SALAMANDRA 51(4) 283–314 30 DecemberNew 2015 TepuihylaISSN from 0036–3375 Pantepui Out of taxonomic limbo: a name for the species of Tepuihyla (Anura: Hylidae) from the Chimantá Massif, Pantepui region, northern South America Philippe J. R. Kok1,2,*, Sebastian Ratz1,*, Marco Tegelaar1, Fabien Aubret3 & D. Bruce Means4 1) Amphibian Evolution Lab, Biology Department, Vrije Universiteit Brussel, 2 Pleinlaan, 1050 Brussels, Belgium 2) Department of Vertebrates, Royal Belgian Institute of Natural Sciences, 29 rue Vautier, 1000 Brussels, Belgium 3) Station d’Ecologie Expérimentale du CNRS à Moulis, USR 2936, 09200 Moulis, France 4) Coastal Plains Institute and Land Conservancy, 1313 Milton Street, Tallahassee, Florida 32303, USA Corresponding author: Philippe J. R. Kok, e-mail: [email protected] Manuscript received: 25 May 2015 Accepted: 23 July 2015 by Stefan Lötters Abstract. We describe a new hylid species of the genus Tepuihyla from Pantepui, northeastern South America. The new species inhabits the Chimantá Massif, Bolívar state, Venezuela. The new species is likely part of a recent non-adaptive ra- diation, and was confused for more than a decade with T. edelcae, a morphologically similar species occurring on the sum- mit of Auyán-tepui, Bolívar state, Venezuela. The new species is mainly distinguished from known congeners by phyloge- netic data, as well as a medium size (37.1 mm maximum SVL in males, 38.4 mm maximum SVL in females), diameter of eye greater than distance from nostril to eye, skin on dorsum smooth in females, with scattered, fine, white-tipped spicules in males, skin on flanks smooth to faintly granular, presence of a pale labial stripe and a dark band or stripe from nostril to eye, a dorsal ground colour from pale grey to dark brown, usually suffused with small to minute dark brown or black mark- ings, no transverse bars on limbs, rear of thighs patternless, axillary membrane poorly developed, breeding males with conspicuous, usually black, nuptial pads extending beyond thenar tubercle, iris dark brown to copper with gold flecks and sometimes fine dark brown reticulation, and white limb bones. The new species inhabits open, mostly flat areas on tepui summits, between ca 1,800 and 2,600 m altitude, where it is intimately associated with carnivorous bromeliads of the genus Brocchinia. The species breeds in deep pools in marshy areas and small shallow rocky pools; its tadpole and advertisement call are described. The IUCN conservation status of the new species is considered Least Concern (LC) because population size still seems relatively large, the species occurs in a number of locations, and is apparently not declining fast enough to qualify for any of the threat categories. Differentiation in morphological, acoustic, and genetic traits of species endemic to tepui summits are briefly discussed. Finally, Tepuihyla rimarum is considered a junior synonym of T. rodriguezi. Key words. Amphibia, cryptic species, genetic divergence, Guyana, morphology, tepui, new species, T. rimarum, Ven- ezuela. Introduction remains highly debated, the so-called “integrative taxono- my”, i.e., the use of multiple lines of evidence to distinguish Species delineation is a core problem in the study of bio- between species (reviewed in Padial et al. 2010), has be- diversity. The task is complicated by a lack of consensus come increasingly popular and been demonstrated to be among biologists on the exact definition of a species (e.g., effective in diverse challenging taxonomic groups (e.g.,Vi - Wheeler & Meier 2000). Delimiting species and species’ cente et al. 2013, Diaye et al. 2014, Soldati et al. 2014). distributions as accurately as possible is nevertheless criti- Whatever the species concept being applied, it appears cally important for conservation, particularly in moun- that the perceived ease of distinguishing between species tainous areas that have been reported as highly sensitive based on external morphology strongly varies among lin- to global warming and threatened with habitat loss by up- eages. New species are also more complex to identify in ward displacement (Rull & Vegas-Vilarrúbia 2006, some geographic areas compared to others. This seems to Nogué et al. 2009). Even though the concept of species be the case when allopatric speciation occurred “recent- * These authors contributed equally to the paper © 2015 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany All articles available online at http://www.salamandra-journal.com 283 Philippe J. R. Kok et al. ly” in comparable habitats that impose strong and simi- morphology in spite of substantial genetic divergences (see lar eco logical constraints. In any case, “cryptic species” are also Kok 2013). Both situations could potentially lead to obviously more common than initially thought (Funk et taxonomic chaos. al. 2012, Gehara et al. 2014), and the danger of leaving The genus Tepuihyla was introduced by Ayarzagüena considerable parts of biodiversity unaddressed (Jörger & et al. (1993b) to accommodate six species these authors had Schrödl 2013) is blatant. previously included in the Osteocephalus rodriguezi group The tepuis of the Pantepui biogeographic region of (Ayarzagüena et al. 1993a). Tepuihyla species, at that time northern South America (Fig. 1) are among those are- all from the Venezuelan Guayana, were reported to differ as where delineating species boundaries is often particu- morphologically from Osteocephalus sensu stricto main- larly difficult for reasons that have not yet been properly ly in osteological characters (Ayarzagüena et al. 1993b). explained. The term tepui has been widely used to char- Since then, Tepuihyla has been regarded as sister to Osteo­ acterize the tabletop mountains made of Precambrian cephalus (Faivovich et al. 2005, Pyron & Wiens 2011), or sandstone that rise above the savannah and tropical for- more recently, to a clade composed of Osteocephalus and est, mainly in the Guayana region of southern Venezuela Dryaderces (Jungfer et al. 2013). The genus Tepuihyla is (states of Bolívar and Amazonas), in west-central Guyana currently restricted to Pantepui, with seven recognized (district of Cuyuni-Mazaruni), and in extreme northern species occurring from eastern and southeastern Venezue- Brazil (states of Amazonas and Roraima). Because of their la to western Guyana (Jungfer et al. 2013). Several popula- ancient origin and their physiographic, edaphic, and cli- tions of Tepuihyla have been reported as single tepui sum- matic isolation, tepui summits have for long been thought mit-endemics (see Gorzula & Señaris 1999, McDiarmid of harbouring old endemic lineages, with some even pos- & Donnelly 2005). Kok et al. (2012) demonstrated that sibly predating the separation of Africa and South America genetic divergence between geographically distant popu- (e.g., McDiarmid & Donnelly 2005). However, Kok et lations (some from different tepui summits) of Tepuihyla al. (2012), using a broad sampling of amphibian and rep- galani, T. rodriguezi, and T. talbergae was extremely low, tile taxa, demonstrated that genetic diversity among most even in a fragment of the fast-evolving protein-coding mi- tepui summit species and populations is much lower than tochondrial gene NADH dehydrogenase subunit 1 (ND1). expected, suggesting that tepuis were only sporadically im- The intent of Kok et al. (2012) was not to make any taxo- permeable barriers to gene flow within the Pantepui region nomic decision, but this was done shortly thereafter by throughout history. Kok et al. (2012) also indicated that in Jungfer et al. (2013) who considered Tepuihyla galani and spite of low genetic distances, a number of tepui summit- T. talbergae to be junior synonyms of T. rodriguezi based populations recognized as distinct species exhibit conspic- on the absence of reliable diagnostic morphological char- uous phenotypic differences (in colouration for example), acters and a very low genetic distance among populations while some tepui summit-populations exhibit identical in a fragment of the mitochondrial gene 16S rDNA (here- Figure 1. Northern part of Auyán-tepui, Bolívar state, Venezuela, showing typical tepuian sheer cliffs and lower forested slopes. Pho- tograph taken while flying by helicopter over the Devil’s Canyon (17 June 2012). Photo: PJRK. 284 New Tepuihyla from Pantepui inafter 16S). Jungfer et al. (2013) placed Osteocephalus ex­ summit of Amurí-tepui (05°08’ N, 62°07’ W, ca 2,200 m ophthalmus, O. phasmatus, and “Hyla” warreni in Tepui­ a.s.l.; Figs 2–3) where four adult males, and one subadult hyla to resolve the non-monophyly of Osteocephalus as were collected; and (3) from the summit of Chimantá- suggested by their phylogenetic tree topology, and consid- tepui (05°19’ N, 62°12’ W, ca 2,200 m a.s.l.; Figs 2–3), where ered T. phasmata to be a junior synonym of T. exophthal­ nine adult males, one subadult, one juvenile (not used in ma. Two single tepui summit-endemics, Tepuihyla rima­ the morphological analyses), and six tadpoles (5 additional rum and T. luteolabris, could not be included in Jungfer larvae were preserved in ethanol for DNA barcoding, and et al.’s (2013) molecular phylogenetic analysis due to the 1 additional, poorly preserved tadpole was not used in the lack of tissue samples. To date, the phylogenetic position of morphological analyses) were secured. These individuals these two microendemic species remains unknown. were compared in detail with the holotype of T. edelcae An additional puzzling taxon is Tepuihyla “aff. edelcae” (MHNLS 10626), as well as with 19 freshly collected adult from the Chimantá Massif. Although Tepuihyla popula- specimens (12 males, 7 females) from two geographically tions from that massif (more specifically from Amurí- close locations on the summit of Auyán-tepui, the type tepui, Abakapá-tepui, Akopán-tepui, Apakará-tepui, Chi- locality of T. edelcae (05°45’ N, 62°32’ W, between 2,200– mantá-tepui, Churí-tepui, and Murei-tepui, but also from 2,300 m a.s.l.; Fig. 2), and seven tadpoles collected on Cer- Tereke-Yurén-tepui in the Los Testigos Massif) have been ro El Sol (06°06’ N, 62°32’ W, ca 1,800 m a.s.l.; Fig.