derived feature. Members of the myersi assembly examined are united by their reduced size (maximum size of adult females 20.2-27.2 mm SVL versus 38.5-48.8 mm in the putative outgroups). The presence of bright inguinal coloration might also be a synapomorphy for these species. In summary, the following relationships were assumed for outgroup comparison: (E. unistrigatus + (E. devillei + E. vertebralis) + ((E. buckleyi + E. curtipes) + myersi assembly)). Character state polarities were optimized using the algorithm of Maddison et al. (1984).