Zootaxa 3731 (3): 345–370 ISSN 1175-5326 (print edition) www.mapress.com/zootaxa/ Article ZOOTAXA Copyright © 2013 Magnolia Press ISSN 1175-5334 (online edition) http://dx.doi.org/10.11646/zootaxa.3731.3.5 http://zoobank.org/urn:lsid:zoobank.org:pub:6D082EF3-DA05-41FE-8D7D-662F800907C8 Rediscovery of the Earless Microteiid Lizard Anotosaura collaris Amaral, 1933 (Squamata: Gymnophthalmidae): A redescription complemented by osteological, hemipenial, molecular, karyological, physiological and ecological data MIGUEL TREFAUT RODRIGUES1,3, MAURO TEIXEIRA JR1, FRANCISCO DAL VECHIO1, RENATA CECÍLIA AMARO1, CAROLINA NISA1, AGUSTÍN CAMACHO GUERRERO1, ROBERTA DAMASCENO2, JULIANA GUSSON ROSCITO1, PEDRO M. SALES NUNES1, & RENATO SOUSA RECODER1 1Departamento de Zoologia, Instituto de Biociências, Universidade de São Paulo, Caixa Postal 11.461, CEP 05508-090, São Paulo, SP, Brazil 2University of California, Berkeley, Museum of Vertebrate Zoology and Department of Integrative Biology, 3101 Valley Life Sciences Building, Berkeley, CA 94720–3140, USA 3Corresponding author. E-mail: [email protected] Abstract More than a century after its discovery by Ernest Garbe, and almost 80 years after its original description, we obtained a series of specimens of the earless gymnophthalmid Anotosaura collaris, the type species of the genus, up to now known only by a single specimen. On the basis of the material obtained at and close to the type locality we redescribe the species, adding information about the external and hemipenial morphology, osteology and karytoype. Molecular data confirm its sister relationship with Anotosaura vanzolinia as well as the close relationship of Anotosaura with the Ecpleopodini Colo- bosauroides and Dryadosaura. We supplement this information with thermophysiological, ecogeographical, karyotypic and ecological data. Key words: Anotosaura collaris, Gymnophthalmidae, Brazil, earless lizard, Caatinga Introduction In the thirties of the last century, after a relatively long period of stasis in the taxonomic knowledge of Brazilian lizards, Afrânio do Amaral, the Director of Instituto Butantan, turned his attention to the biological collections of the Museu Paulista, latter dismembered into several institutions, one of which came to be the present Museu de Zoologia, Universidade de São Paulo (MZUSP). Although his main research focus was on uncovering undetected lizard parasites, he found several apparently undescribed lizards collected by the Museum’s naturalists in different parts of Brazil. Ernest Garbe was one of those. He was hired in 1901 and since then was actively engaged in collecting specimens from unexplored parts of Brazil and sending the samples to the Museum. Amaral eventually published an important paper on this material, in which he describes 17 new lizard species, some of which as new genera (Amaral 1933). Although several were later considered synonyms of previously known species, others remain valid and are still extremely rare. Examples are the microteiid Anotosaura collaris Amaral, 1933, only known by its type specimen, Anolis nasofrontalis Amaral, 1933 and A. pseudotigrinus Amaral, 1933, known from only a few individuals (Williams & Vanzolini 1980), all collected by Garbe. In 1908, in one of the first systematic explorations of the Brazilian Caatingas, Garbe was sent to the interior of state of Bahia, from where he returned in 1909 with an unprecedented amount of new material (Ihering & Ihering 1911). At Villa Nova da Rainha, presently municipality of Senhor do Bonfim, Garbe obtained the first, and up to now the only known, specimen of the earless microteiid lizard described by Amaral as a new genus and species, Anotosaura collaris. Accepted by S. Carranza: 29 Sept. 2013; published: 31 Oct. 2013 345 Besides references in checklists (Amaral 1937, 1938; Ruschi 1966; Peters & Donoso-Barros 1970), Anotosaura collaris is rarely cited in the literature. In his revision of the genus Bachia, Dixon considered Anotosaura as a valid genus and suggested a close relationship with Heterodactylus and Bachia (Dixon 1973). Later on, based on new material, he reviewed Anotosaura and described a new subspecies, Anotosaura collaris vanzolinia Dixon, 1974, from Agrestina, in the state of Pernambuco (wrongly spelled Argestina), and the new species Anotosaura brachylepis Dixon, 1974, from Serra do Cipó in the State of Minas Gerais (Dixon 1974). That was a time when subspecies were widely used in Herpetology Systematics. In view of the minor differences between Amaral’s type and the specimens from Agrestina (presence versus absence of prefrontals), Dixon suggested that hybridization was possible between the two subspecies (Dixon 1974). Contrarily, the distinctiveness of A. brachylepis from its congeners was so striking that he initially hesitated in attributing it to the new genus, but ended up keeping the new species in Anotosaura (Dixon 1974). Two years later, Vanzolini (1976) reexamined the question and kept A. brachylepis in the genus, but raised A. c. vanzolinia to specific rank, a position reinforced later when discussing their distribution (Vanzolini & Ramos 1977). Since then, these three species have been associated with Anotosaura until Pellegrino et al. (2001) removed A. brachylepis from Anotosaura and proposed the new genus Rhachisaurus to accommodate it, based on molecular data. Specimens from the Atlantic Forest of northeastern Brazil that were referred to as Anotosaura sp. n. in the literature (Rodrigues 1990; Pellegrino et al. 2001) were later reallocated into the new genus Dryadosaura (Rodrigues et al. 2005). Of the two species presently recognized in Anotosaura restricted to the semiarid Brazilian Caatingas, our knowledge rely solely on Anotosaura vanzolinia, which occurs only in mesic habitats in this domain (Rodrigues 1986, 1990, 2003; Delfim & Freire 2007; Freire et al. 2009; Oliveira 2011; Gonçalves et al. 2012). In recent analyses based on both molecular and morphological data, Anotosaura vanzolinia has been systematically recovered in close association with Dryadosaura and Colobosauroides, as part of the Ecpleopodini radiation of microteiids (Pellegrino et al. 2001; Castoe et al. 2004; Rodrigues et al. 2005; Peloso et al. 2011). However the relationships as well as ecogeographical differences among these three genera remain unclear. Recently, more than a century after its discovery by Ernest Garbe, and almost 80 years after its original description by Amaral, we obtained new specimens of Anotosaura collaris from the type locality and nearby areas. This new material allowed us to redescribe this earless elongated fossorial species, detailing its external and hemipenial morphology and osteology. Additionally, in order to test the generic allocation of A. vanzolinia and the relationship of A. collaris with Dryadosaura and Colobosauroides we obtained a phylogeny based on mitochondrial and nuclear genes. Finally, we complemented our analysis with a set of ecological, thermophysiological, ecogeographical and karyotype data. Material and methods Field work. We carried out a herpetofaunal survey at the northern portions of the relatively isolated Jacobina mountain range at Chapada Diamantina, a subsection of the Espinhaço range, in northern Bahia. We visited the localities: Senhor do Bonfim (Serra da Maravilha, Coqueiral, and Alto da Rainha mountains, and Missão do Sahy village), Campo Formoso (Serra do Cruzeiro mountain), Antônio Gonçalves (Serra da Gameleira and Serra do Sobradinho mountains), Pindobaçu (Serra da Paciência mountain) and Jaguarari (Beringela and Catuni villages). Specimens were collected using pitfall traps and through active search. The pitfall traps were installed at forest and savanna sites close to Campo Formoso and Antonio Gonçalves towns, covering altitudes between 790 and 980 m a.s.l. Each trap consisted of four 30 L buckets buried on the ground, one central and three radials, connected to each other with a 4 m long and 50 cm high plastic fence. Twenty-five traps were installed and remained opened from 9th December 2012 to 3rd January 2013, totalizing an effort equivalent to 2,500 buckets\day. Active search was performed at the other localities, mainly in Caatinga and campos rupestres (rocky meadows) habitats. Nearly every day, active search sessions were typically performed by 6–7 persons for about 2h. The sessions took place during the morning, early evening, and at the beginning of the night, and involved searching through a diversity of microhabitats at each locality. Thermal physiology and environmental data. We estimated the critical thermal maximum (CTmax) and minimum (CTmin) of six individuals from Senhor do Bonfim and Campo Formoso, right after capture. CTmax and CTmin are indices that represent the temperatures at which locomotory response is compromised yet the lizards 346 · Zootaxa 3731 (3) © 2013 Magnolia Press RODRIGUES ET AL. will recover (Cowles & Bogert 1944). They were measured by heating up or cooling down the lizards until righting response was lost (Brooks & Sassman 1965). For each experiment, a single animal was placed in a deli cup where a thermocouple (type-T) was attached so that its tip was upright inside the cup. The deli cup was placed close to a heating pad (for CTmax) or inside a mini freezer (for CTmin) and righting response was tested by gently displacing the lizard from its normal resting position, flapping the deli cup so the animal would rest in its back and touch the thermocouple. Rate of temperature change was kept around 0.5–0.75 oC/minute, during experiments.