amphibian-reptile-conservation.org rSiasK* . * Board of Directors Craig Hassapakis ARC: Editor, Publisher, and Chairman of the Board of Directors, USA Howard Clark, Jr. ARC: Treasurer & Associate Editor; Garcia and Associates, USA Adel Ibrahim Suez University, EGYPT Franco Andreone Museo Regionale di Scienze Naturali, ITALY Madhava Meegaskumbura University of Peradeniya, SRI LANKA Michael Hutchins former DirectorAVihiam Conway Chair, Department of Conservation and Science, American Zoo and Aquarium Association [1990-2005] and former Executive Director/CEO, The Wildlife Society [2005-2012], USA Ted R. Kahn Neotropical Conservation Eoundation, USA Indraneil Das Universiti Malaysia Sarawak, MALAYSIA Walter R. Erdelen former Assistant Director-General for Natural Sciences of the United Nations Educational, Scientihc and Cultural Organization (UNESCO); FRANCE Bruce Waldman Seoul National University, KOREA Peter Uetz Virginia Commonwealth University, USA Editor Craig Hassapakis Utah Valley University, USA Africa Issue Bill Branch Port Elizabeth Museum, SOUTH AFRICA Associate Editors Erik Wild Belo Horizonte, BRAZIL Howard Clark, Jr. Garcia and Associates, USA Raul Diaz La Sierra University, USA Mayra Oyervides The University of Texas-Pan American, USA Bill Branch Port Elizabeth Museum, SOUTH AFRICA Copy Editor Ruthe Smith California, USA Editorial Board C. Kenneth Dodd, Jr. University of Florida, USA Peter Uetz Virginia Commonwealth University, USA Gunther Kohler Senckenberg Forschungsinstitut und Naturmuseum, GERMANY Aurelio Ramirez-Bautista Universidad Autonoma del Estado de Hidalgo, MEXICO Larry David Wilson Centro Zamorano de Biodiversidad, HONDURAS Melissa Medina-Flores Universidad Nacional Autonoma de Honduras, HONDURAS Rafaqat Masroor Pakistan Museum of Natural History, PAKISTAN Adel Ibrahim Suez University, EGYPT Franco Andreone Museo Regionale di Scienze Naturali, ITALY Indraneil Das Universiti Malaysia Sarawak, MALAYSIA Madhava Meegaskumbura University of Peradeniya, SRI LANKA Javier Sunyer Universidad Nacional Autonoma de Nicaragua-Leon, NICARAGUA Manuel Acevedo Universidad de San Carlos de Guatemala, GUATEMALA Ulisses Caramaschi Museu Nacional, Universidade Federal do Rio de Janeiro, BRAZIL Roman Nazarov Zoological Museum, Moscow State University, RUSSIA Jelka Crnobrnja Isailovic University of Nis, SERBIA Jianping Jiang Chengdu Institute of Biology, Chinese Academy of Sciences, CHINA Bruce Waldman Seoul National University, SOUTH KOREA Daesik Park Kangwon National University, SOUTH KOREA Javier Alvar ado-Diaz Universidad Michoacana de San Nicolas de Hidalgo, MEXICO Rdmulo Romeu da Nobrega Alves Universidade Estadual da Parafba, BRAZIL Antonio Salas Environment and Sustainable Development, PERU Cesar Aguilar Universidad de San Marcos, PERU Bill Branch Port Elizabeth Museum, SOUTH AFRICA Santiago J. Sanchez-Pacheco University of Toronto, CANADA Copyright: © 2014 Lynch et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and educa- tion purposes only provided the original author and source are credited. The of- ficial publication credit source: Amphibian & Reptile Conservation at: amphibian- reptile-conservation. org Amphibian & Reptiie Conservation 8(1) [Special Sec]: 1-7. Rediscovery of Andinophryne o/a//a/ Hoogmoed, 1985 (Anura, Bufonidae), an enigmatic and endangered Andean toad ^Ryan L. Lynch, ^Sebastian Kohn, ^Fernando Ayala- Varela, ^Paul S. Hamilton, and ^Santiago R. Ron ^The Biodiversity Group, Tucson, Arizona, USA ^Rw Manduriacu Cooperative, Quito, ECUADOR ^Museo de Zoologia, Escuela de Biologia, Pontificia Universidad Catolica del Ecuador, Quito, ECUADOR Abstract . — ^We report the rediscovery of Andinophryne olaiiai, an endangered species only known from a single specimen, collected in 1970. At the type locality, Tandayapa, Pichincha Province, numerous follow-up surveys after 1970 failed to record the species suggesting that the population is extinct. The rediscovery of A. oiailai took place in 2012 at Ri'o Manduriacu, Imbabura Province, Ecuador. Two surveys suggest that a healthy population of A. olaiiai survives at the site, with observations of froglets, juveniles, and adults across numerous stream systems. However, the extent of known occupancy of the population is small (<1 km^). Further data are presented to update knowledge of the distribution, ontogeny, morphology, and conservation status of the species. The population at Ri'o Manduriacu is surrounded by logging, mining, and hydroelectric developments that could compromise its future survival. There is an urgent need to establish a monitoring program and to protect its remaining population and habitat in the region. Key words. Andinophryne olaiiai, rediscovery, Tandayapa Andean toad, Andinosapo de Olalla, Bufonidae, Endan- gered species, Ecuador Citation: Lynch RL, Kohn S, Ayala-Varela F, Hamilton PS, Ron SR. 2014. Rediscovery of Andinophryne olaiiai Hoogmoed, 1985 (Anura, Bufonidae), an enigmatic and endangered Andean toad. Amphibian & Reptile Conservation 8(1) [Special Section]: 1-7 (e75). Introduction The small and understudied toad genus Andinophryne (Bufonidae) is restricted to the western slopes of the Andes in Colombia and Ecuador. Three species of An- dinophryne have been described: Andinophryne atelo- poides (Lynch and Ruiz-Carranza 1981), Andinophryne colomai (Hoogmoed 1985), and Andinophryne olaiiai (Hoogmoed 1985). Until recently, all three species were only known from five or fewer adult individuals at the type localities: A. atelopoides (Cauca Department, Co- lombia, 1980), A. colomai (Carchi Province, Ecuador, 1984), and A. olaiiai (Pichincha Province, Ecuador, 1970). The paucity of information available on Andinophryne has led to many questions about the taxonomic and con- servation status of all three species. Andinophryne at- elopoides, the only species endemic to Colombia and only known from two specimens, was originally placed in the genus Bufo by Lynch and Ruiz-Carranza (1981). Four years later, following the discovery of two similar bufonid species (A. colomai and A. olaiiai) in northern Ecuador, and the reexamination of information presented on B. atelopoides by Lynch and Ruiz-Carranza (1981), Hoogmoed (1985) created the genus Andinophryne (Bu- fonidae), and placed all three species within the new ge- nus. Despite numerous attempts by trained scientists and over 150 search hours, subsequent visits to the type lo- calities of A. colomai and A. olaiiai in Ecuador have failed to record either species (Coloma et al. 2004; Ron and Frenkel 2013). Then, in 2005, Murillo et al. (2005) reported a 160 km range extension for A. olaiiai in Rio Nambi, Department of Narino, Colombia. This observa- tion marked the first record of any Andinophryne species in more than two decades. However, as part of our recent work with Andinophryne, a member of our team recently examined a specimen from Rio Nambi and determined that it was not Andinophryne olaiiai but a different spe- cies (Santiago Ron, unpubl. data). This identification has been confirmed by additional fieldwork and specimens collected at Rio Nambi by Paul David Gutierrez-Carde- nas (pers. comm.). Therefore, A. olaiiai is the rarest of all Andinophryne species, with the only known record being the original type specimen from Tandayapa, Pichincha Province, Ecuador in 1970. Forty-three years after the original description of Andinophryne olaiiai, we report the rediscovery of a population of A. olaiiai from Rio Manduriacu (herein Manduriacu), Imbabura Province, Ecuador. We also pro- Correspondence. ^ryan@ biodiversitygroup.org (corresponding author); ^sebastiankohn® hotmail.com; ^fpayala2000@ gmail. com; Hamilton @ biodiversity group, org; ^Santiago, r. ron@ gmail. com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (1) February 2014 | Volume 8 | Number 1 | e75 Lynch et al. -a 1,0 'M.O -?9,0 -7?,0 -B1.0 -BO.a -70.0 -78.0 -77.0 Fig. 1. Known records of Andinophryne olallai in northwest Ecuador; Tandayapa: Type Locality, Manduriacu: New Local- ity. vide the first information on the species’ natural history, geographic range, ontogeny, and conservation status, and present the first published color photos of live individu- als across different age classes. Materials and Methods Our surveys took place in the premontane tropical for- est and cloud forests of Manduriacu in NW Ecuador (1,100-1,400 m), 40 km N of the type locality of A. olal- lai and near the south border of the Cotacahi-Cayapas Ecological Reserve (Eig. 1). Surveys were conducted on 18 November 2012 (original rediscovery) and 13-15 May 2013 using Visual Encounter Surveys (VES) along stream transects between 19:00 and 01:00 h. The objectives of the surveys were: (1) determine the population status of A. olallai', (2) determine the extent of its occupancy in Manduriacu; and (3) obtain informa- tion about the behavior and natural history of the species. Surveys were carried out along small rocky streams with overhanging herbaceous vegetation (Eig. 2). A total of three nights were spent surveying four stream systems neighboring the site of initial discovery (approximately 100 m between streams; < 1 km^ area total). Information collected in the field included: air temper- ature (°C), relative humidity (%), time of encounter (24 hr), perch height (cm), snout- vent length (SVL, mm), sex (when possible), and age class (froglet, juvenile, adult). Froglets (i.e., recently metamorphosed individuals) were defined as individuals with heavily patterned dorsum, lack of pronounced parotoid glands, and SVL between 10-20 mm. Juveniles were defined as individuals with faint dorsal patterning, more pronounced parotoid glands, and SVL between 20-30 mm. Adults were defined as Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (2) individuals with no dorsal patterning, very pronounced parotoid glands, presence of large cream- tan colored tu- bercles on the flanks, and SVL above 30 mm. Perch height for each individual was measured using a marked meter stick and SVL measurements were taken using dial calipers. Climate information was recorded us- ing a handheld Kestrel 3500 Weather Meter. Individual toads were only handled when necessary, and always with use of latex gloves to prevent transferring pathogens such as amphibian chytrid fungus (Batrachochytrium dendrobatidis). Results and Discussion During the first survey of Manduriacu on 1 8 November 2012, two adult A. olallai were encountered perched on leaves overhanging a small running stream. Elevation of the observation site was 1,253 m, and perch heights of the individuals were 1.5 m and 2.0 m above ground. Both individuals appeared to be females, based on size, with SVL of 57 and 58 mm, however sex could not be determined with complete certainty in the field because no secondary sexual characteristics are evident in live Table 1. Reptiles and amphibians associated wiihAndinophyne olallai at Manduriacu, Imbabura Province, Ecuador and their current (August 2013) lUCN and EaunaWebEcuador Red List status (NE - Not Evaluated, DD - Data Deficient, LC - Least Concern, NT - Near Threatened, V - Vulnerable, EN - En- dangered). lUCN Red List available at: http://www.iucnredlist. org/; EaunaWebEcuador Red List available at: http://zoologia. puce.edu.ee/vertebrados/anfibios/EspeciesEstadoConserva- cion.aspx. Species Fauna Web Ecuador Red List lUCN Red List Caecilia guntheri DD DD Centrolene peristictum NT V Epipedobates darwinwallacei EN NE Espadarana prosoblepon LC LC Hyloscirtus alytolylax NT NT Pristimantis achatinus LC LC Pristimantis calcarulatus LC V Pristimantis labiosus NT LC Pristimantis luteolateralis NT NT Pristimantis muricatus V V Pristimantis scolodiscus DD EN Rulyrana orejuela DD DD A lopoglossus festae NT NE Anolis aequatorialis NT NE Anolis gemmosus LC LC Basiliscus galeritus NE NE Bothriechis schlegelii NT NE Cercosaura vertebralis DD NE Diaphorolepis wagneri NT NE Lepidoblepharis conolepis EN NE February 2014 | Volume 8 | Number 1 | e75 Rediscovery of Andinophryne olallai animals. This initial observation yielded two signifieant findings: the first evidence of an A. olallai population in 43 years and the second known locality for the species extending its known range 40 km N from its type locality. During the course of the survey in May 2013 a total of 18 A. olallai were observed across four stream sys- tems. Average nightly environmental conditions during the three nights of surveys in May were: air temperature 18.3 °C and relative humidity 92.8%. We recorded the presence of adults, juveniles, and froglets, indicating on- going population recruitment (Fig. 3). Eleven of the nine- teen individuals encountered were adults, and although their sex could not be determined, eggs were visible in the abdomen of two gravid females. The sex of one pre- served adult male (QCAZ-A 55561) was confirmed by internal gonad examination. The confirmed adult females had SVL of 57 mm and 60 mm, considerably larger than the SVL reported by Hoogmoed (1985) for the holotype ($, 39.6 mm). The single confirmed male had a SVL of 36.5 mm. Mean SVL for adults with unknown sex was 47.1 mm (n = 8). All individuals encountered were perched on branch- es or leaves overhanging or bordering the streams. Mean perch height was 1.4 m (n = 18), with adults generally perching higher than younger individuals. Maximum ob- served perch height was four meters. Although no official surveys were conducted during the day, no individuals were observed along streams during random daytime walks. Although further behavioral work needs to be con- ducted, this observation suggests that A. olallai may be actively foraging during the day in the forests surround- ing streams. At night, they remain immobile perched on leaves overhanging the streams. Lack of movement may protect them from predators. Ontogeny and Morphology All information on A. olallai reported by Hoogmoed (1985) was based on two adult specimens. Our obser- vations of froglets and juveniles mark the first reported information on the species’ pre-adult morphology and ontogeny. Ontogenetic change in color pattern is con- siderable (Fig. 3), and is one of the few reported cases of such an extreme change in bufonids in Ecuador (see Hoffman and Blouin 2000). We observed a total of two froglets (mean SVL 13.1 mm) and five juveniles (mean SVL 26.6 mm). Eroglets have a copper, gold, and white dorsum with a mottling pattern reminiscent of some species of Atelopus (Eig. 3: A, B). This contrasts with the patternless brown dorsum of the adults. The venter of froglets have a series of white undulating lines that extend the length of the body (Eig. 4). The iris in frog- lets and juveniles is more vibrantly red than in adults, which have a yellow copper-colored iris that is darker medially near the horizontally oval pupil. Froglets also differ from adults in lacking tubercles and parotoid Fig. 2. Andinophryne olallai habitat from Rio Manduriacu, Imbabura Province, Ecuador. All individuals encountered were found perched on branches or leaves along streams similar to the stream pictured here. glands. Juveniles retained some of the mottling pattern seen in froglets (primarily posteriorly on the hind legs) and lacked the conspicuous tubercles on the flank (Fig. 3: C, D). However, they begin to show adult traits like pronounced parotoid glands, tan-brown coloration, and strongly webbed fingers. Morphological characteristics of the adults match those of the holotype of A. olallai (comparisons based on photographs of the holotype, available at Link/URL: Amphibiaweb Ecuador, and Hoogmoed 1985). The holo- type and the observed specimens of the population from Manduriacu differ from the other species of the genus in having more developed parotoid glands, larger body size, strongly webbed fingers, and conspicuous yellow- ish glands scattered on the flanks and arranged in rows or in irregular patterns (Fig. 3: E, F) (Hoogmoed 1985). The dorsal texture varies from smooth to mildly tuber- culate. One individual had abundant tubercles on the anterior half of the dorsum and large scattered tubercles on the posterior half. The description of coloration given by Hoogmoed (1985) was of an animal in preservative; however, the color description falls within the variation observed in life at Manduriacu. The only notable differ- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (3) February 2014 | Volume 8 | Number 1 | e75 Lynch et al. Fig. 3. Ontogenetic transformation of color and pattern in Andinophryne olallai from Rio Manduriacu, Imbabura Province, Ecuador. (A) Froglet (11 mm SVL; in situ), (B) Froglet (15.1 mm SVL; in situ), (C) Juvenile (26.3 mm SVL; in situ), (D) Juvenile (28.1 mm SVL; in situ), (E) Adult (44.6 mm SVL; ex situ), (F) Adult (53.3 mm SVL; in situ). Note the progressive ontogenetic change in dor- sal patterning from heavily mottled to no pattern; lack of parotoid glands and tubercles along the flank to presence of conspicuous parotoid glands and tubercles along the flank; a darkening of color from copper, tan, and white to dark brown; and iris color change from vibrant crimson to copper-orange. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (4) February 2014 | Volume 8 | Number 1 | e75 Rediscovery of Andinophryne olallai Fig. 4. Ventral pattern of froglets of Andinophryne olallai. Manduriacu, Imbabura Province, Ecuador. ence is that dorsal and flank coloration is not uniform in all individuals; the head and dorsum were darker brown than the light brown-tan flanks in most live animals ob- served at Manduriacu. Sympatric Species During our herpetofaunal surveys of Manduriacu we re- corded observations of all amphibian and reptile species occurring at the site (Table 1). Most of these species are mid-elevation (1,000-2,500 m) inhabitants of premon- tane and cloud forests of the eastern Andes. A number of the species (i.e., Lepidoblepharis conolepis, Pristimantis scolodiscus) are either nationally or internationally listed as Endangered, and two of the species are categorized as Data Deficient or have not yet been assessed (i.e., Di- aphorolepis wagneri, Epipedobates darwinwallacei) and very little is known about their biology or conservation status due to few available records or localities. Conservation and Threats Andinophryne olallai is currently classified as Data Deficient by the lUCN Red List (Coloma et al. 2010). However, more recent assessments considers A. olallai as Endangered based on its restricted range, the appar- ent extirpation of the species from the type locality and Fig. 5. A recently deforested plot of land that is less than one km from the population of Andinophryne olallai in Manduriacu, Im- babura Province, Ecuador. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (5) February 2014 | Volume 8 | Number 1 | e75 Lynch et al. extensive habitat degradation (Coloma et al. 2011-2012; Ron and Frenkel 2013). The rarity of known distribution and a very small population size likely warrants an lUCN Red List status of Endangered. Although we found evidence of a seemingly healthy population of A. olallai at Manduriacu, with the presence of all size classes across numerous stream systems, the extent of known occupancy remains extremely small (< 1 km^). At present, pristine habitat still exists at Manduria- cu, however, the surrounding forest is rapidly disappear- ing due to a variety of anthropogenic factors (i.e., inten- sive logging, mining, and hydroelectric development). These activities are expanding quickly and resulting in extensive habitat fragmentation and loss (Fig. 5). The apparent extirpation of A. olallai from the type locality, a site where forest has been lost and fragmented, sug- gests that the species is sensitive to anthropogenic habi- tat change. Urgent conservation measures and population monitoring are needed in order to ensure the survival of A. olallai in nature. It is our hope that the rediscovery of A. olallai will result in immediate support for greater protection of the forests in and around Manduriacu, and provide assistance in creating biological corridors be- tween the neighboring reserves of Los Cedros and Co- tacachi-Cayapas. Acknowledgments. — We thank Juan and Monica Kohn for purchasing and protecting the land at Rio Manduriacu. Programa Socio Bosque provides support for conservation of the forests of Manduriacu. Pontifi- cia Universidad Catolica del Ecuador provided logistical support for our fieldwork. The Biodiversity Group pro- vided support for RLL research, and Belisario Cepeda Quilindo gave access to their 2005 publication on A. olallai. Paul Gutierrez-Cardenas provided access to specimens and photographs of A. colomai. This work was conducted under Ministerio del Ambiente permit # 005-12- IC-EAU-DNB/MA. Literature Cited Coloma LA, Guayasamin JM, Menendez-Guerrero P (editors). 2011-2012. Amphibian Red List of Ecua- dor, AnfibiosWebEcuador. Otonga Foundation, Quito, Ecuador. Coloma LA, Ron SR, Cisneros-Heredia DF, Yanez-Mu- noz MH, Gutierrez-Cardenas PD, Angulo A. 2004. Andinophryne olallai. In: lUCN 2011. lUCN Red List of Threatened Species. Version 2011.2. Available: http://www.iucnredlist.org. [Accessed: 27 December 2013]. Hoffman EA, Blouin MS. 2000. A review of colour and pattern polymorphisms in Anurans. Biological Jour- nal of the Linnean Society 70: 633-665. Hoogmoed MS. 1985. A new genus of toads (Amphibia: Anura, Bufonidae) from the Pacific slopes of the An- des in northern Ecuador and southern Colombia, with the description of two new species. Zoologische Med- edelingen 59: 251-274. Link/URL: Amphibiaweb Ecuador. Available: http://zoo- logia.puce.edu.ec/vertebrados/anfibios [Accessed: 25 August 2013]. Lynch JD, Ruiz-Carranza PM. 1981. A new species of toad (Anura: Bufonidae) from the Cordillera Occiden- tal in southern Colombia. Lozania 33: 1-7. Murillo Pacheco J, Cepeda Quilindo B, Elorez Pai C. 2005. Andinophryne olallai (Tandayapa Andes toad). Geographic distribution. Herpetological Review 36: 331. Ron SR, Frenkel C. 2013. Andinophryne olallai. In: Ron SR, Guayasamin JM, Yanez-Munoz MH, Merino- Viteri A (editors). AmphibiaWebEcuador. Version 2013.1. Available: http://zoologia.puce.edu.ee/ver- tebrados/anfibios/FichaEspecie.aspx?ld=l 140 [Ac- cessed: 26 July 2013]. Received: 09 December 2013 Accepted: 24 January 2014 Published: 03 February 2014 Ryan L. Lynch is the lead biologist and photographer for Ecuadorian programs for The Biodiversity Group in Quito, Ecuador. He received his M.S. in wildlife ecology and conservation from the University of Elorida where he used occupancy modeling to determine the status of anurans across the Elorida everglades landscape. Ryan’s current research interests focus on the ecology, distribution, and conservation of rare, threatened, and new species of reptiles and amphibians in Ecuador. Sebastian Kohn is the administrator for the Antisanilla-Sunfohuaico Reserve run by the Jocotoco Eounda- tion in Ecuador. He received his B.A. in biology and environmental studies at Whitman College in Wash- ington State, USA. He currently directs the Rio Manduriaco Cooperative in Imbabura, Ecuador, as well as the llitio Wildlife Rescue Center and Hacienda llitio in Cotopaxi, Ecuador. Sebastian is a founding member of the Andean Condor Conservation Group of Ecuador (Grupo Nacional de Trabajo del Condor Andino) and has been working with, and researching, both wild and captive condors for ten years. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (6) February 2014 | Volume 8 | Number 1 | e75 Rediscovery of Andinophryne olallai Fernando Ay ala- Varela is the director of the herpetology collection at the Pontificia Universidad Catolica del Ecuador in Quito. He received his diploma at the Pontificia Universidad Catolica del Ecuador, Quito in 2004. He has been interested in herpetology since childhood and has dedicated a lot of time studying the lizards of Ecuador, specifically the taxonomy and ecology of Anolis species. His current research interests include reproductive biology and ecology of lizards and snakes in Ecuador. Paul S. Hamilton is the founder and executive director of The Biodiversity Group in Tucson, Arizona, USA. He holds a master’s degree in biology from the University of California, Riverside, and a Ph.D. in biology from Arizona State University, and has conducted field studies in evolutionary, behavioral and conservation ecology both in the tropics and the desert southwest. In addition to his research interests in ecology and conservation of overlooked species such as amphibians, reptiles, and invertebrates, he is also , __ a well published scientific and artistic photographer. t Santiago R. Ron is the curator of amphibians and professor at the Pontificia Universidad Catolica del Ecuador in Quito. His research focuses on the evolution and diversity of neotropical amphibians with emphasis on Ecuador. Areas covered include evolution of animal communication, sexual selection, sys- tematics and taxonomy. In the area of conservation biology Santiago is interested in the study of amphibian extinctions in the Andes. Santiago also oversees the ex situ amphibian conservation project Balsa de los Sapos at the Pontificia Universidad Catolica del Ecuador in Quito. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (7) February 2014 | Volume 8 | Number 1 | e75 Copyright: © 2014 Ayala- Varela et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education purposes only provided the original author and source are credited. The official publication credit source: Amphibian & Reptile Conservation at: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation [Special Section] 8(1) : 8-24. A new Andean anole species of the Dactyloa clade (Squamata: Iguanidae) from western Ecuador Ternando P. Ayala- Varela, ^Diana Troya-Rodn'guez, ^Xiomara Talero-Rodn'guez and ''Omar Torres-Carvajal ^■^■^■^Escuela de Ciencias Biologicas, Pontificia Universidad Catolica del ECUADOR, Avenida 12 de Octubre 1076 y Roca, Apartado 17-01-2184, Quito, ECUADOR Abstract . — ^We describe a new species of Anolis from the western slopes of the Andes of Ecuador, province of Bolivar. It is referred to (1) the aequatoriaiis series based on its moderate size and narrow toe lamellae, and (2) the eu/aemus sub-group based on having a typical Ano/is digit, in which the distal lamellae of phalanx III distinctly overlap the proximal subdigital scales of phalanx II. The new species is most similar morphologically to A. otongae and A. gemmosus, both from similar elevations on the western Andean slopes of Ecuador, but differs from these species in morphology and color patterns. We present a phylogeny based on DNA sequence data as additional evidence supporting delimitation of the new species. The new species and A. gemmosus are sister taxa within the “western Dactyloa clade.” Key words. Clade Dactyloa, DNA, lizard, phylogeny, South America, systematics Citation: Ayala-Varela FP, Troya-Rodriguez D, Talero-Rodriguez X, Torres-Carvajal O. 2014. A new Andean anole species of the Dactyloa clade (Squa- mata: Iguanidae) from western Ecuador. Amphibian & Reptile Conservation 8(1) [Special Section]: 8-24 (e76). Introduction With nearly 490 described species, anole lizards {Anolis) have proliferated impressively in the Americas (Nich- olson 2002; Poe 2004), possibly prompted by ecologi- cal opportunity (Losos 2009). Although the diversity of these lizards has been extensively studied in the West Indies (Losos 2009), the same is not true for the main- land radiation, which is probably greater than previously thought. For example, all but two — Anolis ruibali Navar- ro & Garrido 2004 and 4. sierramaestrae Holahova et al. 2012 — of the 31 new species of Anolis described during the last decade (2003-2013) occur in mainland Central and South America (Uetz and Hosek 2014). Improving knowledge concerning the diversity of mainland anoles is crucial to understanding the nature of this radiation. Anole lizards represent the most species-rich clade traditionally recognized as a genus in Ecuador, with 37 species reported to date (Torres-Carvajal et al. 2014). The diversity of anole lizards in Ecuador is remarkably greater west of the Andes, with more than twice the num- ber of species that occur east of the Andes (25 and 12 species, respectively). Of these, five species have been described during the last six years from both sides of the Andes as a result of both careful examination of exist- ing collections and recent collecting in poorly explored areas. Here we contribute to that growing body of taxo- nomic knowledge with the description of a new species of Anolis endemic to the western slopes of the Andes in Ecuador. We present molecular evidence supporting rec- ognition of the new species by performing phylogenetic analyses of mitochondrial DNA sequence data. Materials and Methods Morphological data All known specimens of the new species described in this paper are included in the type series, and were de- posited in the Museo de Zoologia, Pontificia Universi- dad Catolica del Ecuador, Quito (QCAZ). Specimens of other species of Anolis examined in this study are listed in Appendix 1. We follow previously proposed terminol- ogy (Williams et al. 1995) for measurements and squa- mation. Nine morphological measurements were taken with digital calipers and recorded to the nearest 0.1 ncnn: head length, head width, head height, forelimb length, hindlimb length, snout- vent length, jaw length, axilla- groin length, and snout length. In addition, tail length measurements were taken with a ruler and recorded to the nearest millimeter; regenerated or broken tails were not measured. Sex was determined by noting the pres- ence of hemipenes, which were everted in all male speci- mens during preparation. Statistical analyses Given that the new species is very similar in morphol- ogy to Anolis gemmosus and A. otongae we performed Correspondence. fpayala2000@yahoo.com (Corresponding author); ^dianatrl 7@gmail.com; ^xiomy.talero@gmail.com; ^omartorcar@gmail. com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (8) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. Fig. 1. Head of the holotype (QCAZ 3449) of Anolis poei sp. nov. in dorsal (top), ventral (middle), and lateral (bottom) views [Scale bar =10 mm]. Photographs by F Ayala-Varela. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (9) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador a Principal Component Analysis (PCA) to determinate whether separation in morphological space between those species was statistically significant. Principal com- ponents (PCs) were extracted from a covariance matrix of the raw and rescaled data. The new species is most similar to A. gemmosus, for which we also used Mests to evaluate quantitative differences between both species. One of the assumptions of the t-test for two samples is that the variances of both samples are equal: therefore, F-tests also were performed for each character to test for equality of variances. If the variances were not the same (i.e., P < 0.05), an unequal variance t-statistic was used. Statistical analyses were performed in SPSS Statistics 17 (SPSS Inc. 2008). The distribution map was prepared in ArcMap 9.3 (ESRI, Inc.); WGS84 is the datum for all coordinates presented below. DNA sequence data Total genomic DNA was digested and extracted from liv- er or muscle tissue using a guanidinium isothiocyanate extraction protocol. Tissue samples were first mixed with Proteinase K and a lysis buffer and digested overnight prior to extraction. DNA samples were quantified using a Nanodrop® ND-1000 (NanoDrop Technologies, Inc), re-suspended and diluted to 25 ng/ul in ddH20 prior to amplification. Using primers and amplification protocols from the literature (Folmer et al. 1994; Kumazawa and Nishida 1993; Macey et al. 1997; Schulte and Cartwright 2009) we obtained 2807 nucleotides (nt) representing the nucle- ar gene recombination-activating gene 1 (RAGl, 8 lint), as well as the mitochondrial genes Cytochrome c oxi- dase I (COl, 655nt) and a continuous fragment includ- ing the NADH dehydrogenase subunit 2 (ND2, 1038 nt), tRNATrp, tRNAAla, tRNAAsn, tRNACys (282nt), and the origin of the light-strand replication (Ol, 29nt). The new sequence data were obtained for three individuals of the new species described herein, two of A. gemmosus, and two of otongae. In addition we used sequence data generated by Castaneda and de Queiroz (2011) for 20 in- dividuals of the clade Dactyloa, as well as one sequence of A. occultus, which was used as the outgroup in the phylogenetic analysis. Gene regions of taxa included in phylogenetic analyses along with their GenBank acces- sion numbers are shown in Table 1. Phylogenetic analyses Editing, assembly, and alignment of sequences were performed with Geneious ProTM 5.3 (Biomatters Ltd. 2010). Genes were combined into a single dataset with eleven partitions, three per protein coding gene corre- sponding to each codon position, one with all tRNAs, and one with the Ol. The best partition strategy along with the corresponding models of evolution were obtained in Par- titionFinder 1.1.1 (Lanfear et al. 2012) under the Bayes- ian information criterion. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (10) Phylogenetic relationships were assessed under a Bayesian approach in MrBayes 3.2.0 (Ronquist and Huelsenbeck 2003). Four independent analyses were performed to reduce the chance of converging on a lo- cal optimum. Each analysis consisted of 20 million generations and four Markov chains with default heat- ing values. Trees were sampled every 1,000 generations resulting in 20,000 saved trees per analysis. Stationarity was confirmed by plotting the -In L per generation in the program Tracer 1.6 (Rambaut et al. 2013). Additionally, the standard deviation of the partition frequencies and the potential scale reduction factor (Gelman and Rubin 1992) were used as convergence diagnostics for the posterior probabilities of bipartitions and branch lengths, respec- tively. Adequacy of mixing was assessed by examining the acceptance rates for the parameters in MrBayes and the effective sample sizes (ESS) in Tracer. After analyz- ing convergence and mixing, 2,000 trees were discarded as “bum-in” from each mn. We then confirmed that the four analyses reached stationarity at a similar likelihood score and that the topologies were similar, and used the resultant 72,000 trees to calculate posterior probabilities (PP) for each bipartition on a 50% majority mle consen- sus tree. Systematics The taxonomic conclusions of this study are based on the observation of morphological features and color patterns, as well as inferred phylogenetic relationships. We con- sider this information as species delimitation criteria fol- lowing the general species concept of de Queiroz (1998, 2007). Anolis poei sp. nov. urn:lsid:zoobank.org:act:712687F6-CF33-4969-815D-E4600D01FB4C Proposed standard English name: Telimbela anoles Proposed standard Spanish name: Anolis de Telimbela Holotype QCAZ 3449 (Figs. 1, 2), adult male, Ecuador, Provincia Bolivar, Telimbela, 01.65789°S, 79.15334°W, WGS84 1,354 m, 10 June 2011, collected by Fernando Ayala- Va- rela, Jorge H. Valencia, Diana Troya-Rodriguez, Francy Mora, and Estefama Boada. Paratypes (1 5) ECUADOR: Provincia Bolivar: QCAZ 3444-3448, 3451-3455, 4359, same data as holotype, ex- cept 0.1658440°S, 79.157150°W, 1,310 m; QCAZ 6781-6783 Telimbela, Escuela Elisa Marino de Carva- jal, 0.1665857°S, 79.172096°W, 27 July 2004, collected by Edwin Carrillo-Ponce and Morley Read; QCAZ 9219 Guaranda, Salinas, Recinto Tres Cmces, 01.431380°S, May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. Fig. 2. Anolis poei sp. nov. Holotype, adult male (SVL = 59.67 mm, QCAZ 3449, A), eye close-up (SVL = 60.31 mm, QCAZ 3448, B), subadult male (SVL =52.12 mm, QCAZ 3455, C, D), adult male (SVL = 59.02 mm, QCAZ 3451, E, F), adult male (SVL = 60.31 mm, QCAZ 3448, G, H). Photographs by L. Bustamante (A), and O. Torres-Carvajal (B, C, D, E, F, G, H). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (11) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Fig. 3. Male dewlap of Anolis poei sp. nov. (holotype, QCAZ 3449, A; paratype, QCAZ 3455, B); A. otongae (QCAZ 4661, C; QCAZ 11791, D); and^. gemmosus (QCAZ 4385, E; QCAZ 4352, F; QCAZ 9452, G; QCAZ 11850, H). Photographs byL. Busta- mante (A), O. Torres-Carvajal (B, C, D, E, F, H), and S. R. Ron (G). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (12) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. Table 1. Species of Anolis sequenced in this study, voucher specimen numbers, collecting localities, and GenBank acces- sion numbers. Species Voucher Locality GenBank Number A. gemmosus QCAZ 4385 Ecuador, Car- chi, Rio San Pablo near Chical ND2: KJ854205 COI: KJ854219 RAGE KJ854212 QCAZ 4406 Ecuador, Car- chi, Maldo- nado, Teldibi Ecological Trail ND2: KJ854206 COI: KJ854220 RAGE KJ854213 A. otongae QCAZ 11790 Ecuador, Pichincha, Biological Re- serve Otonga ND2: KJ854207 RAGE KJ854214 COI: KJ854221 QCAZ 11791 Ecuador, Pichincha, Biological Re- serve Otonga ND2: KJ854208 COI: KJ854222 RAGE KJ854215 A. poei QCAZ 3444 Ecuador, BoKvar, Telimbela ND2: KJ854209 COI: KJ854223 RAGE KJ854216 QCAZ 3445 Ecuador, BoKvar, Telimbela ND2: KJ854210 COI: KJ854224 QCAZ 3448 Ecuador, BoKvar, Telimbela ND2: KJ854211 COI: KJ854225 RAGE KJ854217 QCAZ 4359 Ecuador, BoKvar, Telimbela RAGE KJ854218 79.097970°W, 2,628 m, 28 May 2009, collected by Eli- cio E. Tapia, Silvia Aldas- Alarcon, and Eduardo Toral- Contreras. Diagnosis We assign Anolis poei both to the aequatorialis series, based on moderate to large body size, narrow toe lamel- lae, small head scales, smooth ventral scales, and uni- form dorsal scalation; and to the eulaemus-subgroup, based on a typical Anolis digit, in which the distal la- mellae of phalanx III distinctly overlap the first proximal subdigital scale of phalanx II (Williams 1976; Williams and Duellman 1984; Castaneda and de Queiroz 2013). At present ten species are recognized within the eulae- mus-subgroup: Anolis anoriensis Velasco et al. 2010, A. antioquiae Williams 1985, A. eulaemus Boulenger 1908, A. fitchi Williams & Duellman 1984, A. gemmo- sus O’Shaughnessy 1875, yf. maculigula Williams, 1984, A. megalopithecus Rueda-Almonacid 1989, A. otongae Ayala- Varela & Velasco 2010, A. podocarpus Ayala- Varela & Torres-Carvajal 2010, and A. ventrimaculatus Boulenger 1911. Anolis poei differs from them mostly in dewlap features. The dewlap in males of A. poei has a yellowish-green (or both yellow and green) gorgetal re- gion, light blue border, and white sternal and marginal regions (Eig. 3). It has a blackish gorgetal region, and creamy white sternal region with light brown scales in A. anoriensis; brown gorgetal region, and pale brown mar- ginal region in A. eulaemus; bluish-gray gorgetal region, orange stripes, pale bluish-rose anterior third, and white sternal region becoming pale blue toward the belly in A. maculigula; sepia background, with red narrow and ir- regular stripes on each side of rows in A. megalophitecus; white, pale yellow, or greenish-yellow gorgetal region, with white or pale-yellow marginal and sternal regions in A. otongae (Eig. 3); dull yellowish-green or light blue gorgetal region, shading to dull cream, greenish yellow or orange on the marginal region, with white or bluish green gorgetal rows with or without brown spots and with yellowish white, yellow or orange sternal region in A. gemmosus (Eig. 3). The dewlap in males of A. poei has wide rows of 3-7 scales separated by naked skin; the width of these rows is one scale in A. fitchi, 2-5 granular, minute scales in A. podocarpus, 1-2 scales in A. ventri- maculatus, 3-6 scales in A. otongae, and 2-3 scales in A. gemmosus. In addition, females of the new species lack a dewlap, which is present in females of A. anoriensis, A. antioquiae, A. eulaemus, A. fitchi, and A. podocarpus. Anolis poei is most similar morphologically to A. otongae and A. gemmosus (Eig. 4). Erom the former species (character states in parenthesis) A. poei differs in having small dorsal chevrons in females (large dorsal chevrons extending onto flanks), pale yellowish-brown iris (iris dark blue), interparietal scale (if present) sur- rounded by small swollen scales (interparietal scale surrounded by relatively enlarged flat scales), enlarged postanal scales separated by 3-5 scales (postanal scales separated by 1-2 scales), and in lacking a dark stripe on side of head (dark coppery -brown stripe present). Ad- ditionally, PCA analyses suggested that specimens of A. poei have shorter jaws, as well as lower and narrower heads than A. otongae (Table 2, Fig. 5), with PCI (39% of total variation) represented mainly by head height, head width, and jaw length. The new species can be distinguished from A. gem- mosus (Table 3) in having fewer scales between sec- ond canthals (11-14, mean = 12.08 and 12-21, mean = 15.25, respectively; t = 5.31, P<0.005); fewer scales between supraorbital semicircles (1-3, mean =1.62 and 1-5, mean = 3.13, respectively; t = 4.46, P<0.005); more lamellae under phalanges III-IV of fourth toe (18-19, mean = 18.92 and 14-18, mean = 17.33, respectively; t = -7.86, P<0.005); a narrower head (head width = 7.84-8.84, mean = 8.29 and 6.97-17.41, mean = 10.82, respectively; t = -7.03, P<0.005); lower head (head height = 6.54-7.48, mean = 6.92 and 5.42-15.96, mean = 9.51, respectively; t = -6.96, P<0.005); and shorter snout (snout length = 6.75-7.30, mean = 6.92 and 5.79-14.95, mean = 10.58, respectively; t = -11.74, P<0.005). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (13) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Fig. 4. Part 1. Five species of Anolis from western Ecuador. A. aequatorialis: male (QCAZ 11861, A) and female (QCAZ 3443, B); A. binotatus: male (QCAZ 3434, C, D); A.fasciatus: male (QCAZ 3450, E, F); A. otongae: male (QCAZ 11790, G) and female (QCAZ 11791, H). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (14) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. Fig. 4. Part 2. A. gemmosus: male (QCAZ 4352, 1, J), male (QCAZ 4385, K, L), male (QCAZ 11849, M, N), and female (QCAZ 4393, O, P). All photographs by O. Torres-Carvajal, except A, M, N (S. R. Ron). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (15) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Description of hoiotype (scores for para- types in parentheses) Male (Figs. 1, 2); SVL 59.7 mm (46.5-60.3 mm); tail length 150.2 mm (146.2-163.4 mm); head length 15.9 mm (14.8-16.5 mm); head width 8.4 mm (7. 8-8. 8 mm); head height 7.2 mm (6. 5-7. 5 mm); internasal distance 2.0 mm (1. 2-2.1 mm); interorbital distance 2.4 mm (2.2-2. 5 mm); interparietal absent (present, interparietal length 0.8-0. 9 mm; second largest scale length near in- terparietal 0. 3-0.4 mm); ear opening maximum length 1.6 mm (1. 6-2.1 mm); snout length 6.8 mm (6. 8-7.3 mm); jaw length 11.7 mm (11.7-14.4 mm); axila-groin distance 27.7 mm (27.4-30.6 mm); femur length 14.8 mm (14.4-15.6 mm); 4th toe length 12.5 mm (10.6-12.8 mm); 4th toepad width 1.2 mm (1.0- 1.3 mm); forelimb length 36.2 mm (21.8-36.2 mm); hindlimb length 42.6 mm (42.6-52.7 mm). Head scales multicarinate (same, unicarinate, or ru- gose) on frontal region and unicarinate (same, multicari- nate or rugose) on supraocular disc; 11 (10-14) scales between second canthals; 13 (11-15) scales between first canthals; 6 (5-7) scales bordering the rostral posteriorly; anterior nasal in contact with rostral (same or inferior nasal in contact with rostral); supraorbital semicircles separated by two (0-3) scales; supraocular disk with scales heterogeneous in size; one elongate superciliary followed by a series of granules (same or one small scale instead of granules); 6 (5-8) loreal rows on left side; 49 (25-53) loreal scales; interparietal absent (same or, when present, the interparietal smaller than ear opening, with 4-7 scales between interparietal and semicircles on each side, and 8-15 scales between interparietal and nape scales); suboculars in contact with supralabials; 6 (5-7) supralabials counted up to a point below center of eye; 6 (5-7) infralabials counted up to a point below center of eye; 7 (4-7) postmentals; one enlarged sublabial on each side. Table 2. PCA loadings conducted on nine morphological vari- ables of Anolis gemmosus, A. otongae and A. poei. Raw Rotated 1 2 3 1 2 3 Head height -0.96 0.21 -0.02 -0.97 0.16 -0.07 Head length 0.24 0.34 0.05 0.22 0.35 0.05 Head width -0.96 0.20 -0.03 -0.96 0.15 -0.07 Jaw length 0.98 0.06 -0.06 0.98 0.11 -0.03 Snout length 0.82 0.33 -0.07 0.81 0.37 -0.06 Forelimh length -0.01 0.80 0.04 -0.05 0.80 0.00 Hindlimh length -0.01 0.85 0.02 -0.05 0.85 -0.03 Axilla-groin length -0.01 -0.50 -0.02 0.01 -0.50 0.01 Snout- vent length 0.06 -0.04 0.99 0.02 0.01 1.00 Eigenvalue 3.54 1.93 1.00 3.53 1.93 1.01 % van explained 39.31 21.42 11.16 39.23 21.45 11.21 Species O A. gemmosus # A (Aongao G poei Fig. 5. Distribution of Anolis gemmosus, A. otongae and A. poei sp. nov. along the first and second principal components axes. Dorsal crest or enlarged middorsal row absent; dorsal scales keeled, 11 (9-11) dorsal scales in 5% the length of SVL contained in the dorsal midline at the level of the forelimbs; flank scales more or less separated by skin; ventrals smaller than dorsals, 13 (8-13) longitudi- nal rows in 5% the length of SVL; ventrals smooth and granular, arranged in diagonal rows. Toepads overlap the first phalanx in all toes; 19 (18-19) lamellae under phalanges III and IV of fourth toe (character 27 in Williams et al. 1995 and character 9 in Poe 2004); supradigitals multicarinate; tail with a double row of middorsal scales; postanals present (same or absent), with a slightly enlarged scale laterally on each side. Nuchal fold present (absent in females and juveniles); dorsal folds absent; dewlap extending posteriorly to a point halfway between fore and hindlimbs (absent in fe- males); dewlap with five longitudinal rows of 3-7 swol- len scales, similar size to ventrals, separated by naked skin. Sexual variation of meristic and morphometric char- acters in A. poei is presented in Table 4. Color in life Hoiotype (QCAZ 3449; Figs. 2, 3): background of head, body, limbs and tail green; head with light bluish green, dark green, and light grey irregular spots dorsally; dor- sal surface of body with six light grey, small irregular blotches; dorsal surface of neck with two light grey, small irregular blotches; limbs with dark green and yel- lowish-cream spots; lateral surface of head with a white stripe extending posteriorly from loreal region, through subocular region, to a point anterior to the tympanum; white blotch with yellow center above tympanum; eye- lids yellowish green with first row of upper and lower Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (16) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. palpebrals black, second and third rows both yellow and green; lateral surface of neck with dark green dots; body flanks green, with rows of yellow-centered white spots oriented ventro-posteriorly; ventral surface of head yel- lowish green with light yellow blotches; ventral surface of body white with bluish-green reticulations; ventral surface of limbs white with several transparent scales and dark brown reticulations; ventral surface of tail white with dark green spots anteriorly, and yellowish-green transverse bands posteriorly; dewlap skin light blue, dark yellowish green on gorgetal region, light blue on ster- nal region; gorgetal scales light yellowish green; mar- ginals and stemals white; iris dark brown with a white inner ring. When stressed, the dorsal background color switched from green to yellowish brown. Subadult male (QCAZ 3455, Figs. 2,3, differences from holotype): head with dark green and white irregu- lar small spots dorsally; dorsal surface of body and neck with white and dark green small spots, and larger pale yellow spots; lateral surface of head yellowish green with a white stripe extending posteriorly from loreal region, through subocular region, to upper border of tympanum; lateral surface of body with rows of white and dark green small spots, and larger pale yellow spots; ventral surface of head with white blotches and light blue spots; ventral surface of body with dark green reticulations; ventral sur- face of limbs with brown or green reticulations; ventral surface of tail with blackish green reticulations anteri- orly; dewlap skin white, yellow on gorgetal region, white on sternal region; throat, edge of mouth, and tongue pink- ish white (Fig. 6). When stressed, rust-colored blotches appeared on dorsal surface of head, body, limbs and tail. Adult female (QCAZ 3454, Fig. 7): dorsal surface of head, body and tail yellowish green; dorsal surface of body with six narrow brown chevrons, each one de- limited posteriorly by a grayish white blotch; limbs yel- lowish green with dark green spots arranged in bands, and pale yellowish spots; tail with two brown chevrons anteriorly; lateral surface of head yellowish green; loreal region yellow; lateral surface of neck and body yellowish green with brown dots; ventral surface of head pale yel- low with yellowish green reticulations, short white lon- gitudinal stripe on throat; ventral surface of body and tail white with black reticulations laterally; ventral surface of limbs white with some transparent scales and brown reticulations on hindlimbs; ventral surface of tail with brownish green reticulations anteriorly; iris brown with a pale white ring. Subadult female (QCAZ 3446, Fig. 7, differences with QCAZ 3454): occipital and temporal regions with brown and white small blotches; dorsal surface of neck with a distinct brown chevron delimited posteriorly by a grayish white blotch; lateral surface of body yellowish green dor- sally and light blue ventrally, with white or cream spots; dorsal surface of tail with two brown chevrons, each one delimited posteriorly by a grayish white blotch. Fig. 6. Tongue of Anolis poei sp. nov., subadult male (QCAZ 3455, top); A. gemmosus, adult male (QCAZ 4347, middle); A. otongae, adult male (QCAZ 4661, bottom). Photographs by S. R. Ron (top), O. Torres-Carvajal (middle, bottom). Color in preservative Holotype (QCAZ 3449): dorsal background of head, body, limbs and tail grayish brown; dorsal surface of head with metallic green, dark green, blue, gray and white cream irregular spots; dorsal surface of body with six black small chevrons, each delimited posteriorly by a white irregular blotch; limbs with dark brown and white spots; lateral surface of head with a white stripe extend- ing posteriorly from loreal region, through subocular re- gion, to a point anterior to the tympanum; upper border of tympanum with a white spot; eyelids purple with first row of upper and lower palpebrals black, second and third rows white and purple; neck flanks with black dots; body flanks grayish brown, with dark brown diagonal Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (17) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Table 3. Summary of morphological characters of Anolis poei sp. nov. and A. gemmosus from Ecuador. For each quantitative character, the F-value, t-value, and corresponding P-values are given. Range and sample size (in parenthesis) followed by mean + standard deviation are given. Character A. gemmosus A. poei sp. nov. F-value P t-value P Scales between second canthals 12-21 (24) 15.25 + 1.98 11-14(13) 12.08 + 1.12 2.59 0.12 5.31 <0.005 Postrostrals 5-7 (24) 5.79 + 0.72 5-7 (13)5.92 +0.64 1.37 0.25 -0.55 0.59 Row of loreals 6-10(24)7.25 + 1.15 5-8(13) 6.31 + 1.18 0.03 0.86 2.36 0.02 Scales between supraorbital semicircles 1-5 (24)3.13 + 1.23 1-3 (13) 1.62 + 0.77 5.27 0.03 4.46 <0.005 Scales between interparietal (if present) and semicircles 3-8 (24) 5.67 + 1.27 4-7(6)5.83 + 1.17 0.18 0.67 -0.29 0.77 Supralabials 5-7 (24) 6.08 + 0.50 5-7(13)6 + 0.41 1.29 0.27 0.51 0.61 Postmentals 4-8 (24) 6.13 + 1.03 4-7(13)5.77 + 0.93 0.18 0.67 1.03 0.31 Lamellae under phalanges III-IV of fourth toe 14-18 (24) 17.33 + 0.92 18-19(13) 18.92 + 0.28 8.71 0.01 -7.86 <0.005 Head length 13.23-18.12 (94) 15.46+ 1.07 14.79-16.5 (7) 15.67 + 0.51 4.67 0.03 0.93 0.37 Head width 6.97-17.41 (94) 10.82 + 3.24 7.84-8.84 (7) 8.29 + 0.36 32.16 <0.005 -7.03 <0.005 Head height 5.42-15.96 (94) 9.51 +3.32 6.54-7.48 (7) 6.92 + 0.38 31.04 <0.005 -6.96 <0.005 Jaw length 7.31-17.43 (94) 12.32 + 3.02 11.73-14.36 (7) 12.44 + 0.91 19.25 <0.005 0.26 0.80 Snout length 5.79-14.95 (94) 10.58 + 2.93 6.75-7.30 (7) 6.92 + 0.19 41.30 <0.005 -11.74 <0.005 Forelimb length 23.41-34.34 (94) 29.43 + 2.28 21.84-36.18 (7) 28.57 + 4.25 0.02 0.89 -0.12 0.90 Hindlimb length 41.51-63.80 (94) 52.82 + 4.13 42.56-52.68 (7) 49.01 + 3.33 1.00 0.32 -2.38 0.02 Axilla-groin length 20.73-33.51 (94) 26.74 + 2.07 27.35-30.61 (7) 28.54+ 1.30 0.95 0.33 2.26 0.03 Snout- vent length 46.71-66.21 (94) 58.34 + 3.65 46.47-60.31 (7) 56.87 + 4.85 0.35 0.56 -1.00 0.32 Tail length 94.94-191 (94) 154.59 + 18.66 146.21-163.37 (7) 154.74 + 6.32 3.82 0.05 0.02 0.98 Table 4. Sexual variation in lepidosis and measurements (mm) of Anolis poei sp. nov. Range followed by mean + standard devia- tion are given. Character Males Females n = 4 n = 3 Scales between second canthals 11-13 11.75 + 0.96 12-13 12.67 + 0.58 Postrostrals 5-6 5.75 + 0.5 6-7 6.33 + 0.58 Row of loreals 6-8 7 + 1.15 5-6 5.33 + 0.58 Scales between supraorbital semicircles 1-2 1.75+0.5 1-2 1.67 + 0.577 Scales between interparietal and semicircles Interparietal absent 6-7 6.50 + 3.78 Supralabials to below center of eye 6 6 Postmentals 4-7 5.25 + 1.5 6-7 6.33 + 0.58 Lamellae under phalanges II-III of fourth toe 19 19 Head length 15.8-16.5 15.95 + 0.38 14.8-15.62 15.29 + 0.44 Head width 7.84-8.84 8.31 +0.41 8.05-8.66 8.26 + 0.34 Head height 6.67-7.48 7.02 + 0.39 6.54-7.27 6.8 + 0.41 Jaw length 11.73-12.65 12.25 + 0.38 11.86-14.36 12.70 + 1.43 Snout length 6.75-7.04 6.87 + 0.12 7.82-7.30 7 + 0.26 Forelimb length 27.94-36.18 30.75 + 3.72 21.84-28.19 25.68 + 3.37 Hindlimb length 42.56-52.68 49.35 + 4.59 47.50-49.56 48.57 + 1.03 Axilla-groin length 27.35-28.17 27.76 + 0.33 27.94-30.61 29.57 + 1.43 Snout- vent length 58.80-60.31 59.45 + 0.68 46.47-58.48 53.43 + 6.22 Tail length 150.20-163.37 157.89 + 5.89 146.21-155.38 150.53 +4.60 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (18) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. A ^ B 'I7\ f *- 9 c D E F Fig. 7. Anolis poei sp. nov. Adult female (SVL = 46.47 mm, QCAZ 3454, A, B), subadult female (SVL = 47.99 mm, QCAZ 3446, C, D), juvenile male (SVL = 26.85 mm, QCAZ 3453, E, F). Photographs by O. Torres-Carvajal bands oriented ventro-posteriorly and intercalated with Adult male (QCAZ 6783): dorsal surface of head and white spots; ventral surface of head white with light blue reticulations; ventral surface of body white with faint grayish purple reticulations; ventral surface of limbs grayish cream with dark brown reticulations; ventral sur- face of tail white anteriorly with a metallic green tint and grayish purple spots, and gray posteriorly; dewlap skin with a turquoise gorgetal region and white sternal region; gorgetal scales light brown with a gold tint internally, and dark brown externally; dewlap marginals and sternals white; throat, edge of mouth and tongue white. body dark brown with gray dots; dorsal surface of limbs dark brown, with gray dots on forelimbs; lateral surface of head dark brown with white cream dots dorsal and an- terior to tympanum; body flanks dark brown with faint white dots arranged on diagonal lines that reach venter; ventral surface of head with bluish-purple infralabial and sublabial regions, and light purple gular region with white irregular spots; ventral surface of body white with faint purple reticulations; limbs creamish gray with dark brown reticulations; ventral surface of tail white with Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (19) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador purple mottling anteriorly, and gray posteriorly; dewlap skin with a light blue gorgetal region and white ster- nal region; gorgetal scales purple; sternal and marginal scales white. Adult female (QCAZ 3454): dorsal surface of head brown with metallic blue and green frontal and supraocu- lar regions; dorsal surface of body brown with six nar- row black chevrons, each one delimited posteriorly by a white blotch; forelimbs bluish brown with white spots arranged in stripes; hindlimbs brown with dark brown bands and dots; tail with two black chevrons anteriorly; lateral surface of head brown with purple tint; labial re- gion light purple; lateral aspect of neck and body purple with black dots; ventral surface of head white with purple brown stripes; ventral surface of body white with dark brown dots laterally; ventral surface of limbs grayish cream with dark brown reticulations on hindlimbs; ven- tral surface of tail white with dark brown dots. Subadult female (QCAZ 3446, differences with QCAZ 3454): occipital and temporal regions with dark brown, small blotches; dorsal surface of neck with a dis- tinct dark brown chevron; dorsal surface of body with six distinct, dark brown chevrons; dorsal surface of tail with two dark brown chevrons. Phylogenetic relationships The data matrix analyzed in this study contained 1,065 unique site patterns. Of the 2,807 nucleotide characters included in our analysis 1,703 were constant, 224 par- simony uninformative, and 880 were parsimony infor- mative. The 50% majority rule consensus tree resulting from the Bayesian analysis (Fig. 8) is generally congru- ent with the phytogeny of the clade Dactyloa presented by Castaneda and de Queiroz (2011). Both the new spe- cies described here and A. otongae are members of the aequatorialis series of Castaneda and de Queiroz (2013), which corresponds roughly to the “western clade” of Castaneda and de Queiroz (2011). Our phytogeny sup- ports strongly (PP = 0.99) a sister taxon relationship be- tween Anolis poei and A. gemmosus, as well as the ex- clusivity (de Queiroz and Donoghue 1990; de Queiroz 1998) of both species. They form a clade sister (PP = 0.89) to A. otongae. The clade formed by the three spe- cies is sister (PP = 1) to a clade formed by A. aequatoria- lis and^. anoriensis. Distribution and ecology Anolis poei inhabits low montane evergreen forest (Sierra 1999) on the western slopes of the Andes in central Ecua- dor, Provincia Bolivar, between 1,310-1,354 m (Fig. 9). f c u SN M 1 xy 1 A. agasshi KEN2 0.99. Voucher information is presented in Castaneda and de Queiroz (2011) and Table 1. C A. p^mcccte QC AZ6&69 A- pcrvccac QC A Z6R79 A fusuiti QCAZ6930 A.L'hhris MRCI26 A. chhris QCAZftR77 A. cftloiis OCA26920 'A. venirifnucitlufns MRC09I A. venirimciL'nlcitiix MRC1J2 ■ A. anqiiatoiiaiis QCAZ(iS55 — A, (tctfuuifn'iijtix QCAZ6SK3 - A. afitiriatixix MiiUAT5l7 r A. cinurKttxfx MHUAl 156S T- A. anoriensis MHUAT5I6 r A. (itonpae QCAZ] I79<) 1= A. oion^ae QCAZ] i7y] A. poei QCAZ3445_4359 A. poei QC A Z3444 A. poei QCAZ344S rl n.na 4 A - petnnuisiss QCAZ43S5 A. gemmosus QCAZ4406 A. gemmosus QCAZ6X51 A. gemmosus QC A ZtSHB 4 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (20) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. The new species occurs in sympatry with A. aequa- torialis, A. binotatus, and A. fasciatus at its type locality (Fig. 4). Specimens of the new species were collected along the border of a road, close to rivers, in second- ary forest, and on shrubs within pastures. All individu- als were found between 20h00 and 22h00 sleeping with their heads up, or in a horizontal position on branches or vines, 0.5^.5 m above ground or streams. The smallest individual QCAZ 3453 (SVL = 26.9 m; TL = 67.6 mm) was collected on 11 June 2011. Etymology The specific name is a noun in the genitive case and is a patronym for Steve Poe, who has published important contributions to the systematics and evolution of Ano- lis lizards (Poe 2004, 2011). During his collecting trips to Ecuador in 2009 and 2010, Poe trained several young herpetologists in field collecting techniques and inspired them to explore the diversity of anole lizards. This paper is one of the products resulting from that inspiration. Acknowledgments. — We thank Jorge H. Valencia, Francy Mora, and Estefama Boada for assistance in the field; Santiago Ron and Lucas Bustamante for the pho- tographs; Paulina Santiana and Andrea Varela for assem- bling some of the figures; Melissa Rodriguez for helping with the map. Special thanks to Kevin de Queiroz and two anonymous reviewers for commenting on previous versions of this manuscript. OTC received funds from Secretaria de Educacion Superior, Ciencia, Tecnologia e Innovacion (SENESCYT). Specimens were collected under collection permit OOl-IC-FAU/FLO/DRZCHI/MA and 008-09 IC-EAU-DNB/MA issued by Ministerio de Ambiente del Ecuador. Literature Cited Biomatters. 2010. Geneious version 5.3. Available: http://www.geneious.com/ Castaneda MR, de Queiroz K. 2011. Phylogenetic rela- tionships of the Dactyloa clade of Anolis lizards based on nuclear and mitochondrial DNA sequence data. Molecular Phylogenetics and Evolution 61: 784-800. Castaneda MR, de Queiroz K. 2013. Phylogeny of the Dactyloa clade of Anolis lizards: new insights from combining morphological and molecular data. Bulle- tin of the Museum of Comparative Zoology 160(7): 345-398. de Queiroz K, Donoghue MJ. 1990. Phylogenetic sys- tematics or Nelson’s version of cladistics? Cladistics 6(1): 61-75. de Queiroz K. 1998. The general lineage concept of spe- cies, species criteria, and the process of speciation. Pages 57-75 In: Howard DJ, Berlocher SH (editors). Endless Forms: Species and Speciation. Oxford Uni- versity Press, Oxford, United Kingdon and New York, New York, USA. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (21) Fig. 9. Distribution of Anolis gemmosus (triangles), A. otongae (circles) and A. poei sp. nov. (squares) in Ecuador. de Queiroz K. 2007. Species concepts and species de- limitation. Systematic Biology 56: 879-886. Eolmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochon- drial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3: 294-299. Gelman A, Rubin DB. 1992. Inference from iterative simulation using multiple sequences. Statistical Sci- ence 7: 457-511. Kumazawa Y, Nishida M. 1993. Sequence evolution of mitochondrial tRNA genes and deep-branch animal phylogenetics. Journal of Molecular Evolution 37: 380-398. Lanfear R, Calcott B, Ho SY, Guindon S. 2012. Partition- Einder: Combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution 29(6): 1695-1701. Losos JB. 2009. Lizards in an Evolutionary Tree: Ecol- ogy and adaptive radiation of anoles (Volume 10). University of California Press, Berkeley and Los An- geles, California, U.S.A. Macey JR, Larson A, Ananjeva NB, Pang Z, Papenfuss TJ. 1997. Two novel gene orders and the role of light- strand replication in rearrangement of the vertebrate mitochondrial genome. Molecular Biology and Evo- lution 14: 91-104. Nicholson KE. 2002. Phylogenetic analysis and a test of the current infrageneric classification of Norops (Beta May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Anolis). Herpetological Monographs 16: 93-120. Poe S. 2004. Phylogeny of anoles. Herpetological Mono- graphs 18: 37-89. Poe S, Giermakowski JT, Latella I, Schaad EW, Hulebak EP, Ryan MJ. 2011. Ancient colonization predicts re- cent naturalization in Anolis lizards. Evolution 65(4): 1195-1202. Rambaut A, Suchard MA, Xie W, Drummond AJ. 2013. Tracer vl.6. Available: http://tree.bio.ed.ac.uk/soft- ware/tracer Rieppel 0.2010. Species monophyly. Journal of Zoolog- ical Systematics and Evolutionary Research 48: 1-8. Ronquist E, Huelsenbeck JR 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioin- formatics 19: 1572-1574. Schulte JA II, Cartwright EM. 2009. Phylogenetic re- lationships among iguanian lizards using alternative partitioning methods and TSHZl: A new phylogenet- ic marker for reptiles. Molecular Phylogenetics and Evolution 50(2): 391-396. Sierra R. 1999. Propuesta preliminar de un sistema de clasificacidn de vegetacion para el Ecuador conti- nental. Proyecto INEFAN/GEF-BIRE y EcoCiencia. Quito, Ecuador. SPSS Inc. 2008. SPSS Statistics for Windows 17.0. Chi- cago, USA. Torres-Carvajal O, D. Salazar- Valenzuela y A. Me- rino-Viteri. 2014. ReptiliaWebEcuador. Ver- sion 2014.0. Museo de Zoologia QCAZ, Pontifi- cia Universidad Catolica del Ecuador. Available: http://zoologia.puce.edu.ee/Vertebrados/reptiles/ reptilesEcuador [Accessed: 04 May 2014]. Uetz P, Hosek J. (Editors). The Reptile Database. Avail- able: http://www.reptile-database.org [Accessed: 08 January 2014]. Williams EE. 1976. South American anoles: The species groups. Papeis Avulsos de Zoologia 29: 259-268. Williams EE, Duellman WE. 1984. Anolis fitchi, a new species of the Anolis aequatorialis group from Ecua- dor and Colombia. University of Kansas Publications of the Museum Natural History 10: 257-266. Williams EE, Rand H, Rand AS, O’Hara RJ. 1995. A computer approach to the comparison and identifica- tion of species in difficult taxonomic groups. Breviora 502: 1^7. Received: 28 April 2014 Accepted: 20 May 2014 Published: 28 May 2014 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (22) May 2014 | Volume 8 | Number 1 | e76 Ayala- Varela et al. Appendix 1 Additional specimens examined Anolis gemmosus - Ecuador: Carchr. Chilma Bajo, Finca de Anibal Pozo, 0.86397°N, 78.04723°W, 2,022 m, QCAZ 8681-82; Chilma Bajo, Finca de Anibal Pozo, 0.86495 °N, 78.04979'W, 2,071 m, QCAZ 8683; La Centella, 0.89318°N, 78.13471 'W, 1,800-2,400 m, QCAZ 11784; Maldonado, Sendero Ecologico Teldibi, 0.91301 °N, 78.10782'W, 1,477- l, 635 m, QCAZ 12272, 12278, QCAZ 12279-80, 4360, 4406, 4408; Rio San Pablo, cerca a Chical, 0.90302°N, 78.1 6284 'W, 1 ,399 m, QCAZ 4377, 4382, 4385-86,4388; Rio San Pablo, cerca a Chical, 0.90327°N, 78.1 6201 °W, 1 ,429 m, QCAZ 4393; Rio Verde and Rio Pablo, Rio Estrellita, Guapil, 1 ,428-1 ,466 m, QCAZ 12289, 12294, 12302; Cotopaxi: 1 1 5 km Qeste de Pilalo, 0.928 °S, 79.057 'W, 1 ,500 m, QCAZ 4072; 1 8.2 km de Quillutuha, via a Pucayacu, 0.67843 °S, 79.01 565 'W, 1 ,420 m, QCAZ 8845-49; Alrededores de San Francisco de Las Pampas, 0.42371 °S, 78.96765 °W, 1 ,800 m, QCAZ 1440-47, 2123; Bosque Integral Qtonga , 0.4194 °S, 79.00345°W, 1,720-2,143 m, QCAZ 2758, 2809-10, 3121, 3126-27, 3131, 3133, 3174, 3180-90, 3863-3866, 3869-71 , 3940, 3974-76, 4028-34, 4224-25, 4657, 4663, 4785, 5060, 5063, 5371, 5477-79, 5482-83, 6770-73, 9888, 10424, 10438-39, 10441-42, 10452, 12057, 12060-65, 12067, 12072-73, 12075, 12077-82, Bosque Integral Qtonga, a lo largo del rio Esmeraldas, 0.46333 °S, 79.05027 °W, QCAZ 7281 -89; Bosque Integral Qtonga, alrededores de la estacion, 0.41 933 °S, 79.00336 'W, 1 ,980 m, QCAZ 1 0697 ; Bosque Integral Qtonga, arriba de la estacion, 0.41 478 °S, 79.00073 'W, QCAZ 3867-68; Bosque Integral Qtonga, orillas del rio Esmeraldas, 0.41932°S, 78.99396 'W, 1 ,719 m, QCAZ 10393, 10395, 10399; Bosque Integral Qtonga, sendero a la Es- tacion, 0.41933 °S, 79.00336 °W, 1 ,646 m, QCAZ 10696; Cerca a Naranjito, 0.41944 °S, 79.00333 'W, QCAZ 7825; San Francisco de Las Pampas, 0.42371 °S, 78.96765 °W, 1 ,600-1 ,800 m, QCAZ 63, 68-70, 72-79, 3134-53, 3155, 3175; Via a Qtonga, 0.33183 °S, 78.93791 'W, 1,476-1,700 m, QCAZ 8412; Imbabura: 6 de Julio de Cuellaje, 0.4°N, 78.525 °W, QCAZ 4346-47; 6 de Julio de Cuellaje, 0.401 07 °N, 78.51 81 'W, 1 ,886 m, QCAZ 4349; 6 de Julio de Cuellaje, 0.401 02°N, 78.51 ZZO'W, 1,897 m, QCAZ 4350; 6 de Julio de Cuellaje, punto 8, 0.4°N, 78.525 'W, QCAZ 4348; 6 de Julio de Cuel- laje, San Antonio, Cordillera deToisan, 0.45803 °N, 78.54722 'W, QCAZ 9450-53; Carretera nueva via a Cuellaje, Sector de Santa Clara, Reserva Alto Choco, 0.37603°N, 78.45857°W, 2,062 m, QCAZ 4352-54; La Mina, Junin, 0.2754 °N, 78.6603 °W, 1,715 m, QCAZ 3071; Manduriaco, 0.277 °N, 78.873 °W, 1,330 m, QCAZ 5328; Manduriacu, 7.5 km NE of Bellavista, 0.31006°N, 78.85757'W, 1,177-1,227 m, QCAZ 11606, 12305-314; 12322, 12324, 12326, 12328, 12331; Reserva Siempre Verde, NE de Cotacachi, 0.37167°N, 78.421 86 °W, 2,468 m, QCAZ 8837; Reserva Alto Choco, Santa Rosa, 0.36939 °N, 78.44942'W, 2,109 m, QCAZ 7330-31 ; Pichincha: 1-2 km oeste de Tandayapa, 0.004 °S, 78.663 °W, 2,000 m, QCAZ 2070-71 ; 2.9 km de Tandayapa, 0.00952 °S, 78.65698 'W, 1 ,820 m, QCAZ 406-10; 5 km E Tandayapa, 0.02°S, 78.651 °W, 1,975 m, QCAZ 2066-69; A orillas del Rio Chisinche, en la carretera a Conchacato, 0.448°S, 78.76423 'W, 1,693 m, QCAZ 6884-89; 30 km E de Santo Domingo, hacia la Reserva de Bosque Integral Qtonga, 0.3884 °S, 78.92995 'W, QCAZ 9769-70; 9775; Bosque Protector Mindo - Nambillo, refugio, 0.106°S, 78.687°W, 1,700 m, QCAZ 2910; Cooperative El Porvenir, finca El Cedral, 0.114°N, 78.56993°W, 2,297 m, QCAZ 10501-502; Desviacion a Mindo, 1-5 km de la interseccion hacia abajo, 0.02853°S, 78.75861 °W, 1,661 m, QCAZ 9724-31; Estacion Cientifica Rio Guajalito, 0.22676 °S, 78.82171 °W, 1,791-1,814 m, QCAZ 1330, 1333, 1500, 1645, 2682-84, 2786, 2813, 2815-16, 3040-45, 3056-57, 3373, 3385, 4123-25, 4210, 4214, 6413-14, 8859, 8864-65, 9974, 11404, 12088-101; Las Tolas, 0.7281 8 °N, 78.77792 'W, 1,200-1,600 m, QCAZ 11848-49; Manuel Cornejo Astorga (Tandapi), frente a la planta de agua potable "El Placer" via a Conchacato, 0.42471 °S, 78.78905 'W, 1,500 m, QCAZ 6882; Manuel Cornejo Astorga (Tandapi), via Atenas a 5 km de la carretera principal, 0.40625 °S, 78.83621 °W, 1 ,671 m, QCAZ 5365-70; Mindo, 1 ,342- l, 560 m, QCAZ 12350-53, 12356, 12358, 12365, 12370, 12375-76; Mindo Biology Station, 0.07805°S, 78.73194 °W, QCAZ 7518-20, 7522; Mindo, camino entre Mariposas de Mindo y Mindo Garden, 0.06753°S, 78.7535 'W, 1,361 m, QCAZ 6851-53, 6858; Mindo Garden, 4 km de Mindo, 0.06901 °S, 78.801 66 °W, QCAZ 2787; Mindo, El Monte, Road to Mindo Garden, 0.07805 °S, 78.7319°W, QCAZ 7521; Mindo, Sachatamia Lodge, 0.02638 °S, 78.75944 °W, 1,700 m, QCAZ 11857-59; Nanegalito, Finca El Cedral, 0.1141 °N, 78.57007'W, 2,272 m, QCAZ 9462-63; Pachijal, via Nanegali- to-Los Bancos, 0.13°S, 78.72644 'W, 1,741 m, QCAZ 5494-500; Palmeras, 0.244 °S, 78.794 °W, 1,800 m, QCAZ 871, 881-83,1351-52, 2244, 3004-06; Recinto Chiriboga, Estacion La Favorite , 0.21307°S, 78.78421 °W, 1,680 m, QCAZ 5383-84; Reserva Ecologica Bosque Nublado "Santa Lucia," 0.11928°N, 78.59647°W, 1,624-1,927 m, QCAZ 10664, 11850-52, 11888-93,11897, 11899; Tandayapa, 0.00591 °N, 78.67455'W, 1,670 m, QCAZ 4086. Locality in error. Pich- incha, San Antonio de Pichincha, 0.00905 °S, 78.44581 °W, QCAZ 724. Anolis otongae - Ecuador: Cotopaxi: Alrededores de San Francisco de Las Pampas, 0.42371 °S, 78.96765 'W, 1,800 m, QCAZ 2128; Bosque Integral Qtonga , 0.41944 °S, 79.00333 'W, 1,900-2,300 m, QCAZ 1721 , 2050-52, 3129, 3706, 3796, 3872-73, 4025, 4661, 5481, 6219, 11790-91, 12035, 12056, 12058, 12070-71; Los Libres, QCAZ 2781; Penas Coloradas, 0.52343°S, 79.05908°W, QCAZ 1696; Pichincha: La Victoria, 0.47747°S, 79.05336'W, 2,104 m, QCAZ 6394-96. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (23) May 2014 | Volume 8 | Number 1 | e76 A new species ofAnolis lizard from western Ecuador Fernando Ay ala- Varela is the director of the herpetology collection at the Pontificia Universidad Catolica del Ecuador in Quito. He received his diploma at the Pontificia Universidad Catolica del Ecuador, Quito in 2004. He has been interested in herpetology since childhood and has dedicated a lot of time studying the lizards of Ecuador, specifically the taxonomy and ecology ofAnolis species. His current research interests include reproductive biology and ecology of lizards and snakes in Ecuador. Diana Troya-Rodnguez received a B.Sc. in Biology from Pontificia Universidad Catolica del Ecuador (PUCE) in 2013. As a student, she joined the Museo de Zoologia QCAZ, Pontificia Universidad Catolica del Ecuador in Quito, where she developed a great interest in reptiles. She has been studying anole lizards for the last four years. Eor her undergraduate thesis, Diana worked on the “Comparative phylogeography of two sympatric species of Anolis (Squamata: Iguanidae) and the impact of global warming on their dis- tribution.” Xiomara Talero-Rodnguez is an undergraduate biology student at Pontificia Universidad Catolica del Ecuador in Quito. She joined Museo de Zoologia QCAZ last year and has been helping with several anole lizard projects ever since. She is currently interested in studying ecology and behavior of anoles. Omar Torres-Carvajal graduated in Biological Sciences from Pontificia Universidad Catolica del Ecua- dor (PUCE) in 1998, and in 2001 received a Master’s degree in Ecology and Evolutionary Biology from the University of Kansas under the supervision of Dr. Einda Trueb. In 2005 he received a Ph.D. degree from the same institution with the thesis entitled “Phylogenetic systematics of South American lizards of the genus Stenocercus (Squamata: Iguania).” Between 2006-2008 he was a postdoctoral fellow at the Smithsonian Institution, National Museum of Natural History, Washington DC, USA, working under the supervision of Dr. Kevin de Queiroz. He is currently Curator of Reptiles at the Zoology Museum QCAZ of PUCE and an Associate Professor at the Department of Biology in the same institution. He has published more than 30 scientific papers on taxonomy, systematics and biogeography of South American reptiles, with emphasis on lizards. He is mainly interested in the theory and practice of phylogenetic systematics, particularly as they relate to the evolutionary biology of lizards. In accordance with the International Code of Zoological Nomenclature new rules and regulations (ICZN 2012), we have deposited this paper in publicly accessible institutional libraries. The new species described herein has been registered in ZooBank (Polaszek 2005a, b), the official online registration system for the ICZN. The ZooBank publication LSID (Life Science Identifier) for the new species described here can be viewed through any standard web browser by appending the LSID to the prefix “http://zoobank.org/”. The LSID for this publication is: urn:lsid:zoobank.org:pub:61380956-FlAC-46C0-84F3-ClED545C46DC. Separate print-only edition of paper(s) (reprint) are available upon request as a print-on-demand service. Please inquire by sending a request to: Amphibian & Reptile Conservation (amphibian-reptile-conservation.org; arc.publisher@gmail.com). Amphibian & Reptile Conservation is a Content Partner with the Encyclopedia of Life (EOL); http:///www.eol.org/ and submits information about new species to the EOL freely. Digital archiving of this paper are found at the following institutions: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Ernst Mayr Library, Museum of Comparative Zool- ogy, Harvard University, Cambridge, Massachusetts (USA); Elorida Museum of Natural History, Gainesville, Elorida (USA). Complete journal archiving is found at: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Florida Museum of Natural History, Gainesville, Florida (USA). Citations ICZN. 2012. Amendment of Articles 8,9,10,21 and 78 of the International Code of Zoological Nomenclature to expand and refine methods of publication. Zootaxa 3450: 1-7. Polaszek A et al. 2005a. Commentary: A universal register for animal names. Nature 437 : 477. Polaszek A et al. 2005b. ZooBank: The open-access register for zoological taxonomy: Technical Discussion Paper. Bulletin of Zoological Nomenclature 62(4): 210-220. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (24) May 2014 | Volume 8 | Number 1 | e76 Copyright: © 2014 Ron et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education purposes only provided the original author and source are credited. The official publication credit source: Amphib- ian & Reptile Conservation at: amphibian-reptile-conservation.org Reproduction and spawning behavior in the frog, Engystomops pustulatus (Shreve 1941) ^Santiago R. Ron, ^’^Andrea E. Narvaez, and ^Giovanna E. Romero ^Museo de Zoologia, Escuela de Biologia, Pontificia Universidad Catolica del Ecuador, Av. 12 de Octubre y Roca, Aptdo. 17-01-2184, Quito, ECUADOR ^La Trobe University, Department of Zoology, Bundoora VIC 3086, AUSTRALIA ^Museo Ecuatoriano de Ciencias Naturales, Herbario Nacional del Ecuador, Av. Rio Coca E6-115 e Is la Fernandina, Quito, ECUADOR Amphibian & Reptiie Conservation [Special Section] 8(1): 25-32. Abstract— The study of reproductive strategies is central to understand the demography of populations and the energetic relationships of the species with their ecosystem. Documenting the reproductive natural history of the species is pressing in groups, like amphibians, that are threatened with extinction at a global scale. Herein, we describe the reproductive ecology and spawning behavior of the leptodactylid frog Engystomops pustulatus. In addition, we report observations that suggest the existence of an alternative mating strategy. Our results show that reproduction in E. pustulatus is characterized by high maternal investment (15% egg mass relative to body mass). We found evidence of size-assortative mating with a tendency of larger females to mate with larger males. Clutch size was correlated with female weight, female condition and male size. Larger females showed a tendency to lay larger foam nests and larger nests contained more eggs. At reproductive choruses, there was a male-biased operational sex ratio, indicative of high variance in male reproductive success. We observed an amplectant couple spawning while an additional male was embedded in the foam. We hypothesize that this behavior is evidence of an alternative mating strategy where a small non-amplectant male attempts to fertilize the eggs that are extruded by the amplectant female. Resumen. — El estudio de las estrategias reproductivas es fundamental para entender la demografia de las poblaciones y las relaciones energeticas de las especies con su ecosistema. Documentar la historia natural reproductiva de las especies es apremiante en grupos, como los anfibios, que estan amenazados con extincion a nivel mondial. Aqui, describimos la ecologia reproductiva y el comportamiento de anidacion en la rana leptodactilida Engystomops pustulatus. Ademas, reportamos observaciones que sugieren la existencia de una estrategia reproductiva alterna. Nuestros resultados indican que la reproduccion en E. pustulatus esta caracterizada por una alta inversion energetics de la hembra (15% de masa de huevos en relacion a la masa corporal). Se evidencia que el apareamiento es selective con respecto al tamaho, con una tendencia de hembras grandes a aparearse con machos grandes. El tamaho de la puesta estuvo correlacionado con el peso de la hembra, la condicion de hembra y el tamaho del macho. Las hembras mas grandes mostraron una tendencia de poner nidos de espuma mas grandes y los nidos mas grandes tuvieron un mayor numero de huevos. En cores reproductivos, hubo una tasa sexual operativa sesgada hacia los machos, lo que indica una alta varianza en el exito reproductive de los machos. Se observe una pareja en amplexus construyento un nido mientras un macho adicional estaba incrustado en el nido de espuma. Hipotetizamos que este comportamiento evidencia una estrategia de apareamiento alterna en la que un macho pequeho intents fertilizar huevos puestos por una hembra en amplexus con otro macho. Key words. Alternative mating strategy, clutch size, clutch piracy, fertilization rates, nesting behavior, testis size Citation: Ron SR, Narvaez AE, Romero GE. 2014. Reproduction and spawning behavior in the frog, Engystomops pustulatus (Shreve 1941). Amphibian & Reptile Conservation 8(1) [Special Section]; 25-32 (e79). Correspondence. Emails: ^ santiago.r.ron@gmail.com (Corresponding author), ^aenarvaezg@gmail.com, ^giaromev@gmail. com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (25) August 2014 | Volume 8 | Number 1 | e79 Ron et al. Introduction Understanding the natural history of reproduction is essential to characterize the ecological niche and the survival prospects of amphibians. Acquiring a better understanding of amphibian reproduction will assist con- servation efforts in the vertebrate class with the higher number of species threatened with extinction (Chanson et al. 2008). The deposition of eggs in foam nests characterizes the reproduction of most species of the Neotropical fam- ily Leptodactylidae, which has 201 species distributed from southern Texas to southern Chile (Frost 2014). In most species, males call in choruses that are visited by receptive females, which then actively choose a mate (Ryan 1985). Amplectant pairs build foam nests where hundreds of eggs are laid and fertilized (Crump 1974; Heyer 1969; Ryan 1985). The foam is formed when the male kicks the jelly surrounding the eggs while the fe- male discharges them. The foam may protect the eggs from dehydration and/or predation (Duellman and Tmeb 1994; Menin and Giaretta 2003) or from excessive heat (Gorzula 1977). Foam nests may facilitate multiple paternity by retain- ing sperm (Kusano et al. 1991). There is a high propor- tion of foam-nesting species among known cases of mul- timale spawning in anurans, eight out of 15 (Byrne and Roberts 1999; Kaminsky 1997; Prado and Haddad 2003). Although several reproductive characteristics of Lepto- dactylidae should favor multiple male mating strategies, there are only two documented cases, Leptodactylus chaquensis and L. podicipinus (Prado and Haddad 2003). The paucity of records may be partly due to lack of stud- ies. Although several leptodactylid species are abundant and live even in urban areas, little is known about its bi- ology beyond brief accounts of its systematics and mor- phology. Such is the case of the widely distributed and abundant Engystomops pustulatus (Ron and Read 2012). Herein, we describe the reproductive natural history of Engystomops pustulatus including fertilization rates, testis size, clutch size, and relative egg mass to explore factors that influence mate choice and reproductive out- put. We also describe its spawning behavior with obser- vations suggest the existence of a secondary male mating strategy. Materials and Methods Study site and species Engystomops pustulatus inhabits dry shrub, deciduous forest, and lowland moist forest below 300 m in western Ecuador. It can be relatively common during the rainy season, when they reproduce. They are explosive breed- ers that congregate around temporal pools. Males call from the water and amplectant pairs build foam nests to deposit their eggs (Ron and Read 2012). Engystomops Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (26) pustulatus should not be confused with E. pustulosus, a Central American species that has been a model for stud- ies of behavioral ecology (e.g., Ryan 2005). For clarity, hereafter, we refer to E. pustulosus exclusively as “Tun- gara frog.” Operational sex ratio (e.g., the number of males rela- tive to the number of females in breeding aggregations) in Engystomops pustulatus was assessed in western Ecuador at three localities: Reserva Cerro Blanco, (W 80.0214°, S 2.0264°, Provincia del Guayas; 19 March 2003), Patricia Pilar (Provincia Los Rios; 21 Eebmary 2002), and the town of La Mana (Provincia de Cotopaxi; 28 December 2003). Reproductive output, nest size and size assortment were evaluated in La Mana (W 79.265°, S 0.943°, elevation 160 m) between 28 December 2003 and 08 February 2004 and Patricia Pilar (W 79.3707°, S 0.5372°, elevation 200 m) between 23 January and 20 April 2008 during the rainy season. At La Mana and Pa- tricia Pilar, the vegetation is Evergreen Lowland Moist Forest (as deflned by Sierra et al. 1999). Most of the for- est in the region has been converted to pastures and agri- cultural lands. Field observations took place after dusk, between 19:00 and 3:00 h. Breeding occurred in small temporary ponds on the streets of the town. Some sites were under dim artificial light (street poles). Fertilization rates and nest size We estimated fertilization rates from amplectant pairs collected from the field. The amplectant pairs were placed in individual circular plastic containers (10 cm diameter) with water depth of one cm. Most pairs made a nest after few hours. Three or four days later, we washed the foam with a solution of chlorine and water and counted the number of hatched and undeveloped eggs (as described by Ryan 1983). We used this proportion as a proxy for fertilization rates. This methodology does not allow dis- criminating between undeveloped eggs as result of egg unviability or failed fertilization. Therefore, our method- ology may slightly underestimate fertilization rates. To estimate nest size, we measured (with digital cali- pers, to the nearest 0.01 mm) the length of the longest axis, width at the widest point perpendicular to the lon- gest axis, and height of all nests laid in the containers. We estimated nest volume with the formula of Vi ellipsoid: V = —abc 12 where a, b, and c are the length, width, and height, re- spectively. The measurements were taken while the nests were <1 day old. Adult size and egg mass Sex was determined by the presence of nuptial pads, vo- cal sac folds, and/or by gonad inspection. Snout-vent length (SVL) was measured with Fowler digital calipers (nearest 0.01 mm). Body mass was measured in the field August 2014 | Volume 8 | Number 1 | e79 Reproduction and spawning behavior in Engystomops pustulatus (before and after oviposition in females) with a digital balance (nearest 0.1 g). Relative egg mass (maternal in- vestment) was calculated as 1 — the ratio (female mass after oviposition/female mass before oviposition). After being kept in the plastic containers to allow spawning, females were euthanized by immersion in chloretone, fixed in 10% formalin, and preserved in 70% ethanol. Egg mass and body mass were measured after preservation in females that did not spawn. Each female was weighted on a digital balance (to the nearest 0.001 g), after removing excess ethanol. Then, the remaining egg masses (including immature eggs and jelly) were removed from the abdomen and weighted. Relative egg mass was calculated by dividing total egg mass by non- gravid female mass. Estimates of relative egg mass could be influenced by preservation in ethanol. Therefore, com- parisons with relative egg mass in non-preserved nesting females should be interpreted with caution. All preserved specimens are deposited at the amphibian collection of the Zoology Museum of Pontificia Universidad Catolica del Ecuador. Reproductive behavior Behavioral observations were carried out at male cho- ruses in La Mana, Ecuador. Spawning behavior was described from of a single nesting event at La Mana. Spawning was recorded in the field under infrared light with a digital camcorder SONY TRV70. The complete video is available at AmphibiaWeb (http://amphibiaweb. org). Statistical analyses Eor normally distributed variables, we tested the sig- nificance of relationship between them using linear re- gression ANOVAs; for non-normal variables, we tested relationship with Spearman’s rank correlations. Differ- ences between groups were tested with t-tests (assuming non-equal variances). Statistical tests were implemented in software IMP v.5.1 (SAS Institute, 2003). Results Reproductive output, fertilization rates, and nest size Among 77 nests, the mean number of eggs was 320 (SD = 142.6, range 0-747). The average percentage of unfer- tilized eggs was 1.89% (SD = 3.3, 0-19.1, n = 46); Pig. lA); ~l/5 of the nests had a fertilization rate of 100%. Snout- vent length difference between both parents was not correlated with the number of unfertilized eggs (Spearman’s Rho = 0.098, P = 0.524) or the proportion of unfertilized eggs (Rho = 0.145, P = 0.341). Mean nest volume was 37.0 cm3 (SD = 14.4, range 2.8-85.6, n = lA). Nest volume is correlated to the number of eggs (larger nests have more eggs; Table 1, Pig. 2A) and female size (larger females lay larger Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (27) -3 -2 -1 0 1 2 3 4 5 6 Pair SVL difference (mm) 25 26 27 28 29 30 31 32 33 34 35 36 37 SVL female (mm) Fig. 1. Size and fecundity rates for amplectant pairs of Engysto- mops pustulatus. After collected in amplexus in the field, pairs were left in plastic containers where they could spawn. (A) Pro- portion of unfertilized eggs among pairs that successfully built a nest, (B) Female vs. male snout-vent length (SVL) with linear regression and 95% confidence intervals (dashed lines). nests; Table 1, Pig. 2B). A multiple regression of num- ber of eggs, female SVL, and male SVL explains 25% of the variation in nest volume {F = 7.51, df = 66, P < 0.001). However, only number of eggs is signifi- cant for the regression model (F = 18.18, P < 0.001). Number of eggs was significantly correlated with male SVL (Pig. 2C) but not with female SVL. Number of eggs was correlated with female mass before and after oviposition and female condition (Table 1). Non-spawning females had large masses of eggs in their abdomens (mean relative egg mass = 0.354, SD = 0.138, range 0.129-0.621, n = 13). Average maternal investment for spawning females was 15.2% of body weight (SD = 7.77, 1.8-39.4, n = 42). Size assortment and spawning We found size-assortative mating as male and female size of amplectant pairs was correlated (ANOVA’s F = 24.1, P < 0.001, R^ = 0.176; Pig. IB). Overall, females were significantly larger than their mates (n= 115; mean female SVL = 31.0 mm, SD = 1.9, range 25.3-36.5; male August 2014 | Volume 8 | Number 1 | e79 Ron et al. 0 10 20 30 40 50 60 70 80 90 Nest volume (cm3) Female SVL (mm) Male SVL (mm) Fig. 2. Bivariate plots for (A) nest volume vs. number of eggs, (B) females size vs. nest volume, and (C) male size vs. num- ber of eggs in Engystomops pustulatus. Linear regressions with 95% confidence intervals (dashed lines), determination coef- ficients (R^), and ANOVA’s P values are shown. SVL = 28.5, SD = 1.3, 25.2-32.3; paired-f = 14.7, df = 104, P < 0.001). However, in 10 pairs (8.6%) the male was larger. Mean SVL difference between amplectant male and female was 2.5 mm (SD = 1.74, range -2. 6-6.7, n = 105). Reproductive behavior Males began calling immediately after dusk. They called while floating in temporary ponds with water <10 cm deep. Male density at some choruses was high, result- ing in some males calling a few centimeters away from Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (28) each other. Males defended calling sites and aggressive interactions ensued if another male approached within a radius of <10 cm. Aggressive behavior consisted of mew-like vocalizations and attempts to clasp the rival male. Amplexus and egg deposition occurred at the same ponds where choruses were calling. Amplexus is axillary. Operational sex ratio at choruses was male-biased. During a survey at La Mana, we recorded 14 males but only one female; at Cerro Blanco, the ratio was 3:1 (n = 16 individuals); at Patricia Pilar, the ratio was 8.5:1 (n = 19). The average ratio is 8.5:1 (n = 3 surveys). Spawning behavior . — Nests were built while in am- plexus, on shallow water, next to vegetation or muddy banks. The following description is based on an amplec- tant pair found on 04 February 2004 at -1:00 AM (male QCAZ 26672, SVL = 26.2 mm, hereafter referred as a-male; female QCAZ 26671, SVL = 31.7 mm) building a foam nest. At the beginning of the observation, the nest already had a diameter >50 mm. The male remained in amplexus until the couple left the nest (50 minutes later). To form the foam, the male kicked the egg mass while they were being extruded from the female’s vent. Kick- ing occurred in regular bursts with intervening periods during which the couple was inunobile. In a typical burst cycle, the male’s legs move downward, presumably to place his feet next to the female vent. Then, the male’s feet move up until they reach the posterior end of his dorsum. At that moment, usually one or two eggs become visible in the jelly matrix between the feet. This is fol- lowed by a series of -20 rapid kicks on which his legs become partly extended backward and then distended forward until reaching the posterior end of his dorsum. During these kicks, his legs move simultaneously but in opposite directions (forward-backward) and feet momen- tarily touch medially. The burst ends with to 2-A forceful kicks on which his legs are nearly completely extended posterolaterally, partly removing the foam that lies im- mediately behind the couple. Each male burst seems to be triggered by an abdominal movement of the female. Each burst of kicking lasted on average 4.64 s (SD = 0.53, range 2.13-6.22, n = 215); the intervening inuno- bile periods lasted 9.25 s (SD = 12.15, range 0.12-119, n = 215). Total duration of bursts was 16’30” during 50’ of observation. The duration of each burst and the num- ber of bursts decreased during the second half of the se- quence (Fig. 3). Multimale nesting behavior . — Multimale spawn- ing was only observed once, during the spawning event described in the previous section (male QCAZ 26672, female QCAZ 26671). From the beginning of the obser- vation, a peripheral adult male (QCAZ 26673; hereafter referred as p-male) was sitting on the nest edge, directly opposite to the nesting couple and with the posterior Vi of its body embedded in the foam (Fig. 4). On at least flve occasions its body moved slightly from side to side in se- quences that lasted 3-4 s (Fig. 3). The movements were always in concert with the kicking bursts of the a-male. August 2014 | Volume 8 | Number 1 | e79 Reproduction and spawning behavior in Engystomops pustulatus Table 1. Pearson’s correlation coefficients and ANOVA’s P values for linear regressions. Body condition is defined as the residuals between SVL and mass. SVL = snout-vent length. Variable 1 Variable 2 R2 n P Nest volume Female size (SVL) 0.082 lA 0.013* Nest volume Female mass (before oviposition) 0.020 56 0.287 Nest volume Male size (SVL) 0.026 lA 0.168 Nest volume Male condition <0.001 lA 0.796 Nest volume Number of eggs 0.241 70 <0.001* Number of eggs Male size (SVL) 0.051 76 0.049* Number of eggs Female size (SVL) 0.027 76 0.151 Number of eggs Female mass (before oviposition) 0.111 62 0.008* Number of eggs Female mass (after oviposition) 0.128 58 0.006* Number of eggs Female condition 0.125 62 0.005* Number of eggs Male condition 0.011 76 0.352 * Significant at P <0.05 Burst no. 10 ^ 6 Z3 JD N- 4 o ^ 6 2 ! 2 0 Fig. 3. Spawning of Engystomops pustulatus nesting couple (QCAZ 26671-72) and P-male (QCAZ 26673). Above: dura- tion of kicking bursts. Below: number of bursts per minute; as- terisks indicate P-male movements in the foam. Measurements are shown in sequence from the beginning of the observation until the couple left the nest. See text for details. Most likely, the movements were generated by kicking bursts of the p-male legs (hidden below the foam). He left 23 min later, apparently following an amplectant couple (not collected) that approached at a distance of 10 cm from the nest (see below). The p-male (SVL = 25.3 mm) was one of the small- est in the population. Out of 49 calling males measured during the same season, only three were smaller (mean SVL = 27.55 mm, SD = 1.23); out of 59 males found in amplexus, only one was smaller (mean SVL = 27.9 mm, SD = 1.20). Assuming a normal SVL distribution, the probability of drawing a male with equal or lower SVL by chance is 0.020 (z-score = -2.058). On a sample of seven males including the p-male, mean testes mass was 0.47% of total body mass (range 0.24-0.70%; mean body mass = 1.59 g, SD = 0.28). Contrary to our expectations, the p-male had the proportionally smallest testes. Discussion Clutch size, fertilization success, and parental investment Number of eggs/clutch in Engystomops pustulatus is -37% higher than in the tungara frog (Ryan 1985). In several anurans, clutch size is significantly correlated to body size (e.g.. Crump 1974; Ryan 1985; Wells 2007). In E. pustulatus, such a relationship was significant for fe- male condition and gravid and non-gravid female mass. However, the relationship was not significant for female SVL. Interestingly, we also found a significant correla- tion between number of eggs and male SVL suggesting that larger males have a higher reproductive success. This correlation could not be explained by indirect cor- relations with the other measured variables because they were either uncorrelated with male SVL (e.g., nest vol- ume) or uncorrelated with number of eggs (e.g., female SVL). Female SVL and number of eggs are correlated with nest volume. Nest foam also results from intense male physical activity. However, we were unable to find a re- lationship between nest volume and either male size or male body condition (Table 1). We found size assortative mating as large females have a tendency to mate with large males. An adaptive explanation for size assortative mating states that it in- creases fertilization rates because it results in female and male vents being closer during amplexus (Licht 1976). Evidence for this scenario has been reported for the Tun- gara frog (Ryan 1985). Engystomops pustulatus lacks Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (29) August 2014 | Volume 8 | Number 1 | e79 Ron et al. Fig. 4. Engystomops pustulatus nesting couple (QCAZ 26671- 72) and P-male (QCAZ 26673). The couple builds the foam nest as the male kicks the egg masses extruded by the female. Frame from video (infra-red recording). See text for details. Fig. 5. Relationship (in log space) for body and testis mass among 11 species of Leptodactylinae frogs. Except for Engys- tomops pustulatus, data is from Prado and Haddad (2003). Open circles indicate species on which multimale spawning has been reported. Note that E. pustulatus, in which multi-male spawning apparently occurs, also has larger testis than other Leptodactylinae. that relationship as demonstrated by couples with large differences in size {2-A mm) showing high fertilization rates (Fig. lA). The lack of influence of size difference on fertilization may be explained by our observation of spawning behavior because the male uses his feet to drag the eggs from the female’s underside to his own vent. Therefore, the relative position of male and female vents may have a minor influence in the relative position of eggs and released sperm. Fertilization rates are generally high (more than 98% on average) suggesting that size differences between male and female have little influence in individual fitness. Similar results have been reported in other explosive breeding anurans like Lithobates syl- vaticus (Howard and Kluge 1985) and Anaxyrus cogna- tus (Krupa 1988). Size assortative mating could also result from non- adaptive interactions. If small males mating with large females are more easily displaced than large males mat- ing with large females, a size correlation will result (e.g., Howard and Kluge 1985). This mechanism seems unlikely in Engystomops. During our fieldwork with E. pustulatus and with other species of Engystomops in the Choco and the Amazon region, we never saw unmated males attempting to displace amplectant males. Attempts were rare in E. pustulosus and all of them were unsuc- cessful (Ryan 1985). Therefore, an explanation for size assortative mating in E. pustulatus and its sister species, E. puyango (reported by Ron et al. 2010) is pending. Reproductive investment (or effort) is a measure of the allocation of energy in reproduction relative to total energy (Pianka 2011). Theory predicts that a high repro- ductive investment should be more adaptive if females are unlikely to survive to another reproduction event (Williams 1966). Our estimate of mean reproductive in- vestment for Engystomops pustulatus (15.2%; egg mass relative to body mass) is relatively high in comparison to other anurans. For example. Crump (1974) and Prado and Haddad (2005) report investments ranging from 3.1 to 18.2% for 34 Neotropical species (including nine lep- todactylids). The investment of E. pustulatus, however, is not the highest recorded for an anuran. For example, the myobatrachid Crinia signifera invests 25.9% of the gravid female mass in each spawning event (Lemckert and Shine 1993). This high investment was interpreted as resulting from a low probability of survival to additional spawning events (Lemckert and Shine 1993). Similarly, we hypothesize that the observed large investment in E. pustulatus could result from low survival rates. Nesting behavior Overall, nest building behavior was similar to that re- ported for the Tungara frog (Dalgetty and Kennedy 2010; Heyer and Rand 1977) and Physalaemus ephip- pifer (Hodl 1990). The kicking bursts observed in E. pustulatus are comparable to the “rotational movements” described in P. ephippifer except that the legs seem to extend further backwards in E. pustulatus (compared to figure 5 in Hodl 1990). Nest building is an energetically costly task (Ryan 1985) and the observed decrease in the frequency of kicking bursts towards the end of spawning (Fig. 3) was also reported in the tungara frog (Ryan 1985) and Phy- salaemus ephippifer (Hodl 1990). As in Leptodactylus labyrinthicus, the Tungara frog, and P. ephippifer, kick- ing bursts seemed to be triggered by a female abdominal movement (Heyer and Rand 1977; Hodl 1990; Silva et al. 2005). The movie quality did not allow us to deter- mine whether the decrease in burst frequency was male or female-driven in E. pustulatus. Multimale mating behavior Our observation of more than one male spawning with a female during oviposition suggests that multiple pater- nity and alternative reproductive strategies may exist in Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (30) August 2014 | Volume 8 | Number 1 | e79 Reproduction and spawning behavior in Engystomops pustulatus Engystomops pustulatus. Although the P-male was not in amplexus, its movements were similar and in synchrony with those of the amplectant male, suggesting that it was attempting to fertilize eggs (clutch piracy). A similar re- productive behavior (with synchronic leg movements) has been reported in Leptodactylus chaquensis although with up to seven males in addition to the amplectant male (Prado and Haddad 2003). Egg fertilization by periph- eral non-amplectant males has also been demonstrated in Chiromantis xerampelina, a foam-nesting rhacophorid (Jennions and Passmore 1993). The evolution of multimale spawning should be fa- cilitated in reproductive systems where: (1) the opera- tional sex ratio is strongly male biased, (2) fertilization is external, (3) fecundity is high, and (4) eggs are spatially aggregated (Byrne and Roberts 2004; Shuster and Wade 2003). All this characteristics are part of the reproduc- tion of E. pustulatus. Therefore, the occurrence of multi- male spawning was probable. As previously reported in the Tungara frog (Ryan 1983), our data suggests that an individual male is frequently unable to fertilize all the eggs of a clutch, even in the absence of sperm competi- tion. Although the presence of unfertilized eggs suggests the potential for fitness gain of a p-male sneaking into the nest of an amplectant pair, the proportion of unfertil- ized eggs was typically low (1.89% on average). Higher fitness gains for the p-male may result from sperm com- petition. We could not determine the frequency of multimale spawning in the population. We observed monoandrous spawning frequently and multimale spawning was only recorded once, suggesting that it is relatively infrequent. This is consistent with observations across a variety of taxa showing that p-male strategies exist at a low fre- quency in natural populations (Shuster and Wade 2003; but see Byrne 2002; Jennions and Passmore 1993). The low number of reports of multimale spawning among leptodactylids is surprising because the characteristics of the reproductive system of Leptodactylidae should favor the evolution of secondary male mating strategies. The paucity of known cases may be, at least partly, a sam- pling artifact because the reproductive behavior has been described in only few species. Acknowledgments. — Fieldwork in 2003 and 2004 was funded by NSF IRCEB grant 0078150 to D. Can- natella. Fieldwork in 2008 was funded by a DGA grant from Pontificia Universidad Catolica del Ecuador to S. R. Ron. The Ecuadorian Ministerio de Ambiente pro- vided research and collection permits 004-IC-FAU-DPF, and 006-IC-FAU-DBAP/MA. Fieldwork at la Mana was assisted by F. P. Ayala, M. A. Guerra, and S. Padilla and at Patricia Pilar by A. Argoti, P. Arias, F. Camacho, I. Narvaez and A. Teran. Jose R. Ron, G. M. Melo, and R. Valdivieso provided logistic support in Quito. Help- ful comments for the manuscript were provided by X. E. Bernal. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (31) Literature Cited Byrne PG. 2002. Climatic correlates of breeding, simul- taneous polyandry and potential for sperm competi- tion in the frog Crinia georgiana. Journal of Herpe- tology 36: 125-129. Byrne PG, Roberts JD. 1999. Simultaneous mating with multiple males reduces fertilization success in the myobatrachid frog Crinia georgiana. Proceedings of the Royal Society Biological Sciences Series B 266: 717-721. Byrne PG, Roberts JD. 2004. Intrasexual selection and group spawning in quacking frogs {Crinia georgia- na). Behavioral Ecology 15: 875-882. Chanson JS, Hoffman M, Cox NA, Stuart SN. 2008. The state of the world’s amphibians. Pp. 34-52. In: Stuart, SN, Hoffman, M, Chanson, JS, Cox, NA, Berridge, RJ, Ramani, P, Young, BE (Editors) Threatened Am- phibians of the World. lUCN, Conservation Interna- tional, Barcelona, Spain. 758 p. Crump ML. 1974. Reproductive strategies in a tropical anuran community. Miscellaneous Publications Mu- seum of Natural History University of Kansas 61: 1 - 68 . Dalgetty L, Kennedy MW. 2010. Building a home from foam — tungara frog foam nest architecture and three- phase construction process. Biology Letters 6: 293- 296. Duellman WE, Tmeb L. 1994. Biology of Amphibians. The John Hopkins University Press, Baltimore, Mary- land, USA. 672 p. Frost DR. 2014. Amphibian Species of the World: An on- line reference. Version 6.0 (15 April 2014). Museum of Natural History, New York, New York. Available: http://research.amnh.org/vz/herpetology/amphibia/ American [Accessed: 15 April 2014]. Gorzula SJ. 1977. Foam nesting in Leptdactylids: A possible function. British Journal of Herpetology 5: 741-743. Heyer WR. 1969. The adaptive ecology of the species groups of the genus Leptodactylus (Amphibia: Lepto- dactylidae). Evolution 23: 421^28. Heyer WR, Rand AS. 1977. Foam nest construction in leptodactylid frogs Leptodactylus pentadactylus and Physalaemus pustulosus (Amphibia, Anura, Lepto- dactylidae). Jo 11: 225-228. Hodl W. 1990. An analysis of foam nest construction in the Neotropical frog Physalaemus ephippifer (Lepto- dactylidae). Copeia 1990: 547-554. Howard RD, Kluge AG. 1985. Proximate mechanisms of sexual selection in wood frogs. Evolution 39: 260- 277. Jennions MD, Passmore NI. 1993. Sperm competition in frogs: Testis size and a ‘sterile male’ experiment on Chiromantis xerampelina (Rhacophoridae). Biologi- cal Journal of the Linnean Society 50: 211-220. August 2014 | Volume 8 | Number 1 | e79 Ron et al. Kaminsky SK. 1997. Bufo americanus (American toad). Reproduction. Herpetological Review 28: 84. Krupa JJ. 1988. Fertilization efficiency in the Great Plains toad {Bufo cognatus). Copeia 1988: 800-802. Kusano T, Toda M, Fukuyama K. 1991. Testes size and breeding systems in Japanese anurans with special reference to large testes in the treefrog Rhacophorus arboreus Amphibia Rhacophoridae. Behavioral Ecol- ogy and Sociobiology 29: 27-32. Lemckert FL, Shine R. 1993. Costs of reproduction in a population of the frog Crinia signifera (Anura, Myo- batrachidae) from Southeastern Australia. Journal of Herpetology 27: 420-425. Licht LE. 1976. Sexual selection in toads {Bufo ameri- canus). Canadian Journal of Zoology 5 1277-1284. Menin M, Giaretta AA. 2003. Predation on foam nests of leptodactyline frogs (Anura: Leptodactylidae) by lar- vae of Beckeriella niger (Diptera: Ephydridae). Jour- nal of Zoology 261: 239-243. Pianka ER. 2011. Evolutionary Ecology. Addison Wes- ley Longman, Inc., San Francisco, California, USA. 512 p. Prado CPA, Haddad CEB. 2003. Testes size in leptodac- tylid frogs and occurrence of multimale spawning in the genus Leptodactylus in Brazil. Journal of Herpe- tology 37: 354-362. Prado CPA, Haddad CEB. 2005. Size-fecundity relation- ships and reproductive investment in female frogs in the Pantanal, south-western Brazil. Herpetological Journal 15: 181-189. Ron SR, Read M. 2012. Engystomops pustulatus. In: Ron SR, Guayasamm JM, Yanez-Munoz, M Merino- Viteri, A Ortiz, DA (Editors). AmphibiaWebEcuador. Version 2014.0. Museo de Zoologia, Pontificia Uni- versidad Catblica del Ecuador. Available: http://zoo- logia.puce.edu.ecWertebrados/anfibios/FichaEspecie. aspx?Id=1356 [Accessed: 02 March 2014]. Ron SR, Toral E, Rivera M, Teran- Valdez A. 2010. A new species of Engystomops (Anura: Leiuperidae) from southwestern Ecuador. Zootaxa 2606: 25-49. Ryan MJ. 1983. Sexual selection and communication in a Neotropical frog, Physalaemus pustulosus. Evolution 37: 261-272. Ryan MJ. 1985. The Tungara Frog: A study in sexual selection and communication. The University of Chi- cago Press, Chicago, Illinois, USA. 230 p. Shuster SM, Wade MJ. 2003. Mating Systems and Strate- gies. Princeton University Press, Princeton, New Jer- sey, USA. 533 p. Sierra R, Ceron C, Palacios W, Valencia R. 1999. Mapa de vegetacion del Ecuador Continental 1:1 ’000.000. Proyecto INEFAN/GEF-BIRF, Wildlife Conservation Society y Ecociencia, Quito. Silva WR, Giaretta AA, Facure KG. 2005. On the natural history of the South American pepper frog, Leptodac- tylus labyrinthicus (Spix, 1824) (Anura : Leptodac- tylidae). Journal of Natural History 39: 555-566. Wells KD. 2007. The Ecology and Behavior of Amphibi- ans. The University of Chicago Press, London, United Kingdom. 1148 p. Williams GC. 1966. Adaptation and Natural Selection. Princeton University Press, Princeton, New Jersey, USA. 307 p. Received: 16 May 2014 Accepted: 22 July 2014 Published: 08 August 2014 Santiago R. Ron is the curator of amphibians and professor at the Pontificia Universidad Cat61ica del Ecuador in Quito. His research focuses on the evolution and diversity of Neotropical amphibians and the evolution of animal communication and sexual selection. In the area of conservation biology, Santiago is interested in the study of amphibian extinctions in the Andes. He is a founding member of the Ecuadorian Academy of Sciences. Andrea E. Narvaez is a Ph.D. student at La Trobe University (Australia) and holds a Master’s degree in Integrative, Evolutionary Biology and Infectious Diseases from the Universite Eran9ois Rabelais de Tours ^ liW (Erance). She is currently investigating the evolution of visual signaling of Ecuadorian anoles. Her research focus includes ecology and animal behavior, mainly to evaluate the evolution of communication systems, sexual and natural selection. She is interested in the use of quantitative tools to measure animal behavior and has experience working with a variety of animals (frogs, crickets, lizards). Giovanna E. Romero is a Research Associate at Museo Ecuatoriano de Ciencias Naturales, Botany Section (QCNE). She holds a bachelor degree in biological sciences from Pontificia Universidad Catblica del Ecuador and a Master’s degree in Plant Biology from the University of Texas in Austin. She has worked for many years in the Galapagos Islands and has a deep knowledge of their flora and fauna. She has been collaborating with QCNE since 2012. She is interested in taxonomy and digital curation of ferns and lycophytes. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (32) August 2014 | Volume 8 | Number 1 | e79 Amphibian & Reptiie Conservation 8(1) [Special Section]: 33-44. Copyright: © 2014 Guayasamin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education purposes only provided the original author and source are credited. The official publication credit source: Amphibian & Reptile Conservation at: amphibian-reptile-conservation.org High prevalence of Batrachochytrium dendrobatidis in an Andean frog community (Reserva Las Gralarias, Ecuador) ^Juan M. Guayasamin, ^Mngela Maria Mendoza, "^Ana V. Longo, "^Kelly R. Zamudio, and ^^Elisa Bonaccorso ^Centro de Investigacion de la Biodiversidad y Cambio Climdtico (BioCamb), Universidad Tecnologica Indoamerica, Calle Machala y Sabanilla, Quito, ECUADOR ^Laboratorio de Macroecologia, Centro de Investigaciones en Ecosistemas, Universidad Nacional Autonoma de MEXICO ^Grupo de Investigacion en Ecologia y Conservacion Neotropical, SAMANEA Eundacion de Apoyo Educative e Investigative, COLOMBIA "^Department of Ecology and Evolutionary Biology, Cornell University, Ithaca, New York, USA ^Biodiversity Institute, University of Kansas, Lawrence, Kansas, USA Abstract— Wfe report patterns of infection of Batrachochytrium dendrobatidis {Bd) in a cloud forest amphibian community in the Andean Western Cordillera of Ecuador (Reserva Las Gralarias). Data were obtained during the rainy seasons of two consecutive years, using qPCR (year 2012) and end- point PCR (year 2013). We show that average Bd prevalence in this amphibian community is high (2012: 35-49%; 2013: 14-32%), but found no evidence of population declines or that Ed is negatively affecting host populations. We found a significant correlation between Ed prevalence and taxonomy, reproductive mode, and habitat, but no correlation between Ed infection intensity and the same three variables. Contrary to our expectations, frog species with aquatic reproductive modes (glassfrogs, treefrogs) showed lower Ed prevalence than direct-developing frogs {Pristimantis spp.). Although further monitoring is needed to determine long-term population trends, our two-year dataset on disease and population size support the hypothesis that frogs are tolerant to infection, a condition that could potentially have resulted from exposure to previous Ed epidemic outbreaks. Resumen. — En este estudio reportamos dates sobre los patrones de infeccion de Batrachochytrium dendrobatidis {Bd) en una comunidad de anfibios en la Cordillera Occidental de los Andes del Ecuador (Reserva Las Gralarias). Los dates fueron obtenidos durante la estacion Iluviosa en dos ahos consecutivos, utilizando qPCR (aho 2012) y PCR de punto final (aho 2013). Los resultados muestran una alta prevalencia de Eden la comunidad (2012: 35-49%; 2013: 14-32%); sin embargo, no se encontro evidencia de disminuciones poblacionales o de que Ed este afectando negativamente a las especies de anfibios. Existe una relacion significativa entre la prevalencia de Edy la taxonomia, modo reproductive y habitat de los anfibios, pero no hubo correlacion entre la intensidad de infeccion de Ed y las mismas tres variables. Contrario a nuestras predicciones, las especies de anuros con larvas acuaticas (ranas de cristal, ranas arboreas) presentaron prevalencias de Ed mas bajas que los anuros de desarrollo directo {Pristimantis spp.). A pesar de que se requiere de un monitoreo continue para determinar las dinamicas poblacionales a largo plazo, los dates obtenidos hasta el memento apoyan un escenario donde las especies de ranas de la Reserva Las Gralarias parecen tolerar la infeccion de Ed, una condicion posiblemente adquirida mediante la exposicion a brotes epidemicos previos. Key words. Chytridiomycosis, emerging disease, amphibian declines, Andes, conservation Palabras claves. Quitridiomicosis, enfermedad emergente, declinacion de anfibios, Andes, conservacion Citation: Guayasamin JM, Mendoza AM, Longo AV, Zamudio KR, Bonaccorso E. 2014. High prevalence of Batrachochytrium dendrobatidis in an Andean frog community (Reserva Las Gralarias, Ecuador). Amphibian & Reptiie Conservation 8(1) [Special Section]: 33-44 (e81). Correspondence. Emails: ^jmguayasamin® gmail.com (Corresponding author). August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (33) Guayasamin et al. Introduction A third of global amphibian species are threatened with extinction (Stuart et al. 2004; Wake and Vredenburg 2008) and, most concerning, numerous local popula- tion declines and extinctions have occurred in relatively pristine areas, where anthropogenic habitat destruction is low (Lips 1998, 1999; Drost and Fellers 1996; La Marca et al. 2005). In the last two decades, several studies have attributed anuran mass mortality events to the emergence of the pathogenic fungus Batrachochytrium dendroba- tidis (Bd), a pathogen with widespread geographic and ecological distribution (Berger et al. 1998; Daszak et al. 1999, 2003; Lips et al. 2006; Becker and Zamudio 2011; Rodriguez et al. 2014). Alternative explanations to am- phibian declines add a role to global warming and tem- perature variability (Pounds et al. 2006; Rohr and Raff el 2010; Menendez-Guerrero and Graham 2013). Batrachochytrium dendrobatidis infects the keratin- ized skin of amphibians and disrupts the regulatory func- tioning of the integument (Berger et al. 1998; Voyles et al. 2009). Infection inhibits host inunune responses in some species (Fites et al. 2013) and in severe cases of infection, electrolyte depletion and osmotic imbalance may lead to mortality (Voyles et al. 2007, 2009). How- ever, not all amphibian species are equally susceptible to the pathogen. For example, at Santa Fe, Panama, Bd has caused declines or local extinctions of most anurans in the original community, but six species of frogs and toads remain abundant, despite being infected by the fungus (Lips et al. 2006). In laboratory challenge experiments, amphibian mortality rates range from 0% to 100%, de- pending on the species, host age, pathogen genotype, and dosage (Berger et al. 2005a; Daszak et al. 2004; Longo et al. 2014). The reasons for host differences in susceptibil- ity include immunogenic variation (Ellison et al. 2014; Savage et al. 2014), microhabitat use (Kriger and Hero 2007; Griindler et al. 2012), association with water as embryos, tadpoles, or adults (Lips et al. 2003), and host thermoregulatory behavior (Richards-Zawacki 2010). Because Bd transmission may happen through frog-frog contact, or through motile zoospore movement from one host to another, frogs and toads that spend more time in water are expected to have higher exposure and suscep- tibility to infection than species that are primarily terres- trial (i.e, direct developers; Lips et al. 2003; Kriger and Hero 2007). In this study, we report data on infection patterns of Bd obtained during the rainy seasons of two consecu- tive years in the amphibian community of Reserva Las Gralarias, a cloud forest site in the Andean Western Cor- dillera of Ecuador. We found that Bd prevalence in all amphibian species is high, but found no evidence that Bd is negatively affecting amphibians (i.e., no apparent population declines, or records of clinical signs of chy- tridiomycosis). We report on infection prevalence and intensity dynamics for the two-year period, and examine Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (34) patterns of Bd infection in species that vary in their tax- onomy, reproductive mode, and habitat. Materials and Methods Study site: The study was conducted at Reserva Las Gralarias (0°0L S, 78°44’ W; 1822-2400 m), a private reserve covering an area of 1,063 acres (425 ha) located on the Pacific slopes of the Andes, Pichincha Province, Mindo Parish, Ecuador. The study site has an eleva- tional range of 1,825-2,400 m and includes primary and secondary forest, regenerating pasture, and numerous ephemeral and permanent streams and creeks (Hutter and Guayasanfin 2012). Amphibian taxonomy: For generic and suprageneric clas- sification, we follow the taxonomic proposals of Hedges et al. (2008), Guayasamin et al. (2009), and Faivovich et al. (2005), Pyron and Wiens (2011), as sununarized in Frost (2014). Amphibian richness and abundance: During the rainy seasons of 2012 (23 January-29 March) and 2013 (14 March-22 April), we sampled trails of Reserva Las Gralarias during the night, including most of its habitat heterogeneity and elevational gradient, to record the spe- cies richness of the reserve. We placed eight transects, each with an area of 500 x 4 m (Appendix 1), to maxi- mize species detection and to obtain a baseline dataset on population size and Bd prevalence. Each transect was sampled by two people for 3-4 hours during the night (generally starting at 8 pm); temperatures during sam- pling varied between 11-15 °C. All detected amphibians were, when possible, photographed. Calling males were also reported and identified with the aid of photographic and acoustic guides (Arteaga et al. 2013; Centro Jambatu 2011-2014). We used a Student’s t-test to quantify differ- ences in population sizes in transects that were sampled multiple times during the rainy season of 2012 and 2013 (Lucy’s Creek and Kathy’s Creek); the normality of spe- cies abundance was assessed using a Shapiro-Wilk Test. Diagnosis of Batrachochytrium dendrobatidis: We swabbed the ventral regions of all amphibians captured in our survey, following the standard procedures in Hyatt et al. (2007; Fig. 1); dry swabs were stored in -4 °C until analysis. Testing for Bd was carried out using Real-Time PGR (q-PCR) for samples obtained during 2012 and end- point Polymerase Chain Reaction (PCR) for samples ob- tained in 2013; the use of these two methods was contin- gent on access to q-PCR (available during 2012). In both cases, DNA extractions were carried out using guanidin- ium thiocyanate. For samples obtained during 2012, we used a 1:10 dilution of the extract as template in Taqman q-PCR assays for the detection of Bd (Boyle et al. 2004). This assay uses BJ-specific primers ITS 1-3 Chytr and 5.8S Chytr, in addition to the fluorescently-labeled probe Ch)4r MGB2, and amplifies the ITS-1 fragment of the August 2014 | Volume 8 | Number 1 | e81 Batrachochytrium dendrobatidis in an Andean frog community Fig. 1. Swab sample obtained from Centrolene heloderma at Reserva Las Gralarias, Ecuador. Bd genome at the junction of the ITS-1 and 5.8S regions. We used a standard curve that included 1000, 100, 10, 1, and 0.1 zoospore genome equivalents, and followed qPCR conditions described in Boyle et al. (2004). For samples obtained during 2013, Bd presence was tested using the internal transcribed spacer regions (ITS-1, ITS- 2) primers Bdla (5’-CAGTGTGCCATATGTCACG-3’) and Bd2a (5’-CATGGTTCATATCTGTCCAG-3’) de- veloped by Annis et al. (2004); the presence/absence of Bd was determined via the visualization of the amplified band in agarose gel electrophoresis. The two methods to detect Bd have different sensitivities; therefore, direct comparisons of Bd prevalence between years should be considered with caution. However, family and habitat correlates with infection status should not be biased by detection method, and qPCR offers the additional advan- tage of quantifying infection intensity (load). Prevalence and correlates of Batrachochytrium den- drobatidis in amphibians: We estimated prevalence of Bd within each anuran species as the number of frogs that tested positive for Bd, divided by the total number of sampled frogs for that particular species in a given year. We estimated the 95% confidence interval for preva- lence in each species, (Wilson 1927; Newcomb 1998). We modeled Bd presence or absence in each individual by using a logistic regression. We tested for possible as- sociations of Bd prevalence with the following variables: habitat (terrestrial, riparian, lentic), reproductive mode (aquatic, terrestrial), and taxonomy (family). Statistical significance of results was assessed with a chi-square test. Infection intensity of Batrachochytrium dendrobati- dis and correlates in amphibians: We tested for possible associations of Bd intensity (measured as zoospore ge- nomic equivalents) with the following variables: habitat (terrestrial, riparian, lentic), reproductive mode (aquatic, terrestrial), and taxonomy (family, genus, species; Ap- pendix 2). Given the strong right skew of infection load, we used the non-parametric Kruskal- Wallis test. All sta- tistical analyses were performed using R v. 2.15.3 (R CoreTeam 2012). Results Species richness and abundance: During the two sam- pling periods, we recorded a total of 2,450 individuals of 28 species (Appendix 2). The abundance of species at Lucy’s Creek and Kathy’s Creek is summarized in Tables 1 and 2. Because most taxa were scarce, we restricted the comparisons between years to relatively abundant species (glassfrogs). Abundances of glassfrogs at Lucy’s Creek and Kathy’s Creek were not significantly different between years (Tables 1, 2). Prevalence of Batrachochytrium dendrobatidis in amphibians: Swabs of 320 frogs were tested for Bd, and O Fig. 2. Significant Bd infection differences in amphibians according to reproductive modes, habitat use, and taxonomy. P values are reported for 2012 and 2013; significance is noted by ** (p < 0.01) and *** (p < 0.001). August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (35) Guayasamin et al. Table 1. Abundance of amphibian species at Lucy’s Creek, Reserva Las Gralarias. Abundances are presented as minimum-maxi- mum, followed, in parenthesis, by mean + standard error. The Student’s t-test was performed only in species with normally distrib- uted abundances. Lucy’s Creek Mest (p) Population trend Year 2012 2013 Number of nights sampled 10 6 Family/Species Centrolenidae Centrolene lynchi 5 - 20 ( 13 . 1 + 5 . 13 ) 2-18 ( 9.0 + 6 . 54 ) 0.184 No difference Centrolene peristictum 4-35 ( 20.7 + 11 . 68 ) 6 - 26 ( 15.2 + 7 . 00 ) 0.314 No difference Nymphargus grandisonae Hylidae 1 - 10 ( 6.2 + 3 . 12 ) 0-8 ( 3.8 + 2 . 93 ) 0.155 No difference Hyloscirtus alytolylax Craugastoridae 3-7 ( 4.3 + 1 . 42 ) 0 - 6 ( 3.3 + 2 . 16 ) 0.295 No difference Pristimantis achatinus 0-2 ( 0.3 + 0 . 67 ) 0-1 ( 0.2 + 0 . 41 ) — — Pristimantis appendiculatus 0-2 ( 0.8 + 0 . 92 ) 0-1 ( 0.5 + 0 . 55 ) — — Pristimantis calcarulatus 0-2 ( 0.2 + 0 . 63 ) 0-1 ( 0.2 + 0 . 41 ) — — Pristimantis eremitus 0 0-2 ( 0.5 + 0 . 84 ) — — Pristimantis eugeniae 0 0-2 ( 0.3 + 0 . 82 ) — — Pristimantis illotus 0-1 ( 0 . 1 + 0 . 32 ) 0 — — Pristimantis parvillus 0-1 ( 0 . 1 + 0 . 32 ) 0 — — Pristimantis sobetes 0 0-2 ( 0.3 + 0 . 82 ) — — Pristimantis w-nigrum 0-2 ( 0.6 + 0 . 84 ) 0-2 ( 0.7 + 0 . 82 ) — — approximately a third of those were positive. In samples from 2012, prevalence of Bd was relatively high, with 42% of all frogs testing positive for Bd infection. Dur- ing 2013, Bd prevalence was 22%. Differences in preva- lence between the two years are likely caused by detec- tion method. Most species infected in 2012 carried low Bd loads as determined by qPCR; the highest Bd load obtained was in Centrolene ballux with 22.5 genomic equivalents. Prevalence per species per year is summa- rized in Table 3. The logistic regression shows a significant relation- ship (p < 0.001) of Bd infection with species reproduc- tive mode, habitat, and taxonomy (Fig. 2). Frogs with a terrestrial reproductive mode (direct developers; i.e., genus Pristimantis; see Duellman and Trueb 1986) have a higher Bd prevalence than amphibians with aquatic re- production (i.e., glassfrogs and treefrogs). Frog species that are dependent on riverine habitats for reproduction show significantly less infection than anurans that use terrestrial or lentic habitats for reproduction (p < 0.001). Also, species in the Centrolenidae family (glassfrogs) show a lower Bd prevalence than species in Craugastori- dae and Hylidae (Table 3). Although, Bd prevalence dur- ing 2012 was significantly higher than in 2013 (probably as a result of higher sensitivity of qPCR), we found no significant interaction among sampling year and repro- ductive mode, habitat, or taxonomy. Infection intensity of Batrachochytrium dendrobati- dis and correlates in amphibians: We found no relation- ship between Bd infection intensity {Bd load, year 2012) and taxonomy, reproductive mode, or habitat. Discussion Our results show a relatively high mean prevalence of Bd (36%) across both years in the Andean frog concunu- nity of Reserva Las Gralarias (see Hossack et al. 2010 for comparison). From a total of 20 species analyzed, only three (Nymphargus griffithsi, Pristimantis illotus, and P. pteridophilus) tested negative for Bd\ however, sample sizes for non-infected species were low (5, 1, and 3 in- dividuals, respectively). Infected species included frogs with very different reproductive modes, including taxa with terrestrial direct development {Pristimantis spp.), species that deposit eggs in ponds {Dendropsophus car- nifex), and others that place their eggs on vegetation from where hatching tadpoles drop into streams {Centrolene spp., Nymphargus spp., Hyloscirtus spp.). Because Bd is an aquatic pathogen (Berger et al. 2005b) we expected amphibian species with aquatic re- productive modes to show higher infection prevalence (Lips et al. 2005; Brem and Lips 2008). In fact, the most dramatic amphibian declines and extinctions in the Andes have occurred in species with aquatic larvae (La Marca et al. 2005; Bustamante et al. 2005; Merino-Viteri et al. 2005; Coloma et al. 2010). Our results indicate, surpris- ingly, a higher Bd prevalence in frogs with a terrestrial reproductive mode {Pristimantis spp.) than in those that August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (36) Batrachochytrium dendrobatidis in an Andean frog community Table 2. Abundance of amphibian species at Kathy’s Creek, Reserva Las Gralarias. Abundances are presented as minimum-maxi- mum, followed, in parenthesis, by mean + standard error. The Student’s t-test was performed only in species with normally distrib- uted abundances. Kathy’s Creek f-test Population trend Year 2012 2013 Number of nights sampled 10 5 Family/Species Centrolenidae Centrolene ballux 5-37 ( 22.7 + 11 . 6 ) 3-25 ( 11.4 + 8 . 67 ) 0.078 No difference Centrolene peristictum 0-5 ( 2.1 + 1 . 66 ) 0-5 ( 2 . 0 + 1 . 87 ) 0.918 No difference Nymphargus grandisonae 0 - 7 ( 3.7 + 2 . 21 ) 0-6 ( 2.2 + 2 . 28 ) 0.242 No difference Nymphargus griffithsi 0-8 ( 2.3 + 2 . 26 ) 0-3 ( 1 . 4 + 1 . 34 ) — — Nymphargus lasgralarias Hylidae 3-28 ( 19.4 + 8 . 54 ) 7-28 ( 15.0 + 8 . 69 ) 0.366 No difference Hyloscirtus alytolylax Craugastoridae 0-1 ( 0 . 1 + 0 . 32 ) 0-1 ( 0.4 + 0 . 59 ) — — Pristimantis achatinus 0-3 ( 0.3 + 0 . 95 ) 0-1 ( 0.2 + 0 . 45 ) — — Pristimantis appendiculatus 0 - 7 ( 1.2 + 2 . 10 ) 0-1 ( 0.4 + 0 . 59 ) — — Pristimantis calcarulatus 0-3 ( 1 . 1 + 0 . 74 ) 0-3 ( 1 . 0 + 1 . 23 ) — — Pristimantis eremitus 0-1 ( 0 . 1 + 0 . 32 ) 0-1 ( 0.2 + 0 . 45 ) — — Pristimantis eugeniae 0-1 ( 0 . 1 + 0 . 32 ) 1-2 ( 0.8 + 0 . 84 ) — — Pristimantis sobetes 0-1 ( 0.2 + 0 . 42 ) 0-2 ( 0.4 + 0 . 89 ) — — Pristimantis w-nigrum 0-1 ( 0 . 1 + 0 . 32 ) 0-1 ( 0.4 + 0 . 59 ) — — reproduce in water (mainly glassfrogs; Centrolene spp., Nymphargus spp.). This finding supports the idea that even terrestrial breeders may serve as reservoirs for the pathogen in diverse amphibian communities (Longo et al. 2013). Higher prevalence in terrestrial frogs requires that Bd zoospores survive in terrestrial habitats. John- son and Speare (2003) indicated that Bd can survive in moist soil for up to three months. Cloud forests in west- ern Ecuador typically have near constant rain and high levels of humidity during the rainy season (Hutter and Guayasamin 2012; Arteaga et al. 2013), and this may ex- tend zoospore survival in terrestrial environments at Las Gralarias. Higher Bd prevalence in terrestrial frogs com- pared to that in frogs with aquatic reproduction might also be related to intrinsic differences in, for example, the efficacy of immune responses (Rosenblum et al. 2009; Woodhams et al. 2007) or differences in anuran skin mi- crobiota (Flechas et al. 2012). Our results also show that Bd prevalence is signifi- cantly associated with taxonomy (i.e., family). Thus, glassfrogs (family Centrolenidae) might have innnune responses or skin microbiota that work as better barri- ers to the pathogen than those in terrestrial (i.e., Pristi- mantis) frogs. The strong correlation of prevalence with taxonomy, habitat, and reproductive mode (Appendix 2) indicates that further studies need to focus on the specific effects of each of these factors; in other words, phytog- eny (and taxonomy) correlates with reproductive mode and habitat use. A second surprising finding of our study is that, al- though prevalence of Bd is high in most anuran species, we did not observe any sign of population declines or abrupt crashes, nor have we found dead or sick frogs dur- ing four years of intensive fieldwork (2010-2014; JMG pers. obs.). Thus, this frog connnunity persists with an endemic pathogen and with relatively low loads (less than 10 zoospores. Table 3). The apparent increased re- sistance or tolerance of amphibians from Reserva Las Gralarias to Bd infection may be explained by one or several of the following mechanisms: (i) amphibian in- nate and/or acquired defense mechanisms (Savage and Zamudio 2012; Woodhams et al. 2007); (ii) skin bacte- rial connnensals with anti-fungal properties (Harris et al. 2006); (iii) behavioral and ecological factors that reduce the likelihood of infection and disease (e.g., microhabitat selection, reproductive mode; Lips et al. 2003; Rowley and Alford, 2007), and/or (iv) variation in Bd virulence (Berger et al. 2005a; Fisher et al. 2009). Our amphib- ian monitoring took place during the rainy season when most species are active. Therefore, future studies should determine whether or not this apparent tolerance to Bd is stable through longer periods of time or if it fluctuates depending on environmental variables influencing host innnunity, behavior, microbiota, or pathogenicity. After the emergence of an infectious disease, surviv- ing hosts can evolve tolerance or resistance (Retallick et al. 2004; Savage and Zamudio 2011). The earliest known record of Bd in Ecuador is in 1980, in the Harlequin frog Atelopus bomolochos (Ron and Merino-Viteri 2000), a species that is now probably extinct (Coloma et al. 2014). If Bd reached and spread in Ecuador during the early 1980s (Ron et al. 2003; Lips et al. 2008), we hy- August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (37) Guayasamin et al. Table 3. Prevalence of Batrachochytrium dendrobatidis (Bd) in amphibians at Reserva Las Gralarias, Ecuador, during the rainy seasons of 2012 and 2013. Bd prevalence for each species is followed, in parenthesis, by a 95% confidence interval. Bd load sum- mary data includes only samples that tested positive for Bd. 2012 2013 N Positive Bd Prevaience Bd Load (mean ± sd) N Positive Bd Prevaience Family: Centrolenidae Centrolene ballux 17 8 47% (24-71%) 6.5 + 10.7 9 2 22% (39-59%) Centrolene heloderma 6 1 17% (1-63%) 0.6 1 0 0% (0-95%) Centrolene lynchi 6 1 17% (1-63%) - 5 1 20% (1-70%) Centrolene peristictum 21 6 29% (12-52%) 2.1 + 1.5 16 3 19% (5-46%) Nymphargus grandisonae 21 5 24% (9-48%) 4.7 - - - Nymphargus griffithsi 3 0 0% (1-69%) - 2 0 0% (0-80%) Nymphargus lasgralarias 16 6 38% (16-64%) 6.5 + 1.5 10 2 20% (4-56%) Family: Hylidae Dendropsophus carnifex 10 5 50% (20-80%) - - - - Hyloscirtus alytolylax 9 8 89% (51-99%) 2.4 + 2.4 7 2 29% (5-70%) Family: Craugastoridae Pristimantis achatinus 7 4 57% (20-88%) - - - - Pristimantis appendiculatus 23 10 44% (24-65%) 1.9 + 2.4 - - - Pristimantis calcarulatus 15 2 13% (2-42%) 1.1 15 1 7% (4-34%) Pristimantis eremitus 4 4 100% (40-100%) 0.9 + 0.2 4 2 50% (9-91%) Pristimantis eugeniae 18 12 66% (41-86%) 2.5 2 1 50% (3-97%) Pristimantis hectus 8 2 25% (4-64%) - 14 4 29% (10-58%) Pristimantis illotus - - - - 1 0 0% (0-95%) Pristimantis parvillus 9 4 44% (15-77%) - - - - Pristimantis sobetes 8 3 38% (10-74%) - 9 3 33% (9-69%) Pristimantis pteridophilus - - - - 3 0 0% (0-69%) Pristimantis w-nigrum 21 13 62% (39-81%) 1.5 + 0.7 - - - TOTAL 222 94 42% (35-49%) 88 19 22% (14-32%) pothesize that many of the population declines observed in the country at that time (e.g., Coloma 1995, 2002; Coloma et al. 2000; Ron et al. 2003; Bustamante et al. 2005; La Marca et al. 2005; Lips et al. 2008; Coloma et al. 2010) could be attributable to chytridiomycosis. Thus, it is probable that most Andean amphibian communities have been exposed to Bd for more than three decades and that current sampling finds remnant species that are tol- erant to Bd while the susceptible species are already ex- tinct. Under this scenario, selection should have favored the persistence of amphibian species or specific popula- tions that have developed defenses against Bd; therefore. August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (38) Batrachochytrium dendrobatidis in an Andean frog community changing host composition of these communities right after pathogen emergence. We acknowledge, however, that this is a working hypothesis which assumptions depend on our knowledge of the historical distribution of the chytrid. For example, if new data shows that Bd was present in the Ecuadorian Andes before amphibian declines were noticed, such piece of information would support the endemic pathogen hypothesis, which states that environmental changes triggered Bd outbreaks (Ra- chowicz et al. 2005). Reserva Las Gralarias is one of the most studied and species-rich area in the cloud forest of the tropical An- des, containing numerous species considered endangered by the lUCN (2014; see Appendix 2). However, the community (and surrounding areas) lacks at least three groups of species that were conspicuous in Ecuadorian cloud forests: marsupial frogs {Gastrotheca plumbea, G. guentheri), harlequin frogs {Atelopus longirostris, A. mindoensis), and dendrobatid frogs (Hyloxalus lehmani, H. maquipucuna) (Coloma et al. 2011-2014; Arteaga et et al. al. 2013). Marsupial and harlequin frogs are par- ticularly susceptible to Bd (Lips et al. 2003; Elechas et al. 2012; Ellison et al. 2014; DiRenzo et al. 2014) and are the primary species that suffered population declines and extinctions in Ecuador (Lips et al. 2002; La Marca 2005) even in pristine areas. The absence of these lineages at Reserva Las Gralarias supports to the hypothesis that this is a post-decline amphibian community. Understanding the long-term effects of pathogens (eg., chytrid) and tem- perature variability in such a community is essential for the continued effective management of endangered spe- cies in the Andean cloud forests. Considering Ecuador’s high diversity of amphibian species, life history modes, and evolutionary history, our study provides a baseline to study the evolution of de- fense strategies against Bd. We reconnnend further re- search to determine the mechanisms driving the observed differences in pathogen exposure among hosts differing in reproductive modes, habitat, and taxonomy. Acknowledgments. — Previous versions of this article greatly benefited from the reviews of Luis A. Coloma, and five anonymous reviewers. This study was funded by the lUCN Save-Our-Species (SOS) program and Uni- versidad Tecnologica Indoamerica, through the project “Conservation of Endangered Species in the Choco Bio- geographic Zone: Integrating habitat management, bio- logical monitoring, and community outreach.” SOS is a joint initiative of lUCN, the Global Environment Eacility and the World Bank; its objective is to ensure the long- term survival of threatened species and their habitats. AMM’s research was supported by a scholarship from the program “Becas Mixtas del Consejo Nacional de Ciencia y Tecnologfa, CONACYT,” Mexico. Idea Wild kindly granted equipment for AMM’s field work. Reser- va Las Gralarias (Jane A. Lyons) provided invaluable lo- gistic support throughout the project. Special thanks to Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (39) Henry Imba, Italo G. Tapia, Lucas Bustamante, and Jai- me Garcia for assistance during fieldwork. JMG thanks AA for the one-day delay in Miami, where the final ver- sion of the ms was finished. Diana Flores provided as- sistance with molecular diagnosis of Bd. Special thanks to John Kelly for his help in statistical analyses. Research permits were issued by the Ministerio de Ambiente, No 05-2013-IC-FAU-DPAP-MA. Literature Cited Annis SL, Dastoor FP, Ziel H, Daszak P, Longcore JE. 2004. A DNA-based assay identifies Batrachochytri- um dendrobatidis in amphibians. Journal of Wildlife Diseases 40: 420^28. Arteaga A, Bustamante L, Guayasamin JM. 2013. Am- phibians and Reptiles ofMindo: Life in the Cloud For- est. Serie de Publicaciones Cientificas, Universidad Tecnologica Indoamerica, Quito, Ecuador, 1: 1-257. Becker, CG, Zamudio KR. 2011. Tropical amphibian populations experience higher disease risk in natural habitats. Proceedings of the National Academy of Sci- ences of the United States of America 108: 9893-9898. Berger L, Speare R, Daszak P, Green DE, Cunningham AA, Goggin CL, Slocomber R, Ragan MA, Hyatt AD, McDonald KR, Hines HB, Lips KR, Marantelli G, Parkes H. 1998. Chytridiomycosis causes amphib- ian mortality associated with population declines in the rain forests of Australia and Central America. Pro- ceedings of the National Academy of Sciences of the United States of America 95: 9031-9036. Berger L, Marantelli G, Skerratt LL, Speare R. 2005a. Virulence of the amphibian chytrid fungus Batracho- chytrium dendrobatidis varies with the strain. Diseas- es of Aquatic Organisms 68: 47-50. Berger L, Hyatt AD, Speare R, Longcore JE. 2005b. Life cycle stages of Batrachochytrium dendrobatidis Longcore et al. 1999, the amphibian chytrid. Diseases of Aquatic Organisms 68: 51-63. Brem F, Lips KR. 2008. Patterns of infection by Batra- chochytrium dendrobatidis among species, habitats and elevations during epidemic and endemic stages. Diseases of Aquatic Organisms 86: 189-202. Bustamante MR, Ron SR, Coloma LA. 2005. Cambios en la diversidad en siete comunidades de anuros en los Andes de Ecuador. Biotropica 37: 180-189. Centro Jambatu. 2014. Anfibios de Ecuador. Fundacidn Otonga, Quito, Ecuador. Electronic database avail- able : http ://www.anfibioswebecuador. ec/anfibiosec- uador.aspx [Accessed: 01 August 2014]. Coloma LA. 1995. Ecuadorian frogs of the genus Colo- stethus (Anura: Dendrobatidae). University of Kansas Natural History Museum Miscellaneous Publications 87: 17-2. Coloma LA. 2002. Two new species of Atelopus (An- ura: Bufonidae) from Ecuador. Herpetologica 58: 229-252. August 2014 | Volume 8 | Number 1 | e81 Guayasamin et al. Coloma LA, Letters S, Salas AW. 2000. Taxonomy of the Atelopus ignescens complex (Anura: Bufonidae): Designation of a neotype of Atelopus ignescens and recognition of Atelopus exiguus. Herpetologica 56: 303-324. Coloma LA, Duellman WE, Almendariz A, Ron SR, Teran- Valdez A, Guayasamin JM. 2010. Five new (extinct?) species of Atelopus (Anura: Bufonidae) from Andean Colombia, Ecuador, and Peru. Zootaxa 2574: 1-54. Coloma LA, Lotters S, Quiguango-Ubillus A. 2014. At- elopus bomolochos. In: Centro Jambatu. 2011-2014. Anfibios de Ecuador, Fundacibn Otonga, Quito, Ec- uador. Available: www.anfibioswebecuador.ee/fichae- specie.aspx?Id=130 [Accessed: 01 August 2014]. Coloma LA, Guayasamin JM, Menendez-Guerrero P (Editors). (2011-2014) Lista Roja de Anfibios de Ec- uador. AnfibiosWebEcuador, Fundacion Otonga, Qui- to, Ecuador. Available: http://www.anfibioswebecua- dor.ee/listaroja.aspx [Accessed: 01 August 2014]. Daszak P, Berger L, Cunningham A A, Hyatt AD, Green DE, Speare R. 1999. Emerging infectious diseases and amphibian population declines. Emerging Infec- tious Diseases 5: 735-748. Daszak P, Cunningham A A, Hyatt AD. 2003. Infectious disease and amphibian population declines. Diversity and Distributions 9: 141-150. Daszak P, Strieby A, Cunningham AA, Longcore JE, Brown CC, Porter D. 2004. Experimental evidence that the bullfrog (Rana catesbeiana) is a potential car- rier of chytridiomycosis, an emerging fungal disease of amphibians. Herpetological Journal 14: 201-207. DiRenzo GV, Langhammer PE, Zamudio KR, Lips KR. 2014. Fungal infection intensity and zoospore output of Atelopus zeteki, a potential acute chytrid super- shedder. PLoS ONE 9: e93356. doi:10.1371/joumal. pone.0093356 Drost CA, Fellers GM. 1996. Collapse of a regional frog fauna in the Yosemite area of the California Sierra Ne- vada, USA. Conservation Biology 10: 414-425. Duellman WE, Tmeb L. 1986. Biology of Amphibians. McGraw-Hill, New York, New York, USA. 670 p. Ellison AR, Savage AE, DiRenzo GV, Langhammer P, Lips KR, Zamudio K. 2014. Fighting a losing battle: Vigorous immune response countered by pathogen suppression of host defenses in a chytridiomycosis- susceptible frog. G3: Genes, Genomes, Genetics 4(7): 1275-1289. doi: 10.1534/g3.114.010744 Faivovich J, Haddad CEB, Garcia PC A, Frost DR, Camp- bell JA, Wheeler WC. 2005. Systematic review of the frog family Hylidae, with special reference to hylinae: Phylogenetic analysis and taxonomic revision. Bulle- tin of the American Museum of Natural History 294: 1-240. Fisher MC, Bosch J, Yin Z, Stead DA, Walker J, Gow NA, Stajich JE, Gamer TW. 2009. Proteomic and phenotypic profiling of the amphibian pathogen Ba- trachochytrium dendrobatidis shows that genotype is linked to vimlence. Molecular Ecology 18: 415-429. Fites SC, Ramsey JP, Holden WM, Collier SP, Suther- land DM, Reinert LK, Gayek AS, Dermody TS, Aune TM, Oswald-Richter K, Rollins-Smith LA. 2013. The invasive chytrid fungus of amphibians paralyzes lym- phocyte responses. Science 342: 366-369. Flechas SV, Sarmiento C, Cardenas ME, Medina EM, Restrepo S, Amezquita A. 2012. Surviving chytrid- iomycosis: Differential znati-Batrachochytrium den- drobatidis activity in bacterial isolates from three lowland species of Atelopus. PLoS ONE 7: e44832. doi: 10.137 l/journal.pone.0044832 Frost DR. 2014. Amphibian Species of the World: An online reference. Version 6.0 (July 30, 2014). Ameri- can Museum of Natural History, New York, USA. Electronic Database available: http://research.amnh. org/herpetology/amphibia/index.html [Accessed: 01 August 2014]. Grtindler MC, Toledo LF, Parra-Olea G, Haddad CEB, Giasson LOM, Sawaya RJ, Prado CPA, Araujo OGS, Zara FJ, Centeno EC, Zamudio KR. 2012. Interaction between breeding habitat and elevation affects preva- lence but not infection load of Batrachochytrium den- drobatidis in anuran assemblages. Diseases of Aquatic Organisms 91 '. 173-184. Guayasamin JM, Castro viejo-Fisher S, Tmeb L, Ayar- zagtiena J, Rada M, Vila C. 2009. Phylogenetic sys- tematics of glassfrogs (Amphibia: Centrolenidae) and their sister tsixon Allophryne ruthveni. Zootaxa 2100: 1-97. Harris RN, James TY, Lauer A, Simon MA, Patel A. 2006. Amphibian pathogen Batrachochytrium den- drobatidis is inhibited by the cutaneous bacteria of Amphibian species. Ecohealth 3: 53-56. Hedges SB, Duellman WE, Heinicke P. 2008. New world direct-developing frogs (Anura: Terrarana): Molecu- lar phylogeny, classification, biogeography, and con- servation. Zootova 1737: 1-182. Hossack BR, Adams MJ, Campbell Grant EH, Pearl CH, Bettaso JB, Barichivich WJ, Lowe WH, Tme K, Ware JL, Com PS. 2010. Low prevalence of chytrid fun- gus {Batrachochytrium dendrobatidis) in amphibians of U.S. headwater streams. Journal of Herpetology 44:253-260. Hutter CR, Guayasamin JM. 2012. A new cryptic spe- cies of glassfrog (Centrolenidae: Nymphargus) from Reserva Las Gralarias, Ecuador. Zootaxa 3257: 1-21. lUCN: Red List of Threatened Species. 2014. Version 2014.2. Available: www.iucnredlist.org [Accessed: 01 August 2014]. Johnson ML, Speare R. 2003. Survival of Batrachochy- trium dendrobatidis in water: quarantine and disease control implications. Emerging Infectious Disease Journal 9(8). Available: http://wwwnc.cdc.gov/eid/ article/9/8/03 -0145_article [Accessed: 10 August 2014]. August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (40) Batrachochytrium dendrobatidis in an Andean frog community Kriger KM, Hero JM. 2007. The chytrid fungus Batra- chochytrium dendrobatidis is non-randomly distribut- ed across amphibian breeding habitats. Diversity and Distributions 14: 781-788. La Marca E, Lips KR, Lotters S, Puschendorf R, Ibanez R, Rueda-Almonacid JV, Schulte R, Marty C, Cas- tro F, Manzanilla-Puppo J, Garcia-Perez JE, Bolanos F, Chaves G, Pounds JA, Toral E, Young BE. 2005. Catastrophic population declines and extinctions in Neotropical harlequin frogs (Bufonidae: Atelopus). Biotropica 37:1 90-20 1 . Lips KR. 1998. Decline of a tropical montane amphibian fauna. Conservation Biology 12: 106-117. Lips KR. 1999. Mass mortality and population declines of anurans at an upland site in western Panama. Con- servation Biology 13: 117-125. Lips KR, Brem F, Brenes R, Reeve JD, Alford RA, Voyles J, Collins JR 2006. Emerging infectious dis- ease and the loss of biodiversity in a Neotropical amphibian connnunity. Proceedings of the National Academy of Sciences of the United States of America 103: 3165-3170. Lips KR, Reeve JD, Witters LR. 2003. Ecological traits predicting amphibian population declines in Central America. Conservation Biology 37: 222-226. doi: 10.1046/j.l523-1739.2003.01623.x Lips KR, Diffendorfer J, Mendelson III JR, Sears MW. 2008. Riding the wave: Reconciling the roles of dis- ease and climate change in amphibian declines. PLoS Biology 6: e72. doi:10.1371/journal.pbio.0060072 Longo AV, Ossiboff RJ, Zamudio KR, Burrowes PA. 2013. Lability in host defenses: terrestrial frogs die from chytridiomycosis under enzootic conditions. Journal of Wildlife Diseases 49: 197-199. Longo AV, Burrowes PA, Zamudio KR. 2014. Genomic studies of disease-outcome in host-pathogen dynam- ics. Integrative and Comparative Biology {In press). Menendez-Guerrero PA, Graham CH. 2013. Evaluating multiple causes of amphibian declines of Ecuador using geographical quantitative analyses. Ecography 36:756-769. Merino-Viteri A, Coloma LA, Almendariz. 2005. Los Telmatobius (Leptodactylidae) de los Andes del Ec- uador y su declive poblacional. Pp. 9-37 In: Estudios sobre las ranas andinas de los generos Telmatobius y Batrachophrynus (Anura: Leptodactylidae). Editors, Lavilla EO, de La Riva I. Asociacion Herpetologica Espanola, Monografias de Herpetologia 7. Valencia, Espana. Newcombe RG. 1998. Two-sided confidence intervals for the single proportion: comparison of seven meth- ods. Statistics in Medicine 17: 857-872. Pounds JA, Bustamante MR, Coloma LA, Consuegra JA, Fogden MPL, Foster PN, La Marca E, Masters KL, Merino-Viteri A, Puschendorf R, Ron SR, Sanchez- Azofeifa, GA, Still CJ, Young BE. 2006. Widespread amphibian extinctions from epidemic disease driven Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (41) by global warming. Nature 439:161-167. Pyron RA, Wiens JJ. 2011. A large-scale phylogeny of Amphibia with over 2,800 species, and a revised clas- sification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution 61: 543-583. R CoreTeam. 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/ Retallick RW, McCallum H, Speare R. 2004. Endemic infection of the amphibian chytrid fungus in a frog community post-decline. PLoS Biology 2: e351. Richards-Zawacki C. 2010. Thermoregulatory behaviour affects prevalence of chytrid fungal infection in a wild population of Panamanian golden frogs. Proceedings of the Royal Society B 211 \ 519-528. Rohr JR, Raffel TR. 2010. Linking global climate and temperature variability to widespread amphibian de- clines putatively caused by disease. Proceedings of the National Academy of Sciences of the United States of America 107: 8269-8274. Ron SR, Merino-Viteri A. 2000. Amphibian declines in Ecuador: Overview and first report of chytridiomyco- sis from South America. ErogLog 42: 2-3. Ron SR, Duellman WE, Coloma LA, Bustamante MR. 2003. Population decline of the jambato iodid Atelopus ignescens (Anura: Bufonidae) in the Andes of Ecua- dor. Journal of Herpetology 37: 116-126. Rodriguez D, Becker CG, Pupin NC, Haddad CEB, Za- mudio KR. 2014. Long-term endemism of two highly divergent lineages of the amphibian-killing fungus in the Atlantic Forest of Brazil. Molecular Ecology 23:774-787. Rosenblum EB, Poorten TJ, Settels M, Murdoch GK, Robert J, Maddox N, Eisen MB. 2009. Genome- wide transcriptional response of Silurana (Xenopus) tropi- calis to infection with the deadly chytrid fungus. PLoS ONE 4: c6494 doi: 6410.1371/joumal.pone.0006494 Savage AE, Zamudio KR. 2011. MHC genotypes associ- ate with chytridiomycosis resistance in a threatened North American frog. Proceedings of the National Academy of Sciences of the United States of America 108: 16705-16710. Savage AE, Kiemnec-Tyburczy KM, Ellison AR, Fleischer RC, Zamudio KR. 2014. Conservation and divergence in the frog innnunome: Pyrosequencing and de novo assembly of immune tissue transcrip- tomes. Gene 542(2): 98-108. Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues AS, Fischman DL, Waller RW. 2004. Status and trends of amphibian declines and extinctions worldwide. Sci- ence 306: 1783-1786. Voyles J, Berger L, Young S, Speare R, Webb R, Warner J, Skerratt LF. 2007. Electrolyte depletion and osmot- ic imbalance in amphibians with chytridiomycosis. Diseases of Aquatic Organisms 77: 113-118. Voyles J, Young S, Berger L, Campbell C, Voyles WE, August 2014 | Volume 8 | Number 1 | e81 Guayasamin et al. Dinudom A, Speare R. 2009. Pathogenesis of chy- tridiomycosis, a cause of catastrophic amphibian de- clines. Science 326: 582-585. Wake DB, Vredenburg VT. 2008. Are we in the midst of the sixth mass extinction? A view from the world of amphibians. Proceedings of the National Academy of Sciences of the United States of America 105: 11466- 11473 Wilson EB. 1927. Probable inference, the law of succes- sion, and statistical inference. Journal of the Ameri- can Statistical Association 22: 209-212. Woodhams DC, Ardipradja K, Alford RA, Marantelli G, Reinert LK, Rollins-Smith LA. 2007. Resistance to chytridiomycosis varies among amphibian species and is correlated with skin peptide defenses. Animal Conservation 10: 409^17. doi: 10.1371/joumal. pbio.0020351. Received: 19 June 2014 Accepted: 04 August 2014 Published: 28 August 2014 Appendix 1. Transects sampled at Reserva Las Gralarias. Each transect has an area of 500 x 4 m. Latitude and longitude are in decimal degrees. Transect Elevation (m) Latitude Longitude Habitat Lucy’s creek 1822-1858 At start: -0.00492 At end: -0.00342 At start: -78.73344 At end: -78.74051 Riverine vegetation along creek Kathy’s creek 2041-2066 At start: -0.01696 At end: -0.0156 At start: -78.7314 At end: -78.73386 Riverine vegetation along creek Santa Rosa river 1884-1882 At start: -0.0133 At end: -0.01054 At start: -78.72368 At end: -78.7211 Riverine vegetation along river Waterfall trail 1897-2107 At start: -0.0135 At end: -0.01379 At start: -78.72461 At end: -78.7269 Primary and secondary terra firme forest Five-Frog creek 2141-2156 At start: -0.03166 At end: -0.03098 At start: -78.70421 At end: -78.70853 Riverine vegetation along creek Osoverde & Guarumo trail 2141-2156 At start: -0.03166 At end: -0.03098 At start: -78.70421 At end: -78.70853 Primary and secondary terra firme forest Puma trail 1923-2031 At start: -0.00954 At end: -0.00708 At start: -78.7346 At end: -78.73662 Primary and secondary terra firme forest TKA trail 2192-2216 At start: -0.0275 At end: -0.02516 At start: -78.70477 At end: -78.70353 Primary and secondary terra firme forest Peccary trail 1803-1896 At start: -0.00750 At end: -0.0076 At start: -78.72635 At end: -78.72862 Primary and secondary terra firme forest Appendix 2. Amphibians at Reserva Las Gralarias, with corresponding lUCN (2014) conservation status. The list includes three potential new species (Pristimantis sp. 1, Pristimantis sp. 2, and Pristimantis sp. 3). Reproductive modes are sensu Haddad and Prado (2005). Species Reproductive mode Habitat for reproduction Conservation status Family: Centrolenidae (7 spp.) Mode 25: Eggs hatching into exotrophic tadpoles that drop in lotic water Centrolene ballux Mode 25 Vegetation along fast-flowing streams Critically Endangered Centrolene heloderma Mode 25 Vegetation along fast-flowing streams Critically Endangered Centrolene lynchi Mode 25 Vegetation along fast-flowing streams Endangered Centrolene peristictum Mode 25 Vegetation along fast-flowing streams Vulnerable Nymphargus griffithsi Mode 25 Vegetation along fast-flowing streams Vulnerable Nymphargus grandisonae Mode 25 Vegetation along fast-flowing streams Least Concern Nymphargus lasgralarias Mode 25 Vegetation along fast-flowing streams Data Deficient August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (42) Batrachochytrium dendrobatidis in an Andean frog community Appendix 2 (continued). Amphibians at Reserva Las Gralarias, with corresponding lUCN (2014) conservation status. The list includes three potential new species (Pristimantis sp. 1, Pristimantis sp. 2, and Pristimantis sp. 3). Reproductive modes are sensu Haddad and Prado (2005). Species Reproductive mode Habitat for reproduction Conservation status Family: Craugastoridae (16 spp.) Mode 23: Direct development of terrestrial eggs Pristimantis achatinus Mode 23 Terrestrial, mainly in pastures and modified environments Least Concern Pristimantis appendiculatus Mode 23 Terrestrial, mainly primary and secondary forests Least Concern Pristimantis calcarulatus Mode 23 Terrestrial, mainly primary and secondary forests Vulnerable Pristimantis crenunguis Mode 23 Terrestrial, mainly primary and secondary forests Endangered Pristimantis eremitus Mode 23 Terrestrial, mainly primary and secondary forests Vulnerable Pristimantis eugeniae Mode 23 Terrestrial, mainly primary and secondary forests Endangered Pristimantis hectus Mode 23 Terrestrial, mainly primary and secondary forests Data deficient Pristimantis illotus Mode 23 Terrestrial, mainly primary and secondary forests Near Threatened Pristimantis parvillus Mode 23 Terrestrial, mainly primary and secondary forests Least Concern Pristimantis pteridophilus Mode 23 Terrestrial, mainly primary and secondary forests Endangered Pristimantis sobetes Mode 23 Terrestrial, mainly primary and secondary forests Endangered Pristimantis verecundus Mode 23 Terrestrial, mainly primary and secondary forests Vulnerable Pristimantis w-nigrum Mode 23 Terrestrial, mainly primary and secondary forests Least Concern Pristimantis sp. 1 Mode 23 Terrestrial, mainly primary and secondary forests Not evaluated Pristimantis sp. 2 Mode 23 Terrestrial, mainly primary and secondary forests Not evaluated Pristimantis sp. 3 Mode 23 Terrestrial, mainly primary and secondary forests Not evaluated Family: Hylidae (3 spp) Dendropsophus carnifex Mode 1 : Eggs and exotrophic tadpoles in lentic water Ponds Least Concern Hyloscirtus alytolylax Mode 25 Vegetation along fast-flowing streams Near Threatened Hyloscirtus criptico Mode 25 Vegetation along fast-flowing streams Not evaluated Family: Caeciliidae (1 sp.) Caecilia buckleyi Family: Rhinatrematidae (1 sp.) Unknown Unknown Not evaluated Epicrionops bicolor Unknown Unknown Least Concern August 2014 | Volume 8 | Number 1 | e81 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (43) Guayasamin et al. Juan M. Guayasamin is head of the Centro de Investigacion de la Biodiversidad y Cambio Climatico (Bio- Camb) and Professor at Universidad Tecnologica Indoamerica, Quito, Ecuador. He obtained his Master's de- gree and Ph.D. in Ecology and Evolutionary Biology (University of Kansas, EEUU). His research includes am- phibian phylogenetic systematics, taxonomy, biogeography, ecology, and conservation biology {Photographed by Lucas Bustamante). Angela M. Mendoza is a biologist from Universidad del Valle (Colombia) with a Master’s in biological sci- ences at the Universidad Nacional Autonoma de Mexico (UNAM). She is a research assistant in the Conserva- tion Genetics Laboratory at the Alexander von Humboldt Institute, Colombia. Her main interest is to apply molecular tools in ecology and conservation, with emphasis in terrestrial vertebrates, mainly Neotropical am- phibians {Photographed by Angela M. Mendoza). Ana V. Longo is a doctoral student in the Department of Ecology and Evolutionary Biology at Cornell Uni- versity, Ithaca, New York. Her main interests are amphibian disease ecology and evolution. Her research aims to understand the mechanisms underlying seasonal and ontogenetic patterns of disease susceptibility in direct- developing frogs {Photographed by Alberto L. Lopez-Torres). Kelly R. Zamudio is a Professor in the Department of Ecology and Evolutionary Biology at Cornell Univer- sity. She received her B.A. from UC Berkeley in Zoology in 1991, and her Ph.D. from University of Washing- ton, Seattle, in 1996. Her research focuses on the origin and maintenance of vertebrate biodiversity (especially reptiles and amphibians). Her lab integrates field research in population biology, demography, and habitat change with lab research on the genomic underpinnings of population diversification, speciation, and conserva- tion. {Photographed by Karen Lips). Elisa Bonaccorso is a biologist, Ph.D. in Ecology and Evolutionary Biology (University of Kansas, EEUU), and Licenciada en Biologia (Universidad Simon Bolivar, Venezuela). Her research is broad and includes mo- lecular systematics, ecology of wildlife diseases, and conservation biology {Photographed by Juan M. Guayas- amin). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (44) August 2014 | Volume 8 | Number 1 | e81 Anolis podocarpus. Photo by A. Almenddriz. September 2014 | Volume 8 | Number 1 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (45) e82 Amphibian & Reptiie Conservation 8(1) [Special Section]: 45-64. Copyright: © 2014 Almendariz et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education purposes only provided the original author and source are credited. The official publication credit source: Amphibian & Reptile Conservation at: amphibian-reptile-conservation.org Overview of the herpetofauna of the unexplored Cordillera del Condor of Ecuador ^Ana Almendariz, ^John E. Simmons, ^ ^*Jorge Brito, and ^’^Jorge Vaca-Guerrero ^Instituto de Ciencias Biologicas, Escuela Politecnica Nacional, Casilla 17-01-2759, Quito, ECUADOR ^Museologica, 128 Burnside Street, Bellefonte, Pennsylvania 16823 USA Abstract. — ^The Cordillera del Condor is an area rich in unique vegetation assemblages and endemic faunal elements; the herpetofauna is especially diverse, particularly the anurans. The montane forest and sandstone tepuis, located atop large andesite and quartz formations, provide a variety of habitats and microhabitats in which the herpetofauna finds food, shelter, and reproductive sites, such as terrestrial and arboreal bromeliads and a soil type termed “bamba” that is covered with mosses and roots. Information compiled from publications and recent studies has revealed the presence of 120 species of amphibians and 59 species of reptiles, including 41 probable new species (36 amphibians and five reptiles) in the genera Centrolene, Dendrobates, Pristimantis, Lynchius, Chiasmocieis, Boiitogiossa, Anoiis, Erythroiamprus, Tantilia, and Dipsas. Resumen. — La Cordillera del Condor es un area rica en formaciones vegetales unicas y elementos faum'sticos endemicos; presenta una singular diversidad herpetofaunistica, particularmente de la anurofauna. Los bosques montanos y los de “tepuy,” asentados sobre piedras grandes de andesita y cuarzo crean variedad de habitats y microhabitats, en donde la herpetofauna encuentra alimento, refugio y lugares para la reproduccion, como por ejempio las bromelias terrestres y arboreas y un suelo denominado “bamba” que esta cubierto de musgos y raices. La informacion recopilada de material publicado y de los estudios realizados en los ultimos ahos revela la presencia de 120 especies de anfibios y 59 especies de reptiles. Los resultados incluyen 41 especies posiblemente nuevas (36 anfibios y cinco reptiles) de los generos: Centroiene, Dendrobates, Pristimantis, Lynchius, Chismocieis, Boiitogiossa, Anoiis, Erythroiamprus, Tantiiia, y Dipsas. Key words. Ecuador, Cordillera del Condor, amphibian, reptile Citation: Almendariz A, Simmons JE, Brito J, Vaca-Guerrero J. 2014. Overview of the herpetofauna of the unexplored Cordillera del Condor of Ecuador. Amphibian & Reptile Conservation 8(1 ) [Special Section]: 45-64 (e82). Introduction The fauna of Ecuador, in general, has not been extensive- ly studied, despite a notable increase in research activ- ity in recent years (Albuja et al. 2012). In particular, the Cordillera del Condor region, in southern Ecuador along the border with Pern (Figure 1), is a very poorly known area. The purpose of this paper is to suncnnarize and re- view herpetofaunal studies of the Cordillera del Condor region. Studies of the avian and mammalian fauna have been published elsewhere (e.g., Albuja and Patterson 1996; Brito and Arguero 2012; Freile et al. 2014). The long-running border conflicts between Ecuador and Pern and the difficulty in accessing the region have maintained the ecosystems of the Cordillera del Condor almost intact. It has only been since the end of the con- flicts known as the Pasquisha War (which ended in Feb- mary 1981) and the Alto Cenepa War (which ended in Febmary 1995), that roads into the area have begun to open, which has resulted in incipient colonization and an awakened interest in mineral prospection in the region. Nevertheless, there are still some parts of the Cordillera del Condor that remain unaltered. The Cordillera del Condor is part of a biologically di- verse, discontinuous, sub-Andean cordillera that has sev- eral characteristics that distinguish it from the rest of the Andes. Whereas the main Andes are of metamorphic and igneous origin, the Cordillera del Condor is sedimentary, composed largely of limestone and sandstone (Schul- enberg and Awbrey 1997). The region is dominated by *Current address: Museo Ecuatoriano de Ciencias Maturates, Rumi- pamba y Avenida de los Shyris, Quito, ECUADOR Correspondence. Emails: ^ ana. almendariz® epn.edu.ee, ^simmons.johne® gmail.com (Corresponding author, John E. Sim- mons), ^jorgeyakuma® yahoo.es, "^gheovak® hotmail.com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (46) September 2014 | Volume 8 | Number 1 | e82 Almendariz et al. i 77-D'O-W ■— CD _|5 ^Hivnpabhokn COLOMBIA ECUADOR 1 1 ECUAI>0|SE 6 fKta r Mr 1 d Of 1 1 PEHU yiadiriizA tA ■k ^in rungirtzi PERU 0 25 50 100 - - TS-o'™ T7-o'o'W Fig. 1. Map of the Cordillera del Condor region. geologically complex mountains topped with sandstone plateaus at elevations of 300 m to almost 3,000 m that support habitats similar to the sandstone mountains of the Guyana Shield (Figure 2); many of the plateaus have vegetation similar to that of the tepuis. Due to its loca- tion just northeast of the Huancabamba Depression, the lowest point in the Andes (Duellman 1999), the Cordil- lera del Condor receives moisture from both the Atlantic and the Pacific sides of the Andes — moisture from the Western slopes of the Andes as well as moisture moving east across the Amazon basin drops over the Cordillera del Condor, providing frequent, year-round precipitation (Schulenberg and Awbrey 1997). The first systematic botanical studies of the region were carried out in 1990 and 1991 in the Rfo Nangaritza basin (in the southern region of the Cordillera del Con- dor), under the auspices of the Proyecto Promobot and the Tratado de Cooperacion Amazonica, with the partici- pation of both Ecuadorian and foreign scientists. These explorations were limited to areas below about 1,700 m Fig. 2. Alto Paquisha, 2,400 m. Photo by A. Almendariz. in altitude. In 1993, A. H. Gentry collected plants on one of the highest points of the mountain range (2,100 m) as part of a Rapid Assessment Program (RAP) survey, organized by Conservation International (Cl, a non-gov- ernmental organization). Gentry found that the vegeta- tion structure and families of plants were similar to those of the sandstone tepuis of the Guiana Highlands (Schul- enberg and Awbrey 1997). In recent years, D. Neill and his collaborators have surveyed the flora at several points in the cordillera, publishing descriptions of new species and studying the environmental heterogeneity associated with variable types of soils (Neill 2005; Neill and Asan- za 2012; Neill and Ulloa 2011; Riina et al. 2014; Ulloa et al. 2012). The diversity of plant assemblages on the sandstone plateaus produces a variety of microhabitats that provide food, shelter, and reproductive sites for the herpetofauna, particularly terrestrial and arboreal brome- liads; the “bamba” soils of many of these tepuis is thickly covered with mosses and roots, and serves to Alter the tannins that darken the turbid water in creeks and streams (Figure 3). According to the ecosystem classification for conti- nental Ecuador (Ministerio del Ambiente del Ecuador 2012), the following ecosystems have been identified in the Cordillera del Condor: • Evergreen piedmont forest in the Condor-Kutuku ranges • Evergreen forest on the sandstone plateaus of the Condor range in the lower Ecuadorian Amazon Fig. 3. Vegetation in the interior of a tepui forest. Photo by A. Almendariz. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (47) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador • Lower montane evergreen forest in the Condor- Kutuku ranges • Evergreen piedmont forest on the sandstone mesas of the Condor-Kutuku ranges • Evergreen lower montane forests on the sandstone mesas of the Condor-Kutuku ranges • Montane humid shrub in the Condor range • Evergreen montane forests on the sandstone mesas of the Condor-Kutuku ranges • Montane humid shrub with herbaceous rosette thickets (herbazales) in the Condor range Materials and Methods Herpetological surveys of the region have been few and limited (Figure 1). The information presented below is drawn from an extensive survey of the literature and recent field work. The majority of the studies have em- ployed the Rapid Ecological Assessment strategy or RAP developed by Cl (Sayre et al. 2002), in habitats where the presence of herpetofaunal elements was anticipated. The northern zone of the Cordillera del Condor is known from three studies. The first was carried out in 1972 in conjunction with a privately funded orchid col- lecting expedition (accompanied by personnel from the Missouri Botanical Garden and the University of Kansas Museum of Natural History), at elevations of 870-2,000 m at the headwaters of the Rio Piuntza, Rio Chuchum- bleza, Rio Numpatacaimi, and Rio Santa Agueda in Mo- rona Santiago Province (Duellman and Simmons 1988). The second survey was a RAP assessment conducted by Cl, the Escuela Politecnica Nacional, Fundacion Fedima, and the Universidad Nacional Mayor de San Marcos. The areas surveyed included the Ecuadorian flank of the Cordillera del Condor (Coangos and Achupallas in Mo- rona Santiago Province [Figure 4], Miazi and Shaimi in Zamora Chinchipe Province). The Peruvian flank of the Cordillera del Condor was surveyed at the base of Cerro Machinaza, Alfonso Ugarte-PV3, Falso Paquisha-PV22, and Puesto de Vigilancia Comainas. Subsequently, a third survey was carried out by Fundacion Natura (FN 2000) to establish the Parque El Condor, which inven- Fig. 4. Achupallas sector, 2,100 m. Photo by A. Almenddriz. toried the Comunidad Numpatakaime and confluence of the Rio Tsuirim and the Rio Coangos. Another survey, conducted as part of the Proyecto Paz y Conservacion Binacional en la Cordillera del Condor Ecuador-Peru by the Organizacion Intemacional de las Maderas Tropicales, Conservation International, Fun- dacion Natura, and the Institute Nacional de Recursos Naturales (INRENA) in 2005 (Organizacion Intemacio- nal de las Maderas Tropicales and Fundacion Natura y Conservacion Intemacional 2005), collected data from several localities in the southern sector on the Ecuadorian flank of the Cordillera del Condor, including Condor Mi- rador and Herradura. The corresponding Pemvian flank survey was focused on the Zona Reservada Santiago Co- maina. Between March 2008 and July 2012, the Escuela Poli- tecnica Nacional team, under a contract with the Cardno- Entrix Corporation, carried out 16 expeditions to survey the herpetofauna of Alto Manchinaza. In 2009, a Cl RAP survey was conducted by personnel from the Pontificia Universidad Catolica del Ecuador, Louisiana State Uni- versity, and Fundacion Ecologica Arcoiris, with support from Secretaria Nacional de Ciencia y Tecnologia del Ecuador (SENACYT) of the tepuyes of the upper basin of the Rio Nangaritza (Guayasamm et al. 2011). In 2012, the Fundacion Naturaleza y Cultura Intemacional and the Universidad Estatal Amazonica organized an expedition to Cerro Plateado (Figure 5), the southern point of the Cordillera del Condor, which included researchers from the Escuela Politecnica Nacional. Results The 1972 survey of the northern zone of the Cordillera del Condor resulted in the capture of 30 species, includ- ing nine new species (e.g., Duellman and Simmons 1988, Lynch 1974, 1976, 1979; Lynch and Duellman 1980). Specimens and additional records from this survey are deposited in the Biodiversity Research Institute at the University of Kansas, along with additional related spec- imens accounting for 47 species total (Reynolds 1997; Schulenberg and Awbrey 1997). Fig. 5. Peak of Cerrro Plateado, 2,900 m. Photo by V. Carvajal. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (48) September 2014 | Volume 8 | Number 1 | e82 Almendariz et al. Fig. 6. Pristimantis sp. Photo by A. Almendariz. Fig. 7. Centrolene condor. Photo by A. Almendariz. Fig. 8. Excidobates condor. Photo by A. Almendariz. Fig. 9. Enyalioides rubrigularis (female). Photo by A. Fig. 10. Hyloscirtus condor. Photo by J. Brito. Almendariz. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (49) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador The second survey of the northern Zone of the Cordil- lera del Condor recorded a total of 34 species — 27 am- phibians and seven reptiles (Almendariz 1997a, 1997b). Although the survey was conducted under adverse en- vironmental conditions, geographic range extensions for several species were recorded. The Peruvian flank sur- veys recorded 58 species (35 anurans and 23 reptiles); the data from the Peruvian surveys was collected by per- sonnel from the Museo de Historia Natural of the Univer- sidad Nacional Mayor de San Marcos in 1987 (Reynolds and Icochea 1997a, 1997b) and expanded the ranges of two species for Peru, Rhinella festae and Hemiphractus bubalus. The Parque El Condor survey lists a total of 36 species (22 anurans, eight lizards, and six snakes), in- cluding nine species new to the Cordillera del Condor herpetofauna, and makes reference to the difficulty of identifying some of the material. The report summarizes information known up to the year 2000, and lists a total of 95 species for the region (56 amphibians and 39 rep- tiles). The survey of the southern sector of the Cordillera del C6ndor collected specimens that were not identified to the species level of the genus Pristimantis (Figure 6), and a glass frog provisionally identified as Centrolene cf. cro- ceopodes that in 2008 was named as Centrolene condor (Cisneros-Heredia and Morales-Mite 2008; Figure 7). The surveys of Alto Manchinaza recorded 70 species of amphibians and 43 species of reptiles (Almendariz et al. in prep.). The results of these surveys revealed the Fig. 11. Herpetofaunal assemblages and endemics from the Cordillera del Condor. ua UO AMPHSBIANS RtPHlLES □ TdIhI N*5ppcl^ M Pntpntialty Nfw Speaps Fig. 12. Documented and predicted species diversity in the Cordillera del Condor. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (50) presence of several little known or new species, includ- ing a new species of highland poison dart frog, Excido- bates condor (Almendariz et al. 2012; Figure 8) and geo- graphic range extensions for Phyllomedusa ecuatoriana, Centrolene condor, Chimerella mariaelenae, Hyloxalus mystax, and Enyalioides rubrigularis (Figure 9), among others. In addition, ecological data and information on reproduction, vocalizations, and other aspects of the life history for these species was collected (Almendariz and Batallas 2012a, 2012b; Batallas and Brito 2014; Brito et al. in prep.); at least 28 probably new species from differ- ent genera {Centrolene, Bolitoglossa, Pristimantis, Ano- lis, Atractus, Erythrolamprus, Tantilla, and Dipsas) were obtained, which are in the process of being described. This project included environmental education and com- munity outreach work (Almendariz 2012). The 2009 Pontificia Universidad Catolica del Ecua- dor survey recorded 27 species of amphibians and 17 reptiles, including a new species of anuran, Pristiman- tis minimus (Teran- Valdez and Guayasamm 2010). This project included the publication of a field guide to plants and animals of the tepuis of Nangaritza (Almendariz 2010; Freile et al. 2010). The 2012 expedition to Cerro Plateado (Almendariz and Brito 2013) recorded 19 species. Of these, 14 were anurans and salamanders, including nine species of the family Craugastoridae (most were members of the genus Pristimantis). Based on the ecology of the area, it is as- sumed that more species occur at this locality, including members of the family Centrolenidae. A new species of torrent frog, Hyloscirtus condor (Figure 10), was de- scribed based on specimens obtained on this expedition (Almendariz et al. 2014), and at least eight new species in the genera Lynchius, Pristimantis, and Bolitoglossa were obtained that will be described in the future. Discussion The Cordillera del Condor is of particular importance due to its high biodiversity and the presence of several unique ecosystems (e.g., the sandstone formations simi- lar to tepuis). The Condor region, with its diverse range of habitats, contains numerous species that correspond to three faunal components: (1) Amazonian lowlands; (2) eastern flanks of the Andes, and (3) an endemic fauna limited to the southern part of Ecuador (Duellman and Lynch 1988). As shown in Figure 11, the Baja Amazonia herpetofaunal assemblage contains more reptile species (63%) than amphibian species (28%). In the herpetofau- nal assemblages associated with the eastern slopes of the Andes there are slightly more amphibian species (34% and 32%). The percentages relative to the endemic com- ponent of the Cordillera del Condor are notably higher in amphibians than in reptiles (41% and 7%). A summary of the herpetofaunal diversity of the region is provided in Tables 1 and 2; a comparison of known species diversity and predicted species diversity September 2014 | Volume 8 | Number 1 | e82 Almendariz et al. Fig. 13. Lynchius sp. Photo by J. Brito. Fig. 15. Cercosaura dicra. Photo by G. Gallardo. Fig. 14. Pristimantis muscosus. Photo by A. Almendariz. Fig. 16. Erythrolamprus sp. Photo by A. Almendariz. Fig. 17. Tantilla sp. Photo by A. Almendariz. Fig. 18. Anolis podocarpus. Photo by A. Almendariz. for the region is provided in Figure 12. Some of the more distinctive species found in the region include frogs of the genera Lynchius (Craugastoridae; Figure 13) and Pristimantis (Craugastoridae; Figure 14), the gynmoph- thalmid lizard Cercosaura dicra (Figure 15), and the colubrid snakes Erythrolamprus (Figure 16) and Tantilla (Figure 17). During the last five years, the following species have been described based on material from the Cordillera del Condor: Enyaliodes rubrigularis (Torres-Carvajal et al. 2009; Figure 9), Anolis podocarpus (Ay ala- Varela and Torres-Carvajal 2010; Figure 18), Pristimantis minimus (Teran- Valdez and Guayasanun 2010), Excidobates con- dor (Almendariz et al. 2012; Figure 8), Hyloscirtus con- dor (Almendariz et al. 2014; Figure 10), and Siphlophis ayauma (Sheehy et al. 2014). The work has expanded the known geographic distribution of Anolis soini (Ayala- Varela et al. 2011; Figure 19) and revealed new distribu- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (51) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador tion and natural history information for two other species of the genera Centrolene and Hyloxalus (Almendariz and Batallas 2012a, 2012b). It is also noteworthy that several new species are in the process of being described in the genera Pristimantis and Chiasmocleis (Almendariz et al. in prep.). A detailed publication about the herpetofauna of the Condor region is in preparation (Almendariz et al. in prep.). Based on information published in the most recent studies conducted in the Cordillera del Condor, there are a total of 120 amphibian species (11 families, 31 gen- era), and 59 reptile species (nine families, 28 genera), not including those found at elevations below 850 m (Fig- ure 20). In addition, based on the specimens discussed above, the area contains approximately 41 potentially new species (36 amphibians and five reptiles; see Figure 12). These numbers indicate that the region has signifi- cant endemic diversity (see Table 1, Table 2, and Figure 11 ). Conclusion Within the Cordillera del Condor, four areas protected by the Sistema Nacional de Areas Protegidas have been es- tablished: (1) Reserva Biologica El Condor; (2) Reserva Biologica El Quimi; (3) Reserva Biologica Cerro Platea- do; and (4) Refugio de Vida Silvestre El Zarza. On the eastern flank, the Peruvian government has concentrated its efforts to create Parque Nacional Ichigkat Muja-Cor- dillera del Condor (SERNANP 2012). Nevertheless, the ecosystems in the Cordillera del Condor are threatened by imminent human colonization and settlement, the in- troduction of agriculture and livestock, and mining; the latter activity poses the greatest threat to the conserva- tion of the tepui-like forests and the health of the aquatic ecosystems, which are the reproductive habitats of many species of anurans, including hylids and centrolenids. On the other hand, the fact the amphibians of the Andean re- gion have limited distribution makes them susceptible to extinction, and in some cases, the protection of their hab- itat does not improve their chance of survival (Guayas- amin et al. 2011). This situation warrants intensified research and conservation studies of these vertebrates, especially in little explored areas as in the case of the Cordillera del Condor. The preliminary results of surveys of Alto Machinaza and Cerro Plateado have revealed the presence of possibly new species in these areas, indicat- ing that future interventions in these areas should comply strictly with the measures to protect ecosystems, environ- mental mitigation, and management plans. Acknowledgments. — We thank the Kinross and Card- no-Entrix Corporation and their administrative and field staff for use of facilities to carry out recent held stud- ies, and Eundacion Naturaleza y Cultura Internacional and David Neill of the Universidad Estatal Amazonica for the invitation to participate in the expedition to Cerro Fig. 19. Anolis soini. Photo by J. Vaca G. Fig. 20. Familial, generic, and specific diversity of amphibians and reptiles in the Cordillera del Condor. Plateado. Thanks also to Bruce MacBryde and the late Milan D. Eiske for the opportunity to participate in the 1972 expedition into the Cordillera del Condor. Literature Cited Albuja L, Almendariz A, Barriga R, Montalvo ED, Ca- ceres E, Roman JS. 2012. Fauna de Vertebrados del Ecuador. Instituto de Ciencias Biologicas, Escuela Politecnica Nacional, Quito Ecuador. Albuja L, Patterson B. 1996. A new species of northern shrew-opossum (Paucituberculata: Caenolestidae) from the Cordillera del Condor, Ecuador. Journal of Mammalogy 77: 41-53. Almendariz A. 1997a. Reptiles and amphibians of the Cordillera del Condor. Pp. 80-82 In: The Cordillera del Condor Region of Ecuador and Peru: A Biologi- cal Assessment. Editors, Schulenberg T, Awbrey K. RAP Working Papers number 7, Conservation Inter- national, Washington, D.C., USA. 234 p. Available: https://library.conservation.org/Published%20Docu- ments/2009/07%20RAP%20Working%20Papers.pdf [Accessed: 30 August 2014]. Almendariz A. 1997b. Amphibian and reptile species recorded in the Northern and Western Cordillera del Condor. Pp. 199-201 In: The Cordillera del Con- dor Region of Ecuador and Peru: A Biological As- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (52) September 2014 | Volume 8 | Number 1 | e82 Table 1. Species of amphibians recorded from the Cordillera del Condor. Almendariz et al c/3 CD T3 CD C/3 CO GO 0 CD o 0 z 2 0 o o 4— » 0 C] Q_ LU -•— » (— O S’o CD o 00 E c c/3 E X X o CO N c 0 E .c 0 c be c '3 0 >4 _l Q_ 0 C] o3 00 o CO 00 o LO LO X X Q_ < DC c c/3 0 T3 DC O CO 0 ■F^ 05 8 2 o o LO LO to to X X X N 0 , '0 0 c CM E Q_ ^ T3 03 CM 0 < < C 03 A 9 X 0 Q F 0 ^ E C33 X Q. E Cfl F E '0 o -i ^ ^ 3 Z 0 ,0 0 o D3 c 0 O o C 0 _ c '9 ^ 8 ^ ^ -S o LL 0 « 8 O LO o o CO C33 X X X 0 T3 0 .Q 0 3 o Z 2 (f) o N 0 "5 X 3 C 0 o ii 2 o N 0 - Q_ 0 > LO o o 0 CM 0 c o O o is o ^ o E o c 0 - E XXX X c 0 0 N O 0 § “ Cfl 0 C O 3 CM 0 O LO 00 ci> LO C33 X XX X X XXX o J- 0 0 B O i5 Q_ N "0 T3 C 0 CO T3 c 0 CO O CM O CD 00 CM ci> O CD X 0 N 0 C Z o 0 O < 0 0 N "0 T3 C 0 o o CM ci> LO 00 o LO 00 CD o CO X X X X X c _o 'iS 0 O O > o c _o 'ip 0 > 0 LU Z, 03 at .'2 'b s -C) •S L. (30 K S L. -s; •2 C3 S a C3 C3 CL, a C3 .S l! 03 CO 1 s C3 L. ■§ s 0 CJ « 03 R 0 -C) s Ch Ch Co 't? < cs '2 .2 "o S s A 0 .'2 s s Co s Co s -2 C3 s a -C) '2 0 u § g ■4.^ 0 u 0 s a a 03 s r'“ s r'“ < tt os os os U 0 0 C3 S i*i<. •2 C3 2 s Cl, CJ 5 ■ 4 .^ C3 -C) O •S C3 tt: C3 CJ S C3 CJ Co s C3 'Ch a K 5 -s; CJ 0 CJ s s? a -s; 1 g- "§ 0 Ch s a -s; 1 a a o e s & cs e 5 s 0=; L. .a CJ a K 5 s OS JL» o K g ■+.«* SS Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (53) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador Rio Puintza Duellman & Lynch 1988 1550-1830 msnm X X X X X X X X RAP 7 Reyn- olds & Icochea 1997 665-1750 msnm X X RAP 7 Al- mendariz 1997 900-2200 msnm X X X X X X X X X Nump- atkaim / rios Tsuirim y Coangos Fun- dacion Natura 2000 930-1050 msnm X X X Quebrada Shinga- natza Proy. Paz y Conserva- cion, 2005 (Peru) 850-1200 msnm X 0 T3 0 0 CO CO 0 z o Q_ CO LJJ Q ^ 0 0 0 C O) f N CD .y 0 o ■I— » 0 o T- C A ^ O ^ 00 0 0 0 0 o o 0 0 o N Q- Q_ 0 > LO o o 0 CM 0 c o O o is O o E o c 0 - E X X X X c 0 0 N O 0 5 “ C« 0 C O O CM ^ I 0 O LO 00 cb LO 05 X X X X o N o o CO o T3 0 0 "0 T3 c CO CX5 CM E c 0 o 0 0 F T3 c o CM cb o 0 E Q_ 0 CO X X X 0 N 0 C !c o 0 O < 0 0 N "0 T3 C 0 E < o o CM cb LO 00 o LO 00 cb o 00 XXX X X XXX X X s> c _o 'ip 0 O o >. o •t-> 3 < c _o 'p 0 > 0) LU >■> o fl 05 CQ K o > o d t/3 "C o iW CJ -a s •1 fl d t/3 i-<* % CQ -a •3 •3 g I9JD •4.^ K CQ CJ s; CJ s; bl u L. a s -s; Cl a •2 K a L. a. s 05 •2 s CJ _a •3 _a '3 o K? i-<* 05 s 3 •3 a 3 a g o "ts K o 3 05 05 g a o 05 •2 •2 •2 • S2 K SS K SS a a .§ .1 .§ .1 •Ct •2 •2 "C ‘C a. a. a a 3 ..3 *>2 a *>2 |h 05 .5 *>2 *>2 *>2 C«*5 Co s s .5 S s *>2 a a *>2 § g' *>2 .a •S a .Si 05 05 a. p. 05 a, g a, g a, • S2 •2 •2 •2 •2 •2 SS Si Si Si ■4.^ s Si Si ■4.^ s ■4.^ s Si Si a a a .§ .§ .§ .§ .3 .§ .§ .3 .3 .§ .§ ■4.^ .*>2 .*>2 ■4.^ ■4.^ .*>2 .*>2 ‘C '£! '£! a. a. a. a. s 5 c pi a s a s s a L. a. s -s; .c •C^ "§ o "S L. a. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (54) September 2014 | Volume 8 | Number 1 | e82 Almendariz et al Rio Puintza Duellman & Lynch 1988 1550-1830 msnm X RAP 7 Reyn- olds & Icochea 1997 665-1750 msnm RAP 7 Al- mendariz 1997 900-2200 msnm X Nump- atkaim / rios Tsuirim y Coangos Fun- dacion Natura 2000 930-1050 msnm X Quebrada Shinga- natza Proy. Paz y Conserva- cion, 2005 (Peru) 850-1200 msnm 0 T3 0 0 0 CO 0 z o Q_ 0 LJJ Q ^ 0 0 0 C O) f N CD .y 0 o ■I— » 0 o T- C A ^ O ^ 00 o O 0 0 o o 0 0 o N 0- Q_ 0 > LO o o 0 CM 0 c o O o is O o E o c 0 - E c 0 0 N O 0 5 “ 0 0 C O 0 CM ^ I 0 O LO 00 cb LO 05 X X X X o N O O CO o T3 0 0 "0 T3 c CO 00 CM E c 0 O 0 0 F T3 C O CM cb o 0 E Q_ 0 CO 0 N 0 C !c o 0 O < 0 0 N "0 T3 C 0 E < o o CM cb LO 00 o LO 00 cb o 00 X XXX X X XXX X X X ‘C a 5 o -s; s .a. -o a '*»<» s o s g -s; CJ o K -s; CJ a a < PQ (N m m CD r-- 00 Oc o (N m g Oh cL cL cL d cL cL cL cL cL cL cL d d d a O o O O _ >t c •+5 •+5 •+5 •+5 •+5 •+5 •+5 c o s: s: s: s: s: s: s: sr sr sr sr sr s s s s s s s s s s s o ?i ?i ?i ?i ?i ?i ?i ?i ?i ?i ?i ?i C3 C3 C3 C3 o (0 s s s s s s s s s s s s s s s s s s s s s s s (w > o *>2 *>2 *>2 *>2 *>2 *>2 *>2 *>2 *>2 *>2 *>2 *>2 o 3 _l < LU a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. a. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (55) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador Rio Puintza Duellman & Lynch 1988 1550-1830 msnm X RAP 7 Reyn- olds & Icochea 1997 665-1750 msnm RAP 7 Al- mendariz 1997 900-2200 msnm X X Nump- atkaim / rios Tsuirim y Coangos Fun- dacion Natura 2000 930-1050 msnm X Quebrada Shinga- natza Proy. Paz y Conserva- cion, 2005 (Peru) 850-1200 msnm T3 CO Z to m CO 0 CO c O) f N CD .y CO o ■I— » CO o T- C A ^ O ^ 00 X X O O 0 CO o O CO CO >, =- o N 0- Q_ CO o ^ o -g 0 < k>< k>< X 0 N 0 C !c o 0 O » < 0 0 N "0 T3 C 0 E < o o CM O LO 00 o LO 00 o o 00 X X X X X X X X c _o 'ip 0 O o >. o •t-> 3 < c _o 'p 0 > 0 ) LU lo CO r- OO On o (N m m CD r-- 00 2 *>2 *>2 *>2 *>2 *>2 a o a. a. a. a. a. a. a. a. a. a. a. a. a. a. Q U Oh c/3 C3 -C) I S Q L. s 0 Cj -H 1 *>2 a 5 ST I s i-<* o i-<* tt: s Co s i-<* o i-<* tt: L. C3 S s i-<* o i-<* tt: -C) •2 C3 pi C3 S C3 OS Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (56) September 2014 | Volume 8 | Number 1 | e82 Almendariz et al 0 T3 0 0 0 CO 0 z o Q_ 0 LJJ Q ^ 0 0 0 C O) f N CD .y 0 o ■I— » 0 o o o o 00 c 0 0 CO N c 0 E 0 c be C ’3 0 Q_ 0 Q o3 00 o 00 00 o LO LO c 0 XX X X h- c o3 0 0 h- 0-) LO E Q_ >, 0 05 c < 0 T3 0 05 LO CO 0 DC DC 0 0 0 E X X CO N 0 , "0 r^ 0 ^ CM t Q_ ^ T3 05 CM C < < C 05 A ^ DC CD c 0 ^ X X X XXX X 05 Q. E 0 p E 0 o -E ^ ^ '0 Z 0 ,0 0 o D) C 0 O o C 0 _ E ^9 ^ 8 u_ 0 ^ 8 o LO o o 00 05 c 0 X X X 0 T3 0 0 0 o to z ^ (f) ^ LO o N 0 LO o o 0 CM 0 c o O o is O o t o c 0 X X X X c 0 0 N O 0 5 “ 0 0 C O 0 CM ^ I 0 O LO 00 cb LO 05 c 0 X XX XX X X X 0 N 0 0 CO 0 T3 0 0 "0 T3 c 00 00 CM E c 0 0 0 0 F T3 C 0 CM cb 0 0 E Q_ 0 CO X X 0 N 0 C Z o 0 O < 0 0 N "0 T3 C 0 E < o o CM cb LO 00 o LO 00 cb o 00 X X c 0 XXX c 0 'ip 0 o o >. o 3 < c 0 'p 0 > 0) LU 05 CQ 2 u CQ bi P A s 05 M a .s a *>2 a a c> *>2 e> "ts K •S a Cl c g ■+.«* s a ic) s ic) a c g -s; •Sh •a tt: s a c o ic) g CLh a c g -s; a c g -s; •Sh .Sh a tt: a tt: 05 CQ 2 M g iC) s -s; I "ts K c Q "ts 0 "g s -s; 1 "ts K c Q K C S c a r a s -s; I "ts K c Q s ■4^ s s s *>2 I s c Q L. ■§ s o c s c o i-<* tt: Co s "S a s o Cl, s c o i-<* a> tt: c a a s c a o tt: Co s c g c s s a o -C) Co a o rS Cl, cs, tt: tt: a o -C) s c a o -C) Co s a a Cl, g Clo O c Clo Co a o -C) Co '<1 Cl, cs, tt: tt: c s a a o a tt: C S -C) Co s ,-<* a Cl, 5r c o a o c a *>2 s ,-<* a Cl, 5r c o c s •S s a s ">»<» a g- c c o c CL, t/3 s a -s; c c o c ^ ^ ^ 000 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (57) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador Rio Puintza Duellman & Lynch 1988 1550-1830 msnm RAP 7 Reyn- olds & Icochea 1997 665-1750 msnm X X RAP 7 Al- mendariz 1997 900-2200 msnm X X X Nump- atkaim / rios Tsuirim y Coangos Fun- dacion Natura 2000 930-1050 msnm X Quebrada Shinga- natza Proy. Paz y Conserva- cion, 2005 (Peru) 850-1200 msnm c/3 0 T3 0 C/3 CO CO CD CD 0 0 Z E 0 0 0 •t—t 0 Q Q_ LU *-> (— 0 :y r*** cb 0 00 E c c/3 E X X O O CD 5® o 1 - o N Q- Q_ 0 > LO o o 0 CM Ui c o O o is o ^ o E o c Ui ^ E X c 0 0 N O 0 CD 0 C O 3 -- CM ^ I 0 O LO 00 cb LO C03 X X 0 N 0 0 CO 0 T3 0 0 "0 T3 c CO 00 CM E c 0 0 4— » 0 0 F T3 C 0 CM cb 0 0 E Q_ 0 CO X 0 N 0 c !n o 0 o < 0 0 N "0 T3 C 0 E < o o CM cb LO 00 o LO 00 cb o 00 X X X X X X X X c 0 'is 0 o o > o c 0 'ip 0 > 0) LU *>2 C3 S -a S •2 *>2 -a a' a' cj .a a C3 S CJ 03 at CJ C3 a CL a s CL C3 2 a *>2 *>2 IS "SC 5 a' a' 0 •S at -O 0 ft 03 0 s CJ 0 CJ 0 CL Co L. s C3 Co cj o ■4^ Co a C3 S *>2 a o 03 CQ 2 JS 2 u L. o K a K a •2 > O fl d t/3 CJ O •2 (N CO Q U> U 03 CQ 2 a o -« o JS 03 a t/3 a o (30 o o cq a t/3 a o (30 o o cq a t/3 (3 O (30 O o cq a t/3 (3 O (30 <3 O cq < Z, o HH K &H o z o 03 CQ -« U 03 CQ U (3 S (5^ •2 5 Cj Cj 03 C Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (58) September 2014 | Volume 8 | Number 1 | e82 Table 2 . Species of amphibians recorded from the Cordillera del Condor. Almendariz et al oQ 1^ CJ) 1^ CO p CJ) Q_ o CO < c CD CC CD QC o o o o LO 1^ I LO CD CD c CO X X X N o N- "CO o Q_ ■o c 2? cvj cvj < O O’ 1 CC E ^ o < (T> c CO X X X C/) CO CO -1:- "O CO CO C ^ CO -Q O) CD ^ O c N 'o Jr CO CO CL > t LO 2 ^ CL O O C\J o o cvj I o LO CD c CO X X X X X X CO CO N 2 C ^ CO CD O) O C O ^ c/) c „ CO ^ ^ CL > 9^ 0 ) E o CO in CL o C\J o o cvj I o LO CD c CO X X O CO c ■D 3 E -O >^ "O N 'O o "O CO ;= CO CO CO ^ b 0- ^ o 1- CD >; 0) LU O I '■ — ^ E ^ O —1 2 c LD CL O O O O O C\J c CO o o 1 ^ X 1 c CO CO N o LO CD Z CO C\J CO 1 o CO S' “> o < O) LO (T> c CO X X X XX XX X O ■D CO _N CD ^CO CO T3 Q_ C CD O CD O E < CO 2 o m T3 c CO o CD CD CVJ I o o CD X X CO N CO _c !c o CO o < I Ci. c CO 0) N "CO T3 c CD O O cvj I o o CD O o CD I o o CO c CO X X X X K> K> X X XXX X c _o 'iS (0 o o >. o 3 < c _o 'i3 (0 > 0) LU & CO « 'I' *^ll ^ Cj *>! ■ 4 ^ >2 S ■C 3 1 c 53 ‘C 1 JS > 'S Cj -S 1 2 -s: K 2 1 & 2 ■ 4 >«d 5 ^ .a a a s -Cj CJ Cj S o s: -I sii >2 2 CL .55) c JS *>2 *>2 *>! *>! g 2 ’>a a o !>2 >2 *>2 CL S -Cj O S tjo o ’>a >2 S s 'i Cj a :§ s ’S •2 [5 •2 S & 2 a s s o CO >■. o o o -s: s: s: s: O u u X X X s VJU hH X X < ci ! O '"..i o s I s o CJ o c s C) o s "§ o s: >2 S a CJ S s CL '"..i o s o o I o s « 'O u « 'O 2 « JS CL CO a >2 -Cj ■§ O) a I .CJ 'S 0 tN? s S 1 ■4.^ e X Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (59) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador oQ 1^ CJ) 1 ^ CO p CJ) CL 0 CO < c CD cc CD QC 0 0 0 o LO 1^ I LO CD CD c CO X X X X X XX N 0 N- "CO 0 CL ■0 c 2? cvj cvj < 0 O’ 1 cc E ^ 0 < (T> c CO X X C/) CO CO -1:- "O CO CO O) JD CD O c N 'o Jr CO CO CL > t LO 2 ^ CL O O C\J o o cvj I o LO CD c CO X X X XX CO CO N 2 C ^ CO CD o U) c „ CO ^ ^ CL > 9^ 0) b O CO in CL o C\J o o cvj I o LO CD c CO X o CO 1- CD O N CO CL >L o ^ -O 2 ns CO D ^ L) O LU CO C LO o o O o c\j c CO o o 1^ 1 C CO CO N 0 LO CD z CO C\J CO 1 0 CO S' “> 0 < O) LO (T> c CO X X X X XX 0 ■D CO _N CD ^CO CO T3 CL C CD 0 CD 0 E < CO 2 o CO T3 C CO o CD CD CVJ I o o CD X X CO N CO _c !c o CO o < I Ci. c CO 0) N "CO T3 c CD O O cvj I o o CD O o CD I o o CO c CO X X X X X X X X s> S> s>s>s>s>s> c _o 'is (0 O O > O 3 < C _o 'i3 (0 > 0) LU CO H X EL CO « ■© EQ .Cj C CJ c CJ •2 C CJ c CJ 5 •2 "S, CJ Qi a 'O "C _s "o u o 'S’ CJ 5 Dh V!J >2 S 2 Cj •S C Cj CJ *>2 2 5 o •2 ■K s o s 2 5 2 ’>a 2 5 •2 -Cj Cj CJ Cj CJ •S •2- .& Q Q •S S o !2 .2" a s a s CL C3 •S" a Cj 2 Cj .2. e « !L .2" a >2 s 5^ 2 *>2 c 1 2 L. 2" •S &0 2 L. 2" »2 2 : Q Q Q Q t2 2^ 2 2^ 2 Dh V2 >2 s 5^ 2" 2 t2 Cj o CJ K Cj ■C3 s K Cj S L. S *>! s s S I s e •i "i S 2" >2 s >2 s "2 o *>2 s: 0 s: s *>2 1 o a !L *>! I "2 ~s; ~s; CO Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (60) September 2014 | Volume 8 | Number 1 | e82 Almendariz et al oQ 1^ CJ) 1^ CO p CJ) CL o CO < c CD cc >N CD QC o o o o LO 1^ I LO CD CD E c CO E N o N- "CO o CL ■o c 2? cvj cvj < O O’ 1 cc E ^ o < c/) CO ■D CO JD CD o CO N CO c CO C3) c N 'o Jr CO CO 0- > t LO 2 ^ CL O O C\J o o cvj I o LO CD c CO CO CO N ■D CO CO _ c ^ CO CD O) O C O CD c „ CO ^ ^ CL > 9^ -,L 0) b O CO in 0- o c\j o o cvj I o LO CD c CO o ■D CO o ■D c o O CO V ZJ ■D CO V V CD j: CO N CO CL 5^ O C '9 o CO > o T3 CO D O 0) LU CO ' LO O o C\J c CO o o 1^ c 1 c CO CO N o LO CD z CO C\J CO 1 o CO S' “> o < O) LO X X X o CO ■D CO CD N II "i— ^ "CO C\J o CD CD CO "2 o CVJ CL O’ ir 1 o O V V CD E “ 0) LU o .CJ -s; & s CO Dh e s C3 »s: CJ -Cj C3 S o s X L c ■§ o c X X X s> s> s> s> s> s> s> On On On On On On On a ■ft § § ^ >2 *>! ■4.^ -s; s .2 o g s: •2 C O) n; 5 a 2 a >2 •2 P -s; CJ *2 ft Cl a S*.i ■♦-4 *>2 Sy *>! o 2 2 "C g 1 1 c ft -s; -s; -s; o ft ■4.^ ■4.^ ■4.^ eS X o cq o =q o 03 •2 S I ■s o «a Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (61) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador sessment. Editors, Schulenberg T, Awbrey K. RAP Working Papers number 7, Conservation Interna- tional, Washington, D.C., USA. 234 p. Available: https://library.conservation.org/Published%20Docu- ments/2009/07%20RAP%20Working%20Papers.pdf [Accessed: 30 August 2014]. Almendariz A. 2010. Reptiles. Pp. 40^4 In: La Magia de los Tepuyes del Nangaritza. Una Gma para Con- ocer a sus Habitantes. Editors, Freile J, Moscoso P, Felix C. Conservation Intemacional Ecuador Quito, Ecuador. Available: http://bibdigital.epn.edu.ee/han- dle/1 5000/6683 [Accessed: 30 August 2014]. Almendariz A. 2012. Anfibios y Reptiles. Pp. 20-25 In: Animales de la Concesion La Zarza y Alrededores. Estribaciones Occidentales de la Cordillera del Con- dor. Editor, Gilda Gallardo. Kinross-Fruta del Norte, Ecuador. Available: http://bibdigital.epn.edu.ee/han- dle/1 5000/6684 [Accessed: 30 August 2014]. Almendariz A, Batallas D. 2012a. Nuevos datos sobre la distribucion, historia natural y el canto de Centrolene condor Cisneros-Heredia y Morales-Mite 2008 (Am- phibia: Anura: Centrolenidae). Revista Politecnica 30(2): 42-53. Almendariz A, Batallas D. 2012b. Ecologia y distribu- cion geografica de Hyloxalus mystax Duellman y Sim- mons, 1988. Revista Politecnica 30(2): 54-61. Almendariz A, Brito J. 2013. Anfibios y reptiles registra- dos en Cerro Plateado-Cordillera del Condor. Report for Naturaleza y Cultura Intemacional, Quito, Ecua- dor. 1-13. Almendariz A, Brito J, Batallas D, Ron S. 2014. Una especie nueva de rana arborea del genero Hyloscir- tus (Amphibia: Anura: Hylidae) de la Cordillera del Condor. Papeis Avulsos de Zoologia 54(4): 33-49. Almendariz A, Brito J, Vaca J. Una sinopsis de la herpe- tofauna de la Cordillera del Condor. {In prep.). Almendariz A, Ron SR, Brito J. 2012. Una especie nueva de rana venenosa de altura del genero Excidobates (Dendrobatoidea: Dendrobatidae) de la Cordillera del Condor. Papeis Avulsos de Zoologia 52:387-399. Almendariz A, Ron SR, Brito J. Una especie nueva de microhflido (Anura, Mirohylidae), de la Cordillera del Condor. {In prep.) Ayala- Varela F, Torres-Carvajal O. 2010. A new species of dactyloid anole (Iguanidae, Polychrotinae, Anolis) from the southeastern slopes of the Andes of Ecuador. ZooKeys 53: 59-73. Ay ala- Varela F, Poe S, Carvajal-Campos A, Gray L, Davis J, Almendariz A. 2011. Anolis soini Poe and Yanez-Miranda, 2008 (Squamata: Iguanidae: Poly- chrotinae): Distribution extension, first records for Ecuador and notes on geographic variation. Check List! {5): 629-631. Batallas D, Brito J. 2014. Description of the advertise- ment call of Noblella lochites (Anura: Craugastori- dae). Avances en Ciencias e Ingenierias 6(1): B6-B8. Brito J, Almendariz A, Batallas D. Phyllomedusa ecu- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (62) atoriana Cannatella 1982 (Amphibia: Hylidae): vari- acion, descripcion del renacuajo, vocalizacion y an- otaciones sobre la historia natural. Papeis Avulsos. {In prep). Brito J, Arguero A. 2012. Nuevos datos sobre la distribu- cion de Scolomys ucayalensis (Rodentia: Cricetidae) y Phylloderma stenops (Chiroptera: Phylostomidae) en Ecuador. Mastozoologia Neotropical 19(2): 293-298. Cisneros-Heredia DE, Morales-Mite M. 2008. A new species of glassfrog from the elfin forests of the Cor- dillera del Condor, southeastern Ecuador (Anura: Centrolenidae). Herpetozoa 21(1/2): 49-56. Duellman WE. 1999. Distribution patterns of amphibians in South America. Pp. 255-328 In: Duellman, WE. (Editor). Patterns of Distribution of Amphibians. A Global Perspective. The Johns Hopkins University Press, Baltimore, Maryland, USA. Viii -i- 633 pages. Duellman WE, Eynch JD. 1988. Anuran amphibians from the Cordillera de Cutucu, Ecuador. Proceedings of the Academy of Natural Sciences, Philadelphia 140(2): 125-142. Duellman WE, Simmons JE. 1988. Two new species of dendrobatid frogs, genus Colostethus, from the Cordillera del Condor, Ecuador. Proceedings of the Academy of Natural Sciences, Philadelphia 140(2): 115-124. Freile J, Krabbe N, Piedrahita P, Buitron-Jurado G, Rodrfguez-Saltos C, Ahlman F, Brinkhuizen DM, Bonaccorso E. 2014. Birds, Nangaritza River Valley, Zamora Chinchipe Province, southeast Ecuador: Up- date and revision. Check List 10(1): 54-71. Freile J, Moscoso P, Felix C (Editors). 2010. La Magia de los Tepuyes del Nangaritza. Una Gma para Con- ocer a sus Habitantes. Conservation Intemacional Ec- uador, Quito, Ecuador. 68 p. Fundacion Natura. 2000. Parque El Condor: Estudios y Propuesta. Fundacion Natura, Quito, Ecuador. 202 p. Guayasamm JM, Tapia E, Aldas S, Deichmann J. 2011. Anfibios y Reptiles de los Tepuyes de la Cuenca Alta del Rfo Nangaritza, Cordillera del Condor. Pp. 56-61 In: Evaluacion Ecologica Rdpida de la Biodiversidad de los Tepuyes de la Cuenca Alta del Rio Nangaritza, Cordillera del Condor, Ecuador. Conservacion Inter- nacional, Quito, Ecuador. Lynch JD. 1974. A new species of leptodactylid frog {Is- chnocnema) from the Cordillera del Condor in Ecua- dor. Journal of Herpetology 8(1): 85-87. Lynch JD. 1976. Two new species of frogs of the genus Euparkerella (Amphibia: Leptodactylidae) from Ec- uador and Peru. Herpetologica 32(1): 48-53. Lynch JD. 1979. Leptodactylid frogs of the genus Eleu- therodactylus from the Andes of southern Ecuador. Miscellaneous Publication Museum of Natural His- tory, University of Kansas 66: 1-62. Lynch JD, Duellman WE. 1980. The Eleutherodactylus of the Amazonian slopes of the Ecuadorian Andes (Anura: Leptodactylidae). Miscellaneous Publication September 2014 | Volume 8 | Number 1 | e82 Almendariz et al. Museum of Natural History, University of Kansas 69: 1 - 86 . Ministerio del Ambiente del Ecuador. 2012. Sistema de clasificacidn de los ecosistemas del Ecuador conti- nental. Subsecretaria de Patrimonio Natural, Quito, Ecuador. 136 p. Available: http://www.ambiente.gob. ec/wp-content/uploads/downloads/20 1 2/09/LE YEN - DA-ECOSlSTEMAS_ECUADOR_2.pdf [Accessed: 30 August 2014]. Neill D. 2005. Cordillera del Condor. Botanical treasures between the Andes and the Amazon. Plant Talk 41: 17-21. Neill D, Asanza M. 2012. Lozania nunkui (Lacistema- taceae), a New Species from the Sandstone Plateaus of the Cordillera del Condor in Ecuador and Peru. Novon: A Journal for Botanical Nomenclature 22: 207-211. Neill D, Ulloa C. 2011. Adiciones a la Elora del Ecua- dor: Segundo Suplemento, 2005-2010. Eundacion Ja- tun Sacha, Quito, Ecuador. 202 p. Organizacion Intemacional de las Maderas Tropicales and Eundacion Natura y Conservacion Intemacional. 2005. Paz y Conservacion Binacional en la Cordil- lera del C6ndor Ecuador-Peni. Quito, Ecuador. 112 p. Available: http://www.itto.int/files/itto_project_db_ input/197 3/Competition/Final%20Report%20PD%20 2-00%20R.2%20E.pdf [Accessed 30 August 2014]. Peloso PLY, Sturaro MJ, Eorlani M, Gaucher P, Motta AP, Wheeler WC. 2014. Phylogeny, taxonomic revi- sion, and character evolution of the genera Chiasmo- cleis and Syncope (Anura, Microhylidae) in Amazo- nia, with descriptions of three new species. Bulletin of the American Museum of Natural History 386. 112 p. Reynolds RP. 1997. Simmons’ herpetological collection from the Western slopes of the Cordillera del Condor. Pp. 202-203 In: The Cordillera del Condor Region of Ecuador and Peru: A Biological Assessment. Editors, Schulenberg T, Awbrey K. RAP Working Papers num- ber 7, Conservation International, Washington, D.C., USA. 234 p. Available: https ://library.conservation. org/Published%20Documents/2009/07%20RAP%20 Working%20Papers.pdf [Accessed: 30 August 2014]. Reynolds, R.P. and J. Icochea M. 1997a. Amphibians and reptiles of the upper Rfo Comainas, Cordillera del Condor. Pp. 82-84 In: The Cordillera del Condor Re- gion of Ecuador and Peru: A Biological Assessment. Editors, T. Schulenberg and K. Awbrey. RAP Working Papers number 7, Conservation International, Wash- ington, D.C., USA. Available: https://library.conser- vation.org/Published%20Documents/2009/07%20 RAP%20Working%20P apers.pdf [Accessed: 30 Au- gust 2014]. Reynolds, R.P. and J. Icochea M. 1997b. Amphibians and reptiles of the upper Rfo Comainas, Cordille- ra del Condor. Pp. 204-206 In: The Cordillera del Condor Region of Ecuador and Peru: A Biologi- cal Assessment. Editors, T. Schulenberg and K. Aw- brey. RAP Working Papers number 7, Conservation International, Washington, D.C., USA. Available: https://library.conservation.org/Published%20Docu- ments/2009/07%20RAP%20Working%20Papers.pdf [Accessed: 30 August 2014]. Riina R, Vigo MA, Ceron C. 2014. Croton condorensis: An enigmatic new species of Euphorbiaceae from southern Ecuador. Phytotaxa 164(2): 154-158. Sayre R, Roca E, Sedaghatkish G, Young B, Keel S, Roca R, Sheppard S. 2002. Un Enfoque en la Naturaleza. Evaluaciones Ecologicas Rdpidas. The Nature Con- servancy, Arlington, Viriginia, USA. 196 p. Available: http://www.google.com/url?sa=t&rct=j&q=&esrc=s &source= web&cd=3 &ved=OCDMQEJ AC&url=http %3A%2E%2Ewww.upse.edu.ec%2Ecigeo%2Eindex. php%3Foption%3Dcom_docman%26task%3Ddoc_ download%26gid%3D8%26Itemid%3D76&ei=YW EDVNugAcJNggTpvIGIAw&usg=AEQjCNGeBb20s TTb446dR0tme2wfhZ 1 5fA&sig2=FtOgiMDn83m_7 6nYHJ6mKg&bvm=bv.741 15972,d.eXY [Accessed: 30 August 2014]. Schulenberg T, and Awbrey K (Editors). 1997. The Cor- dillera del Condor Region of Ecuador and Peru: A Biological Assessment. RAP Working Papers num- ber 7, Conservation International, Washington, D.C., USA. 234 p. Available: https ://library.conservation. org/Published% 20Documents/2009/07 % 20R AP % 20 Working%20Papers.pdf [Accessed: 30 August 2014]. SERNANP. 2012. Parque Nacional Ichigkat Muja- Cordillera del Condor. Diagnostico del Proceso de Elaboracion del Plan Maestro 2011-2016. Servicio Nacional de Areas Naturales Protegidas por el Estado. Editorial SERNANP, Peru. 66 p. Sheehy CM, Yanez-Munoz MH, Valencia JH, Smith EN. 2014. New Species of Siphlophis (Serpentes: Dipsadi- dae: Xenodontinae) from the Eastern Andean Slopes of Ecuador. South American Journal of Herpetology 9(1): 30^5. Teran- Valdez A, Guayasamm JM. 2010. The smallest ter- restrial vertebrate of Ecuador: A new frog of the genus Pristimantis (Amphibia: Strabomantidae) from the Cordillera del Condor. Zootaxa 2447: 53-68. Torres-Carvajal O, de Queiroz K, Etheridge R. 2009. A new species of iguanid lizard (Hoplocercinae, Enyali- oides) from southern Ecuador with a key to eastern Ecuadorian Enyalioides. ZooKeys 27: 59-71. Ulloa C, Neill D, Dudek O. 2012. A new species of Mi- conia (Melastomataceae, Miconieae) from the Ecua- dor-Pem border. PhytoKeys 12: 35^6. Received: 19 May 2014 Accepted: 20 August 2014 Published: 20 September 2014 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (63) September 2014 | Volume 8 | Number 1 | e82 Herpetofauna of the Cordillera del Condor of Ecuador Ana Almendariz is a researcher and the Curator of Herpetology at the Institute of Biological Sciences at the Escuela Politecnica Nacional in Quito, Ecuador. A native of Quito, Almendariz holds an undergraduate degree in biology and a master’s degree in conservation and management of natural resources. She conducts research on amphibians and reptiles throughout Ecuador and has published extensively on her research. John E. Simmons is president of Museologica consulting, and teaches museum studies for Kent State Uni- versity, Juniata College, and the Universidad Nacional de Colombia. Simmons has an undergraduate degree in systematics and ecology and a master’s degree in historical administration and museum studies, and previously was collections manager at the California Academy of Sciences and the Biodiversity Research Institute at the University of Kansas. Jorge Brito is a mammal and amphibian researcher at the Museo Ecuatoriano de Ciencias Naturales in Quito, Ecuador. He has an undergraduate degree in biology from the Universidad Central del Ecuador; his research interests are focused on amphibians and terrestrial micromammals. He has published several contributions on these species principally from southeastern Ecuador. Jorge Vaca-Guerrero is a Junior Investigator at the Institute de Ciencias Biologicas of the Escuela Politec- nica Nacional in Quito, Ecuador. He has an undergraduate degree in biology from the Universidad Central del Ecuador, and experience in studies of the herpetofauna of the Eastern region of Ecuador. His principle area of interest is the evolution of reptiles, particularly vipers. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (64) September 2014 | Volume 8 | Number 1 | e82 Copyright: © 2014 McCracken and Forstner. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education pur- poses only provided the original author and source are credited. The official publication credit source: Amphibian Reptile Conservation at: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 65-75. Herpetofaunal community of a high canopy tank bromeliad {Aechmea zebrina) in the Yasuni Biosphere Reserve of Amazonian Ecuador, with comments on the use of “arboreai” in the herpetologicai iiterature ^’^Shawn F. McCracken and ^’^Michael R. J. Forstner ^Department of Biology, Texas State University, San Marcos, Texas, USA Abstract . — ^Tank bromeliads provide microhabitat that supports a high diversity of organisms in the harsh environment of tropical forest canopies. Most studies of organisms occupying tank bromeliads have focused on invertebrates found within bromeliads near or at ground level. Few investigations of vertebrate communities utilizing this keystone resource are available. We describe the amphibian and reptile community occupying the high canopy tank bromeliad, Aechmea zebrina, in lowland rainforest of the Yasum' Biosphere Reserve in the Amazon Basin of Ecuador. We used single-rope climbing techniques to sample a total of 160 A. zebrina bromeliads from 32 trees, at heights of 18.3 to 45.5 m above ground. We collected 10 metamorphosed anuran species, one gecko, one snake, and observed two species of lizard within bromeliads. Summary statistics for a suite of environmental factors associated with herpetofauna in A. zebrina bromeliads are reported. We estimated the density of anurans occupying A. zebrina communities and contrast these estimates with anuran densities from tropical forest floor anuran studies. Finally, we discuss the use of the term “arboreal” within the herpetologicai literature, and make recommendations for terminology used to describe the vertical space occupied by a species or assemblage. Key words. Amphibian, anuran, epiphyte, forest, microhabitat, rainforest, reptile Citation: McCracken SF, Forstner MRJ. 2014. Herpetofaunal community of a high canopy tank bromeliad {Aechmea zebrina) in the Yasuni Biosphere Reserve of Amazonian Ecuador, with comments on the use of “arboreal” in the herpetologicai literature. Amphibian & Reptile Conservation 8(1) [Special Section]: 65-75 (e83). Introduction Forest canopies provide habitat for approximately 50% of terrestrial species, yet there are few studies specific to canopy herpetofauna (Stewart 1985; Vitt and Zani 1996; Kays and Allison 2001; Guayasamin et al. 2006; McCracken and Forstner 2008; Lowman and Schowal- ter 2012; Scheffers et al. 2013; McCracken and Forstner 2014). Basic ecological knowledge of arboreality (tree- living) and utilization of high canopy microhabitats by amphibians and reptiles remain depauperate in the litera- ture (Moffett 2000; Kays and Allison 2001; Lehr et al. 2007). A canopy microhabitat frequently used by herpe- tofauna in tropical forests are epiphytes, and in particular epiphytic tank bromeliads that are phytotelms capable of holding relatively large volumes of water (Lowman and Rinker 2004; McCracken and Forstner 2008). In lowland Neotropical rainforest, canopy tank bromeliads typically reside in the overstory and emergent canopy trees at ver- tical heights of 5^5 -f meters with ~5 to >150 individuals on a single tree (McCracken and Forstner 2006). These arboreal bromeliad conununities create a three-dimen- sional “wetland in the sky” that have been estimated to impound up to 50,000 liters of water per hectare (Kitch- ing 2000; McCracken and Forstner 2006). Tank bromeli- ads function as a “keystone resource” in the harsh forest canopy environment where the atmosphere meets and in- teracts with 90% of Earth’s terrestrial biomass; providing a climate-buffered refuge, water source, and food source for canopy herpetofauna (Nadkami 1994; Ozanne et al. 2003; Cardeliis and Chazdon 2005). Kays and Allison (2001) found only 4% of 752 arti- cles published between 1988 and 1998 on tropical forest arboreal vertebrates focused on reptiles and amphibians. Many species of herpetofauna are described as being arboreal regardless of whether they are restricted to the vertical stratum a few centimeters to a few meters above ground, or solely inhabit the high forest canopy at 20 or more meters vertical height (Chaparro et al. 2007; Mc- Cracken et al. 2007; Guayasamin and Funk 2009). Forest structure is associated with vertical partitioning or strati- fication of the component plant community (e.g., trees, shrubs, lianas) and accentuates vertical patterns followed by other organisms (Moffett 2000; Lowman and Rinker Correspondence. Emails: ^smccracken® txstate.edu (Corresponding author); ^ mf@txstate.edu Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (65) October 2014 | Volume 8 | Number 1 | e83 McCracken and Forstner clouds Roads Trails Tree sampled Yasuni National Park 20 ■ Kilometers Fig. 1. (A) Map of South America with Ecuador (shaded light blue) and Yasuni National Park (solid dark green) highlighted. The Amazon ecoregion is outlined with light green line. (B) Northeastern section of Yasuni National Park (light gray line) and surround- ing region where trees were sampled forAechmea zebrina bromeliads within the vicinity of the Tiputini Biodiversity Station - Uni- versidad San Francisco de Quito (TBS) and the Yasuni Research Station - Pontificia Universidad Catolica del Ecuador (YRS). (C) Detail of TBS where trees were sampled for A. zebrina bromeliads. Note: Map is modified from Figure 2 in McCracken and Forstner (2014) and used under the Creative Commons Attribution license. 2004). Spatial patterns of forest cohabitants, such as tank bromeliads and their inhabitants, are likewise strongly influenced by forest structure as a result of the funda- mental organization of resources and space (Lowman and Rinker 2004). Identifying the vertical space occupied by a particular amphibian or reptile species in its given habitat will allow greater insight to their ecological role in the system. Herein, we describe amphibians and reptiles occupy- ing the high canopy tank bromeliad, Aechmea zebrina, in lowland rainforest of the Yasum Biosphere Reserve in the Amazon Basin of Ecuador. We report a suite of environ- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (66) October 2014 | Volume 8 | Number 1 | e83 Herpetofaunal community of a high canopy tank bromeliad Fig. 2. (A) A downward vertical view {in situ) of Aechmea zebrina (foreground center left, and at lower elevation in upper right and center right) and a cluster of Aechmea tessmannii (center, with one in hloom) hromeliads in the tree canopy from ~34 m. (B) A community of A. zebrina hromeliads at ~38 m (in situ). (C) An A. zebrina hromeliad (ex situ) inside screen tent being measured and prepared for dismantling, collected from ~44 m in the canopy. Notice the more upright leaves and reddish color because of increased sun exposure due to high canopy location. mental factors associated with herpetofauna in A. zebrina hromeliads. We estimate the density of anurans occupy- ing mean A. zebrina community sizes in two tree size classes, representative of our shortest and tallest trees in the study. We then compared these with anuran densities from tropical forest floor anuran studies by calculating the two-dimensional area (m^) of the tree crowns for the two tree size classes. In completing our review, we feel it is important to discuss the use of the term “arboreal” within the herpetological literature and make recommen- dations for the incorporation of additional terminology to provide a more informative description of the vertical space utilized by a species or assemblage. Materials and Methods The study was conducted in the northwestern portion of the Yasum Biosphere Reserve (Yasum) located in Orel- lana Province, Ecuador. The reserve includes Yasum Na- tional Park, Waorani Ethnic Reserve, and their respective buffer and transition zones (Einer et al. 2009). Yasunf is part of the Napo Moist Eorest terrestrial ecoregion cover- ing approximately 1 .7 million ha of the upper Amazon Basin (Einer et al. 2009; Bass et al. 2010). Yasum has an elevation range of 190-400 m above sea level; the north- western region averages 2,425-3,145 nun of rainfall per year with no less than 100 mm per month, temperature averages 25 °C (15 °-38 °C), and humidity averages 88% (Blandin 1976; Duellman 1978; Balslev et al. 1987; Bass et al. 2010). Yasum holds world record species diversity for several taxa, including the highest documented land- scape scale (lowland tropical rainforest) herpetofauna di- versity with 150 species of amphibians and 121 species of reptiles (Bass et al. 2010). Collections were made in the vicinity of two research stations, the Tiputini Biodiver- sity Station (TBS) (0°38’ 14”S, 76°08’60”W) operated by Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (67) October 2014 I Volume 8 I Number 1 I e83 McCracken and Forstner Fig. 3. A collection of anurans collected from Aechmea zebrina bromeliads. (A) Pristimantis aureolineatus hiding in leaf axil, and (B) on a leaf of A. zebrina. (C) Pristimantis waoranii emerging from leaf axil, and (D) on a leaf of A. zebrina. (E) Ranitomeya ventrimaculata and (F) Scinax ruber collected from A. zebrina bromeliads. the Universidad San Francisco de Quito and the Yasum Research Station (YRS) (0°40’27”S, 76°23’51”W) oper- ated by the Pontificia Universidad Catolica del Ecuador (Fig. 1). Tiputini Biodiversity Station is only accessible by river and surrounded by undisturbed primary lowland rainforest, and YRS is located approximately 27 km west on an oil pipeline road (Maxus road) that has been expe- riencing forest disturbance within its vicinity but is still surrounded by large tracts of undisturbed forest. Sam- pling of A. zebrina bromeliads took place between 0800 and 1800 hours from April to November of 2008. We focused our sampling on a single large epiphytic tank bromeliad species, Aechmea zebrina, that is native to the Amazon regions of Ecuador and southeastern Co- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (68) October 2014 I Volume 8 I Number 1 I e83 Herpetofaunal community of a high canopy tank bromeliad lombia (Smith 1953). Aechmea zebrina occupy vertical heights of approximately 18^5+ m in the overstory and emergent canopy trees, and range between 1 to >150 in- dividuals on a single host tree (SFM, unpublished data). The leaves are upright and arranged in a spiral with their leaf axils tightly overlapping to form water-holding res- ervoirs (Fig. 2). These cavities provide a critical refuge and food source for invertebrate and vertebrate species in the harsh canopy climate (Nadkami 1994). Sampling methodology for A. zebrina bromeliads fol- lowed our previously published methods (McCracken and Forstner 2008). Single-rope technique (SRT) was used to climb trees for canopy access, and five brome- liads were collected haphazardly from each tree at esti- mated even vertical intervals between one another (Perry 1978). Before each bromeliad removal, we checked for active amphibians or reptiles, we recorded the bromeli- ads elevation, measured the air temperature adjacent to the bromeliad, and the temperature and pH of water held in one of the outer leaf axils. Ideally, when the bromeliad is disturbed the response of most animals is a retreat into the bromeliads leaf bracts and thus prevents loss of speci- mens (McCracken and Forstner 2008). Bromeliads were removed and sealed in a 55-gallon (208 L) plastic bag and then lowered to the ground. After bromeliad collections we counted the number of A. zebrina inhabiting the tree and measured tree height. Bromeliads were transported back to camp where we processed them in a screened tent to prevent escape of animals. We first poured all water from the bromeliads through a 1 mm sieve to separate arthropods, leaf litter, and detritus. We then measured the water volume with a graduated cylinder and the pH of the homogenized solution. We counted the number of mature leaves (used as a size metric) and measured the height of bromeliads to nearest centimeter (from base of plant to highest vertical leaf tip). Bromeliads were then dismantled leaf-by-leaf to collect all herpetofauna. We identified and counted all metamorphosed an- urans and reptiles to species level for each bromeliad. Larval anurans were also collected and counted, with the majority identified to genus or species. In an attempt to better identify larval anurans we maintained individual tadpoles outdoors in 12 oz. plastic cups with water and detritus collected from bromeliads. Once tadpoles began to metamorphose the cups were covered with window screen to prevent escape. Upon sufficient development to allow identification the froglets were euthanized and preserved. All herpetofauna were handled and preserved following the guidelines compiled by the American So- ciety of Ichthyologists and Herpetologists (ASIH), and in compliance to the rules overseen by the Texas State University Animal Care and Use Connnittee (Permit #: 0721-0530-7, 05-05C38ADFDB, and 06-01C694AF). Additionally, we report the herpetofauna species ob- served active amongst A. zebrina bromeliads but not collected. We calculated summary statistics of recorded habitat variables for each species and report raw data for Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (69) Fig. 4. The Banded cat-eyed snake, Leptodeira annulata, col- lected in an Aechmea zebrina bromeliad at 43.5 m above the forest floor. singletons and doubletons. Summary statistics were cal- culated for recorded habitat variables across all bromeli- ads sampled, bromeliads occupied by >1 metamorphosed anurans, and bromeliads not occupied by anurans. We then compared an estimated number of anuran individuals in A. zebrina bromeliads per 100 m^ of tree crown area to other published work of tropical frog as- semblages collected at or near ground level. No other studies were available to provide canopy estimates. Mean anuran abundance per tree was calculated by taking the mean number of metamorphosed anurans per A. zebrina bromeliad (x = 0.6) and multiplying by the mean number of bromeliads per tree (x = 66). Based on tree crown di- ameter measurements by Asner et al. (2002) in lowland rainforest of eastern Amazonia we calculated the number of anurans per 100 m^ of a typical tree crown area for the two largest tree size classes. The two largest classes had mean tree heights of 25.3 m (Dominant) and 46.1 m (Super dominant) with a mean crown diameter of 11.6 m and 19.9 m, respectively. Mean tree crown diameters were used to calculate the area of a circle. These two tree size classes were used as we did not measure individual crown diameters and consider these two as representative of the shortest (28 m) and tallest (49 m) trees in our study. We then divided the mean number of anurans per tree October 2014 I Volume 8 I Number 1 I e83 Water temperature (*C) Bromcliad height (cm) Tree height (m) McCracken and Forstner A, zehrina bromdiad host tree height ^ “ O CTi " o 1 1 1 Absent Present All Trees Metamorphosed Anurans A. zebritia brometiad elevation above forest floor 1 1 \ Absent Present All Trees Metamorphosed Anurans Number of A. zi%riiui bromeliads per tree o (N - A. zebritia bromcLiad heiglit « e t o o o O = o w o 1 1 1 Absent Present All Trees Metamorphosed Anurans A, zebrina bromebad leaf number Absent Present All Trees Metamorphosed Anurans A. z^hritui bromeliad water volume £ 0 1 Absent Present All Trees Metamorphosed Anurans A. zebrina bromeliad vrater temperature A. zebrina bromeliad water pH 0 1 1 1 Absent Present All Trees Metamorphosed Anurans A, zebrina bromeliad water pH (post) Absent Present All Trees Metamorphosed Anurans Air temperature at^. zebrina bromeliad collection site Fig. 5. Box plots of recorded habitat variables for Aechmea zebrina bromeliads collected from all trees, bromeliads with >1 metamorphosed anuran, and bro- meliads absent of anurans. Asterisks represent the mean, open circles are outliers, horizontal line inside box is the median, top and bottom lines of the rectan- gle are the 3rd and 1st quartiles (Q3 and Ql), respec- tively, and the top and bottom whiskers are maximum and minimum values excluding outliers, respectfully. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (70) October 2014 I Volume 8 I Number 1 I e83 Herpetofaunal community of a high canopy tank bromeliad in our study by the tree size class crown area calculated from Asner et al. (2002) and multiplied by 100 to gener- ate an estimated density of individuals per 100 m^. All calculations and statistics based on counts of metamorphosed anurans collected (not larval anurans) and conducted in the R statistical software (version 3.0.1) (R Development Core Team 2013). Results We sampled five bromeliads from each of 32 trees for a total of 160 A. zebrina bromeliads sampled. We collected 10 metamorphosed anuran species (Fig. 3), one gecko, one snake (Fig. 4), and two species of lizard were ob- served amongst bromeliad leaves but not collected (Ta- ble 1). A total of 95 metamorphosed anurans (x = 0.6 per bromeliad) were collected from 56 of the 160 bromeliads (35%) sampled. Between one and five individuals (x = 1.7), and up to two species were observed in single A. ze- brina bromeliads occupied by metamorphosed anurans. The species found together include (number of brome- liads with species together): Pristimantis aureolineatus and P. waoranii (7), P waoranii and P acuminatus (1), P. waoranii and P orphnolaimus (1), P. aureolineatus and Ranitomeya ventrimaculata (1). We also collected a minimum of four larval amphibian species from the water-filled leaf axils of A. zebrina bromeliads includ- ing Osteocephalus fuscifacies, O. planiceps, Ranitomeya variabilis, and R. ventrimaculatus. A total of 27 1 larval anurans were collected from 35 of the 160 bromeliads (21.9%) sampled, with 14 of the 35 larval occupied bro- meliads (40%) also occupied by >1 metamorphosed an- urans. Osteocephalus spp. tadpoles account for 60.5% {n = 1 64) of confirmed species identifications for all larval anurans, and these were collected from five bromeliads. A single O. fuscifacies and a single O. planiceps (both adults) were each found in separate bromeliads with lar- vae of same species (identified after rearing). The gecko, Thecadactylus solimoensis (formerly T. rapicauda), was found in an A. zebrina bromeliad amongst the outer leaf axils at 31.5 m above the forest floor in a tree 46.0 m tall (Bergmann and Russell 2007). The Banded cat-eyed snake, Leptodeira annulata, was found in a central leaf axil of an A. zebrina bromeliad at 43.5 m above the for- est floor in a tree 45.5 m tall (Fig. 4). Anolis transversalis was observed twice amongst the leaves of A. zebrina bro- meliads during collections; once on a bromeliad at ~27 m above the forest floor (36 m tall tree) and in another tree at ~35 m above the forest floor (41 m tall tree). A single male Anolis ortonii was observed displaying his dewlap on an outer leaf of an A. zebrina bromeliad at ~20 m above the forest floor in a tree 28 m tall. Summary statistics for all species reported in Table 1. Trees sampled for A. zebrina bromeliads were 28 to 49 m in height (x = 40.4 m + 5.5, n = 32), and 28 to 49 m in height (x = 40.2 m + 5.8, n = 27) for trees with >1 bromeliad occupied by metamorphosed anurans. Aech- mea zebrina bromeliads were collected at above ground elevations of 18.3 to 44.5 m (x = 32.9 m + 5.6, n = 160), and bromeliads occupied by >1 metamorphosed anurans occurred at elevations of 20.5 to 44.5 m (x = 32.1 m + 6.3, n = 56). The number of A. zebrina bromeliads per host tree was 19 to 150 individuals (x = 66 + 40, n = 32), and 19 to 150 individuals (x = 63 + 38, n = 27) for trees with >1 bromeliads occupied by metamorphosed anurans. Aechmea zebrina bromeliads were 45 to 126 cm in height (x = 75 + 14, n = 160), and 51 to 125 cm in height (x = 78 + 15, w = 56) for bromeliads occupied by >1 metamorphosed anurans. The number of mature leaves per A. zebrina was 14 to 46 (x = 28 + 6, w = 160), and 17 to 43 (x = 29 + 6, w = 56) for bromeliads occupied Table 1. Amphibians and reptiles collected or observed within Aechmea zebrina bromeliads. For each species the number observed, height range (bromeliad in tree), and mean height are provided. Only metamorphed anurans at time of collection included. Species Number observed Height range (m) Mean height (m) Osteocephalus fuscifacies 3 24 . 3 - 28.1 25.6 Osteocephalus planiceps 1 31.5 - Osteocephalus taurinus 1 30.6 - Pristimantis acuminatus 1 40.4 - Pristimantis aureolineatus 36 22 - 44.5 35.7 Pristimantis orphnolaimus 2 31 . 5 - 38.3 34.9 Pristimantis waoranii 35 21 . 2 - 43.9 31.9 Ranitomeya ventrimaculata 1 36.5 - Ranitomeya variabilis 9 25 . 7 - 35.2 30.9 Scinax ruber 6 33 . 8-35 34.8 Anolis ortonii 1 20 - Anolis transversalis 2 27-35 31 Thecadactylus solimoensis 1 31.5 - Leptodeira annulata 1 43.5 - Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (71) October 2014 I Volume 8 I Number 1 I e83 McCracken and Forstner by >1 metamorphosed anurans. The water volume of A. zebrina bromeliads was 42 to 3645 mL (x = 1343 + 656, n = 160), and 355 to 3645 mL (x = 1428 + 726, n = 56) for bromeliads oecupied by >1 metamorphosed anurans. Water temperature within an outer leaf axil of A. zebrina bromeliads at time of collection was 22.3 to 32.3 °C (x = 26.2 + 2.1, n = 160), and 22.6 to 31.2 °C (x = 26.2 + 1.9, n = 56) for bromeliads occupied by >1 metamorphosed anurans. Water pH within an outer leaf axil of A. zebrina bromeliads at time of collection was 2.82 to 6.34 (x = 4.18 + 0.66, n = 160), and 3.22 to 6.34 (x = 4.34 + 0.73, n = 56) for bromeliads occupied by >1 metamorphosed anurans. Water pH of sieved homogenized water for each A. zebrina bromeliad was 3.14 to 6.08 (x = 4.44 + 0.53, n = 160), and 3.60 to 6.08 (x = 4.48 + 0.55, n = 56) for bromeliads occupied by >1 metamorphosed anurans. Air temperature adjacent to bromeliads at time of collection was 21.1 to 34.6 °C (x = 27.8 + 2.8, n = 160), and 21.1 to 33.5 °C (x = 27.6 + 2.7, n = 56) for bromeliads occupied by >1 metamorphosed anurans. Summary statistics for bromeliads absent of anurans are contrasted with those given above in Fig. 5. By taking the mean number of metamorphosed an- urans per bromeliad (x = 0.6) and multiplying by the mean number of A. zebrina bromeliads per tree (x = 66), we calculated an estimated mean of 39.6 metamorphosed anurans occupying the A. zebrina bromeliads of an av- erage tree in our study. The Dominant class tree crown area from Asner et al. (2002) was 105.7 m^ (25.3 m tall) with a calculated 37.5 anurans per 100 m^, and the Super dominant class tree crown area was 311 m^ (46.1 m tall) with a calculated 12.7 anurans per 100 m^. Discussion Our study identified 14 species of herpetofauna (10 an- urans and four reptiles) utilizing the tank bromeliad A^c/^- mea zebrina as habitat in the high canopy environment of the northwestern Amazon Basin. A range of 1-5 meta- morphosed anurans per bromeliad, with up to two spe- cies occupying a single bromeliad, were detected in over one-third of the bromeliads sampled. The observation of larval, metamorphs, and adults of Osteocephalus fuscifa- cies confirm that this species is a phytotelm breeder as proposed by Jungfer et al. (2013). The observation of lar- val, metamorphs, and an Sidult Ranitomeya ventrimacula- ta confirm that this species does deposit tadpoles in high canopy bromeliads as proposed by Brown et al. (2011). Our detection of the gecko Thecadactylus solimoensis at 3 1 .5 m vertical height within the leaf axil of an A. zebrina bromeliad confirms this species use of bromeliads in the high canopy (Vitt and Zani 1997; Bergmann and Russell 2007). Our observation of the snake Leptodeira annulata within the leaf axils of an A. zebrina bromeliad at 43.5 m in the canopy is the highest recorded vertical height to our knowledge; L. annulata is described as terrestrial to Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (72) semi-arboreal with a previous maximum observed ver- tical height of 6 m above ground (Duellman 1978; Vitt 1996; Kacoliris 2006; Avila and Morals 2007). In McCracken and Forstner (2014) we analyzed the habitat data for differences among forest disturbance treatments and found no differences in habitat variables between treatments and no relationships between habitat variables and anuran occupancy or abundance. Also, we found differences between forest disturbance treatments for anuran abundance and occupancy; but report the summary statistics of the habitat data here as a resource characterizing the habitat occupied by canopy tank bro- meliad dwelling herpetofauna. Of particular interest in this study was the mildly acidic mean water pH (4.18 in situ in leaf axils, 4.34 in sieved homogenized water) in A. zebrina bromeliads; as this is within the range reported to affect development of embryonic and larval anurans (Beattie and Tyler-Jones 1992). However, bromeliads are a known breeding site for amphibians and we observed an abundance of aquatic invertebrates and larval anurans in our collections (Benzing 2000). Using the two largest tree size classes of Amazonian trees from Asner et al. (2002) as representative crown area for the shortest (28 m) and tallest (49 m) trees in our study, we calculated an estimate of 12.7-37.5 an- urans per 100 m^ of crown area for an average tree in our study. We consider this estimated range of canopy anuran density to be conservative because 1) the height of trees for the tree size classes used from Asner et al. (2002) are shorter than our shortest and tallest trees; 2) it is calculated on the two-dimensional space of the tree crown and does not include the vertical space occupied by a tree; 3) anurans were only collected from A. zebrina bromeliads and not other available habitat; and 4) the mean anuran abundance per tree in our study was used for calculations of both tree size classes, not accounting for the range of tree heights and number of bromeliads per tree. Regardless of these constraints, the estimated high anuran density of 37.5 anurans/100 m^ is the great- est of any reported density for tropical frog assemblages from comparable studies (e.g., 36.1 anurans/100 m2 at La Selva, Costa Rica [Lieberman 1986]; 15.5 an- urans/100 m^ at Rio Llullapichis, Peru [Toft 1980]; see also Allmon 1991 and Rocha et al. 2007 for compiled sites comparison). The low estimate of 12.7 anurans/100 m^ is still amongst the highest densities of reported stud- ies, particularly in South America (Allmon 1991; Rocha et al. 2007). A limitation of this comparison is that these studies rely on the method of quadrat surveys for density calculations, where the majority of observed anurans are going to be leaf-litter inhabitants or those that are within arms reach (~2 m vertical height). Achieving a more ac- curate canopy anuran density will require research sam- pling all available canopy microhabitats and recording crown measurements for all sampled trees. Within the herpetological community the use of the term “arboreal” has deviated from its recognized defini- October 2014 I Volume 8 I Number 1 I e83 Herpetofaunal community of a high canopy tank bromeliad tion of “inhabiting or frequenting trees” and taken on a broader meaning in reference to vertical habitat use by am- phibians and reptiles to simply mean living above ground level (Merriam-Webster.com. 2014. Merriam-Webster Dictionary. Available from http://www.merriam-webster. com [Accessed 27 April 2014]). While this definition suffices to distinguish these species (arboreal) from those occupying fossorial and ground level habitat, it does not adequately clarify the above ground vertical space uti- lized by a particular species. As an example, Doan (2003) reports the visual encounter survey (VES) method as the best way to sample for arboreal herpetofauna in rainfor- ests. The VES method only allows the researcher access to habitat within arms reach (~2 m vertical height) and fails entirely at observing animals within the other ~40-i- m of vertical habitat above in many rainforests. Arboreal herpetofauna may occur at vertical heights between >0 m to 88 m on vegetation and/or trees; simply referring to a species as arboreal provides no information about its oc- cupied vertical range (Spickler et al. 2006). To alleviate confusion and accurately represent the vertical space oc- cupied by a species or assemblage we propose two alter- natives to be used separately or preferably together. Eirst, basic descriptors delineating vertical zones for a defined forest type could be used to accompany “arboreal” (e.g., “arboreal within the understory” where “understory” has been defined as “near-ground nondominant vegetation”) (Dial et al. 2004). Second, authors should specify vertical height ranges when describing or discussing “arboreal” anurans (e.g., “the arboreal frog Pristimantis waoranii is found in the overstory at 20.5^4 m” where “overstory” has been defined as “high, dominant foliage”) (Dial et al. 2004). Providing vertical range data or descriptions is critical to understanding the many aspects of natural history for a species. Conclusion The canopy of tropical forests are among the most species-rich terrestrial habitats on Earth, yet remain a relatively unexplored biotic frontier (Basset et al. 2003; Lowman and Schowalter 2012). Our research has shown the tank bromeliad Aechmea zebrina to support a di- verse and abundant herpetofauna community in the harsh equatorial tree canopy environment of the Yasum Bio- sphere Reserve in the Amazon Basin of Ecuador. Addi- tionally, our canopy work has contributed to the descrip- tion of two new species of bromeliad-inhabiting anurans {Pristimantis aureolineatus [Guayasamin et al. 2006] and P. waoranii [McCracken et al. 2007]), the detection of Batrachochytrium dendrobatidis (Chytrid fungus) on anurans from the forest floor to the canopy in Amazo- nia (McCracken et al. 2009), and identified the use of high canopy bromeliads by the anuran Scinax ruber (Mc- Cracken and Eorstner 2014). While canopy surveys of tank bromeliads are labor intensive, they provide a very Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (73) effective technique for collecting data on canopy inhabit- ing organisms and associated microhabitat factors. Our estimates of canopy anuran densities, based on collections from a single species of bromeliad, demon- strate the potential ecological importance and current lack of knowledge on the canopy herpetofauna compo- nent in tropical systems. Typical inventories of herpeto- fauna in tropical forests are conducted at ground level (~2 m vertical height stratum) where microclimatic variables are more stable (Guayasamin et al. 2006). Sampling such shallow strata within the strongly vertical structure of these forests has likely served to bias metrics of herpeto- fauna assemblages by focusing on a narrow environmen- tal space and neglecting the large available habitat above into the canopy (Guayasamin et al. 2006; Scheffers et al. 2014). Euture inventory studies should routinely include canopy surveys to properly represent the herpetofauna of forested habitat. Use of the term “arboreal” in the herpetofauna litera- ture does not adequately deflne the vertical range of a species or assemblage. This serves to limit compilation and synthesis from the literature for the ecology of many of these tropical reptiles and amphibians. Our proposed amendments to accompany the description of arboreality in herpetofauna functionally serve to give scale and pro- vide a better understanding of the vertical habitat utilized by a species or assemblage. As research on canopy her- petofauna continues to expand, knowledge of the vertical space occupied will be essential to answering hypothe- sis-driven research questions and enacting sufficient con- servation measures to protect all species. Acknowledgments. — SEM sincerely thanks all the fieldwork assistants who have contributed to this work over the years. We thank all the staff at the Tiputini Bio- diversity Station - Universidad San Erancisco de Quito and the Yasuni Research Station - Pontificia Universi- dad Catblica del Ecuador. We also thank the Waorani and Kichwa peoples who allowed us to conduct fieldwork in their territories. Thank you to Bejat McCracken for everything, but especially the photography. Thank you to Jerad Tullis in the Department of Geography at Texas State University who constructed the mosaic satellite im- age in Eigure 1. Lastly, we thank all our funding sources: National Science Eoundation (Graduate Research Eel- lowship Program and a GK-12 grant No. 0742306), Tex- as State University - Department of Biology, the TAD- POLE Organization, Sigma Xi - The Scientific Research Society, Texas Academy of Science, and The Explorer’s Club. This work was conducted under permit numbers 006-IC-EA-PNY-RSO and 012-IC-EA-PNY-RSO issued by the Ministerio del Ambiente, Ecuador. Literature Cited Allmon WD. 1 99 1 . A plot study of forest floor litter frogs, central Amazon, Brazil. Journal of Tropical Ecology October 2014 I Volume 8 I Number 1 I e83 McCracken and Forstner 7: 503-522. Asner GP, Palace M, Keller M, Pereira R, Silva JNM, Zweede JC. 2002. Estimating canopy structure in an Amazon forest from laser range finder and IKONOS satellite observations. Biotropica 34: 483^92. Avila RW, Morals D. 2007. Notes on the ecology of the colubrid snake Leptodeira annulata in the Pantanal, Brazil. Herpetological Review 38: 278-280. Balslev H, Luteyn J, 011egaard B, Holm-Nielsen LB. 1987. Composition and structure of adjacent unflood- ed and floodplain forest in Amazonian Ecuador. Op- era Botanica 92: 37-57. Bass M, Einer M, Jenkins CN, Kreft H, Cisneros-Here- dia DE, McCracken SE, Pitman N, English PH, Swing K, Villa G, Di Eiore A, Voigt CC, Kunz TH. 2010. Global conservation significance of Ecuador’s Yasuni National Park. PLoS ONE 5: 1-22. Basset Y, Novotny V, Miller SE, Kitching RE. 2003. Conclusion: Arthropods, canopies and interpretable patterns. Pp. 394^06 In: Arthropods of Tropical Forests: Spatio-temporal dynamics and resource use in the canopy. Editors, Basset Y, Novotny V, Miller SE, Kitching RE. Cambridge University Press, Cam- bridge, United Kingdom. Beattie RC, Tyler-Jones R. 1992. The effects of low pH and aluminum on breeding success in the frog Rana temporaria. Journal of Herpetology 26: 353-360. Benzing DH. 2000. Bromeliaceae: Profile of an adaptive radiation. Cambridge University Press, Cambridge, United Kingdom. Bergmann PJ, Russell AP. 2007. Systematics and bioge- ography of the widespread Neotropical gekkonid ge- nus Thecadactylus (Squamata), with the description of a new cryptic species. Zoological Journal of the Linnean Society 149: 339-370. Blandin EC. 1976. El Clima y sus Caracteristicas en el Ecuador. Biblioteca Ecuador. XI Asamblea General del Institute Panamericano de Geograffa e Historia, Quito, Ecuador. Brown JL, Twomey E, Amezquita A, Barbosa de Souza M, Caldwell JP, Lotters S, von May R, Melo-Sampaio PR, Mej fa- Vargas D, Perez-Pena P, Pepper M, Poel- man EH, Sanchez-Rodrfguez M, Summers K. 2011. A taxonomic revision of the Neotropical poison frog genus Ranitomeya (Amphibia: Dendrobatidae). Zoo- taxa 3083: 1-120. Cardeliis CL, Chazdon RE. 2005. Inner-crown microen- vironments of two emergent tree species in a lowland wet forest. Biotropica 37: 238-244. Chaparro JC, Pramuk JB, Gluesenkamp AG. 2007. A new species of arboreal Rhinella (Anura: Bufonidae) from cloud forest of southeastern Peru. Herpetologica 63: 203-212. Dial R, Bloodworth B, Lee A, Boyne P, Heys J. 2004. The distribution of free space and its relation to cano- py composition at six forest sites. Forest Science 50: 312-325. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (74) Doan TM. 2003. Which methods are most effective for surveying rain forest herpetofauna? Journal of Herpe- tology 37: 72-81. Duellman WE. 1978. The biology of an equatorial her- petofauna in Amazonian Ecuador. Miscellaneous Publications of the University of Kansas Museum of Natural History 65: 1-352. Einer M, Vijay V, Ponce E, Jenkins CN, Kahn TR. 2009. Ecuador’s Yasuni Biosphere Reserve: A brief modern history and conservation challenges. Environmental Research Letters 4: 1-15. Guayasamin JM, Eunk WC. 2009. The amphibian com- munity at Yanayacu Biological Station, Ecuador, with a comparison of vertical microhabitat use among Pristimantis species and the description of a new spe- cies of the Pristimantis myersi group. Zootaxa 2220: 41-66. Guayasamin JM, Ron S, Cisneros-Heredia DE, Lamar W, McCracken SE. 2006. A new species of frog of the Eleutherodactylus lacrimosus assemblage (Leptodac- tylidae) from the western Amazon Basin, with com- ments on the utility of canopy surveys in lowlands rainforest. Herpetologica 62: 191-202. Jungfer KH, Eaivovich J, Padial JM, Castro vieJo-Eisher S, Lyra ML, Berneck BVM, Iglesias PP, Kok PJR, MacCulloch RD, Rodrigues MT, Verdade VK, Tor- res Gastello CP, Chaparro JC, Valdujo PH, Reichle S, Moravec J, Gvozdik V, Gagliardi-Urrutia G, Ernst R, De la Riva I, Means DB, Lima AP, Senaris JC, Wheeler WC, Haddad CEB. 2013. Systematics of spiny-backed treefrogs (Hylidae: Osteocephalus): An Amazonian puzzle. Zoologica Scripta 42: 351-380. Kacoliris EP, Berkunsky I, Williams J. 2006. Herpetofau- na of Impenetrable, Argentinean Great Chaco. Phyl- lomedusa 5: 149-158. Kays R, Allison A. 2001. Arboreal tropical forest verte- brates: Current knowledge and research trends. Plant Ecology 153: 109-120. Kitching RE. 2000. Food Webs and Container Habi- tats: The natural history and ecology of phytotelma- ta. Cambridge University Press, Cambridge, United Kingdom. Lehr E, Torres C, Suarez J. 2007. A new species of arbo- real Eleutherodactylus (Anura: Leptodactylidae) from the Amazonian lowlands of central Peru. Herpetolog- ica 63: 94-99. Lieberman SS. 1986. Ecology of the leaf litter herpe- tofauna of a neotropical rainforest: La Selva, Costa Rica.. Acta Zoologica Mexicana 15: 1-72. Lowman MD, Rinker HB. 2004. Forest Canopies. El- sevier Academic Press, Burlington, Massachusetts, United States. Lowman MD, Scho waiter TD. 2012. Plant science in for- est canopies - the first 30 years of advances and chal- lenges (1980-2010). New Phytolologist 194: 12-27. McCracken SE, Eorstner MRJ. 2006. Reproductive ecol- ogy and behavior of Eleutherodactylus aureolineatus October 2014 I Volume 8 I Number 1 I e83 Herpetofaunal community of a high canopy tank bromeliad (Anura, Brachycephalidae) in the canopy of the Upper Amazon Basin, Ecuador. Phyllomedusa 5: 135-143. McCracken SF, Forstner MRJ. 2008. Bromeliad patch sampling technique for canopy herpetofauna in Neo- tropical forests. Herpetological Review 39: 170-174. McCracken SF, Forstner MRJ. 2014. Oil Road Effects on the Anuran Community of a High Canopy Tank Bromeliad (Aechmea zebrina) in the Upper Amazon Basin, Ecuador. PLoS ONE 9: 1-12. McCracken SF, Forstner MRJ, Dixon JR. 2007. A new species of the Eleutherodactylus lacrimosus assem- blage (Anura, Brachycephalidae) from the lowland rainforest canopy of Yasuni National Park, Amazo- nian Ecuador. Phyllomedusa 6: 23-35. McCracken SF, Gaertner JP, Forstner MRJ, Hahn D. 2009. Detection of Batrachochytrium dendrobati- dis in amphibians from the forest floor to the upper canopy of an Ecuadorian Amazon lowland rainforest. Herpetological Review 40: 190-195. Moffett MW. 2000. What’s “Up”? A critical look at the basic terms of canopy biology. Biotropica 32: 569- 596. Nadkami NM. 1994. Diversity of species and interac- tions in the upper tree canopy of forest ecosystems. American Zoologist 34: 70-78. Ozanne CMP, Anhuf D, Boulter SF, Keller M, Kitching RF, Komer C, Meinzer FC, Mitchell AW, Nakashi- zuka T, Dias PUS, Stork NE, Wright SJ, Yoshimura M. 2003. Biodiversity meets the atmosphere: A global view of forest canopies. Science 301: 183-186. Perry D. 1978. A method of access into the crowns of emergent trees. Biotropica 10: 155-157. R Development Core Team. 2013. R: A language and en- vironment for statistical computing. In: R Eoundation for Statistical Computing Vienna, Austria. Rocha CFD, Vrcibradic D, Kiefer MC, Almeida-Gomes M, Borges- Junior VNT, Cameiro PCF, Marra RV, Almeida-Santos P, Siqueira CC, Goyannes- Araujo P, Fernandes CGA, Rubiao ECN, Van Sluys M. 2007. A survey of the leaf-litter frog assembly from an At- lantic forest area (Reserva Ecoldgica de Guapia 9 u) in Rio de Janeiro State, Brazil, with an estimate of frog densities. Tropical Zoology 20: 99-108. Scheffers BR, Edwards DP, Diesmos A, Williams SE, Evans TE. 2014. Microhabitats reduce animal’s expo- sure to climate extremes. Global Change Biology 20: 495-503. Scheffers BR, Phillips BE, Faurance WF, Sodhi NS, Diesmos A, Williams SE. 2013. Increasing arboreality with altitude: A novel biogeographic dimension. Pro- ceedings of the Royal Society B: Biological Sciences 280: 20131581. Smith FB. 1953. Notes on Bromeliacea II. Phytologia 4: 355-368. Spickler JC, Sillett SC, Marks SB, Welsh HW. 2006. Evi- dence of a new niche for a North American salaman- der: Aneides vagrans residing in the canopy of old- growth redwood forest. Herpetological Conservation and Biology 1: 16-26. Stewart MM. 1985. Arboreal habitat use and parachuting by a subtropical forest frog. Journal of Herpetology 19: 391^01. Toft CA. 1980. Seasonal variation in populations of Pan- amanian litter frogs anid their prey: A comparisoin of wetter and drier sites. Oecologia 47: 34-38. Vitt FJ. 1996. Ecological observations on the tropical colubrid snake Leptodeira annulata. Herpetological Natural History 4: 69-76. Vitt FJ, Zani PA. 1996. Ecology of the elusive tropical lizard Tropidurus [= Uracentron]flaviceps (Tropidur- idae) in lowland rain forest of Ecuador. Herpetologica 52: 121-132. Vitt FJ, Zani PA. 1997. Ecology of the nocturnal lizard Thecadactylus rapicauda (Sauna: Gekkonidae) in the Amazon region. Herpetologica 53: 165-179. Received: 01 May 2014 Accepted: 08 August 2014 Published: 30 October 2014 B Shawn F. McCracken is a Research Assistant Professor at Texas State University in San Marcos, Texas, USA. He received his B.A. in biology and a Ph.D. in aquatic resources at Texas State University. He is the founder and executive director of the TADPOLE Organization. His research interests include the conservation, ecol- S9F ogy, and systematics of amphibians; with an emphasis on the effects of anthropogenic disturbance to amphib- ian diversity and abundance in tropical rainforests. His current research focuses on tbe effects of deforestation to canopy inhabiting herpetofauna and microclimate in Amazonian Ecuador, with a concentration on epiphytic canopy tank bromeliads. In the USA, he conducts research on the endangered Houston toad (Anaxyrus hous- tonensis) and the state threatened Texas tortoise (Gopherus berlandieri). Michael R. J. Forstner is a Professor in Biology at Texas State University, and the Alexander-Stone Chair in Genetics. He has a B.S. from Southwest Texas State University, M.S. from Sul Ross State University, and a Ph.D. from Texas A&M University. He has broad interests in the effective conservation of rare taxa, par- ticularly reptiles and amphibians. The students and colleagues working with him seek to provide genetic and ecological data relevant to those conservation efforts. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (75) October 2014 I Volume 8 I Number 1 I e83 Copyright: © 2014 Torres-Carvajal et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptile Conservation-, of- ficial journal website . Amphibian & Reptiie Conservation 8(1) [Special Section]: 76-88. A new species of Phoiidoboius (Squamata: Gymnophthalmidae) from the Andes of southern Ecuador ^Omar Torres-Carvajal, ^Pablo J. Venegas, ^Simon E. Lobos, "^Paola Mafla-Endara, and ^Pedro M. Sales Nunes de Zoologia, Escuela de Ciencias Bioldgicas, Pontificia Universidad Catolica del Ecuador, Avenida 12 de Octubre 1076y Roca, Apartado 17-01-2184, Quito, ECUADOR ^Division de Herpetologia-Centro de Ornitologia y Biodiversidad (CORBIDI), Santa Rita N°105 36 Of. 202, Urb. Huertos de San Antonio, Surco, Lima, PERILED epartamento de Ciencias Naturales, Universidad Tecnica Particular de Loja, San Cayetano Alto s/n C.P 11 01 608, Loja, ECUADOR ^Universidade Federal de Pernambuco, Centro de Ciencias Bioldgicas, Departamento de Zoologia, Av. Professor Moraes Rego, s/n. Cidade Universitdria CEP 50670-901, Recife, PE, BRAZIL Abstract . — ^We describe a new species of Phoiidoboius lizard from the Amazonian slopes of the Andes of southern Ecuador. Among other characters, the new species differs from other species of Phoiidoboius in having a distinct diagonal white stripe extending from the fourth genial scale to the fore limb. We present a phylogeny based on mitochondrial DNA sequence data as additional evidence supporting delimitation of the new species, which is sister to all other species of Phoiidoboius. Our phylogeny further supports the south-to-north speciation hypothesis proposed for other lizard clades from the northern Andes. Key words. Clade Phoiidoboius, DNA, lizard, phylogeny. South America, systematics Citation: Torres-Carvajal O, Venegas PJ, Lobos SE, Mafla-Endara P, Nunes PMS. 2014. A new species of Phoiidoboius (Squamata: Gymnophthalmidae) from the Andes of southern Ecuador. Amphibian & Reptile Conservation 8(1) [Special Section]: 76-88 (e84). Introduction The gynmophthalmid lizard clade Phoiidoboius was recently defined by Torres-Carvajal and Mafla-Endara (2013) as the largest crown clade conimnmg Phoiidoboius montium Peters, 1863, but not Macropholidus ruthveni Noble, 1921. This phylogenetic definition (de Queiroz and Gauthier 1994) is based on a phylogenetic tree ob- tained from analyses of mitochondrial DNA nucleotide sequence data (Torres-Carvajal and Mafla-Endara 2013), and is in conflict with previous non-phylogenetic defini- tions of both Phoiidoboius and Macropholidus (Monta- nucci 1973; Reeder 1996) based on morphological data. As defined by Torres-Carvajal and Mafla-Endara (2013), Phoiidoboius contains four species — P. affinis, P mac- brydei, P montium, and P. prefrontalis. Contrary to pre- vious taxonomic arrangements (Montanucci 1973; Reed- er 1996), “P.” annectens was shown to be part of the clade (traditionally ranked as a genus) Macropholidus. In addition, Torres-Carvajal and Mafla-Endara (2013) concluded that the controversial generic allocation of P. anomalus from southern Peru (Montanucci 1973; Reeder 1996) still remains to be established. Phoiidoboius lizards occur between 1,800 and 4,100 m along the southern part of the northern Andes (i.e., Ecuador and southern Colombia). Only one species, P. macbrydei, occurs also in the Huancabamba Depression in extreme southern Ecuador and possibly northern Peru. Herein, we describe a new species of Phoiidoboius from the Andes in southern Ecuador using data on morphology and color pattern. We also present molecular evidence supporting recognition of the new species by performing phylogenetic analyses of nucleotide sequence data. Methods Morphological data: Type specimens and additional specimens examined (Appendix 1) were deposited in the herpetological collection at Museo de Zoologia, Ponti- ficia Universidad Catolica del Ecuador, Quito (QCAZ). The following measurements were taken with a digital caliper and recorded to the nearest 0.1 mm, except for tail length, which was taken with a ruler and recorded to the nearest millimeter: head length (HE), head width (HW), shank length (ShL), axilla-groin distance (AGD), snout-vent length (SVL), and tail length (TL). Sex was determined by dissection or by noting the presence of everted hemipenes. We follow the terminology of Reeder (1996) for description of the holotype and scale counts. Data for other species of Phoiidoboius were taken from Montanucci (1973). The left hemipenis of two type specimens of the new species (QCAZ 4998 and 4999) were prepared following the procedures of Manzani and Abe (1988), as modified by Pesantes (1994) and Zaher (1999), where the retractor Correspondence. ^ omartorcar@gmail.com (Corresponding author); ^sancarranca@yahoo.es ; ^lobossimon@gmail.com; "^paola. mmafen@gmail. com; ^pedro. nunes@gmail. com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (76) November 2014 | Volume 8 | Number 1 | e84 Torres-Carvajal et al. Fig. 1. Holotype (QCAZ 4998; SVL = 45.52 mm) of Pholidobolus hillisi sp. nov. in dorsal (A) and ventral (B) views. Photographs by OTC. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (77) November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador muscle is manually separated and the everted organ is filled with stained petroleum jelly and paraffin. In addi- tion, the hemipenial calcareous structures were stained in an alcoholic solution of Alizarin Red, following the adaptation of the procedures of Uzzell (1973) proposed by Nunes et al. (2012). Description of the hemipenes fol- lows the terminology of Dowling and Savage (1960), Savage (1997), Myers and Donnelly (2001, 2008), and Nunes et al. (2012). DNA sequence data: Total genomic DNA was di- gested and extracted from liver or muscle tissue using a guanidinium isothiocyanate extraction protocol. Tissue samples were first mixed with Proteinase K and a lysis buffer and digested overnight prior to extraction. DNA samples were quantified using a Nanodrop® ND-1000 (NanoDrop Technologies, Inc), re- suspended and diluted to 25 ng/ul in ddH20 prior to amplification. Using primers and amplification protocols from the literature (Pellegrino et al. 2001; Torres-Carvajal and Mafla-Endara 2013) we obtained 1,573 nucleotides (nt) representing mitochondrial genes 12S (344 nt), 16S (549 nt), and ND4 (680 nt) from three individuals of the new species described herein (GenBank accession numbers KP090167-KP090175). Chronophylogenetic analyses: We added the three sequences generated in this study to the mtDNA dataset of Torres-Carvajal and Mafla-Endara (2013). Editing, assembly, and alignment of sequences were performed with Geneious ProTM 5.3 (Biomatters Ltd. 2010). Genes were combined into a single dataset with three partitions, one per gene. The model of evolution for each partition was obtained in jModeltest 2 (Darriba et al. 2012) under the Akaike information criterion. Chronophylogenetic analyses were performed in Beast 2.1.3 (Bouckaert et al. 2014) as described in Torres-Carvajal and Mafla-Endara (2013), except that we performed four independent 108 generation runs with random starting trees, sampling ev- ery 10,000 generations. The resultant 36,000 trees were used to calculate posterior probabilities (PP) for each bi- partition in a maximum clade credibility tree in TreeAn- notator 2.1.2 (Rambaut and Drummond 2014). Systematics: The taxonomic conclusions of this study are based on the observation of morphological features and color pattern, as well as inferred phylogenetic rela- tionships. We consider this information as species de- limitation criteria following a general lineage or unified species concept (de Queiroz 1998, 2007). Pholidobolus hillisi sp. nov. urn:lsid:zoobank.org:act:EB5A9DDD-742C-456F-B5C9-6E57EDEEE698 Proposed standard English name: Cuilanes of Hillis Proposed standard Spanish name: Cuilanes de Hillis Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (78) Holotype: QCAZ 4998 (Figs. 1, 2), adult male, Ecua- dor, Provincia Zamora-Chinchipe, near San Francisco Research Station on Loja-Zamora road, 3°57’57”S, 79°4’45”W, WGS84, 1,840 m,21 July 2012, collected by Santiago R. Ron, Andres Merino, Fernando Ayala, Teresa Camacho, and Martin Cohen. Paratypes (5): ECUADOR: Provincia Zamora- Chinchipe: QCAZ 4999 (adult male), 5000 (juvenile female), same data as holotype; QCAZ 6840 (adult fe- male), 6842 (adult female), 6844 (adult male), San Fran- cisco Research Station, 3°58’14”S, 79°4’41”W, WGS84, 1,840 m, 29 October 2004, 9 June 2005, and 29 Septem- ber 2005, respectively, collected by Kristin Roos, Alban Pfeiffer, Andy Fries, Ulf Soltau, and Florian Werner. Diagnosis: Pholidobolus hillisi is unique among spe- cies of Pholidobolus in having a distinct diagonal white stripe on each side of the chin, extending from the fourth genial to the fore limb (Fig. 3). It further differs from all species of Pholidobolus, except P. affinis, in having three supraoculars (two in P. macbrydei, P montium, and P prefrontalis). Pholidobolus affinis differs from the new species by having flanks with black reticulations on a reddish orange ground color (flanks brown in P. hillisi'. Fig. 4). The new species also can be distinguished from P. montium and P. macbrydei by the presence of prefrontal scales (absent in the last two species). While P. hillisi shares with P. affinis and P. prefrontalis the presence of prefrontal scales, it differs from them in having a dark brown dorsum with a conspicuous light brown vertebral stripe (dorsum pale brown without a vertebral stripe in P. affinis and P. prefrontalis'. Fig. 4). Furthermore, P. hillisi has fewer dorsal scales in transverse rows (28-31) than P. affinis (45-55), P. montium (35-50), P. prefrontalis (37^6), and P. macbrydei (31^3). Pholidobolus hillisi shares with all other recognized species of Pholidobolus the absence of a single trans- parent palpebral disc and the presence of a ventrolateral fold between fore and hind limbs. These characters dis- tinguish members of Pholidobolus from members of its sister clade Macropholidus (Torres-Carvajal and Mafla- Endara 2013). Characterization: (1) Three supraoculars, anterior- most larger than posterior one; (2) prefrontals present; (3) femoral pores present in both sexes; (4) two to five opaque lower eyelid scales; (5) scales on dorsal surface of neck striated, becoming keeled from fore limbs to tail; (6) two or four rows of lateral granules at midbody; (7) 28-31 dorsal scales between occipital and posterior mar- gin of hind limb; (8) lateral body fold present; (9) keeled ventrolateral scales on each side absent; (10) dorsum dark brown with a conspicuous narrow, pale brown, ver- tebral stripe that becomes grayish brown towards the tail; (11) labial stripe white; (12) sides of body dark brown; November 2014 I Volume 8 I Number 1 I e84 Torres-Carvajal et al. Fig. 2. Head of the holotype (QCAZ 4998) of Pholidobolus hillisi sp. nov. in dorsal (A), lateral (B), and ventral (C) views. Photo- graphs by OTC. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (79) November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador Fig. 3. Head of five speeies of Pholidobolus in ventral view. (A) P. affinis', (B) P. hillisi sp. nov.; (C) P. macbryder, (D) P montiunr, (E) P prefrontalis. Photographs by OTC. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (80) November 2014 I Volume 8 I Number 1 I e84 Torres-Carvajal et al. (13) white stripe along fore limb present; (14) a distinct diagonal white stripe on each side of the chin, extending from the fourth genial to the fore limb; (15) adult males with red flecks and ocelli (black with white center) dorsal to insertion of fore and hind lim bs. Description of hoiotype: Adult male (QCAZ 4998); snout- vent length 45.52 mm; tail length 104 mm; dorsal and lateral head scales juxtaposed, finely wrinkled; ros- tral hexagonal, 2.09 times as wide as high; frontonasal pentagonal, wider than long, laterally in contact with na- sal, smaller than frontal; prefrontals pentagonal, nearly as wide as long, with medial suture, laterally in contact with loreal and first superciliary; frontal hexagonal, lon- ger than wide, slightly wider anteriorly, in contact with the prefrontals and supraoculars I and II on each side; frontoparietals pentagonal, longer than wide, with me- dial suture, each in contact laterally with supraoculars II and III; interparietal roughly hexagonal, lateral borders parallel to each other; parietals slightly smaller than in- terparietal, tetragonal and positioned anterolaterally to interparietal, each in contact laterally with supraocular III and dorsalmost postocular; postparietals three, medial scale smaller than laterals; supralabials seven, fourth lon- gest and below the center of eye; infralabials five, fourth below the center of eye; temporals enlarged, irregularly hexagonal, juxtaposed, smooth; two large supratemporal scales, smooth; nasal divided, irregularly pentagonal, longer than wide, in contact with rostral anteriorly, first and second supralabials ventrally, frontonasal dorsally, loreal posterodorsally and frenocular posteroventrally; nostril on ventral aspect of nasal, directed lateroposteri- orly, piercing nasal suture; loreal rectangular; frenocular enlarged, in contact with nasal, separating loreal from supralabials; supraoculars three, with the first being the largest; four elongate superciliaries, first one enlarged, in contact with loreal; palpebral disk divided into two scales, pigmented; suboculars three, elongated and simi- lar in size; three postoculars, medial one smaller than the others; ear opening vertically oval, without denticulate margins; tympanum recessed into a shallow auditory me- atus; mental semicircular, wider than long; postmental pentagonal, slightly wider than long, followed posteri- orly by four pairs of genials, the anterior two in contact medially and the posterior two separated by postgenials; all genials in contact with infralabials; gulars imbricate, smooth, widened in two longitudinal rows; gular fold incomplete; posterior row of gulars (collar) with four scales, the medial two distinctly widened. Scales on nape similar in size to dorsals, except for the anteriormost that are widened; scales on sides of neck small and granular; dorsal scales elongated, imbricate, arranged in transverse rows; scales on dorsal surface of neck striated, becoming keeled from fore limbs to the tail; number of dorsal scales between occipital and poste- rior margin of hind limbs 28; dorsal scale rows in a trans- verse line at midbody 30; one row of smooth, enlarged Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (81) ventrolateral scales on each side; dorsals separated from ventrals by three rows of small scales at the level of the 13th row of ventrals; lateral body fold present; ventrals smooth, wider than long, arranged in 20 transverse rows between the collar fold and preanals; six ventral scales in a transverse row at midbody; subcaudals smooth; limbs overlap when adpressed against body; axillary region composed of granular scales; scales on dorsal surface of fore limb striated, imbricate; scales on ventral surface of fore limb granular; two thick, smooth thenar scales; su- pradigitals (left/right) 3/3 on finger I, 6/6 on II, 8/8 on III, 9/9 on IV, 6/6 on V; supradigitals 3/3 on toe I, 6/6 on II, 9/9 on III, 11/12 on IV, 8/8 on V; subdigital lamellae of fore limb single, 5/5 on finger I, 8/9 on II, 13/13 on III, 14/14 on IV, 8/9 on V; subdigital lamellae on toes I and II single, on toe III paired on the distal half, on toe IV all paired, on toe V paired at the base; number of subdigi- tal lamellae (pairs when applicable) 6/5 on toe I, 9/9 on II, 13/14 on III, 19/20 on IV, 12/12 on V; groin region with small, imbricate scales; scales on dorsal surface of hind limbs striated and imbricated; scales on ventral sur- face of hind limbs smooth; scales on posterior surface of hind limbs granular; six femoral pores on each leg; preanal pores absent; cloacal plate paired, bordered by four scales anteriorly, of which the two medialmost are enlarged. Measurements (mm) and proportions of the hoiotype: HL 12.6; HW 9.3; ShL 5.2; AGD 24.6; TL/SVL 1.72; HL/SVL0.25; HW/SVL0.18; ShL/SVLO.lO; AGD/SVL 0.48. Hemipenial morphology (Fig. 5): Both organs ex- tend along approximately nine millimeters in length. The lobes of the organs are fully everted and each hemipenis is fully expanded. The hemipenial body is roughly conical in shape, with the base distinctly thinner than the rest of the organ, ending in two small lobes with apical folds in the apex. The sulcus spermaticus is central in position, originat- ing at the base of the organ, which bears a fleshy fold partially overlapping the sulcus spermaticus. From this point on, the sulcus proceeds in a straight line towards the lobes, and acquires an S-shape at the first third of the body. The sulcus becomes broader at halfway the length of the hemipenial body, and returns to its regular width at the apical region; it gets divided in two branches at the lobular crotch. Just before the crotch, the central region of the sulcus bears a tiny fleshy fold, which is not part of the sulcus division. From this point on, the two branch- es of the sulcus run on the medial regions of the lobes among conspicuous lobular folds. The sulcate face of the hemipenial body presents two nude areas, parallel to the sulcus spermaticus, which run throughout the hemipenial body, getting thinner and encircling the base of the lobes. The lateral and asulcate faces of the hemipenial body are ornamented with 28-30 rows of roughly equidistant flounces with calcareous spinules. The first four rows are November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador Fig. 4. Five species of Pholidobolus from Ecuador. (A) P. affinis', sp. nov. Photographs by OTC (A, B, C, D) and S. R. Ron (E). straight, with a large series of spinules on the central as- pect of the asulcate face, and small isolated series of 5-6 spinules bordering the nude areas parallel to the sulcus spermaticus. A V-shaped nude area at the central asulcate face of the body separates the remaining flounces. The flfth and sixth flounces are also interrupted laterally by an extension of the basal nude area. From the seventh to the apical-most one, the flounces cross the lateral aspects of the organ from the sulcate to the asulcate face, initially in roughly straight lines, gradually assuming chevron- shapes and getting reduced in length towards the apex of the organ. The region between the asulcate and the lateral sur- faces is marked by a conspicuous unevenness forming a bulge, which is shared by closely related species, such as Macropholidus annectens, M. huancabambae, M. ruth- (B) P. macbrydei; (C) P montiunr, (D) P prefrontalis; (E) P hillisi veni, Pholidobolus affinis, P macbrydei, P montium, and P prefrontalis (Nunes, 2011). Color of holotype in preservative: Dorsal back- ground uniformly dark brown with a narrow light brown vertebral stripe extending from occiput onto tail; ver- tebral stripe slightly wider anteriorly; dorsal surface of head light brown medially (rostral, frontonasal, prefron- tals, frontal and frontoparietals) and dark brown laterally (including supraoculars); white supralabial longitudinal stripe extending from first supralabial to fore limb; lateral aspect of neck dark brown with a dorsolateral light brown stripe that extends posteriorly along the flanks to the hind limbs; ventrolateral aspect of head and neck with a lon- gitudinal white stripe extending posteriorly from fourth genial to insertion of fore limb and then laterally along Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (82) November 2014 I Volume 8 I Number 1 I e84 Torres-Carvajal et al. Fig. 5. Left hemipenis of Pholidobolus hillisi sp. nov. (QCAZ 4999) in sulcate (left), lateral (middle), and asulcate (right) views. Photographs by P Nunes. upper arm; fore limbs with scattered ocelli (black with white center); flanks grayish brown with two dorsolateral stripes, the dorsal one light brown and the ventral one dark brown; tail light brown dorsally and dark brown on the sides; two and three well-defined, small ocelli (black with white center) dorsal to insertion of fore and hind limbs, respectively; ventral surface of head gray, with dirty cream genials and scattered brown marks; chest, belly and ventral surface of limbs and tail dark gray. Variation: Measurements and scale counts of Pholi- dobolus hillisi are presented in Table 1. Superciliaries usually four, five in QCAZ 6840; supralabials usually seven (eight of left side of specimen QCAZ 6840). Rows of lateral granules at midbody two (QCAZ 4999, 6844) to four (QCAZ 6842). Three specimens including the holo- type, with a ventrolateral row of smooth enlarged scales (QCAZ 4999, 6840). Specimen QCAZ 6842 has a tiny scale separating the cloacal scales posteriorly; all four scales bordering the cloacal plate anteriorly are similar in size in two specimens (QCAZ 4999, 6844), whereas the lateralmost scales overlap the cloacal scales in one specimen (QCAZ 6840). No variation was observed in color pattern in preser- vative among adult males. They can be distinguished from females by the presence of ocelli and pale flecks around insertion of fore and hind limbs. Moreover, the characteristic diagonal white stripe on each side of the chin that extends from the fourth genial to the forearm is more conspicuous in males than in females. Females are larger (maximum SVL 55.7 mm, n=3) than males (maxi- mum SVL 51.1 mm, n=3). Coloration in life of an adult male paratype (QCAZ 4999) was similar to the holotype’s coloration in pre- servative described above, except that specimen QCAZ 4999 had small red flecks both at insertion of fore limbs Table 1. Sexual variation in lepidosis and measurements of Pholidobolus hillisi sp. nov. Range followed by mean + standard devia- tion are given. Character Males (n=3) Females (n=3) Dorsal scales between occipital and posterior margin of hind limb 28-30 (29+1) 29-31 (30+1) Dorsal scale rows in a transverse line at midbody 27-34 (30.33+3.51) 29-35 (31+3.46) Ventral scales between collar fold and preanals 18-20 (20.33+1.15) 18-19 (18.67+0.58) Ventral scale rows in a transverse line at midbody 6-7 (6.67+0.58) 6 Subdigital lamellae on Finger IV 14-15 (14.33+43.0) 13-15 (13.67+1.15) Subdigital lamellae on Toe IV 19-20 (19.33+0.58) 19 Femoral pores 5-8 (6.33+1.52) 2-5 (3.5) (n=2) Maximum SVL 51.1 55.7 TL/SVL 1.86 (n=l) 1.84-2.14(1.99) (n=2) Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (83) November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador 0.15 0.10 0.05 0 Neusticurus rudls Placosoma glabellum Bachia flavescens Cercosaura quadrilineata Potamites ecpieopus Proctoporus bolivianus Riama cashcaensis Macropholidus ruthveni M. annectens M. annectens M. huancabambae M. huancabambae M. huancabambae Pholidobolus hillisi sp. nov. P hillisi sp. nov. P. hillisi sp. nov. P macbrydei P. macbrydei P. pre frontalis P pre frontalis P. montium P montium P affinis P. affinis Fig. 6. Maximum clade credibility tree inferred from the analysis of a dataset eontaining three mitochondrial genes under uneor- related, log normally distributed rates; branch lengths are in substitutions per site. Posterior probability values are shown above branches; asterisks correspond to values of 1. extending onto sides of neck and at insertion of hind limbs extending onto base of tail. In addition, the lateral white stripe that starts on first supralabial extends further posteriorly along flanks in specimen QCAZ 4999 (Fig. 4). Phylogenetic relationships: The maximum clade credibility tree resulting from the chronophylogenetic analysis supports inclusion of the new species within the Pholidobolus clade (Torres-Carvajal and Mafla-Endara 2013) with strong support (PP = 0.96; Fig. 6). Phyloge- netic relationships among other species of Pholidobolus and species of Macropholidus are identical to those ob- tained by Torres-Carvajal and Mafla-Endara (2013). Macropholidus ruthveni is sister (PP = 0.99) to a clade containing both M. annectens and M. huancabambae (PP = 1). Pholidobolus macbrydei is sister (PP = 0.91) to a clade with the three remaining species of Pholidobolus; the latter clade included P prefrontalis as sister (PP = 0.99) to a clade containing P. affinis and P. montium as sister taxa (PP = 0.99). In contrast to the results reported by Torres-Carvajal and Mafla-Endara (2013), the chrono- phylogenetic tree inferred in this paper suggests that the diversification of the clades Macropholidus and Pholi- dobolus occurred at about the same time (Fig. 6). Distribution and ecology: Pholidobolus hillisi inhab- its low montane forests in the eastern slopes of the Andes of southern Ecuador. This area represents a weather di- vide between the humid Amazon and the dry Inter- Ande- an regions (Beck et al. 2008). The new species is known from Provincia Zamora-Chinchipe, at 1,840 m (Fig. 7), in the deep valley of the Zamora river. The only gym- 71- w »-V¥ n-w ii-W w Jl- w ta-ff Tf-'^ Fig. 7. Distribution of Pholidobolus in Ecuador. P. affinis (white circles); P. macbrydei (blue circles); P. montium (green circles); P. prefrontalis (orange circles); P hillisi sp. nov (red circle). nophthalmid species known to occur in sympatry with P. hillisi is Alopoglossus buckleyi, although P macbry- dei is parapatrically distributed (Fig. 7). Two specimens (QCAZ 4998, 4999) were found under logs and rocks next to the Zamora river between 1 130 hrs and 1145 hrs. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (84) November 2014 I Volume 8 I Number 1 I e84 Torres-Carvajal et al. whereas another specimen (QCAZ 5000) was basking on a rock next to the road at 1200 hrs. Other specimens (QCAZ 6840, 6842, 6844) were found and captured by a domestic cat around the San Francisco Research Station in pasture with interspersed shrubs. Etymology: The specific epithet hillisi is a noun in the genitive case and is a patronym for David M. Hillis, who has had a great impact in the development of the field of molecular systematics (e.g., Hillis et al. 1996). In par- ticular, he published a classic paper on evolutionary ge- netics of Pholidobolus lizards, where he compared some phylogenetic tree reconstruction techniques and empha- sized the importance of phylogenetics in biogeography (Hillis 1985). Remarks: The Andes of southern Ecuador and northern Peru between 4°S and 7°S consist of relatively low-ele- vation mountains that create a mixture of environments. This region, known as the Huancabamba Depression, has long been recognized as a major biogeographic barrier for Andean organisms (e.g., Cadle 1991; Duellman 1979; Vuilleumier 1969). Although all species of Pholidobolus, except P. macbrydei, are restricted to the southern part of the northern Andes (i.e., Ecuador and southern Co- lombia), the new species described herein occurs on the northern limit of the Huancabamba Depression. The Huancabamba Depression seems to have in- fluenced the radiation of several Andean lizard clades, such as Stenocercus, Riama, Macropholidus, and Pholi- dobolus (Doan 2003; Torres-Carvajal 2007; Torres-Car- vajal and Mafla-Endara 2013). Except for Macropholi- dus, these clades have diversified along the northern Andes, suggesting that common geological or climatic events have influenced these radiations. The phyloge- netic tree presented in this paper further supports the idea of a south-to-north sequence of speciation events (Doan 2003; Torres-Carvajal 2007) which is congruent with the recent south-to-north uplift of the northern Andes (Simp- son 1979; Aleman and Ramos 2000). Acknowledgments. — We thank Santiago R. Ron for photographs and Andrea Varela for assembling some of the figures. Special thanks to Tiffany Doan and an anonymous reviewer for their valuable comments. OTC received funds from Secretaria de Educacion Superior, Ciencia, Tecnologia e Innovacion (SENESCYT). PMSN is grateful to Funda9ao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP) for financial support (Grant # 2012/00492-8). Specimens were collected under collec- tion permit 001-11 IC-FAU-DNB/MA issued by Ministe- rio de Ambiente del Ecuador. Literature Cited Aleman A, Ramos VA. 2000. Northern Andes. Pp. 453- 480 In: Tectonic Evolution of South America. Editors, Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (85) Cordani UG, Milani EJ, Thomaz-Filho A, Campos DA. Brazilian Academy of Science, Rio de Janeiro, Brazil. Biomatters. 2010. Geneious version 5.3. Available: http://www.geneious.com Beck E, Makeschin F, Haubrich F, Richter M, Bendix J, Valerezo C. 2008. The Ecosystem (Reserva Biologica San Francisco). Pp. 1-13 In: Gradients in a Tropical Mountain Ecosystem of Ecuador. Editors, Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R. Spring- er, Berlin and Heidelberg, Germany. Bouckaert R, Heled J, Kiilmert D, Vaughan T, Wu C-H, Xie D, Suchard MA, Rambaut A, Drummond AJ. 2014. BEAST 2: A Software Platform for Bayesian Evolutionary Analysis. PLoS Computational Biology 10(4): el003537. doi:10.1371/joumal.pcbi.l003537 Cadle JE. 1991. Systematics of lizards of the genus Stenocercus (Iguania: Tropiduridae) from northern Peru: New species and comments on relationships and distribution patterns. Proceedings of the Academy of Natural Sciences of Philadelphia 143: 1-96. Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: More models, new heuristics and paral- lel computing. Nature Methods 9: 772-772. de Queiroz K. 1998. The general lineage concept of spe- cies, species criteria, and the process of speciation. Pp. 57-75 In: Endless Forms: Species and Speciation. Editors, Howard DJ, Berlocher SH. Oxford Univer- sity Press, Oxford, United Kingdom. de Queiroz K. 2007. Species concepts and species de- limitation. Systematic Biology 56: 879-886. de Queiroz K, Gauthier J. 1994. Toward a phylogenetic system of biological nomenclature. Trends in Ecology and Evolution 9 : 27-31. Doan TM. 2003. A south-to-north biogeographic hypoth- esis for Andean speciation: evidence from the lizard genus Proctoporus (Reptilia, Gymnophthalmidae). Journal of Biogeography 30: 361-374. Dowling HG, Savage JM. 1960. A guide to the snake hemipenis: A survey of basic structure and systematic characteristics. Zoologica 45: 17-28. Duellman WE. 1979. The herpetofauna of the Andes: Patterns of distribution, origins, differentiation, and present communities. Pp. 371M-59 In: The South American Herpetofauna: Its Origin, Evolution and Dispersal. Editor, Duellman WE. University of Kan- sas Natural History Museum, Eawrence, Kansas, USA. Hillis DM. 1985. Evolutionary genetics of the Andean lizard genus Pholidobolus (Sauna: Gymnophthalmi- dae): phylogeny, biogeography, and a comparison of tree reconstruction techniques. Systematic Zoology 34: 109-126. Hillis DM, Moritz C, Mable BK. 1996. Molecular Sys- tematics. Sinauer, Sunderland, Massachusetts, USA. Manzani PR, Abe AS. 1988. Sobre dois novos metodos de prepare do hemipenis de serpentes. Memorias do November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador Instituto Butantan 50: 15-20. Montanucci RR. 1973. Systematics and evolution of the Andean lizard genus Pholidobolus (Sauria: Teiidae). University of Kansas Museum of Natural History Mis- cellaneous Publications 59: 1-52. Myers CW, Donnelly MA. 2001. Herpetofauna of the Yutaje-Corocoro massif, Venezuela: Second report from The Robert G. Goelet American Museum-terra- mar expedition to the northwestern tepuis. Bulletin of the American Museum of Natural History 261: 1-85. Myers CW, Donnelly MA. 2008. The summit herpeto- fauna of Auyantepui, Venezuela: report from the Rob- ert G. Goelet American Museum - TERRAMAR Ex- pedition. Bulletin of the American Museum of Natural History 308: 1-147. Nunes PMS. 2011. Morfologia hemipeniana dos lagar- tos microteideos e suas implica96es nas rela9oes filo- geneticas da familia Gymnophthalmidae (Teioidea: Squamata). Ph.D. Dissertation, Universidade de Sao Paulo, Departamento de Zoologia, Sao Paulo, Brazil. Nunes PMS, Fouquet A, Curcio FF, Kok PJR, Rodrigues MT. 2012. Cryptic species in Iphisa elegans Gray, 1851 (Squamata: Gymnophitalmidae) revealed by hemipenial morphology and molecular data. Zoologi- calJournal ofLinnean Society 166: 361-376. Pellegrino KCM, Rodrigues MT, Yonenaga-Yassuda Y, Sites JW. 2001. A molecular perspective on the evolution of microteiid lizards (Squamata, Gymn- ophthalmidae), and a new classification for the fam- ily. Biological Journal of the Linnean Society 74(3): 315-338. Pesantes OS. 1994. A method for preparing the hemipe- nis of preserved snakes. Journal of Herpetology 28: 93-95. Rambaut A, Drummond AJ. 2014. Tree Annotator ver- Appendix 1 sion 2.1.2. Available: http://www.beast2.org Reeder TW. 1996. A new species of Pholidobolus (Squa- mata: Gymnophthalmidae) from the Huancabamba depression of northern Peru. Herpetologica 52: 282- 289. Savage JM. 1997. On terminology for the description of the hemipenis of squamate reptiles. Herpetological Journal 7: 23-25. Simpson BB. 1979. Quaternary biogeography of the high montane regions of South America. Pp. 157-188 In: The South American Herpetofauna: Its Origin, Evolu- tion and Dispersal. Editor, Duellman WE. University of Kansas Natural History Museum, Lawrence, Kan- sas, USA. Torres-Carvajal O, Mafla-Endara P. 2013. Evolutionary history of Andean Pholidobolus and Macropholidus (Squamata: Gymnophthalmidae) lizards. Molecular Phylogenetics and Evolution 68: 212-217. Torres-Carvajal O. 2007. Phylogeny and biogeography of a large radiation of Andean lizards (Iguania, Steno- cercus). Zoologica Scripta 36: 311-326. Uzzell T. 1973. A revision of lizards of the genus Pri- onodactylus, with a new genus for P. leucostictus and notes on the genus Euspondylus (Sauria, Teiidae). Postilla 154: 1-37. Vuilleumier F. 1969. Pleistocene speciation in birds liv- ing in the high Andes. Nature 223: 1179-1180. Zaher H. 1999. Hemipenial morphology of the South American Xenodontine snakes, with a proposal for a monophyletic Xenodontinae and a reappraisal of colubroid hemipenis. Bulletin of American Museum of Natural History 240: 1-168. Received: 10 September 2014 Accepted: 31 October 2014 Published: 12 November 2014 Additional specimens examined P/70//C/060/US aff/n/s.— ECUADOR: Provincia Chimborazo: Colta, r41’56”S, 78°46’25”W, 3,215 m, QCAZ 9899-01 ; Sicalpa, 1°42’18”S, 78°46’32”W, 3,212 m, QCAZ 11887. Provincia Cotopaxi: Cutuchi river, San Miguel de Salcedo, 1°2’9”S, 78°35’53”W, 2,640 m, QCAZ 9641 . Provincia Tungurahua: 6 km N Mocha to 400 m Panamerican Highway, 1°22’1”S, 78°39’16”W, 3,205 m, QCAZ 9895-97; Ambato surroundings, 1°14’59,8”S, 78°37’33”W, QCAZ 9340-73, 9375-9443; Chamisa on road to Guadalupe, r21’44”S, 78°30’39”W, 2,348 m, QCAZ 7266; Cotalo on path to Mucubi Community, 1°25’46”S, 78°31’3”W, 2,626 m, QCAZ 9839; Patate, 1°18’42”S, 78°30’36”W, 2,199 m, QCAZ 9847-50; Poatug Hamlet, Aya Samana, 1°16’58”S, 78°29’29”W, 2,573 m, QCAZ 10005, 10008, 10011-13, 10016, 10018; Poatug Hamlet, Terremoto, 1°16’23”S, 78°29’29”W, 2,547 m QCAZ 9997-10000, 10002-10004; San Miguelito on path to Pil- laro, ri3’12”S, 78°31’31”W, 2,689 m, QCAZ 9844; San Miguelito on path to Teran, 1°12’58”S, 78°31’42”W, 2,741 m, QCAZ 9843. Pholidobolus macbrydei.— ECUADOR. Provincia Azuay: 10 km S Cutchil, 3°8’2”S, 78°48’47”W, 2,900 m, QCAZ 823-24;1.2 km E Qs- orancho, 2°46’8”S, 78°38’10”W, 2,390 m, QCAZ 826; 6.2 km S Cutchil, 3°6’32”S, 78°48’4”W, 2,800 m, QCAZ 827; 20 km NE Cuenca, 2°51’0”S, 78°51’14”W, QCAZ 1359; 7 km Sigsig, 2°59’56”S, 78°48’25”W, 2,890 m, QCAZ 1537; 6 km S Qha, 3°29’49”S, 79°9’47”W, QCAZ 3658; 20 km Cuenca-EI Cajas, 2°46’39”S, 79°10’12”W, 3,508 m, QCAZ 9932-34, 9936-38, 10020; Cochapamba, 2°47’50”S, 79°24’56”W, 3,548 m, QCAZ 10133-35; Cochapata, 3°25’47”S, 79°3’35”W, 3,074 m, QCAZ 12605-07; Cuenca, Cuenca-Azoguez Panamerican Highway 2°53’43”S, 78°57’30”W, 2,486 m, QCAZ 6985; El Cajas National Park, path to PatuI Community, 2°44’28”S, 79°14’5”W, 4,092 m, QCAZ 8010-11; El Cajas National Park, PatuI river, 2°41’37”S, 79°13’56”W, 3,610 m, QCAZ 8893; El Cajas Na- tional Park, Zhurcay river, 3°2’30”S, 79°12’56”W, 3,766 m, QCAZ 8900-01; El Cajas National Park, 2°42’21”S, 79°13’32”W, 3,600 m, QCAZ 8946; El Capo, 2°46’43”S, 79°14’43”W, 4,100 m, QCAZ 4997; Giron, San Gregorio Community, Quinsacocha paramo, 3°6’22”S, 79°13’4”W, 3,242 m, QCAZ 8510-11; Giron, San Gregorio Community, Quinsacocha paramo, 3°2’30”S, 79°12’56”W, 3,766 m, QCAZ 8894-99, 8902-05, 8907; Giron, San Gregorio Community, Quinsacocha paramo, 3°2’30”S, 79°12’57”W, 3,766 m, QCAZ 8906; Guablid, 2°46’30”S, 78°4r51”W, 2,453 m, QCAZ 991 3-1 7, 9919-20, 9940-41; Gualaceo-Limon road, 2°56’53”S, 78°42’43”, 3,110 m, QCAZ 819- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (86) November 2014 | Volume 8 | Number 1 | e84 Torres-Carvajal et al. 22; Gualaceo-Limon road, 8.1 km O Azuay-Morona Santiago border, 2°57’50”S, 78°427”W, 3,140 m, QCAZ 825; Gualaceo, 2°52’56”S, 78°46’31”W, 2,298 m, QCAZ 9606; Gualaceo-Plan de Milagro road, 2°54’35”S, 78°44’4”W, 2,624 m, QCAZ 10875; Las Tres Cruces, 2°46’30”S, 79°14”53”W, QCAZ 4136; Maylas, Gualaceo-Macas road, 2°58’25”S, 78°41’41”W, 3,100 m, QCAZ 7269; Mazan Protected Forest, 2°52’29”S, 79°7’26”W, 2,700 m, QCAZ 1296-97; Mazan Protected Forest, 2°52’31”S, 79°7’45”W, 3,189 m, QCAZ 8008, 8013; Qna-La Paz road, 3°22’42”S, 79°11’20”W, 2,969 m, QCAZ 6031; Patacocha hill, 3°7’16”S, 79°3’54”W, 3,340 m, QCAZ 6144; Pucara, Tres Chorreras, 3°12’49”S, 79°28’3”W, QCAZ 11038; Quinoas river, 3°5’14”S, 79°16’40”W, 3,200 m, QCAZ 1564-66; San Antonio, 2°51’40”S, 79°22’43”W, 2,943 m, QCAZ 9668; San Vicente-Cruz path, 2°47’43”S, 78°42’53”W, 3,044 m, QCAZ 11416-17, 11420; Sigsig, 3°7’46”S, 78°48’14”W, 2,969 m, QCAZ 5605-08; Sigsig road, 3°3’17”S, 78°47’19”W, 2,574 m, QCAZ 9605; Tarqui, 3°0’57”S, 79°2’40”W, 2,627 m, QCAZ 8512. Provincia Cahar: Canar, 2°33’39”S, 78°55’51”W, QCAZ 9947; Culebrillas, 2°25’35”S, 78°52’12”W, 4,000 m, QCAZ 1349; Guallicanga ravine, 2°25’56”S, 78°54’8”W, 3,960 m, QCAZ 10048-49; Guallicanga river, 2°28’24”S, 78°58’22”W, 3,048 m, QCAZ 10051-52; Ingapirca, 2°32’43”S, 78°52’28”W, 3,400 m, QCAZ 1551; Juncal, 2°28’24”S, 78°58’22”W, 3,048 m, QCAZ 10050; Mazar Protected Forest, 2°32’48”S, 78°41’54”W, QCAZ 7376-84, 7883; Mazar Reserve, La Libertad, 2°32’45”S, 78°41’46”W, 2,842 m, QCAZ 10970-72. Provincia Chimborazo: Alao, 10 km Fluamboya, 1°52’22”S, 78°29’51”W, 3,200 m, QCAZ 1567-68; Atillo Grande, Magdalena lake, 2°11’15”S, 78°30’25”W, 3,556 m, QCAZ 9214; Atillo Grande, Frutatian lake, 2°12’57”S, 78°30’5”W, 3,700 m, QCAZ 9216-18; Culebrillas, Sangay National Park, 1°57’39”S, 78°25’55”W, 3,345 m, QCAZ 9612; Pungala, Eten Community, Timbo, 1°55’45”S, 78°32’14”W, 3,408 m, QCAZ 9616-21 ; Pungala, Melan Community, 1°52’30”S, 78°32’52”W, 3,564 m, QCAZ 9626-29, 9631; Qzogoche, 2°22’7”S, 78°41’20”W, 4,040 m, QCAZ 6006-07; Shulata, 2°20’22”S, 78°50’36”W, 3,228 m, QCAZ 5597-9;. Provincia El Qro: Guanazan, 3°26’24”S, 79°29’13”W, 2,638 m, QCAZ 7891, 7894. Provincia Loja: 17.1 km S Saraguro, 3°43’45”S, 79°15’53”W, 3,150 m, QCAZ 828; 26 km N Loja, Fluashapamba Native Forest, 3°39’30”S, 79°16’20”W, 2,894 m, QCAZ 8651; Cordillera of Lagunil- las, Jimbura, 4°49’1”S, 79°21’43”W, 3,600 m, QCAZ 3785; Cordillera of Lagunillas, Jimbura, 4°37’42”S, 79°27’49”W, 3,450 m, QCAZ 6145-47; Fierro Urco, 3°42’38”S, 79°18’18”W, 3,439 m, QCAZ 6949-50; Gurudel, 3°39’22”S, 79°9’47”W, 3,100 m, QCAZ 5078-79; Jimbura, Jimbura lake, 4°42’32”S, 79°26’48”W, 3,036 m, QCAZ 6945-48; Jimbura, path to Jimbura lake, 4°42’34”S, 79°26’8”W, 3348 m, QCAZ 10054-62; Military antenna, Saraguro, 3°40’46”S, 79°14’16”W, 3,190 m, QCAZ 3673-75, 9632; San Lucas, 3°43’55”S, 79°15’38”W, 2,470 m, QCAZ 2861; Saraguro, 3°37’13”S, 79°14’9”W, 3,100 m, QCAZ 3606, 3754; Urdaneta, 3°36’6”S, 79°12’31”W, QCAZ 201 9. Provincia Tungurahua: Poatug Flamlet, El Corral, 1°16’21”S, 78°28’5”W, 3,468 m, QCAZ 8047, 9995-96. Provincia Zamo- ra Chinchipe: Loja-Podocarpus National Park road, 3°59’44”S, 79°8’28”W, 2,776 m, QCAZ 10870-71; Valladolid, Podocarpus National Park, 4°29’3”S, 79°8’56”W, 1 ,800 m, QCAZ 3743. Pholidobolus montium. — ECUADQR: Provincia Cotopaxi: 2 km S Chugchilan on road to Quilotoa, 0°48’24”S, 78°56’11”W, 2,917 m, QCAZ 8056-58; Latacunga, 0°52’27”S, 78°38’26”W, 2,857 m, QCAZ 873-74, 1411-12, 9642; Mulalo, 0°46’35”S, 78°34’40”W, 3,030 m, QCAZ 9639; San Juan de Paste Calle, 0°45’4”S, 78°38’51”W, 1,956 m, QCAZ 8053-54; South llliniza, 0°39’43”S, 78°42’40”W, 3,400 m, QCAZ 858-59, 1454. Provincia Imbabura: Atuntaqui, 0°19’59”N, 78°12’50”W, QCAZ 855; Cotacahi, Peribuela, Cuicocha Lake, Cotacachi-Cayapas Reserve, 0°17’34”N, 78°21’5”W, 3,082 m, QCAZ 9683, 9685-86; 0°23’4”N, 78°15’25”W, 2,900 m, QCAZ 6137, 6139; Cotacachi-Cayapas Reserve, Jose Marla Yerovi Islets, 0°18’20”N, 78°2T41”W, 3,093 m, QCAZ 10959-60; El Juncal, 0°26’6”N, 77°57’58”W, QCAZ 6451 . Provincia Pichincha: 16 km W Chillogallo, Quito-Chiriboga road, 0°17’46”S, 78°39’30”W, 3,100 m, QCAZ 797; 5 km E Pifo-Papallacta road, 0°15’3”S, 78°17’58”W, 2,800 m, QCAZ 1107-08; Alambi, 0°1’59”S, 78°34’26”W, 2,727-3,800 m, QCAZ 9691; AlangasI, 0°18’24”S, 78°24’40”W, QCAZ 1453, 1469; Amaguaha, Hacienda San Ignacio, 0°22’22”S, 78°30’14”W, QCAZ 1463-64, 5275; Calacall, Simon Bolivar Street, uphill through secondary road, 0°T1”N, 78°30’49”W, 3,001 m, QCAZ 11674, 11676, 11678-79; Calacall Stadium, 0°0’0,3”S, 78°30’38”W, 2,833 m, QCAZ 11682; Carretas, 0°6’25”S, 78°26’46”W, QCAZ 875; Chillogallo, 0°16’48”S, 78°33’25”W, QCAZ 840-43; Cumbaya, La Primavera, 0°12’6”S, 78°25’40”W, QCAZ 7248; Guayllabamba, 0°3’23”S, 78°20’26”W, QCAZ 7905; Inga, 5.5 km SE La Merced, 0°17’51”S, 78°20’52”W, 2,798 m, QCAZ 5278; Lloa, 0°14’52”S, 78°34’33”W, QCAZ 4109; Lloa Stadium, 0°14’39”S, 78°35’12”W, 3,059 m, QCAZ 11661; Loreto, road to Molinuco, Central Stadium, 0°23’4”S, 78°24’30”W, 2,844 m, QCAZ 11663; Machachi, 0°29’50”S, 78°32’25”W, QCAZ 844-48, 1374-77, 1462; Machachi, The Tesalia Springs Company S.A. surroundings, 0°30’27”S, 78°33’57”W, 2,900 m, QCAZ 1465-67, 830-31, 833, 860-61, 1459-61; None, 0°4’42”S, 78°34’24”W, 2,843 m, QCAZ 11653-55; None School, 0°4’4”S, 78°34’35”W, 2,754 m, QCAZ 11656-58; Pasochoa, 0°26’24”S, 78°30’15”W, 2,850 m, QCAZ 1451-52; Pomasqui, 0°3’3”S, 78°27’21”W, QCAZ 862-68; Pululahua Volcano, 0°2’34”N, 78°30’15”W, QCAZ 1450, 1520; Quito, Bellavista, 0°11’21”S, 78°28’35”W, QCAZ 1099; Quito, Chillogallo, 0°16’26”S, 78°33’23”W, QCAZ 8967; Qui- to, Itchimbla, 0°13’21”S, 78°29’56”W, QCAZ 834, 1455-58, 1643, 2843; Quito, Garden of the Pontificia Universidad Catolica del Ecua- dor (PUCE), 0°12’33”S, 78°29’28”W, 2,800 m, QCAZ 856-57, 7032, 1295, 2853; Quito, Parque Metropolitano, 0°10’35”S, 78°27’40”W, QCAZ 4051; Quito, Universidad Central del Ecuador, 0°1T59”S, 78°30’19”W, 2,800 m, QCAZ 3727; Rio Guajalito Protected Forest, 0°13’44”S, 78°48’22”W, QCAZ 1338-39; San Antonio de Pichincha, 0°0’33”S, 78°26’45”W, QCAZ 580-81 , 790-92, 849, 1119-20, 1368, 1393, 2220, 2223, 2653; Tababela, International Airport, 0°6’21”S, 78°21’4”W, QCAZ 8046, 9044, 10064, 10974-76; Quito, Tumbaco, 0°12’34”S, 78°24’2”W, QCAZ 1113-14; Uyumbicho, 0°22’59”S, 78°31’6”W, QCAZ 870. Pholidobolus prefrontalis . — ECUADQR: Provincia Chimborazo: AlausI, 2°1T54”S, 78°50’42”W, 2359 m, QCAZ 9907-9911; Tixan, 2°9’22”S, 78°48’3”W, 2,908 m, QCAZ 9951-54. Omar Torres-Carvajal graduated in Biological Sciences from Pontificia Universidad Catolica del Ecua- dor (PUCE) in 1998, and in 2001 received a Master’s degree in Ecology and Evolutionary Biology from the University of Kansas under the supervision of Dr. Einda Trueb. In 2005 he received a Ph.D. degree from the same institution with the thesis entitled “Phylogenetic systematics of South American lizards of the genus Stenocercus (Squamata: Iguania).” Between 2006-2008 he was a postdoctoral fellow at the Smithsonian Institution, National Museum of Natural History, Washington DC, USA, working under the supervision of Dr. Kevin de Queiroz. He is currently Curator of Reptiles at the Zoology Museum QCAZ of PUCE and an Associate Professor at the Department of Biology in the same institution. He has published more than 30 scientific papers on taxonomy, systematics, and biogeography of South American reptiles, with emphasis on lizards. He is mainly interested in the theory and practice of phylogenetic systematics, particularly as they relate to the evolutionary biology of lizards. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (87) November 2014 I Volume 8 I Number 1 I e84 A new Pholidobolus from Ecuador Pablo J. Venegas graduated in Veterinary Medieine from Universidad Naeional Pedro Ruiz Gallo, Lam- bayeque, Peru, in 2005. He is eurrently eurator of the herpetologieal eolleetion of Centro de Omitolo- gia y Biodiversidad (CORBIDI) and researcher of the Museo de Zoologia QCAZ, Pontificia Universidad Catolica del Ecuador in Quito. His current research interest is focused on the diversity and conservation of the Neotropical herpetofauna with emphasis in Peru and Ecuador. So far he has published more than 30 scientific papers on taxonomy and systematics of Peruvian amphibians and reptiles. Simon E. Lobos graduated in Biological Sciences from Pontificia Universidad Catolica del Ecuador (PUCE) in 20 1 3 . As a student, he j oined the Museo de Zoologia QCAZ, Pontificia Universidad Catolica del y Ecuador in Quito, where he developed a great interest in reptiles. He has been studying systematics of gym- nophthalmid lizards for the last four years. For his undergraduate thesis, Simon worked on the “Molecular systematics of lizard Alopoglossus (Autarchoglossa: Gymnophthalmidae) in Ecuador.” This manuscript is 1 the second lizard species description coauthored by Simon. Other papers based on his undergraduate thesis ® work are in preparation. Paola Mafla-Endara graduated in Biological Sciences from Pontificia Universidad Catolica del Ecua- dor (PUCE) in 2011. Her undergraduate thesis entitled “Phylogeography of Andean lizards Pholidobolus (Squamata: Gymnophthalmidae) in Ecuador” provided her a gratifying knowledge about phylogenetics systematics, evolution, statistics, and biogeography. Since this time, she has developed a deep interest in molecular biology. Currently she works mostly in systematics and ecology of fungi. She is convinced that the same knowledge can be useful to solve similar questions in different subjects. This manuscript repre- sents the second lizard species description coauthored by Paola. Others are in preparation. Pedro M. Sales Nunes graduated in Biological Sciences from Universidade de Sao Paulo (USP) in 2003, and in 2006 received a Master’s degree in Zoology from the same institution under the supervision of Dr. Hussam Zaher. In 2011 he received a Ph.D. degree from the same institution with the thesis entitled “Hemipenial Morphology of the microteiid lizards (Squamata: Gymnophthalmidae)” under the supervi- sion of Dr. Miguel Trefaut Rodrigues. Between 2012-2014 he was a postdoctoral fellow at the USP, Sao Paulo, Brazil, also working under the supervision of Dr. Miguel Trefaut Rodrigues. He is currently Curator of the Herpetologieal Collection at the Universidade Federal de Pernambuco (UFPE), Recife, Brazil, and an Adjunct Professor at the Department of Zoology in the same institution. His production is focused on taxonomy and systematics of South American reptiles, with emphasis in Squamata. In accordance with the International Code ofZoologieal Nomenclature new rules and regulations (ICZN 2012), we have deposited this paper in publicly accessible institutional libraries. The new species described herein has been registered in ZooBank (Polaszek 2005a, b), the official online registration system for the ICZN. The ZooBank publication LSID (Life Science Identifier) for the new species described here can be viewed through any standard web browser by appending the LSID to the prefix “http://zoobank.org/”. The LSID for this publication is: um:lsid:zoobank.org:pub:41593E9F-6F66-4E60-B073-2E8BF643358F. Separate print-only edition of paper(s) (reprint) are available upon request as a print-on-demand service. Please inquire by sending a request to: Amphibian & Reptile Conservation (amphibian-reptile-conservation.org; arc.publisher@gmail.com). Amphibian & Reptile Conservation is a Content Partner with the Encyclopedia of Life (EOL); http:///www.eol.org/ and submits information about new species to the EOL freely. Digital archiving of this paper are found at the following institutions: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Ernst Mayr Library, Museum of Comparative Zool- ogy, Harvard University, Cambridge, Massachusetts (USA); Florida Museum of Natural History, Gainesville, Florida (USA). Complete journal archiving is found at: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Florida Museum of Natural History, Gainesville, Florida (USA). Citations ICZN. 2012. Amendment of Articles 8,9,10,21 and 78 of the International Code of Zoological Nomenclature to expand and refine methods of publication. Zootaxa 3450: 1-7. Polaszek A et al. 2005a. Commentary: A universal register for animal names. Nature 437: 477. Polaszek A et al. 2005b. ZooBank: The open-access register for zoological taxonomy: Technical Discussion Paper. Bulletin of Zoological Nomenclature 62(4): 210-220. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (88) November 2014 I Volume 8 I Number 1 I e84 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 89-106 (e88). Early development of the glass frogs Hyalinobatrachium fleischmanni and Espadarana callistomma (Anura: Centrolenidae) from cleavage to tadpole hatching Man'a-Jose Salazar-Nicholls and Eugenia M. del Pino* Escuela de Ciencias Biologicas, Pontificia Universidad Catolica del Ecuador, Av. 12 de Octubre 1076 y Roca, Quito 170517, ECUADOR Abstract— Wfe report the characteristics of embryonic development from cleavage to tadpole hatching in two species of glass frogs, Hyaiinobatrachium fieischmanni and Espadarana caliistomma (Anura: Centrolenidae). This analysis of embryonic development in centrolenid frogs enhances comparative studies of frog early development and contributes baseline information for the conservation and management of Ecuadorian frogs. These frogs reproduced in captivity and their embryos were fixed for developmental analysis. The morphology of embryos was evaluated in whole mounts, bisections, thick sections, and fluorescent staining of cell nuclei. Egg clutches contained an average of 23 and 35 eggs for H. fieischmanni and E. cailistomma, respectively. The eggs of both frogs measured approximately 2.1 mm in diameter. The eggs of H. fieischmanni were uniformly pale green. In contrast, the animal hemisphere of E. callistomma eggs was dark brown and the vegetal hemisphere was light brown. The developmental time of H. fieischmanni and E. caliistomma under laboratory conditions was 6 and 12 days, respectively from the 32-cell stage until tadpole hatching. Differences in environmental conditions may be associated with the time differences of early development observed in these frogs. The development of glass frogs from egg deposition to tadpole hatching was staged into 25 standard stages according to the generalized table of frog development. Archenteron elongation began in the early gastrula and notochord elongation began in mid to late gastrula, as in X. laevis. Development of the gastrocoel roof plate (grp) was precocious in the two centrolenid frogs. The grp was detected in the late gastrula of both species; whereas the grp was detected in neurula stages of X. laevis. The presence of the grp in embryos of these frogs suggests that the mechanisms of left-right asymmetry, found in X. iaevis and other amphibians, may be shared by these centrolenid frogs. The development of H. fieischmanni and E. caiiistomma resembles the pattern found in frogs with rapid development such as the aquatic eggs of X. iaevis and the development in floating foam-nests in the genus Engystomops (Leptodactylidae). Differences in egg pigmentation were particularly significant in connection with the divergent reproductive strategies of these glass frogs. Key words. Developmental time, egg pigmentation, embryonic development, gastmlation, gastrocoel roof plate, neu- mla Citation: Salazar-Nicholls M-J, del Pino EM. 2015. Early development of the glass frogs Hyalinobatrachium fieischmanni and Espadarana callis- tomma (Anura: Centrolenidae) from cleavage to tadpole hatching. Amphibian & Reptile Conservation 8(1) [Special Section]; 89-106 (e88). Copyright: © 2015 Salazar-Nicholls and del Pino. This is an open-access article distributed under the terms of the Creative Commons Attribution- NonCommercial-NoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any me- dium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptile Conservation] official journal website . Received: 13 May 2014; Accepted: 19 December 2014; Published: 27 Feburary 2015 Correspondence. Email: *edelpino@puce.edu.ec (Corresponding author, Eugenia M. del Pino); tel: (593 2) 299 1 700 extension 1280; fax: (593 2) 299 1725. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 89 e88 Salazar-Nicholls and del Pino Introduction Centrolenid frogs are commonly known as glass frogs because the internal organs of the adult are visible through the transparent ventral body wall. This transpar- ent region varies in size among species (Fig. 1 A-B, D-E) (Cisneros-Heredia and McDiarmid 2007). Glass frogs are endemic to the tropical regions of South America from Venezuela to northern Argentina and south-eastern Brazil (AmphibiaWeb 2014) and are particularly diverse in the cloud forests of Colombia and Ecuador (Delia et al. 2010; Guayasamin and Tmeb 2007; Ospina-Sarria et al. 2010). These arboreal frogs deposit their eggs in ge- latinous masses on the upper or lower surface of plant leaves bordering stream banks. After hatching, the tad- poles drop into the underlying streams. Tadpoles are fos- sorial and live in the substrate along the shoreline (Delia et al. 2010; Duellman and Tmeb 1986). We studied the early development of the glass frogs Hyalinobatrachium fleischmanni and Espadarana cal- listomma (Anura: Centrolenidae) to compare their de- velopment with frogs that exemplify different reproduc- tive modes and to contribute to the knowledge of frogs from Ecuador. Development of these centrolenid frogs was compared with the embryogenesis of Tungara frogs, Engystomops (Leptodactylidae). Tungara frogs constmct foam nests that float in the water (Romero-Carvajal et al. 2009). In addition, this comparison was extended to the terrestrial embryos of poison arrow frogs (Dendro- batidae), embryos of the Marsupial frog, Gastrotheca riobambae (Hemiphractidae), and the aquatic embryos of Xenopus laevis (Pipidae) and Ceratophrys stolzman- ni (Ceratophryidae) (Elinson and del Pino 2012; Nieu- wkoop and Eaber 1994; del Pino et al. 2004) (Table 1). The analysis of H. fleischmanni and E. callistomma early development was feasible because of the recent success- ful reproduction of centrolenid frogs in captivity at the Balsa de los Sapos, Centre of Amphibian Investigation and Conservation (CICA), Pontificia Universidad Catdli- ca del Ecuador (PUCE). Hyalinobatrachium fleischmanni (Fig. lA-C) occurs from southern Mexico to northern South America, in- cluding Ecuador. The egg clutches consist of 20-40 pale- green eggs, attached to the underside of plant leaves (Eig. 1C). Parental care of the egg clutch is provided by the male to maintain the needed humidity. The male prevents predation by katydids, wasps, ants, and other insects by kicking with his limbs at the predatory insect (Delia et al. 2010; Greer and Wells 1980; Savage 2002). Espadarana callistomma (Guayasamin and Tmeb 2007) (Pig. ID-F) occurs in the lowlands of northeastern Ecuador and southern Colombia (Guayasamin and Tmeb 2007; Ospina-Sarria et al. 2010). Darkly pigmented eggs are deposited on the upper surface of plant leaves (Guayasamin and Tmeb 2007) (Pig. IF). Egg predation by insects has not been reported for this species. The left-right asymmetric location of organs, such as the liver and the heart is established in the X. laevis gastrocoel roof plate (grp) of the neumla by fluid-flow towards the left side, guided by the clockwise rotation of cilia (Blum et al. 2009; Schweickert et al. 2010). The rotation of cilia in the frog grp, or in equivalent stmctures of other vertebrates, determines the asymmetric expres- sion of the gene Nodal in the grp left side (Blum et al. 2014b). The grp of X. laevis derives from the superflcial prospective mesoderm of the early gastmla that becomes internalized during gastmlation, and ends up in the dorsal roof of the primitive gut. The grp can be detected by the presence of exposed mesoderm corresponding to the no- tochord and some paraxial mesoderm in the dorsal roof of the primitive gut, and it is bordered by the lateral en- dodermal crests (lee). As development advances, the lee close to the midline and the primitive gut cavity becomes totally lined with endoderm (Blum et al. 2009). The left- right asymmetry is determined by fluid flow guided by cilia rotation in the grp of frogs and other vertebrates. However, a comparable structure to the grp has not been reported for the chick and pig, and the left-right symme- try breakage in these vertebrates may depend on a modi- fled mechanism (Blum et al. 2014a, b; Saenz-Ponce et al. 2012b). We analyzed the presence of the grp in the gastmla and neumla of glass frogs to provide additional comparison. We characterized the embryos of these glass frogs from cleavage to hatching of the tadpoles. We found that in glass frogs, gastmlation overlapped with body elonga- tion, as in frogs with rapid embryonic development. The grp was detected in the late gastmla of both species of glass frogs. Its presence suggests that the mechanisms of left-right asymmetry, found in X. laevis and other am- phibians, may be shared by these centrolenid frogs. The reproductive mode of these glass frogs is associated with rapid development. The strategy of egg deposition in the underside or upper surface of leaves is associated with differences in developmental time and pigmentation of embryos and tadpoles. Materials and Methods Locality of collection and staging of embryos. Hya- linobatrachium fleischmanni and Espadarana callis- tomma were collected from Esmeraldas Province, San Lorenzo, Durango, along the banks of the Rio Durango and its tributaries in northwest Ecuador. The altitude of this site is 243 m above sea level, and the geographic co- ordinates are W 78.62405, N 1.04186. Erogs of both spe- cies were collected on 04 October 2009 by Elicio Tapia and Santiago Garcia. The adults successfully reproduced at the Balsa de los Sapos, Centre of Amphibian Investi- gation and Conservation (CICA), School of Biological Sciences, Pontiflcia Universidad Catolica del Ecuador February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 90 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma H. fleischmanni E. callistomma Fig. 1. The adults and egg clutches of glass frogs. (A-C) Hyalinobatrachium fleischmanni. (A) Lateral view of an adult male. (B) Ventral view of an adult male. The arrow indicates the border of the transparent body wall. The intestine and a blood vessel are vis- ible. (C) Partial view of an egg clutch at the gastrula stage. The embryos are uniformly pale and the blastocoel roof is translucent (arrowhead). (D-F) Espadarana callistomma. (D) Lateral view of an adult female. (E) Ventral view of an adult female. The arrow signals the pigmented oocytes visible through the transparent body wall. The size of the transparent region is smaller than in H. fleischmanni shown in B. (F) Partial view of an egg clutch. The embryos were at stages 5-6 (Table 2). Photographs of adult frogs by Santiago Ron (A—B, D—E). February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 91 e88 Salazar-Nicholls and del Pino (PUCE). The permit 016-IC-FAU-DNBAP-MA from the Ministry of the Environment, Ecuador, allowed the col- lection and maintenance of these frogs at Balsa de Sapos. Egg clutches were donated to the Laboratory of Develop- mental Biology for analysis of embryonic development. This study was based on the analysis of embryos derived from seven egg clutches of H. fleischmanni and four egg clutches of E. callistomma. The number of eggs of each egg clutch was record- ed and the embryos were cultured in humid chambers at room temperature, as described for embryos of the dendrobatid frog, E. machalilla (del Pino et al. 2004). At various intervals, some embryos were moved to a Petri dish filled with 15% Steinberg’s solution (del Pino et al. 2004) and the egg-jelly was manually removed to study embryogenesis. Embryos were staged according to the general table of frog development (Gosner 1960). Egg diameter was measured in fixed embryos with the measuring tool of the program, Axiovision (Carl Zeiss, Oberkochen, Germany). Fixation, staining and analysis of embryonic devel- opment. Embryos were fixed in Smith’s fixative (del Pino et al. 2004). The procedures for the bisection of embryos, vibratome sectioning, cell nuclei staining with the fluorescent dye Hoechst 33258 (Sigma- Aldrich, St. Louis, MO, USA), and the staining of cell boundaries with silver nitrate were previously described (Moya et al. 2007; del Pino et al. 2004). Sections were mounted in glycerol, and were examined with a Stemi SV6 stereo microscope (Carl Zeiss, Oberkochen, Germany) or with fluorescent optics using a Z1 Axio Observer microscope (Carl Zeiss, Oberkochen, Germany). Embryos were pho- tographed with Axiocam cameras, attached to micro- scopes, and the image capture program, Axiovision (Carl Zeiss, Oberkochen, Germany). The images were edited with Adobe Photoshop CS6. Results and Discussion Internal organs of adult glass frogs can be observed through their transparent belly; however the size of the transparent window varies in the different genera of cen- trolenid frogs as detected for H. fleischmanni and E. cal- listomma (Fig. lA-B, D-E) (Cisneros-Heredia and Mc- diarmid 2007). In contrast with adults, the eggs of these frogs were opaque (Fig. 1C, F). We also detected sig- nificant pigmentation differences as the H. fleischmanni eggs and embryos were pale-green and those of E. cal- listomma were dark brown (Fig. 1C, F). Egg pigmenta- tion is a distinctive character of the different genera of Centrolenidae; moreover some species deposit their eggs in the upperside and others in the underside of leaves. However, some species show no particular preference for the upper or underside of leaves for the deposition of their eggs (Cisneros-Heredia and Mcdiarmid 2007). Clutch size, egg pigmentation and developmental time. The number of eggs ranged from 14-30 eggs, with a mean of 23 eggs per clutch in H. fleischmanni, and 32- 39 eggs, with a mean of 35 eggs per clutch in E. callis- tomma. The eggs of both species measured about 2.1 mm in diameter (Table 1). The embryos of H. fleischmanni were uniformly pale-green (Figs. 1C; 2A-L; 3A-D). In contrast, the animal hemisphere of E. callistomma em- bryos was dark brown, and the vegetal hemisphere was pale-brown (Figs. IF; 4A-L; 5A-F). Dark pigmentation of the animal hemisphere of the egg may provide protection against solar UV radiation and may capture solar heat required to accelerate early development of frog embryos exposed to solar radiation in moist or aquatic environments. In contrast, there is lack of dark pigment in frog eggs and embryos that develop in secluded places (Duellman and Tmeb 1986; Elinson and del Pino 2012). We propose that H. fleischmanni embryos do not require dark pigmentation because the underside of plant leaves may provide protection against solar radiation. In contrast, the presence of dark pigment in eggs and embryos of E. callistomma may be needed, as the egg clutches are directly exposed to UV solar ra- diation on the upper surface of plant leaves. The differences in pigmentation were detectable in eggs and embryos until tadpole hatching (Figs. 1C, F; 2-5). At hatching, the tadpoles of H. fleischmanni were pale green with little dark pigmentation on the dorsum; whereas E. callistomma tadpoles had a brown color (Figs. 3C-D; 5F). The fossorial free-living tadpoles of H. fleischmanni remained nearly unpigmented had elon- gated bodies, and narrow tail fins to enable digging in the sandy stream bottoms. The eyes were reduced in size and were covered by skin characters likely associated with the fossorial habits of H. fleischmanni tadpoles (Delia et al. 2010; Duellman and Tmeb 1986; Savage 2002).The characteristics of the E. callistomma free-living tadpoles are unknown. The differences in tadpole pigmentation at hatching suggest that the larval stages of these two cen- trolenids may occur in dissimilar aquatic environments. The differences in egg pigmentation observed in H. fleischmanni and E. callistomma may depend on differ- ent expression levels of the gene Shroom2 during oogen- esis. Shroom2, an actin-binding protein, controls pig- ment granule localization in the animal cortex ofX. laevis oocytes (Lee et al. 2009). The oocytes of Engystomops pustulosus (Leptodactylidae) contain small amounts of Shroom2 protein and are white in color. However, Engys- tomops embryos have dark pigment granules around nu- clei of blastomeres (Lee et al. 2009; Romero-Carvajal et al. 2009). Embryos of H. fleischmanni are pale and do not have dark pigment around the nuclei of blastomeres; whereas, in E. callistomma embryos dark pigment was observed on the cell surface of animal pole blastomeres, as well as around blastomere nuclei. Embryos of H. fleischmanni and E. callistomma were maintained under identical laboratory conditions with a February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 92 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma Hyalinobatrachium fleischmanni Fig. 2. External morphology of Hyalinobatrachium fleischmanni embryos from cleavage to the tail bud stage. (A) Stage 7: Thirty- two cell stage. Animal micromeres were much smaller than the vegetal macromeres. (B) Stage 8: Mid cleavage. (C) Stage 9: Blastula. (D) Stage 10.5: Early gastrula. The dorsal blastopore lip can be seen in the dorsal subequatorial region. (E) Stage 11: Mid gastrula. The yolk plug was large. (F) Stage 12: Late gastrula. (G) Stage 12.5: Late gastrula with a small yolk plug. Neural groove and neural plate were visible in embryos of this stage. (H) Stage 12.75: Late gastrula. The neural groove was visible. The yolk plug was small. (I) Stage 14: Early neural fold. (J) Stage 15: Mid-neural fold. The neural folds were more elevated. (K) Stage 16. Closure of the neural tube. The neural folds were near each other. (L) Stage 17. Tail bud stage. The branchial arches were visible. In this and the following figures, numbers in the top right-hand comer give the developmental stage, br, branchial arch; c, cleavage furrow; dl, dorsal blastopore lip; hy, hyoid arch; ma, mandibular arch; nf, neural fold; ng, neural groove; np, neural plate; vl, ventral blastopore lip; yp, yolk plug. temperature fluctuation of 18-23 °C. However, develop- mental time diverged greatly between these frogs, as em- bryos of H. fleischmanni required six days and those of E. callistomma required 12 days from the 32-cell stage until tadpole hatching. However in nature, great variation in developmental time was observed in H. fleischmanni, as egg clutches required 8-21 days from oviposition to tadpole hatching (Greer and Wells 1980). In our labora- tory, development of H. fleischmanni and E. callistomma was slower than in the floating foam-nests of Engysto- mops (Leptodactylidae), and faster than in the terrestrial nests of Dendrobatidae. In two species of Engystomops, development from egg deposition until hatching required only three days whereas 19-21 days were required for the same developmental processes by six species of den- drobatid frogs (del Pino et al. 2004, 2007; Romero-Car- vajal et al. 2009) (Table 1). Reproductive strategies. We propose that rapid devel- opment may be favored in H. fleischmanni in compari- son with E. callistomma because eggs deposited on the underside of plant leaves are at a greater risk of desicca- tion in comparison with eggs deposited on the upperside of leaves (Delia et al. 2010; Savage 2002). Moreover, rapid development may be required in all centrolenids, including frogs of the genus Espadarana, to overcome predation from a number of insect families and other arthropods (Cabanzo-Olarte et al. 2013; Duellman and Tmeb 1986; Villa 1977; Vockenhuber et al. 2008). The deposition of eggs on the underside of plant leaves and February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 93 e88 Salazar-Nicholls and del Pino Table 1. Comparison of reproductive and developmental characteristics of glass frogs. Family and Species Reproduction Clutch size and (egg diameter, mm) Gastrulation time (hrs)® Presence of the grp in the neurula Onset of noto- chord elongation Refs“ Rapid Development 1 Centrolenidae Hyalinobatrachium fleischmanni Leaves underside 23 (2.1) 24 Yes mid gastrula'’ 2 Espadarana callistomma Leptodactylidae Leaves upperside 35 (2.1) 23 Yes mid gastrula'’ 2 Engystomops randi Floating foam-nest 110(1.1) 12.5 Yes mid gastrula'’ 3,4 Engystomops coloradorum Floating foam-nest 130(1.3) 12.5 Unknown mid gastrula'’ 3 Ceratophryidae Ceratophrys stolzmanni Pipidae Aquatic 664 (2.2) 5 Yes mid gastrula'’ 4,5 Xenopus laevis Aquatic 1000(1.2) 6 Yes mid gastrula'’ 6,7 Slow Development 1 Dendrobatidae Epipedobates machalilla Terrestrial nest 15 (1.6) 65 Yes After gastrulation^ 4,8 Epipedobates tricolor Hemiphractidae Terrestrial nest 13 (2.0) 36 Yes After gastrulation^ 8,9 Gastrotheca riobambae Egg brooding 128 (3.0) 168 Yes After gastrulation^ 1,4 '‘Time from stages 10-13. Embryo culture temperatures for: X laevis 23 °C, and 18-23 °C for other frogs.; '’Stage 11; ‘‘Stage 13; ‘'References: 1, (del Pino et al. 2007); 2, This work; 3, (Romero-Carvajal et al. 2009); 4, (Saenz-Ponce et al. 2012b); 5, (Ortiz, 2013); 6, (Nieuwkoop and Faber 1994); 7, (Blum, et al. 2009); 8, (del Pino et al. 2004); 9, (Saenz-Ponce et al. 2012a). predation of eggs and embryos by wasps, ants, katydids and other arthropods are likely determining factors in fa- vor of rapid development in H. fleischmanni. Aquatic eggs and embryos characterize the basal mode of frog reproduction, as exemplified by X. laevis and Ceratophrys stolzmanni (Table 1). These frogs re- lease a large number of small eggs in the water. However, frogs have invaded different environments for reproduc- tion due to competition for water resources, predation, and the dangers of desiccation. Accordingly, clutch size, egg size and developmental time vary among species (Table 1) (Duellman and Tmeb 1986). The dissimilar de- velopmental times of H. fleischmanni and E. callistomma may relate to their egg deposition sites and to different predation pressure on eggs and embryos. Egg deposition in the upperside or underside of leaves associated with differences in egg pigmentation and developmental time, as observed in centrolenid frogs, are different reproduc- tive modes that deserve further investigation. Development of H. fleischmanni and E. callistomma. The characteristics of development are detailed in Table 2, and illustrated in Figs. 2-13. It was of interest to docu- ment the characteristics of development of these glass frogs, given the observed differences in embryonic pig- mentation and developmental time. The development from early cleavage to tadpole hatching of H. fleischman- ni and E. callistomma was characterized according to the generalized table of frog development (Gosner 1960) (Table 2). Embryos of H. fleischmanni from fertilization to the sixteen cell stage were not available. Micrographs of the external morphology of embryos illustrate the developmental stages of both species, and clearly demonstrate the pigmentation differences among species (Figs. 1C, F; 2A-F; 3A-D; 4A-F; 5A-F). The internal morphology of embryos from cleavage until the completion of neurulation follows the typical frog pat- tern, as outlined in the generalized table of development (Gosner 1960) (Figs. 6-13). The most notable differenc- es are the overlap between gastrulation and the onset of neural development, and the lack of pigment in embryos of H. fleischmanni in comparison with embryos of E. cal- listomma. In both species cleavage was holoblastic (Figs. 6A-D; 7A-D), and the blastocoel roof was reduced to two-cells in thickness during gastrulation. At gastrula- tion, a conspicuous dorsal blastopore lip developed in the subequatorial dorsal region (Figs. 8A-F; 9A-E). The onset of neurulation began before completion of blasto- pore closure (Figs. lOA-D; llA-F). Developmental time, gastrulation and body elonga- tion. Our comparative analysis includes frog species with rapid and slow development (Table 1). Embryonic development occurs rapidly in frog species with aquatic reproductive modes. The analyzed frogs with rapid de- velopment and embryos suspended on the vegetation in- cluded H. fleischmanni, E. callistomma (Centrolenidae). Frogs with aquatic eggs and embryos included X. laevis February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 94 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma Table 2. Characteristics of development of the glass frogs Hyalinobatrachium fleischmanni and Espadarana callistomma. G Morphology observed in Centrolenid frogs^ 1 Fertilization (not available). 2 Gray crescent (not available). 3 Two cell stage (not available). 4 Four cell stage. The first two cleavage furrows passed from the animal to the vegetal pole. This stage was available only for E. callis- tomma (not shown). 5 Eight cell stage. The third cleavage furrow was latitudinal in some embryos and longitudinal in others. This stage was available only for E. callistomma (not shown). 6 Sixteen cell stage. Cleavage became asynchronous after the eight cell stage, and embryos with variable numbers of blastomeres were observed. This stage was available only iov E. callistomma (not shown). 7 Thirty-two cell stage. Cleavage in both species was holoblastic, and the animal micromeres were much smaller than the vegetal macro- meres, as observed for other frogs. (Figs. 2A; 4A; 6A-B; 7A-B). 8 Mid cleavage. Development of the blastocoel began during cleavage, as shown for 77. fleischmanni. (Figs. 2B; 6C-D). 9 Blastula. The blastocoel roof was thick and consisted of several cell layers (Figs. 2C; 4B; 7C-D). 10 Early gastrula. A conspicuous blastopore groove was observed on the dorsal subequatorial region of the embryo, and there were bottle cells marking cell ingression at the blastopore groove as shown for both species (Figs. 4C; 8A; 9A-B). In slightly more advance embryos, the dorsal blastopore lip was detected in the dorsal subequatorial region, as shown for 77. fleischmanni (Figs. 2D; 8B). 11 Mid gastrula. The blastopore lip surrounded a large yolk plug in embryos of both frogs (Figs. 2E; 4D; 8C). Internally, the archenteron was elongated, without inflation (Figs. 8D; 9C). The blastocoel roof was translucent (Fig. 1C) and consisted of two-cell layers (not shown). 12 Late gastrula and development of the neural plate (Eigs. 2E; 4E; 8E). The neural groove and the neural plate were visible in gastrula stage embryos with a small yolk plug (stage 12.5) (Eigs. 2G; 4E, 11 A). The archenteron was elongated in an anterior direction and it was inflated, and the blastocoel was reduced in size. The cleft of Brachet, that separates the ectoderm from the endomesoderm, was vis- ible in the roof of the primitive gut (Eigs. 8E; 9D-E; lOA; 11 B-C) The notochord was detected in stage 12.5 embryos, as shown for 77. fleischmanni (Eig. lOB). In stage 12.75, the neural plate was visible in both species (Eigs. 2H; 4G; 11 D). The yolk plug was small, the archenteron was fully inflated, and the germ layers were visible (Fig. lOC-D; 1 1 E-F). A triangular dorsal structure, considered to be the gastrocoel roof plate (grp), was located in the roof of the primitive gut, and was exposed to the cavity of the gastrocoel (Fig. 12C). The grp included the ventral surface of the notochord and paraxial mesoderm, and was bordered by the lateral endodermal crests (lec). The grp is illustrated for E. callistomma (Figs. 12D). 13 The closed blastopore and the neural plate. The yolk plug was totally retracted, the blastopore was at the slit blastopore stage, and the neural plate was visible (Figs. 4H; 12A). The grp was located in the roof of the primitive gut, and it was bordered by the lec, shown in whole mount for 77. fleischmanni (Fig. 12B). 14 Early neural fold stage. The neural folds were slightly elevated (Eigs. 21; 41; 13A). The grp included the ventral surface of the notochord, and somites, and it was bordered by the lec, shown forE. callistomma (Eig. 12 E-E). The neural ectoderm, paraxial mesoderm, notochord, and endoderm were visible (Eig. 13B). 15 Mid neural fold stage. The neural folds were elevated (Eigs. 2J; 4J; 13C). In cross sections, the neural ectoderm, notochord, paraxial mesoderm and endoderm were visible, as shown for 77. fleischmanni (Eig. 13 D). 16 Closure of the neural tube. The neural folds were closed (Eigs. 2K; 4K; 13 E). In cross sections, the neural tube was visible dorsal to the notochord. The somites were visible on each side. The endoderm completely lined the archenteron, as shown for E. callistomma (Eigs. 13 E). 17 Tail bud stage. The tail bud and the head region protruded beyond the yolky endoderm. The branchial arches were visible (Eigs. 2L; 4L). 18 Muscular activity. The branchial arches protruded on the sides of the head. The tail became elongated. This stage is only shown for E. callistomma (Eig. 5A). 19 Heart beat. The heart heated, and the gill buds were visible. This stage is only shown for E. callistomma (Eigs. 5B, C). 20 Circulation to the external gills. There were two gill pairs, each with two small branches. This stage is not shown. 21 The gills were larger, the first pair gill had two branches for both species and the second pair gill was unbranched in 77. fleischmanni (Eigs. 3A; 5D). 22 Tail fin circulation. Not observed. 23 The external gills reached their full size. There were five gill branches in the first pair and four branches in the second pair of external gills in embryos of 77. fleischmanni. The opercular fold was developing. There were four gill branches in the first pair and three branches in the second pair of external gills of E. callistomma embryos (Eigs. 3B-5E). 24 Larval stage. Not observed 25 Tadpole at hatching. Only a small portion of the external gills protruded from the opercular aperture in the hatching tadpoles. The eyes were very small. (Eigs. 3C-D; 5E). ^ The development of the Centrolenid frogs (C), H. fleischmanni and E. callistomma, was compared with the general staging table for frogs (G) (Gosner, 1960). February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 95 e88 Salazar-Nicholls and del Pino Hyalinobatrachium fleischmanni Fig. 3. External morphology of Hyalinobatrachium fleischmanni embryos from the development of the gills stage to hatching. (A) Stage 21: The gills were large, and each gill pair had two branches. (B) Stage 23: Full development of the external gills. There were five gill branches in the first pair and four branches in the second pair of gills. (C) Stage 25: Lateral view of a tadpole at hatching. The eyes were very small. (D) Stage 25: Ventral view of a tadpole at hatching. Only a small portion of the external gills protruded from the opercular aperture. The pink color of the embryo in A was an artifact of fixation, e, eye; fg, first pair gills; g, gills; mo, mouth; sg, second pair gills; tf, tail fin. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 96 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma E. callistomma A B ^ Q 10 D St 11 Fig. 4. External morphology of Espadarana callistomma embryos from cleavage to the tail bud stage. (A) Stage 7: Thirty-two cell stage. Animal micromeres were much smaller than the vegetal macromeres. (B) Stage 9: Blastula. (C) Stage 10: Early gastrula. A conspicuous blastopore groove was observed on the dorsal subequatorial region of the embryo. (D) Stage 11: Mid gastrula. The blastopore lip surrounded a large yolk plug. (E) Stage 12: Late gastrula. (F) Stage 12.5: Late gastrula with a small yolk plug. (G) Stage 12.75: Late gastrula with a very small yolk plug. The neural plate was visible. (H) Stage 13: The neural plate was visible. The yolk plug was totally retracted and the blastopore was at the slit blastopore stage. (I) Stage 14: Early neural fold. The neural folds were visible. (J) Stage 15: Mid neural fold. The neural folds were elevated. (K) Stage 16. Closure of the neural tube. The neural folds were closed. (L) Stage 17. Tail bud stage. The branchial arches were visible, bg, blastopore groove; hr, branchial arch; bp, closed blastopore; c, cleavage furrow; dl, dorsal blastopore lip; hy, hyoid arch; ma, mandibular arch; nf, neural fold; ng, neural groove; np, neural plate; vl, ventral blastopore lip; yp, yolk plug. (Pipidae), and Ceratophrys stolzmanni (Ceratophryidae), and frogs with embryos placed in flotating foam-nests were Engystomops randi and Engystomops coloradorum (Leptodactylidae) (Table 1). In contrast, embryonic de- velopment was much slower in embryos of frogs with terrestrial adaptations. Frogs with slow development included the Marsupial frog Gastrotheca riobambae (Hemiphractidae) that broods its embryos in a dorsal pouch of the mother and the dendrobatid frogs Epipedo- bates machalilla and Epipedobates tricolor (Dendrobati- dae) that deposit their eggs in terrestrial nests (Table 1) (del Pino et al. 2007; Elinson and del Pino 2012). Gastrulation characteristics vary among frogs accord- ing to their developmental speed. Gastrulation and body elongation, as detected by the onset of notochord elonga- tion, overlapped in embryos of X. laevis, C. stolzmanni. E. randi, and E. coloradorum, frogs with rapid develop- ment (Table 1). Similarly, elongation of the notochord overlapped with gastrulation in the rapidly developing embryos of the centrolenid frogs H. fleischmanni and E. callistomma (Figs. 8D, F; 9C-F; lOA-D; IIB-C, E-F; 12D; Table 2). In contrast, gastrulation movements oc- curred before the onset of notochord elongation in the slowly developing dendrobatids E. machalilla and E. tri- color, and in the Marsupial frog, G. riobambae. Egg size is larger in these slowly developing frogs in comparison with the rapidly developing species (Table 1), (Elinson and del Pino 2012; del Pino et al. 2007). The modular nature of gastrulation allows the separa- tion of dorsal convergence and extension, the mechanism that triggers elongation of the notochord and the body, from gastrulation in the slowly developing frogs, and February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 97 e88 Salazar-Nicholls and del Pino E. callistomma Fig. 5. External views of Espadamna callistomma embryos from the stage of muscular activity to hatching stages. (A) Stage 18: Muscular activity. The branchial arches protruded on the sides of the head. (B) Stage 19: Heartbeat. The gill buds were visible. (C) Stage 19.5: Two gill pairs were visible, each with two small branches. (D) Stage 21: The gills were larger, and each gill pair had two branches. (E) Stage 23: Full development of the external gills. There were four gill branches in the first pair and three branches in the second pair of gills. (F) Stage 25: Tadpole at hatching. The eyes were very small, hr, branchial arch; e, eye; fg, first gill pair; gb, gill bud; hy, hyoid arch; ma, mandibular arch; sg, second gill pair; tf, tail fin. the overlap of these two processes in rapidly develop- ing frog species (Elinson and del Pino 2012). Overlap of gastmlation and body elongation is associated with rapid development in the unstable conditions of the reproduc- tive modes that involve aquatic reproduction of X. laevis and C. stolzmanni, floating foam-nest development in Engystomops, and suspension of eggs on the vegetation, in the case of centrolenids frogs (Table 1), (Elinson and del Pino 2012). The distinct modes of gastmlation likely relate to the reproductive mode of frogs, rather than to phylogenetic relationships. The gastrocoel roof plate (grp) and left-right asym- metry. It was of interest to determine whether frogs with different reproductive modes, and different onset of no- tochord elongation share the pattern of left-right asym- metry determination by cilia driven fluid flow towards the left side in the grp, described for X. laevis (Blum et al. 2014b; Saenz-Ponce et al. 2012b). The question is particularly important because the mechanism of sym- metry breakage by cilia driven fluid flow in the grp or equivalent stmctures is universal among vertebrates with exception of the chick and the pig (Blum et al. 2014a,b). In all frogs analyzed, the gastrocoel roof plate (grp) had a triangular shape and was detected in the dorsal lining of the primitive gut of the late gastmla and neumla, as detected in H. fleischmanni and E. callistomma embryos (Pig. 12A-P; Table 1). As in A laevis and other frogs, the grp of H. fleischmanni and E. callistomma embryos con- sisted of the ventral surface of the posterior notochord and paraxial mesoderm, and it was bordered by the later- al endodermal crests (lee), illustrated for E. callistomma, (Pigs. 12D-E). However, in a more rostral region, only the notochord was exposed to the cavity of the primitive gut because the paraxial mesoderm was already covered by the closing lee (Fig. 12F). The major difference de- tected among frogs was the presence of the grp already in the late gastmla of the centrolenid frogs, as shown for E. callistomma (Fig. 12D), whereas the grp developed in the neumla of X. laevis (Blum et al. 2014b). The preco- cious onset of grp formation may relate to the overlap of neumlation and gastmlation in centrolenid frogs, another example of the modular nature frog gastmlation. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 98 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma H. fleischmanni Fig. 6. Cleavage in Hyalinobatrachium fleischmanni. (A) Stage 7: Animal view of a 32-cell embryo. (B) Stage 7: The blastocoel of a 32-cell embryo, observed in a sagittal bisection. (C) Stage 8: Animal view of an embryo at mid-cleavage. (D) Stage 8: The blastocoel of a mid-cleavage embryo, observed in a sagittal bisection, bl, blastocoel. The grp was detected in the neumla of eight frog spe- cies with a wide range of reproductive adaptations, and belonging to six different frog families, (Table 1) (Saenz- Ponce et al. 2012a, b). The presence of the grp in this wide range of frogs suggests that determination of left- right asymmetry may follow mechanisms similar to those described for X. laevis. Moreover, cilia were detected in the grp epithelium that lines the dorsal roof of the primi- tive gut of these various frogs (Saenz-Ponce et al. 2012a, b). The presence of cilia in the grp in centrolenid frogs was not analyzed. Conclusions. The reproductive and developmental strat- egies of the two centrolenid frogs, analyzed in this work, differ from each other. The eggs of E. callistomma, de- posited on the upper sides of plant leaves, contain dark pigment, and take twice as long to reach the hatching stage in comparison with H. fleischmanni embryos. In contrast, the H. fleischmanni development on the under- side of plant leaves is accompanied by the lack of dark pigment in the egg and embryos and reduced develop- mental time. As in other frogs with rapid development, there was overlap between gastrulation and body elonga- February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 99 e88 Salazar-Nicholls and del Pino E. callistomma Fig. 7. Cleavage in Espadarana callistomma. (A) Stage 7: Animal view of a 32-cell embryo. (B) Stage 7: The blastocoel of a 32-cell embryo, observed in a sagittal bisection. (C) Stage 9: Animal view of a blastula. (D) Stage 9: The blastocoel of a blastula, observed in a sagittal section. The blastocoel roof consisted of several cell layers, bl, blastocoel. tion. Moreover, the process of neurulation already started during gastrulation, and the grp became visible in the late gastmla. Presence of the grp in embryos of these cen- trolenid frogs suggests that the mechanisms of left-right asymmetry is likely similar with the cilia-driven pattern of the X. laevis grp. Acknowledgments. — We thank the Centre of Am- phibian Investigation and Conservation, Balsa de los Sa- pos, Pontificia Universidad Catolica del Ecuador (PUCE) for the donation of embryos of the two species analyzed in this work. We express gratitude to the members of the Laboratory of Developmental Biology of PUCE for their assistance in the conduction of this study, and in particu- lar we express gratitude to Natalia Saenz-Ponce, Alexan- dra Vargas, and Andres Garces for their help. We thank Santiago Ron for providing the photographs of the adults of both species, and Clifford Keil for critical analysis of the manuscript and language revision. This study re- ceived the support of a 2013 research grant from PUCE. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 100 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma Hyalinobatrachium fleischmanni Fig. 8. Gastrulation of Hyalinobatrachium fleischmanni (Stages 10-12). Embryos in A, C, E were stained for cell borders. (A) Stage 10: Early gastrula. Dorsal subequatorial region. The dorsal blastopore groove was visible between the small cells of the animal region with clearly delineated borders, and the vegetal cells, whose borders were not as clear. (B) Stage 10.5: Sagittal section of an early gastrula. The dorsal blastopore lip was visible. (C) Stage 1 1 : Mid gastrula. Higher magnification of the dorsal blastopore lip region. There was difference in size of animal and vegetal cells. (D) Stage 1 1 : Sagittal section of a mid gastrula. The archenteron was elongated, and the blastocoel roof was reduced to about two cell layers. (E) Stage 12: Late gastrula. Higher magnification of the yolk plug region. (F) Stage 12: Sagittal section of late gastrula. The arrow indicates the cleft of Brachet. a, archenteron; bg, blastopore groove; bl, blastocoel; dl, dorsal blastopore lip; vl, ventral blastopore lip; yp, yolk plug. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 101 e88 Salazar-Nicholls and del Pino E. callistomma Fig. 9. Gastrulation of Espadarana callistomma (Stages 10-12). (A) Stage 10: Sagittal section of an early gastrula. The dorsal blasto- pore groove was visible. (B) Stage 10. Higher magnification of the embryo in A, stained for cell nuclei. The arrow indicates a bottle cell of the blastopore groove area. (C) Stage 11: Sagittal section of mid gastrula. (D) Stage 12: Sagittal bisection of late gastrula. (E) Stage 12: Sagittal section of the late gastrula shown in D. The single cavity is an artifact of sectioning, it corresponds to the blastocoel and archenteron, as shown in D. (F) Higher magnification of the archenteron roof from the embryo in E, stained for cell nuclei. The arrow indicates the cleft of Brachet. a, archenteron; bl, blastocoel; dl, dorsal blastopore lip; ec, ectoderm; vl, ventral blastopore lip; yp, yolk plug. H. fleischmanni A st12.5|B St 12.5 Fig. 10. Gastrulation of Hyalinobatrachium fleischmanni (Stages 12.5-12.75). (A) Stage 12.5: Sagittal section of a late gastrula. (B) Stage 12.5: Cross section through the rostral region of a late gastrula, stained for cell nuclei. The endoderm covered the notochord in this rostral section. (C) Stage 12.75: Sagittal section of a late gastrula. (D) Stage 12.75: Higher magnification of the archenteron roof from the embryo in E, stained for cell nuclei. The three germ layers were visible, a, archenteron; dl, dorsal blastopore lip; ec, ectoderm; en, endoderm; m, mesoderm; pm; paraxial mesoderm, vl, ventral blastopore lip; yp, yolk plug. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 102 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma E. callistomma Fig. 11. Gastrulation of Espadarana callistomma (Stages 12.5-12.75). (A) Stage 12.5: Late gastrula with a small yolk plug. (B) Stage 12.5: Sagittal section of a late gastrula. (C) Stage 12.5: Sagittal section of the archenteron roof through the rostral region of a late gas- trula, stained for cell nuclei. (D) Stage 12.75: Late gastrula. The neural plate was visible. (E) Stage 12.75: Parasagittal section of a late gastrula. (F) Stage 12.75: Higher magnification of the archenteron roof from the embryo in E, stained for cell nuclei. The arrows in C, E and E indicate the cleft of Brachet. a, archenteron; cbc, circumblastoporal collar; dl, dorsal blastopore lip; ec, ectoderm; en, endoderm; m, mesoderm; ng, neural groove; np, neural plate; vl, ventral blastopore lip; yp, yolk plug. Literature Cited AmphibiaWeb [web application]. Berkeley California. 2014. Available: http://amphibiaweb.org/ [Accessed: 2014]. BlumM, Feistel K, ThumbergerT, Schweickert A. 2014a. The evolution and conservation of left-right pattern- ing mechanisms. Development 141: 1,603-1,613. Blum M, Schweickert A, Vick P, Wright C, Danilchik M. 2014b. Symmetry breakage in the vertebrate embryo: When does it happen and how does it work? Develop- mental Biology 39^. 109-123. Blum M, Weber T, Beyer T, Vick P. 2009. Evolution of leftward flow. Seminars in Cell & Developmental Bi- ology 20: 464^71. Cabanzo-Olarte LC, Ramirez-Pinilla MP, Serrano-Car- dozo VH. 2013. Oviposition, site preference, and evaluation of male clutch attendance in Espadarana Andina (Anura: Centrolenidae). Journal of Herpetol- ogy Al\ 314-320. Cisneros-Heredia DF, Mcdiarmid RW. 2007. Revision of the characters of Centrolenidae (Amphibia: Anura: Athesphatanura), with connnents on its taxonomy and the description of new taxa of glassfrogs. Zootaxa 1572: 1-82. del Pino EM, Avila ME, Perez O, Benitez MS, Alarcdn I, Noboa V, Moya I. 2004. Development of the den- drobatid frog Colostethus machalilla. International Journal of Developmental Biology 48: 663-670. del Pino EM, Venegas-Ferrin M, Romero-Carvajal A, Montenegro-Larrea P, Saenz-Ponce N, Moya IM, Alarcdn I, Sudou N, Yamamoto S, Taira M. 2007. A Comparative analysis of frog early development. Proceedings of National Academy of Sciences of the United States of America 104: 11,882-11,888. del Pino EM. 1996. The expression of Brachyury (T) during gastrulation in the marsupial frog Gastrotheca riobambae. Developmental Biology 177: 64-72. Delia J, Cisneros-Heredia DE, Whitney J, Murrieta- Galindo R. 2010. Observations on the reproductive behavior of a Neotropical glassfrog, Hyalinobatra- chium fleischmanni (Anura: Centrolenidae). South American Journal of Herpetology 5: 1-12. Duellman WE,Trueb L. 1986. Biology of Amphibians. McGraw Hill Inc, New York, New York, USA. 670 p. Elinson RP, del Pino EM. 2012. Developmental diversity of Amphibians. Wiley Interdisciplinary Reviews: De- velopmental Biology 1: 345-369. Gosner KL. 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Her- petologica 16: 183-190. Greer BJ, Wells KD. 1980. Territorial and reproductive behavior of the Tropical American frog Centrolenella fleischmanni. Herpetologica 36: 318-326. Guayasamin JM,Trueb L. 2007. A new species of glass- frog (Anura: Centrolenidae) from the lowlands of Northwestern Ecuador, with connnents on centrolenid osteology. Zootaxa 1447: 27^5. Lee C, Le MP, Cannatella D, Wallingford J. 2009. Chang- es in localization and expression levels of shroom2 February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 103 e88 Salazar-Nicholls and del Pino H. fleischmanni Fig. 12. The gastrocoel roof plate (grp) in embryos of Hyalinobatrachium fleischmanni and Espadarana callistomma. (A) Stage 13: External view of the neural plate of H. fleischmanni. (B) Stage 13: Internal view of the neural plate region of H. fleischmanni. The grp was visible in the midline. Arrows indicate the lateral endodermal crests (lee) at the border of the grp in B-F. (C) Stage 12.75: Internal view of the neural plate region of E. callistomma. The grp was visible in the midline. (D) Stage 12.75: Cross section through the caudal region of a late gastrula of E. callistomma, stained for cell nuclei. The grp was exposed in the midline and bordered by the lee. The grp consisted of the ventral surface of the notochord and paraxial mesoderm. (E) Stage 14: Early neural fold of E. callistomma. Cross section through the caudal region. The grp was exposed in the midline and bordered by the lee. The grp consisted of the ventral surface of the notochord and paraxial mesoderm, as in stage 12.75 embryos (shown in D). (F) Stage 14: Early neural fold of E. callistomma. Cross section through the rostral region, stained for cell nuclei. The grp included only the ventral surface of the notochord, due likely to the rostral closure of the lee. Only the notochord was exposed in the midline, and bordered by the lee. a, archenteron, ec, ectoderm; en; endoderm; grp, gastrocoel roof plate; m, mesoderm; n, notochord; np, neural plate; pm, paraxial mesoderm; ng, neural groove. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 104 e88 Early development of Hyalinobatrachium fleischmanni and Espadarana callistomma H. fleischmanni E. callistomma Fig. 13. Neurulation of Hyalinobatrachium fleischmanni and Espadarana callistomma. (A) Stage 14. Early neural fold stage of H. fleischmanni. (B) Stage 14; Cross section through the rostral region of the embryo in A, stained for cell nuclei. The notochord is totally covered by endoderm. (C) Stage 15: Mid neural fold of H. fleischmanni. (D) Stage 15: Cross section through the rostral region of the embryo in C, stained for cell nuclei. (E) Stage 16: Mid neural fold stage of E. callistomma. (F) Stage 16: Cross section through the rostral region of E. callistomma, stained for cell nuclei. The neural folds were closed, a, archenteron; en, endoderm; n, notochord; nf, neural fold; np, neural plate; nt, neural tube; pm, paraxial mesoderm; s, somite. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 105 e88 Salazar-Nicholls and del Pino and spectrin contribute to variation in amphibian egg pigmentation patterns. Developmental Genes and Evolution 219: 319-330. Moya IM, Alarcon 1, del Pino EM. 2007. Gastrulation of Gastrotheca riobambae in comparison with other frogs. Developmental Biology 304: 467^78. Nieuwkoop PD, Faber J. 1994. Normal Table ofXenopus Laevis (Daudin). Garland Publishing, New York, New York, USA. 243 p. Ortiz DA, Almeida-Reinoso DP, Coloma LA. 2013. Notes on husbandry, reproduction and development in the Pacific horned frog Ceratophrys stolzmanni (An- ura: Ceratophryidae), with comments on its amplexus. International Zoo Yearbook Al\ 151-162. Ospina-Sarria JJ, Bolivar-G W, Mendez-Narvaez J. 2010. Amphibia, Anura, Centrolenidae, Espadarana callis- tomma (Guayasamin and Trueb, 2007): First country records from Colombia. Check List 6: 244-245. Romero-Carvajal A, Saenz-Ponce N, Venegas-Ferrin M, Almeida-Reinoso D, Lee C, Bond J, Ryan M, Wall- ingford J, del Pino EM. 2009. Embryogenesis and laboratory maintenance of the foam-nesting tiingara frogs, GQmxs Engystomops {=Physalaemus). Develop- mental Dynamics 1,444-1,454. Saenz-Ponce N, Arias-Contreras M, del Pino EM. 2012a. La placa del techo del gastrocele de Epipedobates tri- color (Anura: Dendrobatidae). Revista Ecuatoriana de Medicina y Ciencias Bioldgicas 33: 11-19. Saenz-Ponce N, Santillana-Ortiz JD, del Pino EM. 2012b. The gastrocoel roof plate in embryos of differ- ent frogs. Differentiation 83: 62-66. Savage JM. 2002. The Amphibians and Reptiles of Coast Rica: A Herpetofauna between two continents, and two seas. The University of Chicago Press, Chicago, Illinois, USA. 954 p. Schweickert A , Vick P, Getwan M, Weber T, Schneider I, Eberhardt M, Beyer T, Pachur A, Blum M. 2010. The nodal inhibitor coco is a critical target of leftward flow in Xenopus. Current Biology 20\ 738-743. Villa J. 1977. A symbiotic relationship between frog (Amphibia, Anura, Centrolenidae) and fly larvae (Drosophilidae). Journal of Herpetology 1 1 : 317-322. Vockenhuber EA, Hodl W, Karpfen U. 2008. Reproduc- tive behaviour of the glass frog Hyalinobatrachium valerioi (Anura: Centrolenidae) at the Tropical Stream QuebradaNegra ( La Gamba, Costa Rica ). Stapfia 80: 335-348. Maria- Jose Salazar-Nicholls is research assistant in the Laboratory of Developmental Biology at the Pon- tificia Universidad Catblica del Ecuador (PUCE) in Quito. She graduated with a Licenciatura in Biological Sciences from PUCE in 2013. Her research centers on the characterization of early development in centro- lenid frogs. She is currently investigating the mode of somitogenesis in Hyalinobatrachium fleischmanni and Espadarana callistomma. She is interested in climate change and its impacts on conservation. Eugenia M. del Pino is Professor of Biological Sciences (retired) at PUCE in Quito. She studies the re- production and development of Marsupial frogs (Hemiphractidae) in comparison with Xenopus laevis, the model organism of frog developmental biology and with tropical frogs from Ecuador. Her studies are done in collaboration with PUCE students. The analyses of development reveal important variation in develop- mental speed according to the reproductive mode of the various frogs. The developmental data is signifi- cant for the comparative analysis of frog early embryonic development and provide base line information about the biology of several frog species. February 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 106 e88 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 107-120 (e89). A new species and country record of threadsnakes (Serpentes: Leptotyphlopidae: Epictinae) from northern Ecuador i>2,5David Salazar-Valenzuela, ^Angela Martins, "^Luis Amador-Oyola, and ^Omar Torres-Carvajal '^Department of Evolution, Ecology and Organismal Biology, Ohio State University, 300 Aronojf Laboratory, 318 W. 12th Avenue, Columbus, Ohio 43210-1293, USA ^Museo de Zoologla, Escuela de Ciencias Biologicas, Pontificia Universidad Catolica del Ecuador, Avenida 12 de Octubre y Roca, Apartado 17-01-2184, Quito, ECUADOR ^Universidade Eederal do Rio de Janeiro, Museu Nacional, Departamento de Vertebrados, Rio de Janeiro, Rio de Janeiro 20940-040, BRAZIL “^Departamento de Investigacion Cientlfica, Tecnologica e Innovacion, Universidad Laica Vicente Rocafuerte, Avenida de Las Americas, Apartado 11-33, Guayaquil, ECUADOR Abstract . — ^We describe a new species of Triiepida Hedges 2011 from cloud forests of the extreme northern Ecuadorian Andes, Carchi province. Among other characters, the new species is distinguished from all congeners by having a subhexagonal ocular with its anterior border barely rounded at eye level, rostral reaching the anterior border of ocular scales in dorsal view, three supralabials, four or five infralabials, thicker body width, 203-214 middorsal scales, 12 scales around middle of tail, uniform gray dorsum, and gray venter with interspaces between scales cream. Morphologically, the new species is most similar to T. guayaquilensis and T Joshuai from Ecuador and Colombia, respectively. We also report the first records of T. macroiepis ior the country from the lowland and foothill evergreen forests of northwestern Ecuador. Key words. Andes, Choco, cloud forest, fossorial, external morphology, osteology, Triiepida macrolepis; Triiepida pastusa, new species Citation: Salazar-Valenzuela D, Martins A, Amador-Oyola L, Torres-Carvajal O. 2015. A new species and country record of threadsnakes (Serpentes: Leptotyphlopidae; Epictinae) from northern Ecuador. Amphibian & Reptile Conservation 8{t) [Special Section]: 107-120 (e89). Copyright: © 2015 Salazar-Valenzuela et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non- CommercialNoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptile Conservation] official journal website . Received: 12 January 2015; Accepted: 08 February 2015; Published: 02 March 2015 Introduction Fossorial snakes of the family Leptotyphlopidae are among the least known terrestrial vertebrates (Adal- steinsson et al. 2009). Even though some species in the family may be locally abundant and the group has a wide distribution from sea level to mountaintops in Africa, the Americas, and parts of Asia, their secretive habits make them rarely encountered in the field (Curcio et al. 2002; McDiarmid et al. 1999; Passes et al. 2005; Pinto et al. 2010). Most leptotyphlopids are small (150 to 250 mm snout- vent length), thin, and burrowing animals that feed on social insects (termites are probably the main food source for some species) (Vitt and Caldwell 2013). Re- cent phylogenetic analyses based on molecular data parti- tioned the 1 12 species now recognized in the family in 12 genera (Adalsteinsson et al. 2009; Wallach et al. 2014). Correspondence. Email: ^davidsalazarv® gmail.com Amphib. Reptile Conserv. As a combination of limited morphological variation in fossorial squamates and paucity of specimens, morpho- logical synapomorphies for these lineages have not been clearly established yielding differences in researchers’ opinions as to which genus some species should be allo- cated. Especially problematic has been the classification of threadsnakes in the Neotropical genera Rena, Siagon- odon, and Triiepida (Pinto and Curcio 2011; Pinto and Eernandes 2012). In Ecuador, four species of threadsnakes have been confirmed on the basis of voucher specimens: Epic- tia signata, E. subcrotilla, Triiepida anthracina, and T guayaquilensis (Cisneros-Heredia 2008; Pinto et al. 2010; Salazar-Valenzuela et al. 2010; Torres-Carvajal et al. 2014; Wallach et al. 2014). With the exception of E. subcrotilla, which seems to be a common taxon in natural history collections (Cisneros-Heredia 2008; Purtschert 107 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. 2007), most of the leptotyphlopid species reported for the country are known from a few specimens or exclu- sively from the holotype in the case of T. guayaquilensis (Cisneros-Heredia 2008). Cryptozoic species of snakes are difficult to find and the usefulness of regularly including digging techniques during herpetological surveys has been proposed for fos- sorial reptiles (Measey 2006). While performing field- work on the extreme northern Andes of Ecuador, we found specimens of an undescribed species of Trilepida with the help of local people who usually dig for archae- ological remains of Los Pastos pre-hispanic culture. Here we recognize this species based on morphological data, increasing the number of species of Trilepida to 14 (Uetz and Hosek 2014; Wallach et al. 2014). While reviewing material for this study we came across specimens assign- able to Trilepida macrolepis, which constitute the first record of this species for Ecuador and are also reported herein. Materials and Methods We examined specimens housed in the Museo de Zo- ologia, Pontificia Universidad Catolica del Ecuador (QCAZ) and the Museo Ecuatoriano de Ciencias Natu- rales (DHMECN), Quito, Ecuador. In addition, we ana- lyzed photographs of specimens deposited in the Nation- al Museum of Natural History, Smithsonian Institution (USNM), Washington, D.C., USA, and the Museum fur Naturkunde (ZMB) Berlin, Germany. Characters used for description and comparisons were based on internal (skull of a MicroCT Scanned specimen. X-ray plates) and external morphology (meristic and morphometric data, shape of cephalic plates, and color pattern) of examined individuals, as well as published data provided by Ro- jas-Morales and Gonzalez-Duran (2011) and those sum- marized in Pinto and Eernandes (2012). We consider the unique combination of morphological characters present in the new species as delimitation criteria, following the general species concept of de Queiroz (1998, 2007). Ter- minology for cephalic plates, scale features, cloacal sacs, and measurements follows Broadley and Wallach (2007), Kroll and Reno (1971), Passos et al. (2006), Pinto and Curcio (2011), and Pinto and Eernandes (2012). Color description in life is based on analysis of a series of pho- tographs of the holotype and paratypes. Color terminolo- gy and codes follow Kohler (2012). Measurements were taken with a dial caliper to the nearest 0. 1 mm, except for total length (TL) and tail length (TL), which were mea- sured with a ruler to the nearest 1 .0 mm. Measurements and descriptions of paired cephalic scales are provided for the right side. Sex was determined by the presence or absence of hemipenial muscles through a ventral incision at the base of the tail. Characters recorded are: (1) Total length (TL); (2) tail length (TAL); (3) TL/TAL ratio; (4) middorsal scales (rostral and terminal spine excluded); (5) midventral scales (mental scale, cloacal shield, and subcaudals excluded); (6) subcaudal scales (terminal spine excluded); (7) dorsal scale rows around the mid- dle of the tail (DSR); (8) midbody diameter (MB); (9) midtail diameter (MT); (10) head length (HE); (11) head width (HW); (12) relative eye diameter (ocular width at eye level/eye diameter); (13) presence of fused caudals; (14) relative rostral width (rostral width/head width) (Pinto and Curcio 2011; Pinto et al. 2010). The head of one preserved specimen (QCAZ 5778) was scanned on a Skyscan 1176 in-vivo high-resolution micro-CT scan at Universidade de Sao Paulo, Brazil. The specimen was scanned at 40 kV and 533 uA, and the dataset was ren- dered in three dimensions through the use of CTVox for Windows 64 bits version 2.6. Additionally, the skeleton of the holotype and paratypes were examined dorsally and ventrally through X-ray plates. Digital radiographies of QCAZ 5778 were made with a Eaxitron X-Ray EEC MX 20 at Departamento de Vertebrados, Museu Nacio- nal, Universidade Rio de Janeiro, Brazil and X-ray scans of QCAZ 8990 and QCAZ 5846 were made with a Ther- mo Kevex X-ray Imaging System at QCAZ. Terminol- ogy for the braincase, mandible, and vertebral column follows Rieppel et al. (2009), Kley (2006), and Holmann (2000), respectively. Results Trilepida pastusa sp. nov. urn:lsid:zoobank.org:act:E7C8FFlC-07E8-4985-B673-80A52DACA8Dl Eigs. 1-3. Holotype. — Adult female, QCAZ 8690, collected on 23 Eebruary 2009 by O. Torres-Carvajal, S. Aldas- Alarcon, E. Tapia, A. Pozo and local people, surround- ings of Chilma Bajo on the way to Tres Marias waterfall (0°51’53.82” N, 78°2’59.23” W; 2071 m), Tulcan Coun- ty, Carchi province, Ecuador. Paratypes. — Two specimens with same locality data as holotype: one juvenile female (QCAZ 5778) collected on 21 Eebruary 2013 by D. Salazar-Valenzuela, H. Pozo, A. Chalapud, and D. Males, and one juvenile of undeter- mined sex (QCAZ 5846) collected on 20 March 2013 by D. Salazar-Valenzuela and A. Loaiza-Lange. Diagnosis. — Trilepida pastusa is distinguished from all congeners by the following combination of characters: Snout truncate in dorsal and ventral view, rounded in lateral view; supraocular present; ocular subhexagonal with superior border straight and anterior border barely rounded at eye level; rostral subtriangular in dorsal view, reaching anterior border of ocular scales; frontal as long as supraocular and other middorsal cephalic shields, ex- cept for postfrontal which is smaller; temporal distinct; supralabials three (2+1); infralabials four or five; body March 2015 I Volume 8 | Number 1 | e89 Amphib. Reptile Conserv. 108 A new species of threadsnake from northern Ecuador Fig. 1. Dorsal (A), lateral (B), and ventral (C) views of the head of the holotype of Trilepida pastusa sp. nov. (QCAZ 8690). Scale bar =1.5 mm. Drawings by D. Paucar. width relatively thick (TL/MB = 28.6-34.6); middorsal scales 203-214; midventral scales 182-193; subcaudal scales 18-19; fused caudals present; terminal spine coni- cal, slightly longer than wide; scales around middle of tail 12; dorsum uniform gray; venter gray with interspac- es between scales cream. Comparisons (Table 1). — Among all congeners, Trilep- ida pastusa is more similar to T guayaquilensis and T. joshuai in sharing 12 scales around midtail, three supra- labials, and a uniform dark dorsum and pale venter (Pin- to and Fernandes 2012). The new species differs from both by having an ocular with an anterior border barely rounded at eye level, a thick body (TL/MB = 28.6-34.6), and a coloration pattern consisting of a uniform gray dor- sum and a venter with gray on the center and cream on the outside of each scale (Figs. 3, 4A) (vs ocular with rounded anterior border, thin body [TL/MB = 48.6], and uniform brown color dorsally and pale brown ventrally in T. guayaquilensis; ocular with rounded anterior border, moderate to thin body width [TL/MB = 34.0-55.2], and uniform dark brown dorsally and cream ventral scales in T. joshuai) (Orejas-Miranda and Peters 1970; Pinto and Fernandes 2012; Pinto et al. 2010; Rojas-Morales and Gonzalez-Duran 2011). Moreover, T. pastusa dif- fers from T. guayaquilensis in having a lower number of middorsal (203-214 vs 253, respectively) and midventral (182-193 vs 233, respectively) scales (Orejas-Miranda and Peters 1970; Pinto and Fernandes 2012). The new species also differs from T. joshuai by having a higher number of middorsal (203-214 vs 174-199, respective- ly) and midventral (182-193 vs 165-187, respectively) scales, and a higher number of subcaudals (18-19 vs 13- 18, respectively) (Pinto and Fernandes 2012; Pinto et al. 2010; Rojas-Morales and Gonzalez-Duran 2011). Some specimens of T. macrolepis can have 12 scales around Table 1. — Meristic and morphometric variation of the four species of Trilepida occurring in Ecuador. Data are from Pinto and Fer- nandes (2012) and references therein, as well as our own scale counts and measurements. Abbreviations: DO = middorsal scales; VE = midventral scales; SC = subcaudals; TL = total length; TAL = tail length; MB = midbody diameter; MT = midtail diameter; SL = supralabials; IL = infralabials; SO = supraocular. Color pattern adapted from Passos et al. (2006) and Pinto and Fernandes (2012): 1 = uniform violet black dorsally and ventrally; 2 = reticulate dark brown dorsally and reticulate pale brown ventrally; 3 = uniform brown dorsally and pale brown ventrally; 4 = uniform gray dorsally and reticulate gray ventrally. Character Trilepida pastusa sp. nov. Trilepida macrolepis Trilepida guayaquilensis Trilepida anthracina DO 203-214 211-255 253 182-193 VE 182-193 201-237 233 167-176 SC 18-19 16-24 20 15-19 TL/TAL 10.86-13.67 8.2-15.9 13.1 12.2-16.6 TL/MB 28.63-34.55 32.2-68.3 48.6 31.7-43.7 TAL/MT 2.67-3.81 3.8-7.9 - 3.6-10.1 SL 2+1 2+1 2+1 2+1 IL 4-5 4 4 4 SO present present present present Midtail scales 12 10 12 10 Color pattern 4 2 3 1 Amphib. Reptile Conserv. 109 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. Fig. 2. Dorsal (A) and ventral (B) views of the holotype of Tri- lepida pastusa sp. nov. (QCAZ 8690) in preservative. Scale bar = 1 cm. midtail (see Discussion), but T. pastusa differs from this species by having an ocular with barely rounded anterior border, thick body (TL/MB = 28.6-34.6), uniform gray dorsum and venter with gray on the center and cream on the outside of each scale, 203-214 middorsal scales, 182-193 midventral scales, 18-19 subcaudals, non-en- larged eyes, and rostral reaching ocular level in dorsal view (vs ocular with rounded anterior border, moderate to thin body width [TL/MB = 32.2-68.3], reticulate dark brown dorsally and reticulate pale brown ventrally, 211- 255 middorsal scales, 201-237 midventral scales, 16-24 subcaudals, enlarged eyes, and rostral not reaching ocu- lar level in T. macrolepis) (Fig. 4) (Pinto and Fernandes 2012; Pinto et al. 2010). Description of the holotype (Figs. 1, 2 ). — Adult female, TL 315 mm, TAL 29 mm; MB 10.2 mm; MT 7.6 mm; TL/TAL 10.9; TL/MB 30.9; TAL/MT 3.8; HL 6.2 mm, HW 6.0 nun; relative eye diameter 3.1; relative rostral width 0.4. Body subcylindrical, robust, head slightly compressed compared to body and slightly tapered cau- dally near tail. Head subcylindrical, as long as wide and slightly distinguishable from neck. Snout slightly round- ed in dorsal and ventral views, rounded in lateral view; rostral straight in frontal and ventral views, subtriangular in dorsal view but with rounded apex, reaching imagi- nary transverse line between anterior border of oculars; rostral contacting supranasal and infranasal laterally, and Amphib. Reptile Conserv. frontal dorsally; nasal completely divided horizontally by oblique suture crossing nostril and descending pos- teriorly to contact first supralabial; nostril roughly ellip- tical, obliquely oriented and located in middle of nasal suture; supranasal higher than wider, bordering rostral anteriorly, infranasal inferiorly, first and second supral- abials, and ocular posteriorly, and frontal and supraocu- lar dorsally; supranasal ventral margin half the length of upper border of infranasal scale; infranasal about twice as high as wide, longer than any of the supralabials; up- per lip border formed by rostral, infranasal, two anterior supralabials, ocular, and posterior supralabial; temporal distinct in size from dorsal scales of lateral rows; three supralabials, first two anterior to ocular and one posterior (2+1); first supralabial almost twice as high as wide, not reaching nostril and eye levels, second supralabial almost twice as high as wide, higher than first supralabial, reach- ing nostril level; third supralabial trapezoidal, as high as wide, reaching nostril level, its posterior margin in broad contact with temporal; ocular enlarged, subhexagonal, anterior border barely rounded at eye level, higher than wide, contacting posterior margins of supranasal and sec- ond supralabial anteriorly, parietal and third supralabial posteriorly, and supraocular dorsally; eye distinct (diam- eter = 0.7 nun), located in central area of upper part of ocular, displaced above nostril level; supraocular longer than wide, as long as frontal, between ocular and fron- tal, contacting supranasal anteriorly, frontal and ocular laterally, and parietal and postfrontal posteriorly; frontal, interparietal, and interoccipital subequal in size, hex- agonal and imbricate, postfrontal smaller; frontal longer than wide, contacting rostral, supranasal, supraocular, and postfrontal; postfrontal as long as wide, contacting frontal, supraocular, parietals, and interparietal; interpa- rietal as long as wide, contacting postfrontal, parietals, occipitals, and interoccipital; interoccipital wider than long, contacting interparietal, occipitals, and first dorsal scale of vertebral row; parietal and occipital subequal, ir- regularly heptagonal; parietal longer than occipital, twice as high as wide, lower margin contacting upper border of third supralabial, posterior margin contacting tempo- ral, occipital, and interparietal, anterior border in contact with ocular, supraocular, and postfrontal; occipital twice as high as wide, its lower limit attaining upper margin of third supralabial, separated from the latter by temporal; symphysial trapezoidal, anterior border slightly concave and posterior border convex except in the middle, five times wider than high; four infralabials; first infralabial twice as high as wide; second infralabial as high as wide; third infralabial twice as wide as high and not pigment- ed; fourth infralabial as high as wide. Cephalic shields with uniformly scattered sensory pits. Middorsal scales 203; midventral scales 182; scales rows around middle of body 14, reducing to 12 rows in middle of tail; cloacal shield triangular, as wide as long; subcaudals 19; fused caudals present; terminal spine conical, slightly longer than wide; elongated cloacal sacs present. Dorsal scales 110 March 2015 I Volume 8 | Number 1 | e89 A new species of threadsnake from northern Ecuador Fig. 3. Trilepida pastusa in life. Lateral view of body (A) and head (B) of holotype (QCAZ 8690) and lateral (C) and ventral (D) views of body of juvenile paratype (QCAZ 5846). Photographs by O. Torres -Carvajal and S.R. Ron. homogeneous, cycloid, smooth, imbricate, and wider than long. Coloration in preservative of the holotype (Fig. 2 ). — Middorsal scales (i.e., seven longitudinal rows) bluish gray. The remaining seven scale rows forming the ventral and lateral sides of the body are occupied on the center by the same bluish gray color, but the margins of each scale are cream white; the latter pattern is less apparent on the anterior fourth of the body. Border of mouth, men- tal scale, nostrils and eyes are cream. Cloacal shield blu- ish gray except on its posterior margin, which is cream with bluish-gray dots. Color variation. — Dorsal ground color of body similar to that of the holotype in one of the juveniles (QCAZ 5846), the other juvenile (QCAZ 5778) is dark gray; ventral col- oration is similar in all specimens. and they turned Pale Neutral Gray (Color 296) (Fig. 3D). Tongue Smoky White (Color 261). Quantitative variation. — Scale counts in Trilepida pas- tusa vary as follows: middorsal scales 203-214 (x = 206.67 + 6.35, n = 3); midventral scales 182-193 (x = 186 + 6.08, n = 3); subcaudals 18-19 (x = 18.67 + 0.58, n = 3); TL 315 mm (n = 1) in adult and 123-136 nun (x = 129.5 mm + 9.19, n = 2) 'm juveniles; TAL 29 mm (n = 1) in adult and 9-12 mm (x = 10.5 mm + 2.12, n = 2) in juveniles; TL/MB ratio 30.94 (n = 1) in adult and 28.63-34.55 (x = 31.59 + 4.19, n = 2) in juveniles; TAL/ MT ratio 3.81 (n = 1) in adult and 2.67-3.41 (x = 3.04 + 0.52, n = 2) in juveniles; infralabials 4 (n = 1) in adult and 5 (n = 2) in juveniles; relative eye diameter 3.07 (n = 1) in adult and 1.70-1.73 (x = 1.71 + 0.02, n = 2) in juveniles; relative rostral width 0.37 (n = 1) in adult and 0.31-0.38 (x = 0.35 + 0.05, n = 2) in juveniles. Color in life. — Dorsum uniform Dark Blue Gray (Color 194) (Figs. 3 A, 3C, 4A), with upper part of head Brick Red (Color 36) on both juveniles (Figs. 3C, 4A). Venter of body and tail Dark Blue Gray, with interspaces be- tween scales Cream White (Color 52) (Figs. 3B, 3D, 4A). Anal plate entirely Dark Blue Gray. After a few minutes of handling, the borders of each scale became apparent Skull (Fig. 5). — Premaxilla roughly rectangular in frontal and ventral views, edentulous, pierced by six foramina; transverse process of premaxilla absent and vomerian process double; nasals paired, approximately quad- rangular dorsally, and pierced by a pair of foramina in lateral border of contact with prefrontals; nasal septum descending as medial vertical flanges; prefrontals paired, Amphib. Reptile Conserv. 111 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. Fig. 4. Lateral (A) view of body of juvenile paratype of Trilepida pastusa (QCAZ 5778). Lateral (B), dorsal (C), and ventral (D) views of body of T. macrolepis (DHMECN 11400). Dorsal view of head of the holotype of T. pastusa (QCAZ 8690) (E) and T. macrolepis (DHMECN 11400) (E). Photographs by L.A. Coloma, O. Torres-Carvajal, and S.R. Ron. Amphib. Reptile Conserv. 112 March 2015 I Volume 8 | Number 1 | e89 A new species of threadsnake from northern Ecuador subtriangular in dorsal view; septomaxillae paired, com- plex in shape, expanding dorsally into the naris; conchal invagination absent; ascending process of premaxilla pierced by single large foramen; vomers paired, located midventral to vomeronasal cupola, bearing transver- sal arms, and with short posterior arms; frontals paired, nearly rectangular dorsally, without anterolateral projec- tions to attach to prefrontals; frontal pillars absent; optic nerve restricted to lateral descending surface of frontals; maxilla edentulous, irregular in shape, pierced by three large juxtaposed foramina; posterior orbital element ab- sent; parietal single, wide, representing largest bone of braincase; parabasisphenoid arrow-like, with tapered anterior tip lying bellow palatine, and fitting in medial line of vomeronasal cupola; parabasisphenoid with shal- low pituitary fossa; basioccipital single and pentagonal; supraoccipitals fused into single unit, subpentagonal; prootics paired and triangular; large statolythic mass present in cavum vestibuli; crista tubelaris not enclos- ing juxtastapedial recess; stapedial footplate apparently not co-ossified with prootic; otoccipitals paired and rect- angular; palatines paired and triradiate; anterior margin of maxillar process slightly flexing ventrally; pterygoids slender and rod-like, not contacting quadrate posteriorly, and not extending beyond the anterior margin of basioc- cipital; ectopterygoid indistinct; quadrate long and slen- der, about 50% of skull length; dentary supports a series of five teeth ankylosed to the inner surface of the antero- lateral margin of dental concha; mental foramen nearly under the teeth; splenial conical, representing smallest bone in lower jaw; anterior mylohyoid foramen absent on splenial; posterior milohyoid foramen on the ventral surface of angular; angular conical; compound bone pierced by two foramina in the surangular lamina, posterior surangular foramen large and anterior to the articular portion of compound bone, anterior surangular foramen located below the coronoid; coronoid rests on the compound bone. Postcranial data. — Precloacal vertebrae 176-184 (x = 179 + 4.16, n = 3); cervical vertebrae 6 -i- trunk vertebrae 170 {n = 1); caudal vertebrae 23-24 (x = 23.3 + 0.6, n = 3), the last vertebrae representing the fusion of three vertebrae {n = 2). Correlation (n = 3) between middor- sal scales and precloacal -i- subcaudal vertebrae (1:0.99), between midventral scales and precloacal vertebrae (1:1.02), and between subcaudal scales and caudal verte- brae (0.8:1). Pelvic girdle located at the level of the 17P' and 176* precloacal vertebrae (QCAZ 5778), or 176* precloacal and 2"‘^ subcaudal (QCAZ 8690). Pelvic girdle represented by four non-fused bones: ilium, ischium, fe- mur, and pubis. Ilium, ischium, and femur rod-like; ilium represents the longest bone of pelvic girdle; femur stout. Etymology. — The specific epithet is used as a noun in apposition. As explained in Coloma et al. (2010), pas- tuso or pastusa is a Spanish word used to refer to the inhabitants of the Pasto region in northern Ecuador and southern Colombia. Here, we also use it to recognize the presence of Los Pastos pre-hispanic culture (500-1500 CE) (Delgado-Troya 2004), whose vestiges remain in the type locality and allowed the discovery of Trilepida pas- tusa and specimens of another cryptozoic snake species: Atractus savagei (Salazar- Valenzuela et al. 2014). Proposed standard English and Spanish names. — Pas- tuso threadsnakes; Serpientes hebra pastusas. Distribution and natural history (Pigs. 6, 7). — Western versant of the Cordillera Occidental of extreme northern Ecuadorian Andes in Tulcan County, Carchi province. Trilepida pastusa is known only from the type local- ity, which belongs to Montane Cloud Porest (Valencia et al. 1999) at 2,071 m. The holotype was found below mounds of dirt, stones, pasture and moss in conjunction with eggs of Liophis vitti (nomenclature following rec- onnnendations expressed in Curcio et al. 2009) and adult specimens of Atractus savagei (Salazar- Valenzuela et al. 2014). Juvenile paratypes were found below rocks (ca. 40 cm in diameter) in areas of pasture. First records of Trilepida macrolepis for Ecuador. — The big-scaled threadsnake, T macrolepis, is a relatively large (126-322 mm) leptotyphlopid snake with the wid- est geographical distribution of all species in the genus (Pinto et al. 2010). Localities for the species in north- ern South America include Panama, Colombia, Ven- ezuela, Guyana, Suriname, Prench Guiana, Brazil, and Peru (Wallach et al. 2014). It is distinguished from con- geners by having three supralabial and four infralabial scales, 10 rows in the middle of the tail, more than 210 middorsal scales, and more conspicuously because of a dorsal and ventral coloration pattern consisting of dark brown to black scales with white borders (i.e., reticulate) (Passos et al. 2005; Pinto and Pemandes 2012; Pinto et al. 2010). Two specimens from Esmeraldas province in northern Ecuador agree with most of these charac- ters, their scale counts vary as follows (QCAZ 10247, juvenile of undetermined sex and DHMECN 11400, adult male, respectively): middorsal scales 239, 250; midventral scales 220, 228; subcaudals 21, 20; TL 158 mm, 333 mm; TAP 11 mm, 23 mm; SL 2-1-1, 2-1-1; IP 4, 6; SO 1, 1; midtail scales 12, 10. Coloration pattern on both is reticulate dark brown dorsally and reticulate pale brown ventrally (Pigs. 4B^D). Specimen QCAZ 10247 was collected on 13 Pebruary 2010 in Otokiki Reserve, Alto Tambo (0°54’21.6” N, 78°36’21.6” W, 620 m), San Lorenzo County, Esmeraldas province; the snake was found in primary forest 30 cm below ground among fern roots. Specimen DHMECN 11400 was collected on 01 April 2012 near Durango (1°02’30.7” N, 78°37’26.6” W, 243 m), San Lorenzo County, Esmeraldas province; the snake was found in secondary forest one m above ground among leaf litter accumulated on the junction of Amphib. Reptile Conserv. 113 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. Fig. 5. Three-dimensional reconstruction of the skull of Trilepida pastusa based on HRXCT data. Dorsal (A), ventral (B), lateral (C), anterior (D), and posterior (E) views of juvenile paratype (QCAZ 5778). Scale bar = 3.5 mm. Bo, basioccipital; CB, compound bone; De, dentary; Fr, frontal; Ma, maxilla; Na, nasal; Ot, otico-occipital; Pa, parietal; Pal, palatine; Pbs, parabasisphenoid; Pf, prefrontal; Pm, premaxilla; Pr, prootic; Pt, pterygoid; Qd, quadrate; Sm, septomaxilla; So, supraoccipital; Vo, vomer. Amphib. Reptile Conserv. 114 March 2015 I Volume 8 | Number 1 | e89 A new species of threadsnake from northern Ecuador lianas. Both localities belong to Lowland and Foothill Evergreen Forests of northwestern Ecuador (Ceron et al. 1999) (Fig. 6). Discussion The conservation assessment of reptile species belong- ing to families that are completely or primarily fossorial (e.g., Amphisbaenidae, Anomalepididae, Leptotyphlo- pidae, Typhlopidae, Uropeltidae) is incomplete (Santos 2013). Due to their secretive habits and non-inclusion during routine herpetological surveys, knowledge about their distribution and population dynamics is scarce (Measey 2006; Pyron and Wallach 2014). Bohm et al. (2013) estimated that 10.5% (range: 5.6-57%) of species of fossorial reptiles are Threatened; however, the authors recognized that this low estimate and wide confidence intervals reflect the fact that a large number (47% from a subsample of 1,500 reptile species) of the included taxa had been classified as Data Deficient. Therefore, this study re-emphasized the need to target these groups in future research and surveys. Records for Trilepida macrolepis and T. pastusa pro- vided here come from a region where several new spe- cies of snakes have been discovered in the last 15 years (e.g., Dixon 2000; Passos et al. 2009; Salazar- Valenzuela 120 ECUADOR Pacmc Ocean PERU ALTITUDE (m) I I fl.HM m Ml - t.OQO 1 . 1 H 1 - 1 ^ V»1 -3.000 2.001 - 2 ^500 ^.SOl - 3 j000 >^.000 rii' w Fig. 6. Geographic distribution of Trilepida pastusa (circle) and T. macrolepis (triangles) in Ecuador. Fig. 7. Habitat of Trilepida pastusa surrounding the cloud forests near the town of Chilma Bajo, Tulcan County, Carchi province (A); pasture areas where individuals were collected (B); microhabitat of T pastusa (C); and a juvenile individual of the new species in situ (arrow) (D). Photographs by D. Salazar-Valenzuela. Amphib. Reptile Conserv. 115 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. et al. 2014; Torres-Carvajal et al. 2012). The singularity of this region may be attributable to the proposed exis- tence of a habitat transition between northern and cen- tral parts of the Choco bioregion (western Colombia and northwestern Ecuador) and communities found further south along the Pacific coast and adjacent Andean slopes of Ecuador (Anderson and Jarrin-V 2002; Anderson and Martmez-Meyer 2004; Cisneros-Heredia 2006; Salazar- Valenzuela et al. 2014). Trilepida macrolepis has been suggested to represent a complex of species due to its wide distribution, pres- ence on both sides of the Andes, and deep molecular divergence between individuals from two localities in northern Brazil and Guyana (Adalsteinsson et al. 2009; Orejas-Miranda 1967). Our data from Ecuador agree with the diagnosis provided for this species in Pinto et al. (2010), except for the presence of 12 midtail scales in specimen QCAZ 10247. This character has been used in the taxonomy of members of the genus (Orejas-Miranda and Peters 1970; Pinto et al. 2010) and will need to be evaluated in future studies. We tentatively assign QCAZ 10247 to T. macrolepis based mainly on the presence of a high number of middorsal and midventral scales (239 and 220, respectively), an ocular with rounded anterior border, and a reticulate dorsal and ventral coloration pattern, but acknowledge that a detailed revision of this taxon is needed to confirm the taxonomic identity of this specimen. Perez-Santos and Moreno (1991) showed two color photographs of leptotyphlopid snakes of Ecuador without species identification. One of them (picture 148) seems to agree with the dorsal reticulate pattern present in T. macrolepis, suggesting that this taxon was already col- lected in Ecaudor. However, the same image (picture 85) was also included in Perez-Santos and Moreno (1988)’s book on snakes of Colombia rendering questionable the origin of that specimen. The discovery of individual DH- MECN 11400 among leaf litter one m above the forest floor is in agreement with the report of individuals of this species complex from the Amazonian lowlands wrapped two m above ground around small tree trunks and mov- ing their heads back and forth (Vitt and Caldwell 2013). These authors suggested that the snakes were probably trying to detect airborne chemical cues associated with termite nests. Putative synapomorphies for the genus Trilepida in- clude a hemipenis body with a narrow base and a robust terminal portion, middorsal cephalic scales of moderate size (i.e., supraocular scales smaller or equal to frontal and postfrontal scales), and an enlarged terminal spine (Passos et al. 2006; Pinto and Curcio 2011; Pinto and Fernandes 2012). Although we could not examine hemi- penis for T. pastusa since adult males were not avail- able in our sample, we assign this species to the genus Trilepida based on the presence of middorsal cephalic scales of moderate size and a slightly enlarged terminal spine (see Pinto and Curcio, 2011). Also, the paired or unpaired condition of the nasal bone is variable in differ- Amphib. Reptile Conserv. ent members along the Leptotyphlopidae family (Rieppel et al. 2009). Although osteological characters have not been employed in less inclusive phylogenetic analysis on the Renina subtribe (represented by the genus Rena and Trilepida), previous morphological studies on the skull of members pertaining to this subtribe (e.g.. Brock 1932; List 1966; McDowell and Bogert 1954; Rieppel et al. 2009) indicate that the paired condition of the nasal bones may be a feature that could distinguish the genus Trilepida (paired condition) from the genus Rena (fused condition). As many other species currently allocated in the genus Trilepida, T pastusa also has paired nasal bones. Intrageneric phylogenetic relationships have not been established for members of the genus Trilepida', out of the 14 species of the genus, three consistently show the presence of 12 midtail scales: T guayaquilensis, T joshuai, and T pastusa. These species are restricted to Colombia and Ecuador and these shared features may indicate close phylogenetic relationships between them. Trilepida guayaquilensis is still only known from the ho- lotype even though it was described 45 years ago from a specimen collected in Guayaquil, Guayas province, Ec- uador. We analyzed photographs from specimens collect- ed in this locality and misidentified as T guayaquilensis. These specimens actually represent Epictia subcrotilla, which is a leptotyphlopid snake distributed in the low- lands of Ecuador and Peru, relatively common in natural history collections (Cisneros-Heredia 2008; Purtschert 2007). Indeed, specimen QCAZ 12769 collected by us in Guayaquil confirms the presence of this species in ur- ban areas of this city. Trilepida guayaquilensis therefore remains to be confirmed as a species with a distribution that includes Guayaquil but no further records exist from this area because it is either extremely rare in abundance or searching efforts have not been enough to locate this fossorial animal. Alternatively, T. guayaquilensis could be a species whose only known specimen did not origi- nate from Guayaquil but may have been brought there from a nearby locality in the Pacific lowlands or the western versant of the Andes, as has been suggested for a couple of other species of Ecuadorian snakes (Cadle 2005; Curcio et al. 2012). As is common in tropical parts of the world where these groups have significantly diversified (Pyron and Burbrink 2012; Vitt and Caldwell 2013), knowledge about the diversity of fossorial snake fauna in Ecuador is still fragmented. Cisneros-Heredia (2008) clarified much of the confusion present at the time regarding the exis- tence of voucher specimens for species registered in the country and concluded that only three species of the fam- ily Leptotyphlopidae were present in the country. The ad- dition of two species of threadsnakes to the snake fauna of Ecuador reported in this study should bring us closer to the true diversity present in the country, even though it is clear that there are more species that remain to be described in this group (Cisneros-Heredia 2008, pers. observ.). 116 March 2015 I Volume 8 | Number 1 | e89 A new species of threadsnake from northern Ecuador Acknowledgments. — We thank the following cu- rators and their staff for allowing us to examine speci- mens or photographs of specimens under their care: K. de Queiroz and J. Poindexter (USNM) and M. Yanez- Munoz (DHMECN). We are grateful to E. Levy, A. Pozo, and the Chilma Bajo connnunity for assistance with lo- gistics in the field. S. Aldas- Alarcon, G. Buitron, A. Cha- lapud, A. Loaiza-Lange, D. Males, P. Piedrahita, H. Pozo, P. Santacruz-Ortega, and E. Tapia provided assistance in the field. We thank L.A. Coloma for the use of his photo- graphs of Trilepida macrolepis, R. Pinto for kindly shar- ing photographs of the holotype of T. guayaquilensis, D. Paucar-Guerrero and S.R. Ron for illustrations and pho- tographs of T. pastusa, S. Lobos for improving the dis- tribution map, and A. Varela for help with figure edition. Reviews by P. Passes and an anonymous reviewer sub- stantially improved the manuscript. Specimens were col- lected under collection permit 008-09 IC-EAU-DNB/MA and were deposited at Museo de Zoologfa (QCAZ), Pon- tificia Universidad Catolica del Ecuador. OTC received support from Secretarfa de Educacion Superior, Ciencia y Tecnologfa del Ecuador (SENESCYT), project PIC-08- 0000470. Einancial support for AM was provided by the Coordena^ao de Aperfeigoamento de Pessoal de Nfvel Superior (CAPES), Conselho Nacional de Desenvolvim- ento Cientifico e Tecnologico (CNPq), and the Fundagao de Amparo a Pesquisa do Rio de Janeiro (EAPERJ). Literature Cited Adalsteinsson SA, Branch WR, Trape S, Vitt LJ, Hedges SB. 2009. Molecular phylogeny, classification, and biogeography of snakes of the Family Leptotyphlopi- dae (Reptilia, Squamata). Zootaxa 2244: 1-50. Anderson RP, Jarrm-V P. 2002. A new species of spiny pocket mouse (Heteromyidae: Heteromys) endemic to western Ecuador. American Museum Novitates 3382: 1-26. Anderson RP, Martmez-Meyer E. 2004. Modeling spe- cies’ geographic distributions for preliminary conser- vation assessments: An implementation with the spiny pocket mice {Heteromys) of Ecuador. Biological Con- servation 116: 167-179. Bohm M, Collen B, Baillie JEM, Bowles P, Chanson J, Cox N, Hammerson G, Hoffmann M et al. 2013. The conservation status of the world’s reptiles. Biological Conservation 157: 372-385. Broadley DG, Wallach V. 2007. A revision of the genus Leptotyphlops in northeastern Africa and southwest- ern Arabia (Serpentes: Leptotyphlopidae). Zootaxa 1408: 1-78. Brock GT. 1932. The skull of Leptotyphlops. Anato- mischer Anzeizerl3: 199-204. Cadle J. 2005. Systematics of snakes of the Dipsas oreas complex (Colubridade: Dipsadinae) in western Ecua- dor and Peru, with revalidation of D. elegans (Bou- lenger) and D. ellipsifera (Boulenger). Bulletin Mu- Amphib. Reptile Conserv. scum of Comparative Zoology 158: 67-136. Ceron C, Palacios W, Valencia R, Sierra R. 1999. Las formaciones naturales de la Costa del Ecuador. Pp. 55-78 In: Propuesta Preliminar de un Sistema de Cla- sificacion de Vegetacion para el Ecuador Continen- tal. Editor, R Sierra. Proyecto INEFAN/GERF-BIRF y Ecociencia, Ecuador. 174 p. Cisneros-Heredia DE. 2006. Distribution and ecology of the western Ecuador frog Leptodactylus labrosus (Amphibia: Anura: Leptodactylidae). Zoological Re- search 27: 225-234. Cisneros-Heredia DE. 2008. Reptilia, Squamata, Lepto- typhlopidae, Leptotyphlops, Ecuador: Re-evaluation of the species cited for the country. Check List 4: 178-181. Coloma LA, Duellman WE, Almendariz CA, Ron SR, Teran- Valdez A, Guayasamm JM. 2010. Five new (extinct?) species of Atelopus (Anura: Bufonidae) from Andean Colombia, Ecuador, and Peru. Zootaxa 2574: 1-54. Curcio FF, Nunes PMS, Argolo AJS, Skuk G, Rodrigues MT. 2012. Taxonomy of the South American dwarf boas of the genus Tropidophis Bibron, 1840, with de- scription of two new species from the Atlantic Forest. Herpetological Monographs 26: 80-121. Curcio FF, Piacentini VdQ, Fernandes DS. 2009. On the status of the snake genera Erythrolamprus Boie, Lio- phis Wagler and Lygophis Fitzinger (Serpentes, Xen- odontinae). Zootaxa 2173: 66-68. Curcio FF, Zaher H, Rodrigues MT. 2002. Rediscovery of the blind snake Leptotyphlops brasiliensis Laurent, 1949 (Serpentes, Leptotyphlopidae) in the wild. Phyl- lomedusa 1: 101-104. de Queiroz K. 1998. The general lineage concept of spe- cies, species criteria, and the process of speciation: A conceptual unification and terminological recom- mendations. Pp. 57-75 In: Endless Eroms: Species and speciation. Editors, DJ Howard, SH Berlocher. Oxford University Press, Oxford, United Kingdom. 496 p. de Queiroz K. 2007. Species concepts and species de- limitation. Systematic Biology 56: 879-886. Delgado-Troya JM. 2004. Cronica de los Pastos. Edicio- nes, Abya-Yala, Quito, Ecuador. 168 p. Dixon JR. 2000. Ecuadorian, Peruvian, and Bolivian snakes of the Liophis taeniurus complex with descrip- tions of two new species. Copeia 2000: 482^90. Hedges SB. 2011. The type species of the threadsnake genus Tricheilostoma Jan revisited (Squamata, Lepto- typhlopidae). Zootaxa 3027: 63-64. Holman JA. 2000. The Fossil Snakes of North America: Origin, evolution, distribution, paleoecology. Indiana University Press, Bloomington, Indiana, USA. 376 p. Kley NJ. 2006. Morphology of the lower jaw and suspen- sorium in the Texas blindsnake, Leptotyphlops dulcis (Scolecophidia: Leptotyphlopidae). Journal of Mor- phology 267: 494-515. March 2015 I Volume 8 | Number 1 | e89 117 Salazar-Valenzuela et al. Kohler G. 2012. Color Catalogue for Field Biologists. Herpeton, Offenbach, Germany. 49 p. Kroll JC, Reno HW. 1971. Are-examination of the cloa- cal sacs and gland of the blind snake, Leptotyphlops dulcis (Reptilia: Leptotyphlopidae). Journal of Mor- phology 133: 273-280. List JC. 1966. Comparative osteology of the snake fami- lies Typhlopidae and Leptotyphlopidae. Illinois Bio- logical Monographs 36: 1-112. McDiarmid RW, Campbell JA, Toure T. 1999. Snake Species of the World: A taxonomic and geographic reference. Volume 1. The Herpetologists’ League, Washington D.C., USA. Slip. McDowell SB, Bogert CM. 1954. The systematic posi- tion of Lanthanotus and the affinities of the anguino- morphan lizards. Bulletin of the American Museum of Natural History 105: 1-142. Measey GJ. 2006. Surveying biodiversity of soil herpeto- fauna: Towards a standard quantitative methodology. European Journal of Soil Biology 42: S103-S110. Orejas-Miranda BR. 1967. El genero ''Leptotyphlops'" en la region Amazonica. Atas do Simposio sobre a Biota Amazonica 5: 421-442. Orejas-Miranda BR, Peters G. 1970. Eine neue Schlank- blindschlange (Serpentes: Leptotyphlopidae) aus Ec- uador. Mitteilungen aus dem Zoologischen Museum in Berlin 46: 439^41. Passos P, Caramaschi U, Pinto RR. 2005. Rediscovery and redescription of Leptotyphlops salgueiroi Amaral, 1954 (Squamata, Serpentes, Leptotyphlopidae). Bo- letim do Museu Nacional, Nova Serie Zoologia 520: 1 - 10 . Passos P, Caramaschi U, Pinto RR. 2006. Redescription of Leptotyphlops koppesi Amaral, 1954, and descrip- tion of a new species of the Leptotyphlops dulcis group from Central Brazil (Serpentes: Leptotyphlopi- dae). 27: 347-357. Passos P, Mueses-Cisneros JJ, Lynch ID, Eemandes R. 2009. Pacific lowland snakes of the genus Atractus (Serpentes: Dipsadidae), with description of three new species. Zootaxa 2293: 1-34. Perez-Santos C, Moreno AG. 1988. Ofidios de Colombia [Snakes of Colombia]. Museo Regionale di Scienze Naturali, Torino, Italy. 517 p. Perez-Santos C, Moreno AG. 1991. Serpientes de Ecua- dor [Snakes of Ecuador]. Museo Regionale di Scienze Naturali, Torino, Italy. 538 p. Peters WCH. 1857. Vier neue amerikanische Schlangen aus der Eamilie der Typhlopinen vor und machte dar fiber einige vorlausige Mittheilungen. Monatsberichte der koniglich Akademie der Wissenschaften zu Berlin 1857: 402^03. Pinto RR, Passos P, Caicedo-Portilla JR, Arredondo JC, Eemandes R. 2010. Taxonomy of the threadsnakes of the tribe Epictini (Squamata: Serpentes: Leptotyphlo- pidae) in Colombia. Zootaxa 2724: 1-28. Pinto RR, Curcio EE. 2011. On the generic identity of Siagonodon brasiliensis, with the description of a new leptotyphlopid from central Brazil (Serpentes: Lepto- typhlopidae). Copeia 2011: 53-63. Pinto RR, Eemandes R. 2012. A new blind snake species of the genus Tricheilostoma from Espinhago Range, Brazil and taxonomic status of Rena dimidiata (Jan, 1861) (Serpentes: Epictinae: Leptotyphlopidae). Co- peia 2012: 37^8. Purtschert B. 2007. Taxononua y biogeografia de Lep- totyphlops subcrotillus (Serpentes: Leptotyphlopi- dae) en Ecuador. B.Sc. Thesis, Colegio de Ciencias Biologicas y Ambientales, Universidad San Erancisco de Quito, Quito, Ecuador. Pyron RA, Burbrink ET. 2012. Extinction, ecological op- portunity, and the origins of global snake diversity. Evolution 66: 163-178. Pyron RA, Wallach V. 2014. Systematics of the blind- snakes (Serpentes: Scolecophidia: Typhlopoidea) based on molecular and morphological evidence. Zoo- taxa 3829: 1-81. Rieppel O, Kley NJ, Maisano JA. 2009. Morphology of the skull of the white-nosed blindsnake, Liotyphlops albirostris (Scolecophidia: Anomalepididae). Journal of Morphology 270: 536-557. Rojas-Morales JA, Gonzalez-Duran GA. 2011. Descrip- tion of the colouration in life of Tricheilostoma josh- uai (Serpentes, Leptotyphlopidae). A species tolerant of disturbed habitats? Salamandra 47: 237-240. Salazar-Valenzuela D, Carrillo EO, Aldas A S. 2010. Geographic distribution: Tricheilostoma anthraci- num. Herpetological Review 41: 111-112. Salazar-Valenzuela D, Torres-Carvajal O, Passos P. 2014. A new species of Atractus (Serpentes: Didap- sidae) from the Andes of Ecuador. Herpetologica 70: 350-363. Santos LCD. 2013. Discovery of the second specimen of the African amphisbaenian Monopeltis kabindae Witte & Laurent, 1942. African Journal of Herpetol- ogy 2013: 1-8. Torres-Carvajal O, Salazar-Valenzuela D, Merino- Vit- eri A. 2014. ReptiliaWebEcuador. Version 2014.0. Museo de Zoologia. Pontificia Universidad Catolica del Ecuador. Available: http://zoologia.puce.edu.ee/ Vertebrados/reptiles/ReptilesEcuador/ [Accessed: 17 December 2014]. Torres-Carvajal O, Yanez-Munoz MH, Quirola D, Smith E.N., Almendariz A. 2012. A new species of blunt- headed vine snake (Colubridae, Imantodes) from the Choco region of Ecuador. ZooKeys 244: 91-110. Uetz P, Hosek J (Editors). 2014. The Reptile Database. Available: http://www.reptile-database.org [Ac- cessed: 21 December 2014]. Valencia R, Ceron C, Palacios W, Sierra R. 1999. Las formaciones naturales de la Sierra del Ecuador. Pp. 79-108 In: Propuesta Preliminar de un Sistema de Clasificacidn de Vegetacion para el Ecuador Conti- nental. Editor, R Sierra. Proyecto INEEAN/GERE- Amphib. Reptile Conserv. 118 March 2015 | Volume 8 | Number 1 | e89 A new species of threadsnake from northern Ecuador BIRF y Ecociencia, Ecuador. 174 p. World: A catalogue of living and extinct species. CRC Vitt LJ, Caldwell JR 2013. Herpetology: An introductory Press, Boca Raton, Elorida, USA. 1,237 p. biology of amphibians and reptiles. 4th edition. Else- vier Inc., London, United Kingdom. 776 p. Wallach V, Williams KL, Boundy J. 2014. Snakes of the APPENDIX Specimens examined Epictia subcrotilla (n = 4). — ECUADOR: Provincia Guayas: Guayaquil, 7 m, QCAZ 12769; USNM 232401-03 (photographs examined). Trilepida anthracina (n = 1). — ECUADOR: Provincia Zamora Chinchipe: 6.5 km SE of Tundayme, 1,300-1,500 m, QCAZ 7396. Trilepida guayaquilensis {n = 1). — ECUADOR: Provincia Guayas: Guayaquil, ZMB 4508 (holotype) (photo- graphs examined). Trilepida macrolepis (n = 2). — ECUADOR: Provincia Esmeraldas: Alto Tambo, Otokiki Reserve, 620 m, QCAZ 10247; Durango, 243 m, DHMECN 11400. David Salazar- Valenzuela graduated in Biological Sciences from Pontificia Universidad Catdlica del Ec- uador (PUCE) in 2007. He is currently a researcher of the Museo de Zoologia QCAZ of PUCE and a Ph.D. candidate in the Department of Evolution, Ecology and Organismal Biology at The Ohio State University. His doctoral dissertation is focused on systematics, historical demography, and venom variation of the Bothrops asper species complex using genomic and proteomic approaches. So far David has published six scientific papers on taxonomy, ecology, and toxinology of Ecuadorian amphibians and reptiles. Angele Martins received ber Master’s degree in 2012 from Museu Nacional do Rio de Janeiro/UFRJ- Brazil, and is now a Pb.D. student in Zoology at this same institution. She has dedicated her research efforts in the last six years to the study of snake morphology and herpetofauna from the Atlantic Forest in Brazil. In the last three years, she has focused on the study of the comparative anatomy of scolecophidians, with significant interest on threadsnakes (Eeptotyphlopidae), which lead her to investigate the comparative anatomy of this group for her Ph.D. thesis. Luis Amador-Oyola graduated in Biological Sciences from the University of Guayaquil (UG) in 2005. He is currently completing his Master’s thesis on the biogeography and systematics of amphibians from the Chonghn Colonche mountains (equatorial pacific) at the same university. His work has focused on the distribution and biodiversity of amphibians and reptiles of areas on the coast of Ecuador, however Luis is interested in future work on evolution and biogeography of amphibians. This manuscript represents the first description of a species of herpetofauna coauthored by Luis; other works are in preparation. Omar Torres-Carvajal graduated in Biological Sciences from Pontificia Universidad Cathlica del Ecua- dor (PUCE) in 1998, and in 2001 received a Master’s degree in Ecology and Evolutionary Biology from the University of Kansas under the supervision of Dr. Linda Trueb. In 2005 he received a Ph.D. degree from the same institution with the thesis entitled “Phylogenetic systematics of South American lizards of the genus Stenocercus (Squamata: Iguania).” Between 2006-2008 he was a postdoctoral fellow at the Smithsonian Institution, National Museum of Natural History, Washington DC, USA, working under the supervision of Dr. Kevin de Queiroz. He is currently Curator of Reptiles at the Zoology Museum QCAZ of PUCE and an Associate Professor at the Department of Biology in the same institution. He has published more than 30 scientific papers on taxonomy, systematics, and biogeography of South American reptiles, with emphasis on lizards. He is mainly interested in the theory and practice of phylogenetic systematics, particularly as they relate to the evolutionary biology of lizards. Amphib. Reptile Conserv. 119 March 2015 I Volume 8 | Number 1 | e89 Salazar-Valenzuela et al. In accordance with the International Code of Zoological Nomenclature new rules and regulations (ICZN 2012), we have deposited this paper in publicly accessible institutional libraries. The new species described herein has been registered in ZooBank (Polaszek 2005a, b), the official online registration system for the ICZN. The ZooBank publication LSID (Life Science Identifier) for the new species described here can be viewed through any standard web browser by appending the LSID to the prefix “http://zoobank.org/”. The LSID for this publication is: urn:lsid:zoobank.org:pub:3FC7DC45-E3D4-49B6-AEDD-3925A347665E. Separate print-only edition of paper(s) (reprint) are available upon request as a print-on-demand service. Please inquire by sending a request to: Amphibian & Reptile Conservation (amphibian-reptile-conservation.org; arc.publisher@gmail.com). Amphibian & Reptile Conservation is a Content Partner with the Encyclopedia of Life (EOL); http:///www.eol.org/ and submits information about new species to the EOL freely. Digital archiving of this paper are found at the following institutions: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Ernst Mayr Library, Museum of Comparative Zool- ogy, Harvard University, Cambridge, Massachusetts (USA); Florida Museum of Natural History, Gainesville, Florida (USA). Complete journal archiving is found at: ZenScientist (http://www.zenscientist.com/index.php/filedrawer); Florida Museum of Natural History, Gainesville, Florida (USA). Citations ICZN. 2012. Amendment of Articles 8,9,10,21 and 78 of the International Code of Zoological Nomenclature to expand and refine methods of publication. Zootaxa 3450: 1-7. Polaszek A et al. 2005a. Commentary: A universal register for animal names. Nature 437: 477. Polaszek A et al. 2005b. ZooBank: The open-access register for zoological taxonomy: Technical Discussion Paper. Bulletin of Zoological Nomenclature 62(4): 210-220. Amphib. Reptile Conserv. 120 March 2015 I Volume 8 Number 1 e89 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 121-135 (e90). Development and gastrulation in Hyloxalus vertebral is and Dendrobates auratus (Anura: Dendrobatidae) Francisca Hervas, Karina P. Torres, Paola Montenegro-Larrea, and ^Eugenia M. del Pino Escuela de Ciencias Bioldgicas, Pontificia Universidad Catolica del Ecuador, Av. 12 de Octubre 1076 y Roca, Quito 170517, ECUADOR Abstract. — ^We document the embryonic development of Hyloxalus vertebralis, a frog species of the Ecuadorian highlands, declared as Critically Endangered by the International Union for the Conservation of Nature (lUCN) due to significant declines in its populations. Our work may be of value for conservation and management of this endangered frog, especially as it is being bred in captivity to ensure against extinction. We were able to analyze and compare the development of H. vertebralis with Dendrobates auratus (Dendrobatidae), and other frogs, because of the successful reproduction in captivity of Ecuadorian frogs at the Balsa de los Sapos, Centre of Amphibian Investigation and Conservation (CICA), of the Pontificia Universidad Catolica del Ecuador, in Quito. Embryos were fixed, and the external and internal morphology was described from whole mounts, and serial sections. Cellular morphology was analyzed by staining nuclei. Embryos of H. vertebralis and D. auratus developed from eggs that were 2.6 and 3.5 mm in diameter, respectively. In spite of the large size of their eggs, the morphology of H. vertebralis embryos from cleavage to hatching was similar to the morphology of Epipedobates machalilla (Dendrobatidae) embryos. The comparison of gastrulation morphology was extended to six additional species of Dendrobatidae (E. machalilla, Epipedobates anthonyi, Epipedobates tricolor, H. vertebralis, Ameerega bilinguis, D. auratus), and to Xenopus laevis (Pipidae), and Gastrotheca riobambae (Hemiphractidae). We found that elongation of the notochord occurs after blastopore closure in the six species of dendrobatid frogs, as in G. riobambae; whereas gastrulation and notochord elongation overlap during X. laevis development. We propose that the separation of gastrulation from notochord elongation may relate to slower development patterns, probably associated with the terrestrial reproductive strategies of dendrobatid frogs and marsupial frogs. This analysis contributes to the knowledge of frog embryology and gastrulation, and provides developmental information that may be useful for the conservation and management of H. vertebralis. Key words. Ameerega bilinguis, Epipedobates machalilla, Epipedobates anthonyi, Epipedobates tricolor, notochord, neurula Citation: Hervas F, Torres KP, Montenegro-Larrea P, del Pino EM. 2015. Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus (Anura: Dendrobatidae). Amphibian & Reptiie Conservation 8(1) [Special Section]: 121-135 (e90). Copyright: © 2015 Hervas et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptiie Conservation; official journal website . Received: 14 May 2014; Accepted: 12 November 2014; Published: 22 March 2015 Introduction We analyzed the embryonic development of Hyloxalus vertebralis and Dendrobates auratus from cleavage to tadpole hatching and compared these patterns with the development of Epipedobates machalilla (Dendrobati- dae) and Xenopus laevis (Pipidae), frogs with well-stud- ied development. These comparisons were then extended to embryos of other Ecuadorian, neotropical frogs (del Pino et al. 2004, 2007; Moya et al. 2007; Nieuwkoop and Faber 1994). Our aim was to extend the analysis of frog embryonic development to additional species and to pro- vide information that may be useful for the conservation and management of H. vertebralis, an endangered frog. Most of the H. vertebralis population has disappeared, possibly due to chytridiomycosis infection and habitat destruction. For these reasons, the International Union for Conservation of Nature (lUCN) declared H. verte- Correspondence. Email: ^edelpino@puce.edu.ec, tel: (593 2) 299 1 700 extension 1280; fax: (593 2) 299 1725. Amphib. Reptile Conserv. 121 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. Fig. 1. External views of H. vertebralis embryos from cleavage to the neurula stage. (A) Stage 2: Two-cell stage. (B) Stage 4: Eight- cell stage. (C) Stage 5: Sixteen-cell stage. (D) Stage 9: Advanced blastula. (E) Stage 10: Early gastrula. (F) Stage 11: Mid-gastrula. (G) Stage 12: Late-gastrula. (H) Stage 12.5: Late-gastrula with a small yolk plug. (I) Stage 13: Slit-blastopore stage. (J) Stage 13.5: Advanced slit-blastopore stage. The neural plate was visible. (K) Stage 14: Early neural fold stage. (L) Stage 15: Mid-neural fold stage. In this and the following figures, the developmental stage (st) is given in top right-hand comer of each image; b, blastopore; c, cleavage furrow; dl, dorsal blastopore lip; np, neural plate; nt, neural tube; yp, yolk plug. bralis as Critically Endangered (Coloma et al. 2004). It is currently being bred in captivity to guard against ex- tinction. Hyloxalus vertebralis occurs at elevations of 1,770- 3,500 m above sea level in the inter-Andean valleys of Ecuador. In addition, it occurs on the eastern and west- ern slopes of the Andes in central and southern Ecuador, respectively (Coloma 1995). Its habitat is the cloud for- est and it has also been found in ponds, open areas, and streams. The nests consist of 5-12 eggs that are placed on the ground (Coloma 1995). After the tadpoles hatch, the males transport them to streams for further development (Coloma 1995). Dendrobates auratus is distributed from southeastern Nicaragua to northwestern Colombia (Solis et al. 2004). This species does not occur in Ecuador. These frogs de- posit their eggs in terrestrial nests, and embryonic de- velopment occurs inside the egg capsules until tadpole hatches in the leaf litter. Brood care is performed by the male. After hatching, tadpoles are transported individu- ally by the male to small seasonal pools (Solis et al. 2004). Eggs of D. auratus are the largest among the den- drobatids and measure 3.5 mm in diameter (del Pino et al. 2007; Hervas and del Pino 2013). Dendrobatid frogs are of great developmental inter- est because of their great variation in egg size (Table 1), and their modified pattern of gastrulation. Notochord elongation occurs after gastrulation in E. machalilla, and Epipedobates anthonyi, as in the Marsupial frog, Gas- trotheca riobambae (Hemiphractidae); whereas, the on- set of notochord elongation is a feature of the Xenopus laevis mid-gastrula (Benftez and del Pino 2002; Keller and Shook 2004; del Pino et al. 2007; Moya et al. 2007; Montenegro-Larrea and del Pino 2011; Elinson and del Pino 2012). For this reason, we compared the gastrula- tion characteristics of Epipedobates anthonyi, Epipe- dobates tricolor, H. vertebralis, Ameerega bilinguis, and D. auratus with E. machalilla (Dendrobatidae). In a previous study, Ameerega bilinguis was identified as Epipedobates ingeri (del Pino et al. 2007). This analysis Amphib. Reptile Conserv. 122 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus H. vertebralis Fig. 2. External views of H. vertebralis embryos from closure of the neural tube to hatching. (A) Stage 16.5: Closure of the neural tube. (B) Stage 17: Tail bud stage. The brachial arches protruded on the sides of the head. (C) Stage 19: Embryo at the muscular response stage. (D) Stage 19.5: Gill buds of the two external gill pairs were visible. (E) Stage 21: Development of the external gills. There were seven branches in the first gill pair and the second gill pair was unbranched. (F) Stage 25: Embryo at hatching, hr, branchial arch; e, eye; g, gills; gb, gill bud; tf, tail fin. Table 1. Gastrulation in dendrobatid frogs in comparison withX laevis (Pipidae) and G. riobambae (Hemiphractidae). Family and Species Eggs per clutch (No. of clutches) Egg diam- eter (mm) Gastrulation time (hrs)® Archenteron elongation Onset of noto- chord elongation References^ Pipidae Xenopus laevis 1.3 5 Early gastrula'’ Mid gastrula^ 1 Dendrobatidae Epipedobates machalilla 15 (72) 1.6 65 Late gastrula“ After gastrulation® 2 Epipedobates anthonyi 18 (30) 2.0 36 Late gastrula“ After gastrulation® 3 Epipedobates tricolor 13 (34) 2.0 36 Late gastrula“ After gastrulation® 2 Hyloxalus vertebralis 13 (39) 2.6 39 Late gastrula“ After gastrulation® 4 Ameerega bilinguis 10 (04) 3.0 55 Late gastrula'* After gastrulation® 4 Dendrobates auratus 05 (42) 3.5 72 Late gastrula'* After gastrulation® 1 Hemiphractidae Gastrotheca riobambae 87 3.0 168 After gastrulation® After gastrulation® 5 ‘‘Time from stages 10-13. Embryo culture temperatures for: X. laevis 23 °C, and 18-21 °C for other frogs; '’StagelO; “Stage 12.5; ‘^Stage 12; “Stage 13; ‘^Stage 11; ^References: 1, (del Pino et al. 2007); 2, (del Pino et al. 2004); 3, (Montenegro-Larrea and del Pino 2011); 4, This work; 5, (del Pino 1996; Moya et al. 2007). Amphib. Reptile Conserv. 123 March 2015 | Volume 8 | Number 1 | e90 Hervas et al. Fig. 3. Internal morphology of the H. vertebralis and D. auratus early gastrula. Micrographs of H. vertebralis embryos are shown in A, B, E-G, and micrographs of D. auratus embryos are shown in C, D, H-J. Sections shown in E-G, I-J were stained for cell nuclei. (A) Stage 9: Sagittal section of an advanced blastula. (B) Stage 10: Sagittal section of an early gastrula. (C) Stage 10: Sagit- tal section of an early gastrula. (D) Stage 11: Sagittal section of the mid-gastrula. (E) Stage 10: Higher magnification of the dorsal blastopore groove from the embryo in B. The arrow signals a bottle cell. (F) Stage 10: The blastocoel roof of an early gastrula. It was two-three cells in thickness. (G) Stage 12.5: One cell layer in the blastocoel roof of a late-gastrula. (H) Stage 10: Higher mag- nification from the embryo in C. The arrow signals a bottle cell. (I) Stage 10: The blastocoel roof of an early gastrula of two cells in thickness. (J) Stage 11: The blastocoel roof of mid-gastrula with one-two cells in thickness, bl, blastocoel; bg, blastopore groove; dl, dorsal blastopore lip; vl, ventral blastopore lip. of gastrulation in several dendrobatids expands previous studies (del Pino et al. 2007; Montenegro-Larrea and del Pino 2011). The gastrulation pattern of these dendroba- tids is similar to the pattern of E. machalilla, with the no- tochord elongation after completion of gastrulation (del Pino et al. 2004, 2007; Moya et al. 2007). We report the features of development from cleavage to tadpole hatching of H. vertebralis and from gastrula to tadpole hatching of D. auratus. This study expands the report on the mode of myogenesis, neurulation, and internal features of embryos of these two dendrobatids (Hervas and del Pino 2013). In spite of the large size of their eggs, the external and internal morphology from cleavage until tadpole hatching of H. vertebralis, and D. auratus is similar to that of E. machalilla (del Pino et al. 2004 2007; Hervas and del Pino 2013). Moreover, myo- genesis occurs by cell interdigitation, as in embryos of other dendrobatid frogs (del Pino et al. 2007; Hervas and del Pino 2013). Materials and Methods Collection sites Adults of Hyloxalus vertebralis were collected by Per- nando Duenas and Italo Tapia on 10 September 2008. The locality of collection was Azuay Province, Sevilla de Oro, in southern Ecuador at an altitude 2,418 m above sea level. The geographic coordinates of this site are W 78.60097, S 2.63605. The permit 016-IC-FAU-DNBAP- MA from the Ministry of the Environment, Ecuador, al- lowed the collection and maintenance of frogs at Pon- tificia Universidad Catblica del Ecuador (PUCE). The Atlanta Zoo donated adults of Dendrobates auratus to the PUCE. Adults of both species reproduced success- fully at the Balsa de los Sapos, Centre of Amphibian Investigation and Conservation (CICA) of PUCE. Egg clutches were donated to the laboratory of developmental biology for embryonic analysis. Analysis of embryonic development The number of eggs per egg clutch was recorded. Em- bryos were analyzed from cleavage until tadpole hatch- ing, and were staged according to the E. machalilla table of stages (del Pino et al. 2004). Embryos were cultured in humid chambers at room temperature (18-23 °C). Procedures for fixation of embryos in Smith’s fixative, vibratome sectioning, and the staining of sections for cell nuclei with the fluorescent dye Hoechst 33258 (Sigma- Aldrich, St. Louis, MO, USA) were previously described (del Pino et al. 2004; Moya et al. 2007). Sections were mounted in glycerol, and microscopically examined with normal light using a Stemi SV6 stereomicroscope (Carl Zeiss, Oberkochen, Germany) or with fluorescent op- tics using a Z1 Axio Observer microscope (Carl Zeiss, Amphib. Reptile Conserv. 124 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus H. vertebralis D. auratus Fig. 4. Stage 14: Early neural fold stage of H. vertebralis and D. auratus embryos. Micrographs of H. vertebralis embryos are shown in A, C, and micrographs of D. auratus embryos are shown in B, D. Sections shown in C-D were stained for cell nuclei. (A) Lateral view of a neural fold stage embryo. (B) Dorsal view of a neural fold stage embryo. The neural plate was visible in embryos of the two species. (C) Cross section through the region of the trunk (Reproduced from Hervas and del Pino, 2013). (D) Cross sec- tion through the caudal region of an embryo. The notochord was visible in C and D. b, blastopore; e, endoderm; ec, ectoderm; n, notochord; nf, neural fold; pm, paraxial mesoderm. Oberkochen, Germany). Embryos were photographed with Axiocam cameras and the image capture program Axiovision (Carl Zeiss, Oberkochen, Germany). The im- ages were edited with Adobe Photoshop CS6. Egg diam- eter was measured in fixed embryos with the measuring tool of the program Axiovision (Carl Zeiss, Oberkochen, Germany). Results and Discussion Clutch size and developmental time The number of eggs ranged from 2-25 eggs, with a mean of 13 eggs per clutch in H. vertebralis, and 2-8 eggs, with a mean of five eggs per clutch in D. auratus (Table 1). The eggs of H. vertebralis and D. auratus measured about 2.6 and 3.5 mm in diameter, respectively (Table 1). The diameter of H. vertebralis eggs was previously reported to be of about three nun (Coloma 1995). The animal hemisphere of embryos was dark brown and the vegetal hemisphere was pale-yellow in embryos of both frogs (Eig. lA-D; not shown forZ). auratus). Egg clutch- es of H. vertebralis required 18 days from the two-cell stage to tadpole hatching under laboratory conditions; whereas 19-21 days were required from fertilization to tadpole hatching by the six species of dendrobatid frogs (del Pino et al. 2004, 2007; Hervas and del Pino 2013). The similarity of developmental times suggests that pa- rental care allows slow development in all of the spe- cies of dendrobatid frogs examined in comparison with X. laevis. Embryonic development of H. vertebralis and D. auratus The development from early cleavage until tadpole hatch- ing of H. vertebralis and D. auratus was divided into 25 stages, according to the staging criteria for E. machalilla (del Pino et al. 2004) given in Table 2. Micrographs of the external and internal morphology of H. vertebralis Amphib. Reptile Conserv. 125 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. H. vertebralis D. ouratus Fig. 5. Stage 15: Mid-neural fold stage of embryos of H. vertebralis and D. auratus. Micrographs of H. vertebralis embryos are shown in A, C, and micrographs of D. auratus embryos are shown in B, D. Sections shown in C-D were stained for cell nuclei. (A-B) Dorsal views of embryos. The neural tube was open in embryos of both species. (C) Cross section through the caudal region. The notochord was not detected in this caudal region (Reproduced from Hervas and del Pino, 2013). (D) Cross section through the trunk region. The notochord was visible, e, endoderm; ec, ectoderm; n, notochord; nf neural fold; pm, paraxial mesoderm. and D. auratus embryos illustrate these developmental stages (Figs. l-12).Cleavage of H. vertebralis was ho- loblastic as in E. machalilla (del Pino et al. 2004) (Fig. 1 A-C). Cleavage of D. auratus was not observed. The H. vertebralis blastula consisted of small, pigmented cells in the animal hemisphere; whereas, cells of the vegetal hemisphere were larger. The blastocoel developed during cleavage, and was large in blastula and gastrula stage em- bryos (Fig. 3A-D). The blastocoel roof, of the two spe- cies, was several cell diameters in thickness at stage 10 (Fig. 3F and I), and it was reduced to one cell thickness in the late-gastrula stage of H. vertebralis (stage 12.5; Fig. 3G). Similarly the thickness of the blastocoel roof was reduced to one or two cells in thickness in the early gas- trula of D. auratus (stage 11; Fig. 3J). Thickness of the blastocoel roof in the late-gastrula was not documented for this frog species. In E. machalilla, expansion of the blastocoel was accompanied by reduction in its thickness until it was a monolayer of cells in the late-gastrula (del Pino et al. 2004). The onset of gastrulation in H. vertebralis and D. au- ratus was marked by the presence of the dorsal blasto- pore lip in a sub-equatorial location (Fig. IE). A field of bottle cells was observed at the blastopore groove (Figs. 3E, H), as in X. laevis and E. machalilla (Hardin and Keller 1988; Moya et al. 2007). The gastrula developed a conspicuous yolk plug that became smaller during gas- trulation, until it was totally retracted by the end of gas- trulation (Fig. lE-I). The closed blastopore looked like a small slit in stage 13 embryos (Fig. II), as in E. macha- lilla and other frogs (del Pino et al. 2004). Internally, a small dorsal archenteron developed, which did not elon- gate until stage 13 in//, vertebralis (Fig. 12J-L), as 'mE. machalilla (del Pino et al. 2004); whereas in D. auratus, the archenteron was already large and inflated at stage 12 (Fig. 12P) (del Pino et al. 2007). The neural plate developed in late stage 13 (Fig. IJ). In stage 14, the neural folds were elevated (Figs. IK; 4A-B). The notochord was observed underneath the neural plate of stage 14 embryos (Fig. 4C-D). The neu- Amphib. Reptile Conserv. 126 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus Table 2. Stages of development of H. vertebralis and D. auratus in comparison with the E. machalilla table of development. Stage^ Characteristics of embryos D X G 1 1 1 Fertilization (not shown). 1 - 2 Gray crescent (not shown). 2 2 3 Two-cell stage (Fig. lA). This stage was observed only for //. vertebralis. 3 3 4 Four-cell stage (not shown). 4 4 5 Eight-cell stage (Fig. IB). This stage was observed only for H. vertebralis . 5 5 6 Sixteen-cell stage (Fig. 1C). This stage was observed only for H. vertebralis. 6 6 7 Thirty-two cell stage (not shown). 7 7 8 Large-cell blastula (not shown). 8 8 - Medium-cell blastula (not shown). 9 9 9 Advanced blastula (Fig. ID; 3A). 10 10 10 Early gastrula. The dorsal blastopore lip was formed (Eig. IE), had a subequatoiial location (Eigs. 3B, C), and there were bottle cells making the onset of cell ingression at the blastopore, as shown for both species (Eigs. 3E, H).The blas- tocoel was a large cavity, and its roof was several cells in thickness. The thickness of the blastocoel roof was reduced to a single cell in the late gastrula (Eigs. 3E, G, I, J). 11 11 11 Mid-gastrula with a yolk plug that measured about 1/2 of the embryo's diameter (Eig. IE). Internally, the ventral blasto- pore lip was formed as shown for D. auratus (Eig. 3D). 12 12 12 Late gastrula with a yolk plug that was 1/3 of the embryo's diameter or smaller (Eigs. IG). The archenteron of H. verte- bralis was smaller than D. auratus (Eigs. 12J, K, P, Q). 13 13 13 Slit blastopore stage (Eig. 11). Internally, the archenteron was elongated. A large circumblastoporal collar was visible (Eigs. 12L; R). The neural plate became visible in the late stage 13 (Eig. IJ). 14 14 14 Early neural fold stage. Images of H. vertebralis (Eig. IK; 4A, C), and of D. auratus (Eigs. 4B, D). In the trunk region of both species, the neural plate, notochord, and mesoderm were visible (Eigs. 4C, D). 15 16 15 Mid-neural fold stage. The neural folds approached each other. Images of H. vertebralis (Eig. IL; 5A, C), and of D. au- ratus (Eigs. 5B, D). The neural folds were elevated and touched each other in the trunk region {H. vertebralis, Eig. 5C); but were open in the cephalic region {D. auratus, Eig. 5D). 16 20 16 Closure of the neural tube in H. vertebralis (Eig. 2A; 6A, C), and D. auratus (Eigs. 6B, D). Closure of the neural tube was complete in both species. 17 24 17 Tail bud stage. The tail bud and the head region protruded beyond the yolky endoderm in H. vertebralis (Eig. 2B; 7); not shown for D. auratus. The epidermis, neural tube, notochord, somites, and endoderm were visible in the trunk region (Eig. 7B, C). 18 26 18 Muscular activity. The branchial arches protruded on the sides of the head. The eye vesicles were small (not shown). Im- ages of H. vertebralis (Eigs. 8A, C, E), and of D. auratus (Eigs. 8B, D, E). The epidermis, neural tube, notochord, rows of somites, and endoderm were visible in the trunk region of both species (Eigs. 8C-E). 19 33 19 Heart beat and external gill buds. The gill buds of the two pairs of external gills were visible. Images of H. vertebralis (Eigs. 2C, D; 9A, C, E), and of D. auratus (Figs. 9B, D, F). The dorsal fin was visible (Fig. 9C), the notochord was vacu- olated (Fig. 9C), and the pronephros was detectable (Fig. 9D). The number of somites increased (Figs. 9E, F). 20 40 20 Circulation to the external gills. The first gill pair had four or more branches. Images ofH. vertebralis (Fig. lOA). Inter- nally, the otocysts, brain, notochord, and somites were observed, as shown for 77. vertebralis (Fig. IOC). 21 41 21 Development of the external gills. The first pair of external gills had seven branches in H. vertebralis. The second pair of external gills was small and unbranched (Fig. 2E). In D. auratus, the first gill pair had six branches and the second gill pair developed two branches. The tail became elongated (not shown). 22 41 22 The external gills enlarged and eye pigment. The eyes contained pigment. Images of H. vertebralis (Fig. lOB). Inter- nally, the otocysts, eye, internal gills, notochord, and somites were observed, as shown for 77. vertebralis (Fig. lOD). 23 43 23 The external gills reached their full size (Fig. 11). The first pair of external gills had eight and nine branches in 77 ver- tebralis and D. auratus, respectively. The second gill pair was unbranched in 77 vertebralis and had two branches in D. auratus (Figs. 11 A, B). The opercular fold was visible. The eyes and the body were pigmented. Internally the epidermis, eye, otocysts, and somites were detected. Images of 77 vertebralis (Figs. 11 A, C) and of D. auratus (Figs. IIB, D). 24 44 24 The external gills were visible only on the left side. The operculum was closed on the right side (not shown). 25 45 25 The spiracle was formed. The embryos hatched and had the appearance of a tadpole. Internally, the brain, otocysts, somites, and yolky endoderm were observed. Images of 77 vertebralis (Fig. 2F, llE-G). 'D, stages of the dendrobatid frogs, H. vertebralis and D. auratus, according to the E. machalilla standard stages of development (del Pino et al. 2004); X, normal stages of X. laevis development (Nieuwkoop and Faber 1994); G, the generalized table of frog development (Gosner 1960). Amphib. Reptile Conserv. 127 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. H. vertebralis D. auratus Fig. 6. Stage 16: Closure of the neural tube in embryos of H. vertebralis and D. auratus. Micrographs of H. vertebralis embryos are shown in A, C, and micrographs of D. auratus embryos are shown in B, D. Sections shown in C-D were stained for cell nuclei. (A-B) Dorsal views of embryos. The neural tube was closed. The branchial arches were visible in the head region. (C) Cross sec- tion through the rostral region, anterior to the notochord. The neural tube was completely closed. (Reproduced from Hervas and del Pino, 2013). (D) Cross sections through the trunk region of an embryo. The somites were visible, br, branchial arch; e, endoderm; ec, ectoderm; hy, hyoid arch; ma, mandibular arch; m, mesoderm; n notochord; nt, neural tube; s, somite. ral folds became closed along the midline during stages 15-16 (Figs.lL; 2A; 5; 6). The external and internal characteristics of the neumla from stages 14-16 of H. vertebralis and D. auratus were compared (Figs. 4-6) and were found to be similar to E. machalilla embryos (del Pino et al. 2004). The tail bud embryos (stage 17) of H. vertebralis were examined in their external and internal morphol- ogy (Figs. 2B; 7). The body became elongated and the head and tail regions protruded over the large yolky en- doderm. The branchial arches were visible (Fig. 7A), and the brain and neural tube were detected. Embryos of this stage contained numerous somites (Fig. 7B, C). Embryos of stage 18 were characterized by muscular activity, and the embryos were longer. Buds of the external gills were detected in the head region (Eig. 8A, B). Somites, the notochord and neural tube were detected in the trunk region (Eig. 8C, D). A row of somites was detected on each side of the notochord (Eig. 8E, E). Myogenesis in both species occurred by cell interdigitation, as in other dendrobatid frogs and in the Marsupial frog, G. riobam- bae\ whereas, cell rotation is the pattern for X. laevis myogenesis (Gatherer and del Pino 1992; Hervas and del Pino 2013). Gill buds were larger in stage 19 embryos (Pigs. 2C-D; 9), and the external gills were fully devel- oped in embryos of stage 22-23. The first gill pair of H. vertebralis developed eight branches, and the second pair was unbranched; whereas, embryos of D. auratus developed nine and two branches in the first and second gill pairs, respectively (Pigs. 10; 11 A, B). The number of gill branches in the first and second pair of external gills varies among species of Dendrobatidae (del Pino et al. 2004). The tail became longer in embryos of stages 18-25, the brain, spinal cord, somites, and internal or- gans developed and the embryos gradually acquired the tadpole shape in both species (Pigs. 2C-P, 8-11). The processes of neurulation, somitogenesis, and internal em- bryo morphology of H. vertebralis and D. auratus were similar to the patterns described for other species of den- drobatid frogs (del Pino et al. 2004, 2007). Embryos of H. vertebralis hatched at stage 25 (Pigs. 2P, llE-G). The mouth had darkly pigmented teeth (Pig. UP), the body had dark pigment, and the embryo had the appearance of a tadpole (Pigs. 2P, UP, G). Amphib. Reptile Conserv. 128 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus H. vertebralis C St 17 Fig. 7. Stage 17: Tail bud stage of H. vertebralis embryos. (A) Lateral view of an embryo. (B) Cross section through the trunk region of the embryo in A. (C) Horizontal section at the level of the notochord and somites with the rostral region towards the left. A row of somites was detected on each side of the no- tochord. A portion of the neural tube was detected in the rostral region of the section, hr, branchial arch; e, endoderm; ec, ec- toderm; hy, hyoid arch; ma, mandibular arch; n notochord; nt, neural tube; s, somite. Comparative analysis of gastrulation Gastrulation is characterized by common morphogenet- ic events that occur in all of the analyzed frog species. Formation of the dorsal blastopore lip, its development to enclose a yolk plug, and the process of internaliza- tion of cells at the blastopore lip by the movements of involution are among these common morphogenetic pro- cesses (Elinson and del Pino 2012). Other developmental events, however, may be dissociated from gastrulation in some frog species. In particular, dorsal convergence and extension and the onset of notochord elongation are separated from gastrulation in the Marsupial frog, G. riobambae, and in dendrobatid frogs; whereas, these events occur simultaneously with gastrulation in X. lae- vis and in Engystomops (Leptodactylidae) (Table 1) (del Pino 1996; Benitez and del Pino 2002; Keller and Shook 2004; Moya et al. 2007; Elinson and del Pino 2012). The simultaneous occurrence of gastrulation and on- set of notochord elongation may be related to the repro- ductive adaptation of frogs for rapid development under unstable environmental conditions such as the aquatic environment in which embryos of X. laevis develop, or the development in floating foam nests in species of Engystomops (Elinson and del Pino 2012). Embryos of these frogs require from 5 hours to 12.5 hours from the onset of gastrulation to blastopore closure (Stage 10-13) (Nieuwkoop and Eaber 1994; Romero-Carvajal et al. 2009). Elongation of the notochord and gastrulation oc- cur simultaneously in embryos ofX. laevis, Engystomops coloradorum, and Engystomops randi (Leptodactylidae) (Keller and Shook 2004; Romero-Carvajal et al. 2009; Venegas-Eerrm et al. 2010). Early elongation of the no- tochord may be required for embryos to rapidly acquire the elongated tadpole shape in the unstable conditions of their reproductive environments. The most divergent mode of gastrulation was detected in embryos of the Marsupial frog, G. riobambae. Gas- trulation results in the formation of an embryonic disk from which the body of the embryo develops (del Pino and Elinson 1983). Cells that involute during gastrulation accumulate in the blastopore lip, and after blastopore clo- sure give rise to an embryonic disk of small cells, visible on the surface. Internally, the small cells that involuted during gastrulation accumulated in the embryonic disk and in its internal circumblastoporal collar (Moya et al. 2007). Eormation of the embryonic disk of G. riobambae is associated with delayed onset of notochord elongation that only starts once the blastopore is closed (del Pino 1996). Embryos of the Marsupial frog, G. riobambae de- velop slowly, and take a total of 168 hours from the onset of gastrulation to its completion (Table 1). As in G. riobambae, cells that involuted during gas- trulation became accumulated in a large circumblasto- poral collar in embryos of dendrobatid frogs, with sepa- Amphib. Reptile Conserv. 129 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. H. vertebrolis D. auratus Fig. 8. Stage 18: Muscular activity stage of H. vertebmlis and D. auratus embryos. Micrographs of H. vertebralis embryos are shown in A, C, E, and micrographs of D. auratus embryos are shown in B, D, F. Sections shown in C-D were stained for cell nuclei. (A) Lateral view of an embryo. (B) Dorsal view of an embryo. The gill buds were visible on each side of the head in embryos of both species. (C-D) Cross sections through the trunk. The cavity in D corresponds to the gut. (E-F) Horizontal sections. A row of numerous somites was detected on each side of the notochord. The brain and the otocysts were visible in E, and the gut was visible in F. bn, brain; ec, ectoderm; g, gut; gb, gill bud; hy, hyoid arch; ma, mandibular arch; n notochord; nt, neural tube; ot, otocyst; p, pronephros; s, somite. ration of the morphogenetic events of gastmlation and the onset of notochord elongation. However, dendrobatid frogs do not develop an embryonic disk (Elinson and del Pino 2012). Egg size varied from 1.6 to 3.5 mm in diam- eter among dendrobatid frogs (Table 1), and their devel- opment was slow. Embryos of dendrobatid frogs require 36-72 hours from the onset of gastmlation to its comple- tion (Stage 10-13; Table 1). We analyzed the characteris- tics of the gastmla in dendrobatid embryos derived from eggs of different diameters (Table 1; Eig. 12). Protection of embryos in the terrestrial nests of dendrobatids or inside a pouch of the mother in G. riobambae may al- Amphib. Reptile Conserv. 130 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus H. vertebralis D. auratus Fig. 9. Stage 19: Muscular response stage of H. vertebralis and D. auratus embryos. Micrographs of H. vertebralis embryos are shown in A, C, E, and micrographs of D. auratus embryos are shown in B, D, F. (A) Lateral view of an embryo. (B) Dorsal view of an embryo. The developing gills were visible. (C-D) Cross sections through the trunk. The dorsal fin was visible in C, and the pronephros in D. (E) Horizontal section at the level of the gut. (F) Horizontal section at the level of the brain and the gut. bn, brain; df, dorsal fin; fg, first gill pair; g, gut; gb, gill bud; hy, hyoid arch; ma, mandibular arch; n notochord; nt, neural tube; p, pronephros; s, somite. low slow development and the separation of gastrulation from notochord elongation (Elinson and del Pino 2012). Details of the morphology of the H. vertebralis and D. auratus gastrula are illustrated in Fig. lE-I, Fig. 3B-J, and Fig. 12J-F, P-R. The archenteron roof remained relatively thin during gastrulation in H. vertebralis and D. auratus in comparison with stage 13 embryos of X. laevis (Fig. 12C, J-L, P-R). Elongation and inflation of the archenteron varied greatly among dendrobatids. The archenteron remained small during gastrulation and be- Amphib. Reptile Conserv. 131 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. H. vertebralis Fig. 10. Stages 20-22: External gill development in H. vertebralis. (A) Stage 20: Circulation to the external gills. Three branches were visible in the first gill pair. The second gill pair was unbranched. (B) Stage 22: The external gills enlarged. Seven branches were visible in the first gill pair. The second gill pair was unbranched. (C) Stage 20: Horizontal section at the level of the brain. (D) Stage 22: Horizontal section at the level of the internal gills, bn, brain; ey, eye; ec, ectoderm; fg, first gill pair; g, internal gill; nt, notochord; ot, otocyst; s, somite; sg, second gill pair. came elongated and inflated after gastmlation in H. ver- tebralis (Fig. 12J-L); whereas, at stage 12, the archen- teron was already elongated in the very large embryos of A. bilinguis and D. auratus (Fig. 12 M, P). Moreover inflation of the archenteron was already deteeted in stage 12 embryos of D. auratus (Fig. 12 P). In other species of dendrobatids, we detected variation in the level of arch- enteron elongation and inflation (Fig. 12D-R). We con- cluded that in A. bilinguis, and D. auratus, dendrobatids with very large eggs, the elongation of the archenteron begins earlier in comparison with embryos of dendroba- tid frogs with smaller eggs such E. machalilla (Table 1) (del Pino et al. 2007). In spite of the differences detected in the onset of archenteron elongation, the cells that involuted during gastmlation became accumulated in a large circumblas- toporal collar at stage 13 in all of the dendrobatid frogs analyzed, as previously reported for E. machalilla, and shown for E. anthonyi, E. tricolor, H. vertebralis, A. bi- linguis, andZ). auratus, (Fig. 12F, I, L, O, R) (Moya et al. 2007). Notochord elongation is dissociated from gastm- lation in these frogs (Bemtez and del Pino 2002; del Pino et al. 2007; Moya et al. 2007; Venegas-Ferrfn et al. 2010; Montenegro-Larrea and del Pino 2011). The comparative analysis of gastmlation indicates that in spite of the great variation in egg size and onset of Amphib. Reptile Conserv. 132 March 2015 | Volume 8 | Number 1 | e90 arehenteron elongation and inflation, the Dendrobatidae species examined develop a large circumblastoporal col- lar as a result of gastmlation (Fig. 12D-R; Table 1) (del Pino et al. 2007; Moya et al. 2007; Montenegro-Larrea and del Pino 2011). Moreover, notochord elongation is delayed until after blastopore closure as in G. riobambae. In spite of their large circumblastoporal collar, dendro- batid frog embryos did not develop an embryonic disk. Conclusions Development of the dendrobatid frogs, H. vertebralis and D. auratus, shared the developmental characteristics described for E. machalilla (del Pino et al. 2004). Gas- tmlation and notochord elongation occurred as separate morphogenetic events in these frogs in comparison with additional species of Dendrobatidae. Development in a somewhat stable terrestrial environment may be associat- ed with the separation of these developmental events and with comparatively slow development. The developmen- tal analysis of H. vertebralis and other frogs contributes to a better knowledge of their biology and may contribute to the conservation and reproductive management of en- dangered frogs. Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus H. vertebralis D. auratus velopmental Biology of PUCE for their assistance. In particular, we acknowledge the help of Alexandra Vargas with the preparation of illustrations. We thank Clifford Keil for valuable criticism and for language revision. This study received the support of research grants from PUCE. Fig. 11. Stages 23-25: Complete development of the external gills to tadpole hatching in embryos of H. vertebralis and D. auratus. Micrographs of H. vertebralis embryos are shown in A, C, E, F, G, and micrographs of D. auratus embryos are shown in B, D. (A) Stage 23 of H. vertebralis: Full development of external gills. The first gill pair of the external gills had eight branches, which at this stage were fully extended. The second gill pair of external gills was unbranched. (B) Stage 23 of D. auratus: The first gill pair of the external gills had nine branches, which at this stage were fully extended. The second pair of external gills was smaller and had two branches. In embryos of both species the eyes and the body were pigmented. Tbe tail was elongated. (C) Stage 23: Sag- ittal section. The section was done through the embrionary brain and somites. The eyes, notochord, and tail fin were observed. (D) Stage 23: Horizontal section at tbe level of somites. The eyes and otocysts were visible. (E) Stage 25: Head of a tadpole at hatching in dorsal view. The eyes were visible. (F) Stage 25: Ventral view of the head of the tadpole shown in A. The spiracle was visible. (G) Stage 25: Horizontal section of a tadpole at hatching at the level of the otocysts. The eyes, otocysts, and somites were visible, bn, brain; ey, eye; ep, epidermis, fg, first gill pair; n, notochord; ot, otocyst; sg, second gill pair; s, somite; tf, tail fin; y, yolky en- doderm. Acknowledgments. — We express our thanks to the Cen- tre of Amphibian Investigation and Conservation (CICA), Balsa de los Sapos of the Pontificia Universidad Catolica del Ecuador (PUCE) for the donation of embryos. We ex- press gratitude to the members of the Laboratory of De- Literature Cited Benftez MS, del Pino EM. 2002. Expression of Brachy- ury during development of the dendrobatid frog Co- lostethus machalilla. Developmental Dynamics 225: 592-596. Coloma LA. 1995. Ecuadorian frogs of the genus Colo- stethus (Anura: Dendrobatidae). University of Kansas Museum of Natural History, Miscellaneous Publica- tions 87: 1-72. Coloma LA, Ron S, Morales M, Almendariz A. 2004. Hy- loxalus vertebralis. In: lUCN 2013. lUCN Red List of Threatened Species. Version 2013.2. Available: http:// www.iucnredlist.org [Accessed: 04 April 2014]. del Pino EM. 1996. The expression of Brachyury (T) during gastrulation in the marsupial frog Gastrotheca riobambae. Developmental Biology 111: 64-72. del Pino EM, Avila ME, Perez O, Benitez MS, Alarcbn I, Noboa V, Moya I. 2004. Development of the den- drobatid frog Colostethus machalilla. International Journal of Developmental Biology 48: 663-670. del Pino EM, Elinson RP. 1983. A novel development pattern for frogs: Gastrulation produces an embryonic disk. Nature 306: 589-591. del Pino EM, Venegas-Ferrfn M, Romero-Carvajal A, Montenegro-Larrea P, Saenz-Ponce N, Moya IM, Alarcbn I, Sudou N, Yamamoto S, Taira M. 2007. A comparative analysis of frog early development. Proceedings of National Academy of Sciences of the United States of America 104: 11,882-11,888. Elinson RP, del Pino EM. 2012. Developmental diversity of amphibians. Wiley Interdisciplinary Reviews: De- velopmental Biology 1: 345-369 Amphib. Reptile Conserv. 133 March 2015 I Volume 8 | Number 1 | e90 Hervas et al. St 12 St 12,5 St 13 Fig. 12. Gastrulation of dendrobatid frogs in comparison withX laevis. To facilitate the comparison, the stages (st) are given as column headings: Stage 12, late-gastrula; Stage 12.5, advanced late-gastrula; Stage 13, slit blastopore stage. The letters in the upper right-hand comer indicate the species: Ah, Ameerega bi- linguis; Da, Dendrobates auratus; Ea, Epipedobates anthonyi; Et, Epipedobates tricolor, Hv, Hyloxalus vertebralis; XI, Xeno- pus laevis. (A-C) Diagrams ofX. laevis embryos. (D-R) Sagit- tal sections of gastmlae. (D-F) Embryos of E. anthonyi. (The section in E was reproduced from Montenegro-Larrea and del Pino, 2011). (G-I) Embryos of E. tricolor. (J-L) Embryos of H. vertebralis. (M-O) Embryos of A. bilinguis. (P-R) Embry- os of D. auratus. a, archenteron; hi, blastocoel; cbc; circum- blastoporal collar; dl, dorsal blastopore lip; yp, yolk plug. Gatherer D, del Pino EM. 1992. Somitogenesis in the marsupial frog Gastrotheca riobambae. International Journal of Developmental Biology 36: 283-291. Gosner KL. 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Her- petologica 16: 183-190. Hardin J, Keller R. 1988. The behaviour and function of bottle cells during gastrulation of Xenopus laevis. De- velopment 103: 211-230. Hervas F, del Pino EM. 2013. Desarrollo neural, somi- togenesis y morfologfa interna de los embriones de Hyloxalus vertebralis y Dendrobates auratus (Anura: Dendrobatidae). Revista Ecuatoriana de Medicina y Ciencias Bioldgicas 34: 99-112. Keller R, Shook D. 2004. Gastrulation in amphibians. Pp. 171-204 In: Editor, Stem CD. Gastrulation from Cells to Embryo. Cold Spring Harbor Laboratory Press, New York, New York, USA. 73 1 p. Montenegro-Larrea P, del Pino EM. 2011. La gastm- lacion de Epipedobates anthonyi (Anura: Dendroba- tidae). Revista Ecuatoriana de Medicina y Ciencias Bioldgicas 32: 24-32. Moya IM, Alarcon I, del Pino EM. 2007. Gastmlation of Gastrotheca riobambae in comparison with other frogs. Developmental Biology 304: 467^78. Nieuwkoop PD, Faber J. 1994. Normal Table of Xenopus laevis (Daudin). Garland Publishing, New York, New York, USA. 282 p. Romero-Carvajal A, Saenz-Ponce N, Venegas-Ferrfn M, Almeida-Reinoso D, Lee C, Bond J, Ryan M, Wall- ingford J, del Pino EM. 2009. Embryogenesis and laboratory maintenance of the foam-nesting tiingara frogs. Genus Engystomops {=Physalaemus). Develop- mental Dynamics 23'^: 1,444-1,454. Solis F, Ibanez R, Jaramillo C, Chaves G, Savage Y, Kohler G, Jungfer KH, Bolivar W, Bolanos F. 2004. Dendrobates auratus. In: lUCN 2013. lUCN Red List of Threatened Species. Version 2013.2. Available: http://www.iucnredlist.org [Accessed: 04 April 2014]. Venegas-Ferrm M, Sudou N, Taira M, del Pino EM. 2010. Comparison of Liml expression in embryos of frogs with different modes of reproduction. International Journal of Developmental Biology 54: 195-202. Amphib. Reptile Conserv. 134 March 2015 I Volume 8 | Number 1 | e90 Development and gastrulation in Hyloxalus vertebralis and Dendrobates auratus Francisca Hervas was Adjunct Professor and developmental biology researcher at the School of Biologi- cal Sciences, Pontificia Universidad Catolica del Ecuador (PUCE, 2014), in Quito. She holds a Eicencia- tura in Biological Sciences from PUCE, and is enrolled in the PUCE master's degree program in conserva- tion biology. Her Licenciatura thesis is the study of the morphology of the neurula and more advanced embryos of the species Hyloxalus vertebralis and Dendrobates auratus', she also analyzed the mode of myogenesis in the large embryos of these frogs. Her research interests are focused on amphibians, with an emphasis on Ecuadorian species. Karina P. Torres is a graduate of the Eicenciatura in Biological Sciences Program at the Pontificia Uni- versidad Catolica del Ecuador (PUCE) in Quito (2014). For her thesis research she investigated the early development of Hyloxalus vertebralis (Dendrobatidae) in the Eaboratory of Developmental Biology at PUCE. Her research centers in the analysis of the morphological characteristics of the H. vertebralis gas- trula in comparison with other dendrobatid frogs. Paola Montenegro-Larrea is a Ph.D. student at the Interdisciplinary Fife Sciences, Purdue University, West Lafayette, Indiana, USA. She holds a M.S. in molecular genetics and diagnostics from The University of Nottingham, United Kingdom, and a Licenciatura in biology from the Pontificia Universidad Catolica del Ecuador (PUCE), in Quito. Her Licenciatura thesis researched the characterization of gastrula mor- phology in four Ecuadorian species of Dendrobatid frogs with eggs of different sizes. Earlier in her career, she took part in the establishment of the Molecular Genetics Laboratory at the hospital of the Ecuadorian Armed Forces in Quito (Hospital de las Fuerzas Armadas del Ecuador). Eugenia M. del Pino is professor of biological sciences (retired) at the Pontificia Universidad Catolica del Ecuador (PUCE) in Quito. She studied the reproduction and development of marsupial frogs (Hemiphrac- tidae) in comparison with Xenopus laevis, the model organism of frog developmental biology and with several frogs from Ecuador. Her studies are done in collaboration with PUCE students. Her analyses of development reveal important variation in morphology and developmental time among frogs. The devel- opmental data is significant for the comparative analysis of frog early embryonic development, and provide base line information about the biology of several frog species. Amphib. Reptile Conserv. 135 March 2015 I Volume 8 | Number 1 | e90 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 136-140 (e91). Short Communication First records of Anolis ventrimaculatus Boulenger, 1911 (Squamata: Iguanidae) in Ecuador Ternando Ayala- Varela, ^Julian A. Velasco, ^Martha Calderon-Espinosa, "^Alejandro F. Arteaga, ^’ Yerka Sagredo, and ^ ’^Sebastian Valverde ^Escuela de Ciencias Bioldgicas, Pontificia Universidad Catolica del Ecuador, Avenida 12 de Octubre 1076 y Roca, Apartado 17-01-2184, Quito, ECUADOR ^Laboratorio de Andlisis Espaciales, Instituto de Biologia, Universidad Nacional Autonoma de Mexico, MEXICO D.F ^Instituto de Ciencias Naturales, Edificio 425, oficina 111, Universidad Nacional de Colombia, Sede Bogota, COLOMBIA "^Tropical Herping, Av Eloy Alfaro N39-202 y Jose Puerta. Ed Montecatini. Quito, ECUADOR Abstract— \Ne report the first records of Anoiis ventrimaculatus for Ecuador based on twelve specimens from three localities: Chical (Provincia Carchi), El Cristal (Provincia Esmeraldas), and Lita (Provincia Imbabura). The locality in the Provincia Carchi lies approximately 16 km S from the nearest record (Nambi, Department Nariho, Colombia). We also present information on scalation and coloration. Key words. Anole lizards, color, distribution, Ecuador, scalation Citation: Ayala- Varela F, Velasco JA, Calderon-Espinosa M, Arteaga AF, Sagredo Y, Valverde S. 2015. First records of Anolis ventrimaculatus Bou- lenger, 1911 (Squamata: Iguanidae) in Ecuador. Amphibian & Reptile Conservation 80) [Special Section]: 136-140 (e91). Copyright: © 2015 Ayala- Varela et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommer- cialNoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptile Conservation] official journal website . Received: 13 December 2014; Accepted: 02 March 2015; Published: 04 April 2015. Thirty-seven species of Anolis have been reported for- mally from Ecuador (Ayala- Varela et al. 2014). Anolis ventrimaculatus Boulenger 1911 was described from two syntypes, an adult female and a juvenile from Rio San Juan, Department Risaralda, Colombia. Williams and Duellman (1984) designated the adult female syntype (BMNH 1946.8.13.5) as the lectotype. Anolis ventrimaculatus is assigned to the aequatoria- lis series Castaneda and de Queiroz (2013) by having a moderate to large body size, narrow toe lamellae, small head scales, smooth ventral scales, and uniform dorsal scalation. It is assigned to the eulaemus-subgroup based on a typical Anolis digit, in which the distal lamellae of phalanx III distinctly overlap the first proximal subdigital scale of phalanx II (Williams 1976; Williams and Du- ellman 1984; Castaneda and de Queiroz 2013). Eleven species of the eulaemus-subgroup occur on both sides of the Andes {Anolis anoriensis Velasco et al. 2010, A. antioquiae Williams 1985, A. eulaemus Bou- lenger 1908, A. fitchi Williams and Duellman 1984, A. gemmosus O’Shaughnessy 1875, A. maculigula Wil- liams 1984, A. megalopithecus Rueda-Almonacid 1989, A. otongae Ayala- Varela and Velasco 2010, A. podocar- pus Ayala- Varela and Torres-Carvajal 2010, A. pod Aya- la- Varela et al. 2014, and A. ventrimaculatus Boulenger 1911). Specimens examined for comparisons are housed in the herpetological collections of the Museo de Zoologfa, Pontificia Universidad Catdlica del Ecuador, Quito, Ec- uador (QCAZ); Museo de Herpetologia de la Universidad de Antioquia, Antioquia, Colombia (MHUA); Coleccion de Herpetologia, Universidad del Valle, Santiago de Cali, Colombia (UVC); and Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota, Colombia (ICN). External character terminology follows Williams et al. (1995) and Poe and Yanez-Miranda (2008). Lamel- lar number was counted using the method of Williams et al. (1995), i.e., only on phalanges III and IV of the Correspondence. Email: fpayala2000@gmail.com (Corresponding author), duvelas@gmail.com, ^mlcalderone@unal.edu.co, *yevasanu@gmail. com, fycus_8 7@hotmail. es April 2015 | Volume 8 | Number 1 | e91 Amphib. Reptile Conserv. 136 Ayala- Varela et al. fourth toe. Measurements were made with digital cali- pers on preserved specimens and are given in millimeters (mm), usually to the nearest 0.1 nun. Snout- vent length (SVL) was measured from tip of snout to anterior edge of cloaca. Femoral length was measured from midline of venter to knee, with limb bent at a 90-degree angle. Tail length was measured from anterior edge of cloaca to distal point. Herein we report the first records of Anolis ventrimac- ulatus (Fig. 1) for Ecuador based on specimens collected at three localities. Four specimens (QCAZ 3284-3286, 8934) were collected on 16 September 1992 in Lita (0.87°, -78.45°), Provincia Imbabura; four specimens (QCAZ 2666, 3923, 3924, 3929) were collected on Au- gust 1994 in El Cristal (0.83°, -78.49°, 1,200-1,250 m), Reserva Ecolbgica Cotacachi-Cayapas, Provincia Es- meraldas; and four specimens (QCAZ 4376, 4378, 4389, 4390) were collected on 03 July 2011 in Rfo San Pablo, near Chical (0.90°, -78.16°, 1,399-1,451 m), Provincia Carchi. The last locality lies approximately 16 km S from the nearest record (ICN 11981-85, 11987-989, 12097, Nambf, Barbacoas municipality. Department Narino, Colombia) (Pig. 1, Table 1). The individuals from Chical (Provincia Carchi) were captured in secondary forest; all individuals were found on leaves, branches, or ferns from 50-150 cm above ground; a male were found head-down, while two fe- males were found head-down and head-up. The small- est specimen (QCAZ 8934, juvenile, SVL = 31.4 mm) was collected on 16 September 1992. An adult female (QCAZ 4378) collected in July 2011 deposited one white egg (17.11 mm x 6.44 mm). Our collections of Anolis ventrimaculatus in Ecuador were found from 1,200 to 1,451 m above sea level. In Ecuador, this species occurs in sympatry with A. aequatorialis, A. gemmosus, and A. maculiventris in Chical (Provincia Carchi); with A. lyn- chi, A. maculiventris, and A. princeps (pers. obs. Sebas- tian Valverde) in Lita (Provincia Imbabura), and with A. lynchi in El Cristal (Provincia Esmeraldas). Scalation and morphometric characters of Anolis ven- trimaculatus are presented in Table 2. Scale counts are similar between Ecuadorian and Colombian specimens. Our Ecuadorian specimens of Anolis ventrimaculatus are smaller than those from Colombia (maximum SVL 62 mm and 80 mm, respectively). Coloration in life of specimens from Ecuador was re- corded as follows: Adult female (QCAZ 4390, Pigs. 2 A, B): dorsal sur- faces of head, body and tail dark brown; dorsal surface of body with a pale brown longitudinal stripe extending from occipital region to base of tail; limbs pale brown with dark brown reticulation; tail pale brown; lateral sur- face of head with two stripes, one dark brown and ex- tending posteriorly from loreal region, through subocu- lar region, above tympanum to level of the hind limb, the other stripe is pale green and extending posteriorly from loreal region, through subocular region, above tym- Fig. 1. Distribution of Anolis ventrimaculatus in South America (locality numbers are listed in Table 1). panum to level of neck; lateral surface of body brown anteriorly and olive-green near inguinal region; ventral surface of head yellowish green with pale brown reticu- lations; ventral surface of body cream; ventral surface of limbs dark cream with dark brown reticulations; ventral surface of tail dark cream. Adult female QCAZ 4378 (Pigs. 2 C, D) differs from the previous pattern in having the dorsal surface of body brown, with seven dark brown blotches arranged longi- tudinally along the midline. Adult male (QCAZ 4389, Pigs. 2 E, P, G): When stressed, the background of head, body, limbs and tail was yellowish brown; dorsal surface of the neck with two dark brown bands; dorsal surface of body with nine dark brown blotches arranged longitudinally; limbs with dark brown bands; dorsal surface of tail with dark brown transversal bands, and with three dark brown blotches in the proximal part of tail; lateral surface of head with a darker brown first stripe, extending posteriorly from lo- real region, through subocular region, above the tympa- num and bifurcating into branches that continue on nu- chal crest and shoulder, respectively; a yellowish-green second stripe, extending posteriorly from loreal region, through subocular region, above the tympanum to the shoulder; black ocelli with white centers on the shoul- der; lateral surface of body with reddish-brown bands oriented ventroposteriorly; ventral surface of head yel- lowish green with pale brown reticulations; ventral sur- face of neck pale green; ventral surface of body cream; ventral surface of limbs pale brown with dark brown re- April 2015 I Volume 8 | Number 1 | e91 Amphib. Reptile Conserv. 137 First records of Anolis ventrimaculatus in Ecuador Table 1. Localities of Anolis ventrimaculatus in Ecuador and Colombia. Site number Country Locality Latitude Longitude Source 1 Colombia Bosque de San Antonio, km 18 via Cali-Bue- naventura, Valle del Cauca 3.22 -76.65 JAV pers. obs 2 Colombia Bosque de San Antonio, km 18 via Cali-Bue- naventura, Valle del Cauca 3.51 -76.62 UVC 9737, 9749, 9896, MHUA 1671-79, JAV pers. obs. 3 Colombia Antena, Cerro La Horqueta, ca. 28 km de Cali, Valle del Cauca 3.44 -76.52 JAV pers. Obs, ICN 3567 4 Colombia Vereda La Tulia, Mpio. Bolivar, Valle del Cauca 4.42 -76.24 JAV pers. obs 5 Colombia Vereda Chicoral, La Cumbre, Valle del Cauca 3.58 -76.58 JAV pers. obs, UVC 10223 6 Colombia PRN Barbas-Bremen, Mpio. Eilandia, Quindio 4.71 -75.64 JAV pers. obs 7 Colombia Alrededores Lago Calima, Mpio. Darien, Valle del Cauca 3.86 -76.56 JAV pers. obs; UVC 5189-96, ICN 3553-54 8 Colombia Reserva Eaunistica Bosque de Yotoco, Valle del Cauca 3.88 -76.44 JAV pers. obs 9 Colombia Penas Blancas, Pichinde, Valle del Cauca 3.42 -76.66 UVC 223, 224 10 Colombia Alto de Galapagos, carretera Cartago-San Jose del Palmar, Limite Valle-Choco 4.86 -76.22 UVC 9366, UVC 8489-95 11 Colombia PMN Arrayanal, Mpio. Apia, Risaralda 5.29 -75.90 JAV pers. obs 12 Colombia PMN Planes de San Rafael, Mpio. Santuario, Risaralda 5.13 -76.00 JAV pers. obs 13 Colombia PMN Agualinda, Mpio. Mistrato, Risaralda 5.12 -75.94 JAV pers. obs 14 Colombia PMN Verdum, vereda La Secreta, Risaralda 5.01 -76.03 JAV pers. obs 15 Colombia Vereda Buenos Aires, Cuenca Rio Barbo, Pereira, Risaralda 4.73 -75.58 JAV pers. obs 16 Colombia Rio Nambi, Narino 1.30 -78.08 JAV pers. obs 17 Colombia Reserva La Planada, Narino 1.08 -77.88 JAV pers. obs 18 Colombia Nambi, Narino 1.02 -78.07 ICN 11981-85, 11987-889, 12097 19 Ecuador Lita, Imbabura 0.87 -78.45 QCAZ 20 Ecuador El Cristal, Reserva Ecol6gica Cotacachi Cay- apas, Esmeraldas 0.83 -78.49 QCAZ 21 Ecuador Rio San Pablo, cerca de Chical, Carchi 0.90 -78.16 QCAZ ticulations; ventral surface of tail pale brown with small dark brown reticulations; dewlap skin yellowish brown; gorgetals pale green; marginals and stemals yellowish green; iris dark brown with yellowish-brown inner ring. The coloration of populations of Anolis ventrimacula- tus from Colombia display a dorsal surface of the body that is bright emerald green, or greenish-brown with slight darker oblique bars and yellow spots on each side of the dorsal midline; yellow spots fuse forming a series of saddle-shaped bars that cross the back and tail, more visible in the stressed phase. At their stressed phase, dor- sal and lateral surface of body brown with tiny yellow spots; lateral surface of head with a yellow line under the eye and with a prominent pale yellow or green line over the lips extending back over the ear opening and along the sides of the neck; ventral surface of head yellow- green, sometimes with reticulations; ventral surface of body cream to yellow-green, with dark brown spots on the sides; ventral surface of tail orange in male adults. Some females have a dorsal surface of body with a tan longitudinal stripe and dark edges. Anolis ventrimaculatus has a wide range of distri- bution, approximately 570 km in airline between the northern and southernmost localities. However, there is a huge distributional gap between central and southern Colombian populations (approximately 265 km airline between Bosque de San Antonio, Department Valle del Cauca and the Rio Nambi, Department Narino). One of the main reasons for this gap is the lack of extensive her- petological inventories in these areas, particularly in both foothills of the Andes cordilleras. More sampling effort should addressed to these areas with the aim to fill distri- butional gaps in several species, including Anolis lizards. Acknowledgments. — We thank Omar Torres-Carva- jal of the Museo de Zoologfa (QCAZ), Vivian Paez of the Museo Herpetolbgico de Antioquia (MHUA), and John Lynch of the Institute Nacional de Ciencias Naturales (ICN) for the loan of museum specimens, information about localities, and work space; Melissa Rodriguez for helping with the map; O. Torres-Carvajal for critical and valuable comments throughout the development of this April 2015 | Volume 8 | Number 1 | e91 Amphib. Reptile Conserv. 138 Ayala- Varela et al. Fig. 2. Anolis ventrimaculatus from Ecuador: female adult (A-B, QCAZ 4390) in dorsal and ventral view, female adult (C-D, QCAZ 4378) in dorsal and ventral view, male adult (E-F, QCAZ 4389) in dorsal and ventral view, male dewlap (G, QCAZ 4389) in lateral view. Photographs by F Ayala-Varela. manuscript. This work was funded by Secretaria de Edu- cacion Superior, Ciencia, Tecnologia e Innovacion del Ecuador (SENESCYT). Ecuadorian specimens were col- lected under collection permit 008-09 IC-FAU-DNB/MA issued by Ministerio del Ambiente and were deposited at Museo de Zoologia (QCAZ), Pontificia Universidad Catolica del Ecuador. Julian Velasco thanks the Wildlife Conservation Society for the support of fieldwork in Co- lombia under permits provided by CARDER (resolution 1085 of April, 6 2010).. Literature Cited Ayala-Varela E, Velasco JA. 2010. A new species of dac- tyloid anole (Squamata: Iguanidae) from the western Andes of Ecuador. Zootaxa 2577: 46-56. Ayala-Varela EP, Troya-Rodrfguez D, Talero-Rodrfguez X, Torres-Carvajal O. 2014. A new Andean anole species of the Dactyloa clade (Squamata: Iguanidae) from western Ecuador. Amphibian & Reptile Conser- vation 8 [Special Section]: 8-24. Ayala-Varela EP, Torres-Carvajal O. 2010. Anew species of dactyloid anole (Iguanidae, Polychrotinae, Anolis) from the southeastern slopes of the Andes of Ecuador. ZooKeys 53: 59-73. Castaneda MR, de Queiroz K. 2013. Phylogeny of the Dactyloa clade of Anolis lizards: new insights from combining morphological and molecular data. Bulle- tin of the Museum of Comparative Zoology 160(7): 345-398. April 2015 | Volume 8 | Number 1 | e91 Amphib. Reptile Conserv. 139 First records of Anolis ventrimaculatus in Ecuador Table 2. Scale count and measurements (mm) of specimens of Anolis ventrimaculatus from Ecuador and Colombia. Range (sample size) and mean. SVL = snout- vent length. Ecuador QCAZ Colombia UVC, ICN, MHUA Colombia Williams et al. 1995 Number of scales between second canthals 14-17(10) 15.7 12-17(18) 14.6 11-21 (20) Number of scales bordering rostral 5-7 (10) 6.4 6-8 (18) 6.8 6-10 (20) Number of scales between supraorbital semicircles 2-5 (10) 3.6 4-6 (18) 4.8 2-6 (20) Number of scales between interparietal and supraor- bital semicircles 7-11 (8) 6.9 6-11 (18)8.0 5-16 (20) Interparietal +/- +/very small (+/?) Number of loreal rows 5-8 (9) 7.9 7-9 (18) 8.2 7-11 (20) Number of supralabials to center of eye 6-8 (10) 7.2 6-8(18) 7.4 6-8 (20) Number of postmentals 6-8 (9) 6.3 6-9(18) 6.6 4-8 (20) Number of sublabials in contact with infralabials 0-2 (9) 0.5 1-3 (18) 2.5 0-2 (20) Lamellar number 16-18(10) 17.2 17-22 (18) 19.4 16-22 (20) Number of middorsals in 5% SVL 12-15(10) 12.9 14-19(18) 16.6 - Number of midventrals in 5% SVL 7-11 (10) 9.2 9-14(18) 11.9 - Femur length 16.0-20.6(10) 18.5 15.4-23.1 (16) 18.9 - Maximum SVL (male/female) 62/57 75/69 80/62 Poe S, Yanez-Miranda C. 2008. Another new species of green Anolis (Squamata: Iguania) from the East- ern Anders of Peru. Journal of Herpetology 42 (3): 564-571. Williams EE. 1976 South American anoles: The species groups. Papeis Avulsos de Zoologia 29: 259-268. Williams EE, Duellman WE. 1984. Anolis fitchi, a new species of the Anolis aequatorialis group from Ecua- dor and Colombia. University of Kansas Publications, Museum of Natural History 10: 257-266. Williams EE, Rand H, Rand AS, O’Hara RJ. 1995. A computer approach to the comparison and identifica- tion of species in difficult taxonomic groups. Breviora 502: 1-47. Fernando Ay ala- Varela is the director of the herpetology eolleetion at the Pontifieia Universidad Catdliea del Eeuador in Quito. He reeeived his diploma at the Pontifieia Universidad Catdliea del Eeuador, Quito in 2004. He has been interested in herpetology since childhood and has dedicated a lot of time studying the lizards of Ecuador, specifically the taxonomy and eeology of Anolis species. His current research interests include reproductive biology and ecology of lizards and snakes in Ecuador. Julian A. Velasco is a Ph.D. student at Instituto de Biologia, Universidad Naeional Autdnoma de Mexieo. His doe- toral research is focused on understanding the ecological and evolutionary processes responsible for species richness and diversifieation of Anolis lizards. He addresses several evolutionary and ecological topics using a combination of conceptual and methodological approaches as niche modeling, geospatial analysis, historical biogeography, and macroecology. Martha Calderon is the eurator of the reptile eolleetion at the Instituto de Cieneias Naturales, Universidad Na- eional, Colombia. She obtained her doetor degree at the Universidad Naeional Autdnoma de Mexieo (UNAM) in Mexico City. She works on ecomorphology, thermal ecology, reproductive biology, and molecular systematics of lizards. More information can be found here: www.biodiversidadysistematicamolecular.blogspot.com Alejandro Arteaga is a wildlife photographer and undergraduate biology student from Venezuela. In 2009, he co-founded Tropical Herping, an institution striving to preserve tropical reptiles and amphibians through tourism, photography, research, and education. Alejandro is author of The Amphibians and Reptiles of Mindo and several seientifie artieles. He has deseribed three speeies new to seienee and his photographie work has been featured in National Geographic, Anima Mundi, and the Discovery Channel. Yerka Sagredo Nunez is an Assoeiate Researeher at the Museum of Zoology, Pontifieal Catholie University, Ee- uador. Her baehelor’s degree was obtained in biologieal seienees from the Central University of Eeuador. Currently she is working as an assistant in the herpetology eolleetion at the Museo de Zoologia of the Pontifieia Universidad Catdliea del Eeuador (QCAZ). She is interesting in eeology, behavior, and taxonomy of amphibians and reptiles. She is also involved in studies of the genus Pristimantis. Sebastian Valverde is an Assoeiate Researeher at the Museum of Zoology at the Pontifieal Catholie University, Ecuador. He has participated in several herpetology projects across the country and has worked in conservation projects such as the creation of a biological corridor for the Podocarpus National Park, Ecuador. April 2015 | Volume 8 | Number 1 | e91 Amphib. Reptile Conserv. 140 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(1) [Special Section]: 141-142 (e93). Book Review The Amphibians and Reptiles of Mindo: Life in the Cloudforest ^^Howard O. Clark, Jr. and ^''Craig Hassapakis ^Senior Wildlife Ecologist, Garcia and Associates, Clovis, California, USA ^Publisher and Editor, Amphibian & Reptile Conservation; Editor, ErogLog; lUCN SSC Amphibian Specialist Group and Genome Resources Working Group; Provo, Utah, USA Key words. Amphibia, Reptilia, Ecuador, conservation, ecotourism, field researeh, eitizen scientist Citation: Clark HO Jr, Hassapakis C. 2015. Book Review — The Amphibians and Reptiles of Mindo: Life in the Cloudforest. Amphibian & Reptile Conservation 8{t) [Special Section]; 141-142 (e93). Copyright: © 2015 Clark and Hassapakis. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCom- mercialNoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptile Conservation] official journal website . Received: 01 Feburary 2015; Accepted: 14 April 2015; Published: 18 April 2015 Title: The Amphibians and Reptiles of Mindo: Life in the Cloudforest Authors: Alejandro Arteaga, Lucas Bustamante, Juan M. Guayasamin Copyright: 2013 ISBN: 978-9942-13-496-7 Publisher: Universidad Technologica Indoamerica Pages: 258; Price: $49.00 (US) The authors have produced a much needed local field guide for the Mindo parish, located in northwestern Ec- uador and set a high standard for future field guides to follow. The book fills a void with great detail and care. It begins with the Table of Contents, a Forward, and Pref- ace. A brief Symbols and Abbreviation page is followed by the Introduction, which leads into a helpful section on locating and observing reptiles and amphibians in the Mindo region. The authors recommend that those inter- ested in exploring Mindo ’s herpetofauna should do some homework: understand the habitats and environments where herpetofauna can be found, know your subject, keep a low profile, and try not to disturb the sensitive mi- crohabitats in which these unique species are found. Page 1 1 illustrates some principal identification features of the amphibians of Mindo (as a diagram figure; p. 11). On the next few pages are additional figures that show the groin pattern and color of the Mindo rainfrogs (Pristimantis), principal scale types of Mindo lizards, dewlap color of Mindo anoles, and basic terminology for snake scalation. In total the guide features 20 charts and figures, and 228 superb photographs and art work (a special feature and highlight of the book). These figures allow the future ex- plorer of Ecuadoran cloudforests (particularly Mindo) to prepare for an informed and exciting field trip. The crux of the book are the accounts. The guide fea- tures 101 species accounts of Mindo’s unique reptiles and amphibians, with each account accompanied with, as mentioned above, outstanding photos and in addition, a range map. The 228 photos are adequate for identify- ing the target species and have been photographed with a white background, eliminating distracting clutter so the reader can focus on key marks, characteristics, and colors of each species (see Figure 1 for examples). The range maps are up-to-date and reflect the most current research (in total, 4,000 locality records are featured). Each species account has been peer-reviewed by two or more experts (71 total reviewers and hundreds of per- sonal connnunications from experts). The accounts are divided into several key sections: English and Spanish common names, Latin name with describing author and year, recognition information, natural history, distribu- tion, conservation status, etymology, notes, reviewer and contributor information, and references. Prior to the species account sections is the “Plan of the Book” — this section is a must read in that it explains how species ac- counts are set up and discusses the rational of account structure. Additionally, pages 27-29 discusses the Mindo parish; why the area is worthy of continual conservation, and describes the unique characteristics of the region that is home to more than 100 species of reptiles and amphib- ians in an area smaller than the state of Nevada. Correspondence. Email: ^hclark® garciaandassociates.com-, ‘^arc.publisher® gmail.com (Corresponding author). April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 141 e93 Clark and Hassapakis After the species accounts the book presents a de- scription of a new species of Pristimantis found in Mindo (Arteaga et al. 2013). This account illustrates that new species are continually being discovered and regions like Mindo may harbor other new species. Following the new species description is the Glos- sary, Reference section (the book lists 1,935 references throughout), and the final section is “about the authors,” listing the scientific illustrators and geographers (Rita Hidalgo, Silvia Cevallos, and Belen Bans). Overall, the field guide of Mindo is an outstanding contribution to the ever-growing field of herpetology, will help conservation efforts, encourage ecotourism and nature observation, be a high standard for other field guides to follow, among other positive allied outcomes, while linking conservation efforts through its publication. The guide is pleasing to read and should inspire others to write and publish regional guides in species rich areas of the Americas, and, as mentioned several times al- ready, sets a high standard for others to follow. The book emphasizes a warning that species extinction is real, is primarily a result of habitat loss, and areas like Mindo are not necessarily safe. The future is unknown and with the advent of climate change, disease, encroachment, as well as many other detrimental factors not mentioned, we may be witnessing the last sanctuaries for these one- of-a-kind species. We need to do our part to spread the word and conserve what’s left. The publication of a field guide such as this is very important in bringing attention to the great variety of unique species and lending impe- tus to conservation efforts. Field guides like this one are also synergistic in bringing about increased conservation efforts and making a positive impact to curb the unprec- edented rate of habitat loss. We recommend that you sup- port the conservation of Mindo by purchasing the book, learning about Mindo’s amphibians and reptiles, and joining in the conservation efforts of the area (or other similar areas throughout the world) through ecotourism, conservation research (e.g., citizen scientists), and other avenues of endeavors and conservation activism via your individual expertise and enthusiasm to conserve all life on earth, including our own species. No matter who we are (average or exceptional, and all other categories as well) we all can make a positive difference in protecting and conserving earth’s unique and precious life systems and diverse biological life (see also Conrad and Hilchey Fig. 1. Book cover of The Amphibians and Reptiles of Mindo: Life in the Cloudforest. Photo by Howard O. Clark, Jr. 2011; Johnson et al. 2014), for which amphibians and reptiles form an exciting component. Literature Cited Arteaga A, Yanez-Munoz M, Guayasamin JM. 2013. A new frog of the Pristimantis lacrimosus group (An- ura: Craugastoridae) from the montane forests of northwestern Ecuador. Serie de Publicaciones Cienti- ficas 1: 198-210. Conrad CC, Hilchey KG. 2011. A review of citizen sci- ence and community-based environmental monitor- ing: issues and opportunities. Environmental Monitor- ing and Assessment 176(1-4): 273-291. doi: 10.1007/ S10661-010-1582-5 Johnson MF, Hannah C, Acton L, Popovici R, Karanth KK, Weinthal E. 2014. Network environmentalism: Citizen scientists as agents for environmental advo- cacy. Global Environmental Change 29: 235-245. http://dx.doi.Org/10.1016/j.gloenvcha.2014.10.006 Howard O. Clark, Jr. is a Certified Wildlife Biologist® with 20 years of professional wildlife experienee. He focuses his time on the fauna and ecosystems of Northern, Central, and Southern California, and the Mojave Desert. He regularly works with the San Joaquin Kit Fox, Giant Kangaroo Rat, and the Mohave Ground Squirrel. He currently volunteers as the Layout Editor for ']o\xma[ Amphibian & Reptile Conservation. Craig Hassapakis is the publisher and editor of the journal Amphibian & Reptile Conservation (amphibian- reptile-conservation.org); be is also an editor of FrogLog (www.amphibians.org/froglog/) and is a member of tbe lUCN SSC Amphibian Specialist Group (ASG) and volunteer coordinator for the Genome Resources Working Group (ASG/GRWG) in that same organization. April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 142 e93 April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 143 e92 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation 8(2) [Special Section]: 143-162 (e92). Amphibians and reptiles of an agroforestry system in the Colombian Caribbean ^Oscar Angarita-M., ^Andres Camilo Montes-Correa, and ^Juan Manuel Renjifo Grupo de Investigacion en Manejo y Conservacion de Fauna, Flora y Ecosistemas Estrategicos Neotropicales (MIKU), Universidad del Magdalena, COLOMBIA Abstract. — Land-use change is a factor that may alter the assembly of herpetofaunal communities. To determine the effects of land use change, we characterized the herpetofaunal community of “La Gloria Project” in Magdalena, Colombia. Agroforestry crops (Red Gum, Pink Trumpet Tree, Beechwood, and Teak), native forest, wetlands, and built-up zones composing the site. From March to October 2012, we performed eleven field trips, of ten days (eight hours each) for a total sampling effort of 880 hours per observer. We implemented visual encounter surveys and pitfall traps for herpetofauna detection. We recorded 23 amphibian (3,555 individuals) and 37 reptile species (1,088 individuals); the highest diversity for both amphibians and reptiles were found in native forest. Comparing disturbed areas. Teak agroforest presented the highest diversity for both taxa relative to non-natural environments, by factors such as big leaf size, generating conditions to sustenance of some species. However, we demonstrated that short-term differences between natural and non- natural habitats are significant, since there has not been enough time for generalist species to displace the susceptible species and occupy their niches in all vegetation coverages in the study area. Key words. Agroforest, Caribbean lowlands, habitat fragmentation, herpetofaunal concnnunities, tropical dry forest, lower Magdalena River Resumen. — El cambio de usos del suelo es un factor que puede afectar el ensamblaje de las comunidades de herpetofauna. Para determinar los efectos del cambio del uso de suelo, caracterizamos la comunidad de herpetofauna del “Proyecto La Gloria” en el departamento del Magdalena, Colombia. Cultivos agroforestales (eucalipto rojo, roble rosado, melina y teca), bosques natives, humedales y zonas con construccion constituyen el area de estudio. De marzo a octubre de 2012, desarrollamos once salidas de campo de diez di'as (cada uno de echo horas) per un esfuerzo total de muestreo total de 880 horas x observador. Utilizamos busqueda libre per encuentro casual y trampas de caida para la deteccion de herpetofauna. Registramos 23 anfibios (3,555 individuos) y 37 reptiles (1,088 individuos); La mayor diversidad tanto para anfibios como reptiles la encontramos en los bosques natives. Comparando las areas intervenidas, el agrobosque de teca presento la mayor diversidad de ambos taxones con respecto a los otros ambientes no naturales, por factores como el gran tamaho de sus hojas, que generan condiciones para el sostenimiento de algunas especies. Empero, se demuestra que a corto plazo, las diferencias entre los habitats naturales y no naturales son significativas, pues no ha pasado suficiente tiempo para que las especies generalistas desplacen a la especies sensibles y ocupen sus nichos. Palabras clave. Agrobosques, bajo rio Magdalena, bosque seco tropical, comunidades de herpetofauna, fragmentacion de habitat, tierras bajas del Caribe Citation: Angarita-M O, Montes-Correa AC, Renjifo JM. 2015. Amphibians and reptiles of an agroforestry system in the Colombian Caribbean. Am- phibian & Reptiie Conservation 8{2) [Special Section]: 143-162 (e92). Copyright: © 201 5 Angarita-M et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal title Amphibian & Reptiie Conservation] official journal website . Received: 22 January 2015; Accepted: 09 April 2015; Published: 15 April 2015 Correspondence. Email: ^oscarangaritabio@ gmail.conr, ^andresc.montes® gmail.com (Corresponding author); ^juanmanuel. renjifo @ gmail. com April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 144 e92 Angarita-M et al. Introduction Colombia ranks second in taxonomic diversity of am- phibians (785 species) and third in reptiles (593 species) (Acosta-Galvis 2014; Andrade-C. 2011). In the Caribbe- an lowlands 167 reptiles species and 55 amphibians are recorded (Romero-Martmez and Lynch 2012; Carvajal- Cogollo et al 2012). The low diversity of amphibians in the Colombian Caribbean is due to drier conditions of the region, however, the small number of species have morphological, physiological, and behavioral adapta- tions to tolerate drought (Cuentas et al. 2002). Existing surveys include checklists, inventories, and diversity of amphibians and reptiles for the entire region (Dugand 1975; Carvajal-Cogollo et al. 2012; Romero-Martmez and Lynch 2012); as well as the states of Cordoba (Ren- jifo and Lundberg 1999; Carvajal-Cogollo et al. 2007; Carvajal-Cogollo and Urbina-Cardona 2008; Romero- Martmez et al. 2008; Romero-Martmez and Lynch 2010), Sucre (Galvan-Guevara and de la Ossa- Velasquez 2009; Acosta-Galvis 2012b), Bolivar and Atlantico (Cuentas et al. 2002), Cesar (Rueda-Almonacid et al 2008a; b; Me- dina-Rangel 2011; Medina-Rangel et al. 2011), La Gua- jira (Galvis et al. 2011; Blanco-Torres et al. 2013), and Magdalena (Ruthven 1922; Duenez-Gomez et al. 2004; Rueda-Solano and Castellanos-Barliza 2010; Montes- Correa et al. 2015). Many studies were performed in nat- ural areas with wetlands or native forests, with different levels of anthropogenic intervention. Nonetheless, the information on the herpetofauna of dry spots is scarce, and most of the available literature are species descrip- tions, taxonomic reviews of specific groups, or national lists (Acosta-Galvis, 2012a). Deforestation and changes in land-use modify the assembly of amphibian and reptile communities (Cas- tro and Kattan 1991; Garden et al, 2007). The physical transformation of natural environments can cause drastic changes in humidity and temperature, having significant effects in these organisms (Herrera et al. 2004). How- ever, dryland amphibians have several adaptations to sur- vive the lack of water, as the changes in activity patterns and development of wide ranges of dehydration (Thor- son 1995; Cuentas et al. 2002). Moreover, reptiles are more resistant to disturbance as their skin is covered by keratinized scales. Anmiotic eggs make reptiles more tolerant to dehydration and sunstroke (Vargas-Salinas and Bolanos 1999). Even so, the canopy cover, leaf litter cover, and understory density are impor- tant factors for the establishment and distribution of both taxa, since it can determine the movement patterns of these ectothermic animals (Urbina-Cardona et al. 2006). Our goal was to determine the diversity of herpeto- fauna in “La Gloria Project” (Sabanas de San Angel, Magdalena, Colombia), and assess the characteristics and variations of herpetofaunal communities among the various vegetation coverages (Agroforestry crops — Red Gum, Pink Trumpet Tree, Beechwood, and Teak — native forest, wetlands, and built-up zones [any area inhabited by humans] composing the area). Materials and Methods Study site: “La Gloria project” is part of “Reforestadora de la Costa (REEOCOSTA S.A.S.)” organization, within the jurisdiction of the municipality of Sabanas de San Angel, Magdalena department, 30 km from the county seat (10°10’29.2”N; 74°19’38.052”W) (Pig. 1). The study area includes 7,288 hectares, and corresponds to “zonobioma tropical altemohigrico” (tropical dry forest) proposed by Hemandez-Camacho and Sanchez (1992). This locality has a unimodal biseasonal climate with an average annual rainfall of 1,157 mm (Rangel-Ch. and Carvajal-Cogollo 2012). The oldest agroforest is about about 20 years old. Timber is grown in the middle ex- tension of the La Gloria project. The main crop is Teak (Tectona grandis) (21% of the total extent of study area), followed by Red Gum {Eucalyptus tereticornis) (18%). Also grown to a lesser extent is. Pink Trumpet Tree {Ta- bebuia rosea) (7%) and Beechwood (Gmelina sp.) (2%), Caribbean Sea ftVROWA SAMTA M'AkI ik’lDTalj*/ WA.PA. PROYECTO LA GLORIA mi Red Gum □ Pink Trumpet T mi Beechwood ] t Teak Native forest mi Wetland Built-up zones Other Crops CZJi Teak regrowth L_l Pasture ESC: 1:SS.DD0 Leyenda Study area Nationals Parks 1 : 1 . 000,000 Fig. 1. Map of La Gloria Project (taken and modified from Re- focosta 2012). Map developed by HD Granda-Rodriguez. April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 145 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean while the remaining 2% consists of other crops. In addi- tion to agroforests, there is an area of regrowth of Teak (5%), pasture (34%), native forest (10%), and wetlands (1%) (Refocosta 2012). There are also small and scat- tered built-up zones within “La Gloria Project.” Surveys for this study were carried out in areas with agroforest, native forests, wetlands, and urbanized sites. Fieldwork: from March to October of 2012, we made 11 field trips, each one lasting ten days. We used Visual En- counter Surveys (VES) (Crump and Scott 1994). Daily, a single person did random walks for eight hours (09:00- 12:00, 14:00-17:00, and 19:00-22:00 h), for a total sam- pling effort of 880 hours x observer. In addition, we cap- tured cryptic species with terrestrial, semifosorial, and fossorial habits with pitfall traps (Vogt and Mine 1982), eight trap systems per habitat during each survey (56 in total). These traps system consist of two 3.78 liters buck- ets, and a two m interception net between them. Traps remained open for ten days. We used a 10% chlorobutanol solution to euthanize all amphibians captured and intrathoracic lidocaine injec- tions for euthanizing reptiles. No turtles or crocodilians were sacrificed for this study. All vouchered specimens were deposited in the Centro de Colecciones Biologicas de la Universidad del Magdalena (CBUMAG:REP and CBUMAG:ANF acronym). The scientific nomenclature used in this contribution is that accepted by Uetz et al. (2014) and Frost (2014). Data analysis: Relative abundance was calculated as the number of individuals in each sample relative to capture effort, expressed in individuals/hours x observer (RA= Ind/h X obs.) (Lips 1999). Species were qualified accord- ing to their relative abundance in “very rare” (VR) if it was observed between 0.1-0.24 individuals per hour x observer; “rare” (R) if it was observed between 0.25- 0.49; “common” (C) if it was observed between 0.50- 0.74; “abundant” if it was observed between 0.75-0.99; and “very abundant” if it was observed between 1.0 or more (Rueda-Solano and Castellanos-Barliza 2010). Us- ing PRIMER 6 (v 6.1.11) (Clarke and Corley 2001) we calculated Margalef Richness Index (d), Pielou Unifor- mity Index (J’), Shannon- Wiener Diversity Index (H’), and Simpson Dominance Index (X) for each vegetal coverage. We built a Bray-Curtis Similarity Matrix of non-transformed amphibian and reptile abundance data, to generate a nonparametric one-way similarity analysis (ANOSIM) (999 permutations), in order to refute a null hypothesis when there were no significant differences be- tween diversity of amphibians and reptiles among sites. We made dendrograms with the same Bray-Curtis Ma- trix, to evaluate the similarity among vegetal coverages within the study area; likewise, the similarity between La Gloria project and other localities with published inven- tories of amphibians and reptiles in the Colombian Carib- bean. It should be noted that if the similarity was greater than 50%, it was considered a homogenous cluster. We used the software Estimates (v 9.1.0) (Coldwell 2013) to create a species accumulation curve from non-parametric qualitative estimators Chao 2, Bootstraps, Jacknife 1, and Jacknife 2 (randonfized 999 times for each case) to quan- tify the representativeness of the sample. We also calcu- lated the unique and duplicates species. Results and Discussion Representativeness of survey: Bootstraps, Chao 2, Jack- nife 1, and Jacknife 2 estimators show that amphibian survey had representativeness among 24.83% to 28.95%. The Chao 2 curve was only one who got stabilization. The unique and duplicated species were not reduced during the survey (Fig. 2a). Furthermore, the reptile sur- veys had more representativeness, since the estimators reached among 39.79% to 45.94%. The Chao 2 and Jack- nife 2 curve obtained asymptote. In this case, unique and duplicates species neither decreased (Fig. 2b). Jacknife 1 and Jacknife 2 estimators have higher values, suggesting that surveys had a low representativeness in both taxa (Carvajal-Cogollo and Urbina-Cardona 2008). Boot- straps estimator obtained close values with observed species. Taking this as a reliable algorithm to estimate total richness, amphibian and reptile surveys reached a representativeness of 24.83% and 39.79% respectively. A comparison of survey methods used (observational surveys [VES] and trapping) results in a greater number of species and abundance being obtained through VES. (Fig. 3a, b). Using this technique, we detected 92.31% 0 20 40 eo so 100 120 #3 Ivfean(nms) ■UmqufisMfian * Duplicates Mean Chao 2 Mean KJack 1 Mean • Jack 2 Mean -l-BootstrapMfean Fig. 2. Cumulative curve species of la Gloria project. April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 146 e92 Angarita-M et al. Table 1. Relative abundance = RA, VA = very abundant, A = abundant, C = common, R = rare, VR= very rare, NA = not avail- able, and vegetation coverage, RG = Red Gum, PTT = Pink Trumpet Tree, BW= Beechwood, T = Teak, NF = native forest, WL = wetland, BZ = built-up zones. CBUMAG = Centro de Colecciones Biologicas de la Universidad del Magdalena (ANF = amphibian; REP = reptile). TAXA GT PTT BW T NF WL BZ RA Voucher CLASS AMPHIBIA CBUMAG:ANF Order Anura Family Bufonidae Rhinella marina (Linnaeus 1758) X X X X X X R 699 Rhinella humboldti (Gallardo 1965) X X X X X VR 701 Family Ceratophrydae Ceratophrys calcarata (Boulenger 1890) Family Hylidae X VR 672 Dendropsophus microcephalus (Cope 1886) X X VA 713 Dendropsophus ebraccatus (Cope 1874) X VR 00666-67 Hypsiboas pugnax (Schmidt 1857) X X X VA 00697-8 Hypsiboas crepitans (Wied-Neuwied 1824) X VR 30 Scarthyla vigilans (Solano 1971) X X VA 718 Scinax rostratus (Peters 1863) X X VR 00031-32, 49 Scinax “x-signatus” (Spix 1824) X X X X X R 15 Trachycephalus typhonius (Linnaeus 1758) X X VR 696 Phyllomedusa venusta Duellman and Trueb 1967 X VR 676 Pseudis paradoxa (Linnaeus 1758) Family Leptodactylidae X VR Leptodactylus fuscus (Schneider 1799) X X X X X R 00703-4 Leptodactylus insularum Barbour 1906 X X X X R 00693, 695, 700 Leptodactylus poecilochilus (Cope 1862) X X VR 348 Leptodactylus fragilis (Brocchi 1877) X VR Engystomops pustulosus (Cope 1864) X X X X X R 00708,711,716 Pleurodema brachyops (Cope 1869) X X X X C 00702, 705 Pseudopaludicola pusilla (Ruthven 1916) Family Microhylidae X X X X C 00709, 717 Elachistocleis panamensis (Dunn, Trapido, and Evans 1948) X X VR 719 Elachistocleis pearsei (Ruthven 1914) Order Gymnophiona Family Caecilidae X X X X VR 00710, 720 Caecilia subnigricans Dunn 1942 X VR 634 CLASS REPTILIA CBUMAG:REP Order Squamata Family Sphaerodactylidae Gonatodes albogularis (Dumeril and Bibron 1836) X X X X X VR 236 Lepidoblepharis sanctaemartae (Ruthven, 1916) X X VR Family Gekkonidae Hemidactylus frenatus (Dumeril and Bibron 1836) Family Phyllodactylidae X VR 237 Thecadactylus rapicauda (Houttuyn 1782) Family Iguanidae X X VR Iguana iguana (Linnaeus 1758) Family Dactyloidae X VR Anolis auratus Daudin 1 802 X VR 231 April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 147 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean Table 1 (Continued). Relative abundance = RA, VA = very abundant, A = abundant, C = common, R = rare, VR = very rare, NA = not available), and vegetation coverage, RG = Red Gum, PTT = Pink Trumpet Tree, BW = Beecbwood, T = Teak, NF = native for- est, WL = wetland, BZ = built-up zones. CBUMAG = Centro de Colecciones Biologicas de la Universidad del Magdalena. TAXA GT PTT BW T NF WL BZ RA Voucher Family Corytophanidae Basiliscus basilicus (Linnaeus 1758) Family Scincidae X VR Maracaiba zuliae (Miralles, Rivas, Bonillo, Schargel, Barros, Garcia-Perez, and Barrio- Amoros 2009) X X X X VR 235 Family Gymnophthalmidae Leposoma rugiceps (Cope 1869) X X X VR 239 Tretioscincus bifasciatus (Dumeril 1851) Family Teiidae X X X X VR 00232-33 Cnemidophorus gaigei Ruthven 1915 X X R Ameiva praesignis (Baird and Girard 1852) X X R Ameiva bifrontata Cope 1862 Family Anomalepididae X R Liotyphlops albiwstris (Peters 1857) Family Boidae X VR 194 Boa constrictor Linnaeus 1758 X VR Epicrates maurus Gray 1849 Family Colubridae X VR 234 Chironius spixii (Hallowell 1845) X X VR 120 Tantilla melanocephala (Linnaeus 1758) X VR 00208, 210 Leptophis ahaetulla (Linnaeus 1758) Family Dipsadidae X VR 10 Leptodeira annulata (Linnaeus 1758) X X VR 34 Leptodeira septentrionalis (Kennicott 1859) X X X VR Lygophis lineatus (Linnaeus 1758) X X VR Pseudoboa neuwiedii (Dumeril, Bibron, and Dumeril 1 854) X X VR 91 Imantodes cenchoa (Linnaeus 1758) X X VR 16 Thamnodynastes gambotensis Perez-Santos and Moreno 1989 X X NA 232 Thamnodynastes paraguanae Bailey and Thomas 2007 X X NA 38 Helicops danieli Amaral 1938 X VR 128 Oxyrhopus petolarius (Linnaeus 1758) X VR 238 Xenodon rabdocephalus (Wied 1824) Family Viperidae X VR 00170-71 Crotalus durissus Linnaeus 1758 X VR Porthidium lansbergii (Schlegel 1841) X VR 74 Bothrops asper (Garman 1883) Family Elapidae X X VR 165 Micrurus dissoleucus (Cope 1860) X VR Order Testudines Family Chelidae Mesoclemmys dahli (Zangerl and Medem 1957) Family Emydidae X VR Trachemys callirostris (Gray 1855) X VR Eamily Testudinidae Chelonoidis carbonarius (Spix 1824) Order Crocodylia Eamily Alligatoridae X X VR Caiman crocodilus (Linnaeus 1758) X A Amphib. Reptile Conserv. 148 April 2015 | Volume 8 | Number 1 | e92 Angarita-M et al. of amphibian and 68.48% of reptile individuals, respec- tively. With VES, we recorded 21 amphibian species and 35 reptile species. With this method we recorded 25 exclusive species (10 amphibians and 15 reptiles), that are strictly arboreal or aquatic. Conversely, we captured 7.69% and 31.52% of amphibian and reptile individuals respectively, using pitfall traps. This method recorded 12 amphibian species and 19 reptile species. We only found two fossorial species {Elachistocleis pearsei and Micru- rus dissoleucus) with pitfall traps. Amphibians: A total of 3,555 individuals, corresponding to two orders, six families, and 23 species (Table 1), were recorded. Anurans found represented five families and 22 species (37% of the total herpetofauna of the area) (Fig 4); a single caecilian specie was encountered (Fig. 4). Forty-two percent (41.8%) of lowland amphibian species occurring in the Colombian Caribbean were observed at Fa Gloria Project. The absence of expected species is due to a lack of specialized capture methods. For example, Typhlonectes natans is rarely observed due to its cryp- tic aquatic habits despite being distributed throughout the Caribbean region of the upper Magdalena-Cauca River (Tapley and Acosta-Galvis 2010). However, in this study we report the first record of the Clown Treefrog (Dendropsophus ebraccatus) in the lower Magdalena River, for which the nearest known distribution is in Rio Manso, Cordoba (Cochran and Coin 1970). In this con- tribution, we prefer to name Scinax "‘x-signatus’' instead Scinax ‘"rubef' (as was known previously Renjifo and Fundberg 1999; Cuentas et al. 2002). This is due to unre- solved controversy regarding its taxonomy and biogeog- raphy (Barrio- Amoros 2004; Acosta-Galvis et al. 2006; Barrio-Amoros et al. 2011; Acosta-Galvis et al. 2012a). Following Rivero’s (1969) criteria, the absence of dark dorsolateral lines and head equally long as wide place the collected specimens within the x-signatus and ruber groups. Reptiles: We recorded 1,088 specimens corresponding to three orders, 19 fanfilies, and 37 species (Table 1). The most diverse order was Squamata with 15 fanfilies and 32 species, the suborder Facertilia was the richest with nine families and 13 species, 20% of the total herpeto- fauna of the area. The suborder Serpentes represented six fanfilies and 20 species (34%). We observed three families and three species of turtles (5%) and recorded one crocodilian species (2%) (Fig. 3). Fa Gloria Project harbors 21.8% of lowlands reptile species of the Colom- bian Caribbean. We found three endenfic species from Colombia, Helicops danieli, Thamnodynastes gamboten- sis, and M. dahli, the latter with restricted distribution in the Colombian Caribbean (Rossman 2002; Bailey and Thomas, 2007; Carvajal-Cogollo et al. 2012; Forero-Me- dina et al. 2013). The presence of M. dahli in the study area was unexpected, as species distribution models by Forero-Medina et al. (2012) propose a low probability a)Amphibians 100 b) Reptiles so 70 Richness Abundance (%) Exclusive species ■ PitfaUTiape iVES Fig. 3. Comparisons between the methods used for herpeto- fauna recording and capturing. Squainata: Seipentes 34 % Testudines Crocodylia 5 % * 2 % Squamata: Lacertlia 20 % Fig. 4. Herpetofauna composition percentage in la Gloria Proj- ect. of occurrence within this region. However, this area has many first-order streams with abundant riverine vegeta- tion, throughout native forests and agroforests, habitat characteristics of this species (Forero-Medina et al. 2011; Montes-Correa et al. 2014). In addition, we report the first record of Maracaiba zuliae in the lower Magdalena River, an expansion of its currently known distribution. This species was recently reported in Colombia in Reserva Forestal Protectora Montes de Oca, Fa Guajira state (Galvis et al. 2011). Several records by Ruthven (1922) in the Barbacoas Riv- er, the Arenas Stream, and Fas Pavas must correspond with this recently described species. Fikewise, we report the first record of Thamnodynastes paraguanae in the re- gion. In Colombia, this snake is only known from Fa Gu- April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 149 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean a) Richness 30 RG PTT BW T NF WL BZ ■ Amphibiaiis ■ Reptiles Fig. 5. Richness (a) and abundance (b) of amphibians and rep- tiles between habitats (RG = Red Gum, PTT = Pink Trumpet Tree, BW = Beechwood, T = Teak, NF = native forest, WL = wetland; BZ = built-up zones). ajira: Uribia, Riohacha, and Reserva Forestal Protectora Montes de Oca (Bailey and Thomas 2007; Galvis et al. 2011). We must clarify that while the fieldwork was de- veloped, snakes of Thamnodynastes genus were treated as one species, and they are not included in this analysis because their relative abundance is not available. Richness and abundance patterns: in La Gloria Project, the native forest was the habitat that hosted the greatest number of species (Fig. 5a), 18 amphibians and 26 rep- tiles. The wetlands were the second habitat in amphib- ian composition, while the Teak agroforest was second in number of reptiles. Remaining habitats had less or equal to 10 species, both for amphibians and reptiles. We verified the greatest abundance in native forest (Fig. 5b), as 65.63% of amphibians and 48.35% of reptiles were detected in this habitat. All non-natural habitats scored an abundance below 10%. Some studies show that abun- dance patterns of natural and non-natural environments are similar (Gardner et al. 2007; Carvajal-Cogollo and Urbina-Cardona et al. 2008). Over time, composition and abundance tend to homogenize by dominance of the generalist species that displace more sensitive species for their lower habitat requirements and increased toler- ance to disturbance (Offerman et al, 1995; Laurance et al. 2002). Surely, La Gloria Project does not present homog- enization because agroforests are very recent. Regarding the qualitative relative abundance in amphibians, we ob- served three very abundant species, two common, five rare, and 13 very rare. Dendropsophus microcephalus, Scarthyla vigilans, and Hypsiboas pugnax were the most abundant amphibians, while Caecilia subnigricans is represented by a single individual. Moreover, in reptiles we observed one abundant species, three rare, 31 very rare, and two not available. The most abundant species of reptiles were Caiman crocodilus, Ameiva bifrontata, mdAmeiva praesignis. Furthermore, Mesoclemmys dah- li and Micrurus dissoleucus were observed for a single individual. Similarly, other studies of tropical dry forest herpetofauna, found over half species had low relative abundance (Rueda-Solano and Castellanos-Barliza 2010; Pedroza-Banda and Angarita-Sierra 2011). In addition, snakes present a lower detection, possibly due to their cryptic habits or low abundance. Leptodeira annulata and Leptodeira septentrionalis were the most common snakes throughout the study area, supporting Scott and Seigel (1992) and Dodd (1993) hypotheses, where small sized snakes are more tolerant to disturbance, therefore, possibly more abundant. As to the connnunity attributes (Table 2), native forest had the highest Margalef Rich- ness and Shannon-Wiener Diversity for amphibians and reptiles and Beechwood agroforest had the greatest Pielou Uniformity Value. For these three attributes, built- up areas showed the lower values, however, this habi- tat had dominance for the highest values. In this study, the higher value of Margalef Richness, Shannon- Wiener Diversity, and Pielou Uniformity created higher values obtained for the coverage of floristic and structural com- plexity. A similar pattern was observed in Zapatosa re- gion by Medina-Rangel (2011). Habitat comparisons and herpetofaunal autoecology: ANOSIM determined there are global composition and abundance differences between seven evaluated habitats (p- value = 0.502). However, there are specific differ- ences between Red Gum agroforest and Pink Trumpet Tree agroforest (p-value = 0.006), Red Gum and Beech- wood (p-value = 0.038), Red Gum and Teak (p-value = 0.161), Pink Trumpet Tree and Beechwood (p- value = 0.068), Pink Trumpet Tree and Teak (p-value = 0.012), and Beechwood and Teak (p-value = 0.357). These simi- larities among agroforests are due to sharing among pio- neer and generalist species that are able to tolerate condi- tions imposed by the new environment (Luja et al. 2008), e.g.. Nest-building Frogs {Leptodactylus) (Heyer 1969). Some of these can be considered as connnon colonizers (see also, Duenez-Gomez et al. 2004). In La Gloria Project, the herpetofauna composition was quite heterogeneous, thus, all clusters were below 50% similarity (Fig 6). The more similar habitats were the Teak and Pink Trumpet Tree (48.5% similarity). Likewise, native forests and wetlands have a cluster (42.6%) and Beechwood and Red Gum agroforest an- April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 150 e92 Angarita-M et al. BW RG T prr WL NF BZ 0 20 40 60 80 100 Similarity Fig. 6. Bray-Curtis similarity dendrogram between habitats in la Gloria Project (RG = Red Gum, PTT = Pink Trumpet Tree, BW = Beechwood, T = Teak, NF = native forest, WL = wetland; BZ = built-up zones). Universidad del Magdalea Besotes Neguanje Cordoba wetlands La Gloria Montes de Marfa and La Caimanera Zapatosa Atlantico and Bolivar El Botillero Montes de oca Rancherfa Urra Coraza Murrucucu lowlands 1 ^ ^ ^ 1 20 40 60 80 100 Similarity Fig. 7. Similarity of amphibian richness between La Gloria project and others inventories in Caribbean lowlands. Humedales del Cordoba (Romero-Martrnez and Lynch 2010); Montes de Marfa and Cienaga la Caimanera (Acosta-Galvis 2012b); El Botil- lero (Duenez-Gomez et al. 2004); Cienaga del Zapatosa (Medina-Rangel et al. 2011); Atlantico and north Bolivar (Cuentas et al. 2002); Montes de Oca (Galvis et al. 2011); Rancherfa (Blanco-Torres et al. 2013); Urra (Renjifo and Lundberg 1999); Los Besotes (Rueda-Almonacid et al. 2011a); Serranfa de Coraza (Galvan-Guevara and de la Ossa- Velasquez 2009); Universidad del Magdalena (Montes-Correa et al. 2015); Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); Cerro de Murrucucu lowlands (Romero-Martfnez et al. 2008). other (36.8%). The more dissimilar habitat is the built-up zone with 3.1% similarity with respect to other habitats. The species with greater frequency of occurrence was Rhinella marina, which was present in six of the seven evaluated habitats. This species has ecological plastic- ity and is able to tolerate highly degraded environments, including benefiting from human activities (Zug and Zug 1979). On the other hand, we found 28 exclusive species from a single cover. For example, Pseudis para- doxa. Caiman crocodilus, and Trachemys callirostris are strictly aquatic species and only found in wetlands. The exclusivity of Hemidactylus frenatus is due to its strong synanthropy (Caicedo-Portilla and Dulcey-Cala 2011). Phyllomedusa venusta and Trachycephalus typhonius were exclusive of native forests, since these organ- isms have behavioral adaptations to tolerate prolonged drought in these habitats (Cuentas et al. 2002). The richness and abundance of amphibians in La Glo- ria project was higher in native forests and their nearby wetlands. Moreover, in the Red Gum agroforest, amphib- ian richness and abundance was lower due to the sparse canopy of this tree which allows more sunlight to reach the forest floor, similar to what Gardner et al. (2007) reported for Brazil. In Indonesia, Wanger et al. (2009) Amphib. Reptile Conserv. 151 April 2015 | Volume 8 | Number 1 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean Zapatosa Humedales del Cordoba El Botillero La Gloria Urra Coraza Universidad del Magdalena Neguanje Montes de Oca Rancheria Besotes 50 60 70 80 90 Similarity 100 Fig. 8. Similarity of reptile richness between La Gloria project and others inventories in Caribbean lowlands. Humedales del Cordoba (Carvajal-Cogollo et al. 2); El Botillero (Duenez-Gomez et al. 2004); Cienaga del Zapatosa (Medina-Rangel et al. 2011); Montes de Oca (Galvis et al. 2011); Rancheria (Blanco-Torres et al. 2013); Urra (Renjifo and Lundberg 1999); Los Besotes (Rue- da- Almonacid et al. 2011b); Serrama de Coraza (Galvan-Guevara and de la Ossa- Velasquez 2009); Universidad del Magdalena (Montes-Correa et al. 2015); Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010). found that amphibians are more abundant in native rain- forests than in Cacao Tree agroforest. In Gorgona Island, amphibians were more abundant in little disturbed rain- forests than in palm cultivation (Urbina-Cardona and Londono-Murcia 2003). On the other hand, in mountain rainforest, amphibian composition and abundance were higher in open areas that agroforest and native forests (Hoyos-Hoyos et al. 2012). Canopy coverage may not be as important to some reptiles. Wanger et al. (2009) found that reptile richness and abundance was similar in Cacao Tree agroforest, na- tive rainforests and open areas, and even these showed greater richness in open areas than in native rainforests. In Gorgona Island, reptile richness was higher in second- ary forests; nevertheless, were more abundant in dis- turbed areas than in primary forests (Urbina-Cardona and Londono 2003). In La Gloria project, the reptile richness was higher in native forests, although we recorded sev- eral species in agroforests, mainly in Teak; this because large leaves of this tree generate heavy shade and leaf-lit- ter layers able to generate favorable microclimatic condi- tions for herpetofaunal establishment. In other agroforest reptile composition and abundance was low due to thin canopy cover and insufficient leaf-litter depth. In the case of Red Gum agroforest, the leaf-litter layer is very poor, as this tree is perennial. Changes of leaf-litter dynamics can alter amphibian and reptile assembly (Whitfield et al. 2014). In La Gloria project, there are typical species of forest formations, but not necessarily exclusive of native forest. For example, Lepidoblepharis sanctaemartae occurred in native forest and Teak agroforest, being slightly more abundant in the native forest; L. sanctaemartae, as other small leaf-litter geckos, requires a leaf-litter layer con- taining humidity and little light penetration through the canopy, because of their passive thermoregulatory strat- egy (Vitt et al. 2005). Because of this aspect, L. sanc- taemartae was not present in Red Gum agroforest. This species is a good model of Garden et al. (2007) hypoth- eses, since a dense canopy and a humid leaf-litter layer are more important for this species persistence than for- est vegetation composition. Therefore, L. sanctaemartae is abundant both in preserved native forests as agroforest with sufficient coverage canopy and leaf-litter humidity (Montes-Correa pers. obs.). The tortoise Chelonoidis carbonarius was present al- most exclusively in native forest, where there is avail- able fruit, which makes up much of their diet (Rueda- Almonacid et al. 2007). A single individual was recorded in Pink Trumpet Tree agroforest, feeding on flowers of this tree in breeding season, which are also an important part of their diet (Moskovits and Bjomdal 1990). We did not find this species in other agroforests since the tim- ber cultivation does not offer alimentary resources. The slider turtle Trachemys callirostris was more abundant in wetlands with open areas on its banks, as these offered sites for nesting (Moll and Legler 1971). The Spectacled caiman (C. crocodilus) was very abundant, being present in all wetlands in the zone. The low metabolic rate and generalist feeding habits allow them to maintain populations in areas with small and dis- perse wetlands (Castro-Herrera et al. 2013). Likewise, it is possible that the extermination of Crocodylus acutus in the lower Magdalena River has favored the increasing April 2015 I Volume 8 | Number 1 Amphib. Reptile Conserv. 152 e92 Angarita-M et al. Table 2. Attributes of amphibians and reptiles communities in the habitats of La Gloria project (d = Margalef richness, J’ = Pielou Uniformity, H’ = Shannon- Wiener Diversity, X = Simpson Dominance). Red Gum Pink Trumpet Tree Beechwood Teak Native Forest Wetiands Buiit-up Zone Amphibians d 2.03 1.11 0.92 1.33 2.19 2.18 0.8 J' 0.85 0.81 0.93 0.77 0.68 0.55 0.75 H' 0.85 0.57 0.56 0.65 0.85 0.67 0.36 X 0.17 0.31 0.3 0.27 0.2 0.32 0.51 Reptiles d 0.96 1.55 1.12 3.3 3.83 0.69 0.22 J' 0.82 0.92 0.92 0.84 0.7 0.23 0.31 H' 0.39 0.72 0.44 1.01 0.98 0.16 0.09 X 0.47 0.21 0.39 0.12 0.17 0.85 0.89 populations of C. crocodilus. A similar situation occurred in Venezuelan Llanos with Crocodylus intermedius ex- termination (Medem 1981). Compared to other inventory studies in the Colom- bian Caribbean lowlands. La Gloria project presented similarity in richness of amphibians with other inven- tory studies in areas with abundant wetlands (Fig. 6); it showed the highest similarity with the Humedales del Cordoba (Romero and Lynch 2010) (85.7% similarity). Although they agreed in many lowlands species, forest formations are scarce in Cordoba Wetlands, thus, in La Gloria project forest species such as Phyllomedusa ve- nusta were present, while in Cordoba Wetlands it was not reported. There is another great cluster with the lo- calities of La Guajira. Studies made in Urra (Renjifo and Lundberg, 1999), Coraza (Galvan-Guevara and de la Ossa- Velasquez 2009), and Murrucucu (Romero et al. 2008) suggest the area of influence of the Sinu River has many common elements with the Cordillera Occidental, biogeographic Choco, and Central America, (v. gr. Co- lostethus pratti, Strabomantis bufoniformis, Bolitoglossa biseriata, and Oscaecilia polizona). Clustering between Neguanje (Rueda-Solano and Castellanos-Barliza 2010), and Universidad del Magdalena (Montes-Correa et al. 2015) and Besotes (Rueda-Almonacid et al. 2008a) is due to the typical elements of tropical dry forest and the Sierra Nevada de Santa Marta (as Colostethus ruthveni, Cryptobatrachus boulengeri, and Allobates sp.). In reptiles, La Gloria project is very similar to other areas of lowlands with wetlands, presenting the most similarity between Humedales del Cordoba (Carvajal- Cogollo et al. 2007) and Cienaga del Zapatosa (Medi- na-Rangel et al. 2011) (69.8% similarity) (Fig. 7). This evident clustering of the lowlands is very similar to the localities in La Guajira but differs from typical elements from northeastern Caribbean, as Gonatodes vittatus, Ba- chia talpa, and Thamnodynastes paraguanae. The west- ern regions are very dissimilar to La Gloria project by having typical elements of biogeographic Choco, as Che- lydra acutirostris and Anolis vittigerus (Medem 1977; Castro-Herrera and Vargas-Salinas 2008). Conclusions This study shows that connnunities of amphibians and reptiles are affected by structural changes in forests, since cultivated timber does not provide the necessary microhabitats to sustain many elements of herpetofauna species. The introduction of agroforests results in al- terations of the spatial distribution of species, restricting them to small remnants of native forest. A greater problem of studies of amphibians and rep- tiles in the Colombian Caribbean is that the predominant information is unpublished literature and the method- ologies unclear (Blanco-Torres et al. 2013). This study contributes to the state of knowledge of amphibian and reptile richness in the lower Magdalena River, provid- ing three new records for the region and establishes a list from a standardized inventory. Acknowledgments. — We thank the company of Re- focosta S.A.S. for allowing us to conduct our studies. We also thank our friends of the class of Herpetology 2012-1: Katherin Linares, Stefanny Barros, Ricardo Martinez, and Karen Vega and also to our friends of the Herpetology Lab of Magdalena University: Danny Ver- gara, Juan Jimenez, Efram Rada, Miguel Arevalo, Mar- tin Caicedo, Heman Granda Rodriguez, Carlos Villa de Leon, Liliana Saboya, Danilo Vergara, and Caitlin Webb (and for reviewing the manuscript). Special mention goes to colleagues John D. Lynch, Julio Mario Hoyos, Cesar Barrio Amoros, German Lorero Medina, Victor Acosta Chaves, Andres R. Acosta Galvis, Paulo Tigreros, and Luis Duarte and for their contributions to the manuscript. Linally, we thank the Centro de Colecciones Biologicas de la Universidad del Magdalena for their support and protection of all our vouchers. Literature Cited Acosta-Galvis AR, Huertas-Salgado C, Rada M. 2006. Aproximacion al conocimiento de los anflbios en una localidad del Magdalena medio (departamento de Caldas, Colombia). Revista de la Academia Colombi- April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 153 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean ana de Ciencias Exactas, Fisicas y Naturales 30(115): 291-303. Acosta-Galvis AR. 2012a. Anfibios de los enclaves secos en la ecorregion de La Tatacoa y su area de influencia, alto Magdalena, Colombia. Biota Colombiana 13(2): 182-210. Acosta-Galvis AR. 2012b. Anfibios de los enclaves secos del area de influencia de los Montes de Maria y la Cie- naga Caimanera, Departamento de Sucre, Colombia. Biota Colombiana 13(2): 211-231. Acosta-Galvis AR. 2014. Lista de los anfibios de Colom- bia. V.03.2014. Batrachia, Villa de Leyva, Boyaca, Colombia. Available: http://www.batrachia.com/ [Ac- cessed: 30 December 2014]. Andrade-C. MG, 2011. Estado del conocimiento de la biodiversidad en Colombia y sus amenazas. Consider- aciones para fortalecer la interaccion ambiente-polfti- ca. Revista de la Academia Colombiana de Ciencias Exactas, Eisicas y Naturales 35(137): 491-507. Bailey JR, Thomas RA. 2007. A revision of the South American snake genus Thamnodynastes Wagler, 1830 (Serpentes: Colubridae, Tachymenini). II. Three new species from northern South America, with further descriptions of Thamnodynastes gambotensis Perez- Santos and Moreno and Thamnodynastes ramonriv- eroi Manzanilla and Sanchez. Memoria de la Eun- dacion La Salle de Ciencias Naturales 166: 7-27. Barrio-Amoros CL. 2004. Amphibians of Venezuela sys- tematic list, distribution and references, an update. Revista de Ecologia Latinoamericana 9(3): 1^8. Barrio-Amoros CL, Brewer-Carfas C, Fuentes-Ramos O. 2011. Aproximacion preliminar a la herpetocenosis de un bosque pluvial en la seccion occidental de la Sierra de Lema, Guayana Venezolana. Revista de Ecologia Latinoamericana 16(1): 1^6. Blanco-Torres A, Baez L, Patino-Flores E, Renjifo JM. 2013. Herpetofauna del valle medio del rfo Rancherfa, La Guajira, Colombia. Revista de Biodiversidad Neo- tropical 3(2): 113-122. Caicedo-Portilla R, Dulcey-Cala CJ. 2011. Distribucion del gecko Hemidactylus frenatus (Dumeril y Bibron 1836) (Squamata: Gekkonidae) en Colombia. Biota Colombiana 12(2): 45-56. Carvajal-Cogollo JE, Urbina-Cardona IN. 2008. Patrones de diversidad y composicion de reptiles en fragmentos de bosque seco tropical en Cordoba, Colombia. Tropi- cal Conservation Science 1(4): 397^16. Carvajal-Cogollo JE, Castano-Mora OV, Cardenas- Arevalo G, Urbina-Cardona JN. 2007. Reptiles de areas asociadas a humedales de la planicie del de- partamento de Cordoba, Colombia. Caldasia 29(2): 427^38. Carvajal-Cogollo JE, Cardenas- Arevalo G, Castano-Mo- ra OV. 2012. Reptiles de la region Caribe de Colom- bia. Pp. 791-812 In: Colombia diversidad biotica XII: La Region Caribe de Colombia. Editor, Rangel-Ch. JO. Instituto de Ciencias Naturales, Universidad Na- Amphib. Reptile Conserv. cional de Colombia, Bogota, Colombia. 1046 p. Castro H, Kattan G. 1991. Estado de conocimiento y conservacion de los anfibios del Valle del Cauca. Pp. 310-323 In: Memorias Primer Simposio Nacional de Fauna del Valle del Cauca. Editors, Florez E, Kattan G. INCIVA, Cali, Colombia. 381 p. Castro-Herrera F, Vargas-Salinas F. 2008. Anfibios y rep- tiles en el departamento del Valle del Cauca, Colom- bia. Biota Colombiana 9(2): 251-277. Castro-Herrera R, Valencia- Aguilar A, Villaquiran D. 2013. Evaluacion de la poblacion de babillas en los humedales del sur-occidente de la isla Gorgona, Pacf- fico colombiano. Herpetotropicos 9(1-2): 19-23. Clarke KR, Gorley RN. 2001. PRIMER User Manual: Plymouth Routines in Multivariate Ecological Re- search. Plymouth Marine Laboratory, Plymouth, United Kingdom. 91 p. Cochran DM, Coin CJ. 1970. Frogs of Colombia. Smith- sonian Institution, United States National Museum 288. Washington D.C., USA. 665 p. Colwell RK. 2013. Estimates, Version 9.1: Statistical Estimation of Species Richness and Shared Species from Samples. Available: http://viceroy.eeb.uconn. edu/estimates/EstimateSPages/EstSUsersGuide/Esti- mateSUsersGuide.htm [Accessed: 01 January 2015]. Crump ML, Scott NG. 1994. Standard techniques for inventory and monitoring. Pp. 77-171 In: Measur- ing and Monitoring Biological. Standard Methods for Amphibians. Editors, Heyer MA, Donelly RW, McDi- armid LA, Hayek C, Foster MS. Smithsonian Institu- tion Press, Washington D.C, USA. 364 p. Cuentas D, Borja R, Lynch JD, Renjifo JM. 2002. Anuros del Departamento del Atldntico y Norte del Bolivar. Cencys 21, Barranquilla, Colombia. 117 p. Dodd CK, 1993. Strategies for snake conservation. Pp. 363-393 In: Snakes: Ecology and Behavior. Editors, Sigel RA, Collins JT. McGraw-Hill, New York, New York, USA. 414 p. Duenez-Gomez F, Munoz-Guerrero J, Ramfrez-Pinilla MR 2004. Herpetofauna del corregimiento Botillero (El Banco, Magdalena) en la depresion Momposina de la region Caribe colombiana. Actualidades Bi- ologicas 26(81): 161-170. Dugand A. 1975. La serpentifauna de la llanura costera del Caribe. Caldasia 11(35): 63-82. Forero-Medina G, Cardenas-Arevalo G, Castano-Mora OV. 2011. Abundance, home range, and movement patterns of the endemic species Dahl’s Toad-headed turtle {Mesoclemmys dahli) in Cesar, Colombia. Che- lonian Conservation and Biology 10(2): 228-236. Forero-Medina G, Cardenas-Arevalo G, Castano-Mora OV. 2012. Habitat modeling of Dahl’s toad-headed turtle {Mesoclemmys dahli) in Colombia. Herpetolog- ical Conservation and Biology 7(3): 313-322. Forero-Medina G, Castano-Mora OV, Cardenas-Areva- lo G, Medina-Rangel GF. 2013. Mesoclemmys dahli (Zangerl and Medem 1958), - Dahl’s Toad-Headed April 2015 | Volume 8 | Number 1 | e92 154 Angarita-M et al. Turtle, Carranchina, Tortuga Montanera. Pp. 069.1- 069.8 In: Conservation Biology of Freshwater Turtles and Tortoises: A Compilation Project of the lUCN/ SSC Tortoise and Freshwater Turtle Specialist Group. Editors, Rhodin AGJ, Pritchard PCH, van Dijk PP, Saumure RA, Buhlmann KA, Iverson JB, Mittermeier RA. Chelonian Research Monographs No. 5. 1,182 p. Frost D. 2014. Amphibian Species of the World: An On- line Reference. Version 6.0. Available: http://research. amnh.org/herpetology/amphibia/index.html [Acces- sed: 30 December 2014]. Galvan-Guevara S, de la Ossa- Velasquez J. 2009. Her- petofauna registrada para el area de influencia de la reserva forestal protectora serrama de coraza, Coloso, Sucre, Colombia. Revista Colombiana de Ciencias Animales 1(2): 250-258. Galvis PA, Mejia-Tobon A, Rueda-Almonacid JV. 2011. Fauna Silvestre de la Reserva Forestal Protectora Montes de Oca, La Guajira, Colombia. Corpoguajira, Riohacha, Colombia. 822 p. Garden JG, McAlpine CA, Possingham HP, Jones DN. 2007. Habitat structure is more important than veg- etation composition for local-level management of native terrestrial reptile and small mammal species living in urban renmants: A case study from Brisbane, Australia Austral Ecology 32(6): 669-685. y Gardner TA, Ribeiro- Junior MA, Barlow J, Avila-Pires TCS, Hoogmoed MS, Peres C. 2007. The value of primary, secondary, and plantation forests for a Neo- tropical herpetofauna. Conservation Biology 21(3): 775-787. Hemandez-Camacho J, Sanchez H. 1992. Biomas ter- restres de Colombia. Pp. 153-174 In: La Diversi- dad Biologica de Iberoamerica. Editors, Halffter GI, CYTED-D. Acta Zoologica Mexicana, Mexico D.F., Mexico. 389 p. Herrera A, Olaya LA, Castro-Herrera F. 2004. Incidencia de la perturbacion antropica en la diversidad, riqueza y la distribucion de Eleutherodactylus (Anura: Lep- todactylidae) en u bosque nublado del suroccidente colombiano. Caldasia 26(1): 265-274. Heyer WR. 1969. The adaptive ecology of the species groups of the genus Leptodactylus (Amphibia, Lepto- dactylidae). Evolution 23(3): 421^28. Hoyos-Hoyos JM, Isaac s-Cubides P, Devia N, Galindo- Uribe DM, Acosta-Galvis AR. An approach to the ecology of the herpetofauna in agroecosystems of the Colombian coffee zone. South American Journal of Herpetology 7(1): 25-34 Laurance WE, Lovejoy TE, Vasconcelos HE, Bruna EM, Didham RK, Stouffer PC, Gascon C, Bierregaard RO, Laurance SG, Sampiao E. 2002. Ecosystem decay of Amazonian forest fragments: A 22-year investigation. Conservation Biology 16(3): 605-618. Lips K. 1999. Mass Mortality and Population Declines of Anurans at an Upland Site in Western Panama. Con- servation Biology 13(1): 117-125. Amphib. Reptile Conserv. Luja VH, Herrando-Perez S, Gonzalez-Solis D, Luiselli L. Secondary rain forests are not havens for reptile species in tropical Mexico. Biotropica 40(6): 747- 757. Medem F. 1977. Contribucion al conocimiento sobre la taxonomia, distribucidn geografica y ecologia dela tortuga “Bache” {Chelydra serpentina acutirostris). Caldasia 13(53): 41-101. Medem F. 1981. Los Crocodylia de Sur America. Volu- men 1. Los Crocodylia de Colombia. Conciencias, Bogota, Colombia. 356 p. Medina-Rangel GF. 2011. Diversidad alfa y beta de la comunidad de reptiles en el complejo cenagoso de Za- patosa, Colombia. Revista de Biologia Tropical 59(2): 935-968. Medina-Rangel GF, Cardenas-Arevalo G, Castano-Mora OV. 2011. Anfibios y Reptiles de los alrededores del complejo cenagoso de Zapatosa, departamento del Cesar, Colombia. Editors, Rangel-Ch. JO. Colom- bia Diversidad Biotica, Publicacion Especial No. 1, Gmpo de Biodiversidad y Conservacion, Institute de Ciencias Naturales, Universidad Nacional de Colom- bia-CORPOCESAR, Bogota. D.C., Colombia. 105 p. Moll EO, Legler JM. 1971. The life history of a Neo- tropical slider turtle, Pseudemys scripta (Schoepff) in Panama. Bulletin of the Los Angeles County Museum of Natural History Science 11: 1-102. Montes-Correa AC, Saboya- Acosta LP, Paez V, Vega K, Renjifo JM. 2014. Distribucion de tortugas del Ca- ribe colombiano. Acta Biologica Colombiana 19(3): 341-350. Montes-Correa AC, Jimenez-Bolano JD, Vergara-Rios D, Avila-Silva Y, Saboya-Acosta LP, Renjifo JM. 2015. Herpetofauna del campus de la Universidad del Magdalena, Santa Marta, Colombia. Revista de Biodi- versidad Neotropical 5(1): 54-63. Moskovits D, Bjomdal KA 1993. Diet and food prefer- ences of the tortoises Geochelone carbonarius and G. denticulata in northwestern Brazil. Herpetologica 46(2): 207-218. Pedroza-Banda R, Angarita- Sierra T. 2011. Herpeto- fauna de los humedales La Bolsa y Charco de Oro, Andalucia, Valle del Cauca, Colombia. Revista de la Academia Colombiana de Ciencias Exactas, Fisicas y Naturales 35(135): 243-260. Offerman HE, Dale VH, Pearson SM, Bierregaard RO, O’Neil RV. 1995. Effects of forest fragmentation on neotropical fauna: current research and data availabil- ity. Environmental Reviews 3: 191-211. Rangel-Ch JO, Carvajal-Cogollo JE. 2012. Clima de la region Caribe colombiana. Pp. 67-129 In: Colombia Diversidad Biotica XII: La region Caribe de Colom- bia. Editors, Rangel-Ch. JO. Institute de Ciencia Na- turales, Universidad Nacional de Colombia, Bogota, Colombia. 1,042 p. Reforestadora de la Costa SAS. 2012. Resumen Plan Forestal 2012-2020. Colombia. Ill p. April 2015 | Volume 8 | Number 1 | e92 155 Herpetofauna of an agroforestry system in the Colombian Caribbean Renjifo JM, Lundberg M. 1999. Anfibios y Reptiles de Urrd. SKANSKA, Editorial Colina, Medellm, Co- lombia. 96 p. Rivero JA. 1969. Sobre la Hyla rubra Laurenti y la Hyla x-signata Spix (Amphibia, Salientia). Memoria de la Sociedad de Ciencias Natur ales La Salle 83: 109-118. Romero-Martinez HJ, Lynch JD. 2010. Anfibios de los humedales del departamento de Cordoba. Pp. 349- 360 In: Colombia Diversidad Biotica IX: Cienagas del Cordoba: Biodiversidad, Ecologia y Manejo Am- biental. Editors, Rangel-Ch. JO. Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Cor- poracion Autonoma Regional de los Valles del Sinu y San Jorge CVS, Bogota, Colombia. 1,042 p. Romero-Martinez HJ, Lynch JD. 2012. Anfibios de la Region Caribe. Pp. 677-701 In: Colombia, Diversi- dad Biotica XII: La region Caribe de Colombia. Edi- tors, Rangel-Ch. JO. Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota, Colom- bia. 1,046 p. Romero-Martinez HJ, Vidal-Pastrana CC, Lynch JD, Duenas PR. 2008. Estudio preliminar de la fauna Am- phibia en el cerro Murrucucu, Parque Natural Nacio- nal Paramillo y zona amortiguadora, Tierralta, Cor- doba, Colombia. Caldasia 30(1): 209-229. Rossman DA. 2002. Morphological variation in the endemic Colombian Water Snake, Helicops danieli Amaral, 1937 (Serpentes: Xenodontidae). Revista de la Academia Colombiana de Ciencias Exactas, Eisi- cas y Naturales 26(100): 589-594. Rueda-Almonacid JV, Carr JL, Mittermeier RA, Rodri- guez-Mahecha JV, Mast RB, Vogt RC, Rhodin AGJ, de la Ossa- Velasquez J, Rueda JN, Mittermeier CG. 2007. Las Tortugas y los Cocodrilianos de los Raises Andinos de Tropico. Serie de guias tropicales de cam- po N° 6, Conservacion Internacional, Editorial Pana- mericana, Eormas e Impresos, Bogota, Colombia. 537 P- Rueda-Almonacid JV, Velasquez AA, Galvis PA, Gual- dron-Duarte J. 2008a. Anfibios. Pp. 169-192 In: Guia ilustrada de fauna del Santuario de Vida Silvestre Los Besotes, Valledupar, Cesar, Colombia. Editors, Rodri- guez-Mahecha JV, Rueda-Almonacid JV, Gutierrez- Hinojosa TD. Serie de guias tropicales de campo N° 7, Conservacion Internacional, Editorial Panamericana, Formas e Impresos, Bogota, Colombia. 574 p. Rueda-Almonacid JV, Velasquez AA, Galvis PA, Gual- dron-Duarte J. 2008b. Reptiles. Pp. 193-268 In: Guia ilustrada de fauna del Santuario de Vida Silvestre Los Besotes, Valledupar, Cesar, Colombia. Editors, Rodri- guez-Mahecha JV, Rueda-Almonacid JV, Gutierrez- Hinojosa TD. Serie de guias tropicales de campo N° 7, Conservacion Internacional, Editorial Panamericana, Formas e Impresos, Bogota, Colombia. 574 p. Rueda-Solano LA, Castellanos-Barliza J. 2010. Herpe- tofauna de Neguanje, Parque Nacional Natural Tay- rona, Caribe colombiano. Acta Biologica Colombiana Amphib. Reptile Conserv. 15(1): 195-206. Ruthven AG. 1922. The amphibians and reptiles of Si- erra Nevada de Santa Marta, Colombia. Miscella- neous Publications, Museum of Zoology, University of Michigan 8: 1-69. Scott NJ, Seigel RA. 1992. The management of amphib- ian and reptile populations: Species priorities and methodological and theoretical constraints. Pp. 343- 368 Wildlife 2001: Populations. Editors, McCollough DR, Barret HR. Elsevier Applied Science, Londres, United Kindom. 1,163 p. Tapley B, Acosta-Galvis AR. 2010. Distribution ofTyph- lonectes natans in Colombia, environmental param- eters and implications for captive husbandry. Herpe- tological Bulletin 113: 23-29. Uetz P, Hosek J, Hallerman J. The reptile database. Avail- able: http://reptile-database.reptarium.cz/ [Accessed: 30 December 2014]. Urbina-Cardona JN, Londono-Murcia MC. 2003. Distri- bucion de la comunidad de herpetofauna asociada a cuatro areas con diferente grado de perturbacion en la Isla Gorgona, Pacifico colombiano. Revista de la Academia Colombiana de Ciencias Exactas, Eisicas y Naturales 27(102): 105-113. Urbina-Cardona JN, Olivares-Perez M, Reynoso VH. 2006. Herpetofauna diversity and microenvironment correlates across a pasture-edge-interior ecotone in tropical rainforest fragments in the Los Tuxtlas Bio- sphere Reserve of Veracruz, Mexico. Biological Con- servation 132(1): 61-75. Urbina-Cardona JN, Londono-Murcia MC, Garcia- Avila DG. 2008. Dinamica espacio-temporal en la diversi- dad de serpientes en cuatro habitats con diferente gra- do de alteracion antropogenica en el Parque Nacional Natural Isla Gorgona, Pacifico colombiano. Caldasia 30(2): 479^93. Thorson TB. 1955. The relationship of water economy to terrestrialism in amphibians. Ecology 36 (1): 100- 116. Vargas-Salinas F, Bolanos ME. 1999. Anfibios y reptiles presentes en habitats perturbados de selva Iluviosa tropical en el Bajo Anchicaya, Pacifico colombiano. Revista de la Academia Colombiana de Ciencias Ex- actas, Eisicas y Naturales 23 (Suplemento especial): 499-511. Vitt LJ, Sartorius SS, Avila-Pires TCS, Zani PA, Esposito MC. 2005. Small in a big world: Ecology of leaf-letter geckos in the New World tropical forest. Herpetologi- cal Monographs 19: 137-152. Vogt RC, Hine RL. 1982. Evaluation of techniques for assessment of amphibian and reptile population in Wisconsin. Pp. 201-217 In: Herpetological commu- nities: A symposium of the Society for the Study of Am- phibians and Reptiles and the Herpetologists' League, August 1977. Editors, Scott NJ. U.S. Fish & Wildlife Research Report 13, Washington, D.C., USA. 239 p. Wanger TC, Iskandar DT, Motzke I, Brook BW, Sodhi April 2015 | Volume 8 | Number 1 | e92 156 Angarita-M et al. NS, Clough Y, Tschamtke T. 2010. Effects of land-use change on connnunity composition of tropical am- phibians and reptiles in Sulawesi, Indonesia. Conser- vation Biology 24(3): 795-802. Whitfield SM, Reider K, Greenspane S, Donelly MA. 2014. Litter dynamics regulate population densities in a declining terrestrial herpetofauna. Copeia 2014(3): 454^61. Zug GR, Zug PB. 1979. The marine toad, Bufo marinus: A natural history resume of native populations. Smith- sonian Contributions 284: 1-284. Oscar Angarina-M. is an undergraduate biology student at Universidad del Magdalena, Santa Marta, Colombia. His interests are in the study of herpetofaunal community ecology, habitat fragmentation, and environment legisla- tion. Andres Camilo Montes-Correa is a biology student at Universidad del Magdalena, Santa Marta, Colombia. Since its inception, be joined the Herpetology Lab, where he began to develop interest in ecological, taxonomic, and systematic studies. Among his current research projects are feeding ecology of leaf-litter tropical dry forest lizards, taxonomy of Caribbean Dwarf Geckos (Lepidoblepharis), and habitat use of Orinoquian Freshwater Turtles (chelids and kinosternids). Juan Manuel Renjifo is a Colombian herpetologist, wildlife photographer, and biologist at Pontificia Universi- dad Javeriana, having received his M.Sc. degree at the University of Kansas. He has developed studies in ecology and taxonomy of Colombian herpetofauna, ophidism, and snakebite. Juan has served as director of Laboratorio de Sueros Antioffdicos of Instituto Nacional de Salud for 30 years and is dedicated to teaching (ad honorem) at the Universidad Nacional de Colombia and Universidad del Magdalena universities. April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 157 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean Appendix I. Amphibian Caribbean lowlands inventories used for Bray-Curtis Similarity Analyses. A = La Gloria Project; B = El Botillero (Duenez-Gomez et al. 2004); C = Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); D = Medio Rancheria (Blanco-Torres et al. 2013); E = Reserva Eorestal Protectora Montes de Oca (Galvis et al. 2011); E = Serrama de Coraza (Galvan-Guevara and De la Ossa- Velasquez 2011); G = los Montes de Maria y la Cienaga La Caimanera (Acosta-Galvis 2012b); H = Represa de Urra (Renjifo and Lundberg 1999); I = Murrucucu lowlands (sensu Romero-Martmez et al. 2008); J = Humedales del Cordoba (Romero-Martmez and Lynch 2010); K = Atlantico and North Bolivar (Cuentas et al. 2002); L = Santuario de Vida Silvestre Los Besotes (Rueda-Almonacid et al. 2008a); M = Cienaga del Zapatosa (Medina-Rangel et al. 2011); N = Universidad del Magdalena (Montes-Correa et al. 2015). Species A B c D E F G H 1 J K L M N Rhinella humboldti 1 1 0 1 1 1 1 1 0 1 1 1 1 1 Rhinella margaritifera 0 0 0 0 0 0 1 0 1 0 0 0 0 0 Rhinella marina 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Rhinella sternosignata 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Rhaebo haematiticus 0 0 0 0 0 1 0 1 0 0 0 0 0 0 Hyalinobatrachium collymbiphyllum 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Hyalinobatrachium fleischmanni 0 0 0 0 0 0 1 1 0 0 0 0 0 0 Ceratophrys calcarata 1 1 0 1 1 1 1 1 0 1 1 1 1 0 Craugastor raniformis 0 0 0 0 0 0 1 1 1 1 1 0 1 0 Pristimantis taeniatus 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Pristimantis viejas 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Strabomantis bufoniformis 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Eleutherodactylus johnstonei 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Colostethus pratti 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Colostethus ruthveni 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Dendrobates truncatus 0 0 1 0 0 1 1 1 1 1 1 0 1 0 Cryptobatrachus boulengeri 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Dendropsophus ebraccatus 1 0 0 0 0 0 0 1 0 1 0 0 0 0 Dendropsophus microcephalus 1 1 0 1 1 1 1 1 0 1 1 0 1 0 Hypsiboas boans 0 0 0 0 0 1 0 1 1 0 1 0 1 0 Hypsiboas crepitans 1 1 1 0 1 0 1 0 0 1 1 1 1 0 Hypsiboas pugnax 1 0 1 1 1 1 1 1 1 1 1 1 1 1 Hypsiboas rosenbergi 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Phyllomedusa venusta 1 1 0 0 0 0 1 1 1 0 1 1 0 0 Pseudis paradoxa 1 1 0 0 0 0 1 1 0 1 1 0 0 0 Scarthyla vigilans 1 1 0 1 0 1 1 1 0 1 1 0 0 0 Scinax boulengeri 0 0 0 0 0 0 0 1 0 0 1 0 0 0 Scinax elaeochrous 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Scinax rostratus 1 1 0 0 0 0 1 0 0 1 0 0 1 0 Scinax ruber 0 1 0 1 1 1 1 1 1 1 1 0 0 0 Scinax x-signatus 1 0 0 0 0 0 0 0 0 1 0 0 0 0 Smilisca phaeota 0 0 0 0 0 0 0 1 1 0 0 0 0 0 Smilisca sila 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Trachycephalus typhonius 1 0 0 1 1 1 1 1 0 1 1 0 1 0 Engystomops pustulosus 1 1 1 1 1 0 1 1 1 1 1 1 1 1 Pleurodema brachyops 1 1 1 1 1 0 1 1 0 1 1 1 1 1 Pseudopaludicola pusilla 1 1 0 1 1 0 1 1 0 1 1 0 1 0 Leptodactylus colombiensis 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Leptodactylus fragilis 1 1 1 1 0 0 1 1 0 1 1 0 1 0 Leptodactylus fuscus 1 1 0 1 1 1 1 1 0 1 1 1 1 1 Leptodactylus insularum 1 1 1 1 1 1 1 1 0 1 1 1 1 1 Leptodactylus poecilochilus 1 1 0 1 1 0 1 0 1 1 1 1 1 0 April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 158 e92 Angarita-M et al. Appendix I (Continued). Amphibian Caribbean lowlands inventories used for Bray-Curtis Similarity Analyses. A = La Gloria Project; B = El Botillero (Duenez-Gomez et al. 2004); C = Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); D = Medio Rancheria (Blanco-Torres et al. 2013); E = Reserva Eorestal Protectora Montes de Oca (Galvis et al. 2011); E = Serrama de Coraza (Galvan-Guevara and De la Ossa- Velasquez 2011); G = los Montes de Maria y la Cienaga La Caimanera (Acosta-Galvis 2012b); H = Represa de Urra (Renjifo and Lundberg 1999); I = Murrucucu lowlands (sensu Romero-Martmez et al. 2008); J = Humedales del Cordoba (Romero-Martmez and Lynch 2010); K = Atlantico and North Bolivar (Cuentas et al. 2002); L = Santuario de Vida Silvestre Los Besotes (Rueda-Almonacid et al. 2008a); M = Cienaga del Zapatosa (Medina-Rangel et al. 2011); N = Uni- versidad del Magdalena (Montes-Correa et al. 2015). Species A B c D E F G H 1 J K L M N Leptodactylus savagei 0 0 0 0 0 0 0 1 1 0 1 0 0 0 Lithodites lineatus 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Elachistocleis panamensis 1 0 1 1 1 0 0 1 0 1 1 0 1 0 Elachistocleis pearsei 1 0 0 0 0 1 1 1 0 1 1 0 1 0 Pipa parva 0 0 0 1 1 0 0 0 0 0 0 0 0 0 Lithobates vaillanti 0 0 0 0 1 0 0 1 0 0 1 1 1 0 Caecilia isthmica 0 0 0 0 0 0 1 0 0 0 0 0 0 0 Caecilia caribea 1 0 0 0 0 0 0 0 0 0 1 0 0 0 Caecilia subnigricans 0 0 0 0 0 0 0 1 0 1 0 0 0 0 Oscaecilia polyzona 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Typhlonectes natans 0 1 0 0 0 1 1 1 0 1 0 0 0 0 Bolitoglossa biseriata 0 0 0 0 0 1 0 1 1 0 0 0 0 0 April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 159 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean Appendix II. Reptile Caribbean lowlands inventories used for Bray-Curtis Similarity Analyses. A = La Gloria Project; B = El Botil- lero (Duenez-Gomez et al. 2004); C = Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); D = Medio Rancheria (Blanco-Torres et al. 2013); E = Reserva Eorestal Protectora Montes de Oca (Galvis et al. 2011); E = Serrama de Coraza (Galvan- Guevara and De la Ossa- Velasquez 2011); G = Represa de Urra (Renjifo and Lundberg 1999); H = Humedales del Cordoba (Car- vajal-Cogollo et al. 2007); I = Santuario de Vida Silvestre Los Besotes (Rueda-Almonacid et al. 2008b); J = Cienaga del Zapatosa (Medina-Rangel et al. 2011); K = Universidad del Magdalena (Montes-Correa et al. 2015). Species A B c D E F G H 1 J K Amphisbaena alba 0 0 0 0 1 0 0 0 1 0 0 Amphisbaena fuliginosa 0 0 0 0 1 0 0 0 1 0 0 Amphisbaena medemi 0 0 0 1 0 0 0 0 1 0 0 Gonatodes albogularis 1 1 1 1 1 1 1 1 1 1 1 Gonatodes vittatus 0 0 0 1 1 0 0 0 0 0 0 Lepidoblepharis sanctaemartae 1 0 1 1 1 0 0 1 1 1 1 Sphaerodactylus heliconiae 0 0 0 0 0 0 0 0 0 1 0 Phyllodactylus ventralis 0 0 1 1 1 0 0 0 2 0 1 Thecadactylus rapicauda 1 1 1 1 1 0 1 1 0 1 1 Hemidactylus brookii 0 0 1 1 1 1 1 1 1 0 1 Hemidactylus frenatus 1 0 0 0 1 0 0 0 0 1 1 Basiliscus basilicus 1 1 0 1 1 1 1 1 0 1 0 Basiliscus galeritus 0 0 0 0 0 0 0 1 0 0 0 Corytophanes cristatus 0 0 0 0 0 0 1 0 0 0 0 Anolis auratus 1 1 1 1 1 1 1 1 1 1 1 Anolis biporcatus 0 0 0 0 1 0 0 0 0 0 0 Anolis pentaprion 0 0 0 0 0 0 1 0 0 0 0 Anolis onca 0 0 0 0 1 0 0 0 0 0 0 Anolis tropidogaster 0 0 0 0 1 0 1 1 0 1 0 Anolis vittigerus 0 0 0 0 0 1 0 1 0 0 0 Iguana iguana 1 1 1 1 1 1 1 1 1 1 1 Polychrus marmoratus 0 0 1 1 1 0 0 0 1 1 0 Stenocercus erythrogaster 0 0 1 0 1 0 0 0 1 1 0 Maracaiba zuliae 1 0 0 0 1 0 0 0 0 0 0 Mabuya sp. 0 1 1 1 0 1 1 1 1 1 0 Bachia bicolor 0 0 1 0 0 0 0 0 0 1 1 Bachia talpa 0 0 0 1 1 0 0 0 1 0 0 Gymnophthalmus speciosus 0 0 0 1 1 0 0 1 1 1 1 Leposomoma rugiceps 1 1 1 0 0 0 1 1 0 1 0 Tretioscincus bifasciatus 1 1 1 1 1 0 0 1 1 1 1 Ameiva praesignis 1 1 1 1 1 1 1 1 1 1 1 Ameiva bifrontata 1 0 1 1 1 0 0 0 0 0 1 Cnemidophorus lemniscatus 1 1 1 1 1 1 1 1 1 1 1 Holcosus festivus 0 0 0 0 0 1 1 1 0 1 0 Tupinambis teguixin 0 1 0 1 1 1 1 1 1 1 0 Liotyphlops albirostris 1 0 1 1 1 1 1 1 0 0 1 Epictia goudotii 0 0 0 1 0 0 0 0 1 0 1 Trilepida macrolepis 0 0 0 0 0 0 1 0 0 0 0 Trilepida dugandi 0 0 0 0 0 0 0 0 1 0 0 Boa constrictor 1 1 1 1 1 1 1 1 1 1 1 Corallus batesi 0 0 0 0 0 0 1 1 0 0 0 Corallus ruschenbergerii 0 0 1 1 1 1 1 1 0 1 0 Epicrates maurus 1 1 0 1 1 1 1 1 0 1 0 Chironius carinatus 1 0 0 0 1 1 1 1 0 1 0 Amphib. Reptile Conserv. 160 April 2015 1 Volume 8 | Number 1 1 e92 Angarita-M et al. Appendix II (continued). Reptile Caribbean lowlands inventories used for Bray-Curtis Similarity Analyses. A = La Gloria Project; B = El Botillero (Duenez-Gomez et al. 2004); C = Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); D = Medio Rancberia (Blanco-Torres et al. 2013); E = Reserva Eorestal Protectora Montes de Oca (Galvis et al. 2011); E = Serrama de Coraza (Galvan-Guevara and De la Ossa- Velasquez 2011); G = Represa de Urra (Renjifo and Lundberg 1999); H = Humedales del Cordoba (Carvajal-Cogollo et al. 2007); I = Santuario de Vida Silvestre Los Besotes (Rueda-Almonacid et al. 2008b); J = Cienaga del Zapa- tosa (Medina-Rangel et al. 2011); K = Universidad del Magdalena (Montes-Correa et al. 2015). Species A B c D E F G H 1 J K Coluber mentovarius 0 0 0 0 1 0 0 0 0 0 0 Dendrophidion bivittatus 0 0 0 0 0 1 1 0 0 0 0 Dendrophidion percarinatus 0 0 0 0 0 0 1 0 0 0 0 Drymarchon caudomaculatus 0 0 0 1 0 0 0 0 0 0 0 Drymarchon melanurus 0 0 0 0 1 0 0 0 1 0 0 Leptophis ahaetulla 1 0 0 1 1 1 0 1 1 1 0 Mastigodryas boddaertii 0 0 0 1 1 0 1 0 1 0 1 Mastigodryas pleei 0 1 1 1 1 0 1 1 1 1 0 Oxybelis aeneus 0 1 1 1 1 1 1 0 0 1 1 Oxy bells fulgidus 0 0 1 0 1 0 0 0 1 0 0 Pliocercus euryzonus 0 0 0 0 0 0 1 0 0 0 0 Pseustes poecilonotus 0 0 0 0 0 0 1 0 0 0 0 Pseustes shropshieri 0 0 0 0 0 0 1 0 0 0 0 Spillotes pullatus 0 1 0 1 0 1 1 1 1 1 0 Stenorrhina degenhardtii 0 0 0 0 0 0 1 0 0 0 0 Tantilla melanocephala 1 0 0 0 1 0 0 1 1 0 1 Tantilla semicincta 0 0 0 1 1 0 0 0 1 0 1 Clelia Clelia 0 0 0 0 1 1 1 1 1 0 0 Enulius flavitorques 0 0 0 0 1 0 0 1 1 1 1 Erythrolamprus melanotus 0 0 0 1 1 0 1 1 1 0 0 Erythrolamprus bizona 0 0 0 0 0 0 0 0 1 0 0 Helicops danieli 1 1 1 0 0 1 1 1 0 1 0 Imantodes cenchoa 1 0 0 0 1 1 1 0 1 1 0 Leptodeira annulata 1 0 1 1 1 0 0 1 1 0 1 Leptodeira septentrionalis 1 1 0 0 0 1 0 0 0 1 1 Lygophis lineatus 1 0 0 0 1 1 1 1 1 1 0 Ninia atrata 0 0 0 0 0 0 1 0 0 0 0 Oxyrhopus petolarius 1 0 0 0 1 0 1 0 0 0 0 Phimophis guianensis 0 1 1 1 1 0 0 0 1 1 1 Pseudoboa neuwiedii 1 1 1 1 1 0 1 1 0 1 0 Sibon nebulatus 0 0 0 0 1 0 1 0 0 0 0 Thamnodynastes paraguanae 1 0 0 1 1 0 0 0 0 0 0 Thamnodynastes gambotensis 1 1 0 0 0 0 1 1 0 1 0 Xenodon severus 0 0 0 1 1 0 0 0 0 0 0 Xenodon rabdocephalus 1 0 0 0 1 0 0 0 0 0 0 Micrurus camilae 0 0 0 0 0 0 1 0 0 0 0 Micrurus dissoleucus 1 0 0 1 1 0 1 0 1 0 1 Micrurus dumerili 0 0 0 0 1 0 0 1 1 0 0 Bothriechis schlegelii 0 0 0 0 0 0 1 0 0 0 0 Bothrops asper 1 1 1 0 1 0 1 1 1 1 0 Crotalus durissus 1 0 1 1 1 0 0 0 1 1 1 Porthidium lansbergii 1 1 1 1 1 0 0 1 1 1 1 Porthidium nasutum 0 0 0 0 0 0 1 0 0 0 0 Mesoclemmys dahli 1 0 0 0 0 0 0 1 0 1 0 Amphib. Reptile Conserv. 161 April 2015 1 Volume 8 | Number 1 1 e92 Herpetofauna of an agroforestry system in the Colombian Caribbean Appendix II (continued). Reptile Caribbean lowlands inventories used for Bray-Curtis Similarity Analyses. A = La Gloria Project; B = El Botillero (Duenez-Gomez et al. 2004); C = Ensenada Neguanje (Rueda-Solano and Castellanos-Barliza 2010); D = Medio Rancberia (Blanco-Torres et al. 2013); E = Reserva Eorestal Protectora Montes de Oca (Galvis et al. 2011); E = Serrama de Coraza (Galvan-Guevara and De la Ossa- Velasquez 2011); G = Represa de Urra (Renjifo and Lundberg 1999); H = Humedales del Cordoba (Carvajal-Cogollo et al. 2007); I = Santuario de Vida Silvestre Los Besotes (Rueda-Almonacid et al. 2008b); J = Cienaga del Zapa- tosa (Medina-Rangel et al. 2011); K = Universidad del Magdalena (Montes-Correa et al. 2015). Species A B c D E F G H 1 J K Podocnemis lewyana 0 0 0 0 0 0 1 0 0 1 1 Chelydra acutirostris 0 0 0 0 0 0 1 0 0 0 0 Cryptochelys leucostomum 0 0 0 0 1 0 1 0 0 0 0 Kinosternon scorpioides 0 0 0 0 1 0 1 1 1 1 1 Rhinoclemmys melanosterna 0 0 0 0 1 0 0 1 0 1 0 Trachemys calliwstris 1 1 0 1 1 0 1 1 0 1 1 Chelonoidis carbonaria 1 1 0 1 1 0 1 1 1 1 1 Crocodylus acutus 0 0 0 0 1 0 1 0 0 1 0 April 2015 | Volume 8 | Number 1 Amphib. Reptile Conserv. 162 e92 Copyright: © 2014 Pierson et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and educa- tion purposes only provided the original author and source are credited. The of- ficial publication credit source: Amphibian & Reptile Conservation at: amphibian- reptile-conservation. org Amphibian & Reptiie Conservation 8(1) [Gen Sec]: 1-6. A survey for the Chinese giant salamander {Andrias davidianus', Blanchard, 1871) in the Qinghai Province Todd W. Pierson, ^yan Fang, ^WANG Yunyu, and Theodore Papenfuss ^University of Georgia, 150 East Green Street, Athens, Georgia, 30602, USA ^State Key Laboratory of Genetic Resources and Evolution, and Yun- nan Laboratory of Molecular Biology of Domestic Animals, Kunming Institute of Zoology, Chinese Academy of Sciences, 32 Jiaochang Donglu, Kunming, Yunnan, 650223, CHINA ^Museum of Vertebrate Zoology, University of California, Berkeley, California, 94720, USA Abstract . — ^The Chinese giant saiamander {Andrias davidianus) was once common, but it has de- ciined precipitousiy in the past severai decades. An enigmatic specimen coiiected in 1966 repre- sents the oniy historicai record of the species from the Qinghai-Tibetan Piateau. From June-Juiy 2012, we conducted opportunistic community inquiries and fieid surveys in Qinghai to attempt to iocate Andrias. We received anecdotai evidence that additionai Andrias have been found in recent years, but we faiied to discover any Andrias during our fieid surveys. We suspect that reiict popuia- tions persist in Qinghai, but the significant degradation of stream quaiity in the region iikeiy threat- ens the iong-term survivai of any remaining Andrias. Here, we provide a brief overview of Andrias conservation, a summary of our surveys, and emphasize the importance of continued searches for this geographicaiiy disjunct popuiation. Key words. Cryptobranchidae, Qinghai-Tibetan Plateau, conservation Citation: Pierson TW, Van F, Wang Y, Papenfuss T. 2014. A survey for the Chinese giant salamander {Andrias davidianus] Blanchard, 1871) in the Qing- hai Province. Amphibian & Reptile Conservation 8{t) [General Section]; 1-6 (e74). Introduction The Chinese Giant Salamander {Andrias davidianus) was once widely distributed throughout the Yangtze, Yellow, and Pearl River drainages. However, dramatic declines since 1950 have restricted the species to twelve frag- mented regions across seventeen provinces (Zhang et al. 2002). These declines are due largely to habitat degrada- tion and harvest for food (Dai et al. 2009). In response to these declines in the wild, the 2004 International Union for Conservation of Nature Red List evaluated A. davidi- anus as Critically Endangered, and the recognition of the conservation needs of the species has attracted national and international attention. Additionally, at least thirty preserves have been established in China to conserve A. davidianus, and captive breeding for human consump- tion has increased in prevalence and success (Dai et al. 2009; Zhang et al. 2002). One challenge for the conser- vation of A. davidianus is the preservation of genetic diversity, and several studies have examined variation between and among populations of A. davidianus. Sig- nificant substructuring exists among populations (Mur- phy et al. 2000; Tao et al. 2006), although results may be confounded by translocations of animals through the food trade. However, the overall genetic diversity of A. davidianus is relatively low compared to other salaman- ders (Tao et al. 2005; Yang et al. 2011). A single specimen of A. davidianus was collected in the headwaters of the Yangtze River in the Qinghai Prov- ince in August 1966 (33.898 96.522; Fig. 1; Trap Loca- tion 9, Figure 2; Fig. 3). The specimen was a gravid fe- male caught on hook-and-line near the town of Bagan at approximately 4,200 m, representing the highest known distribution record of A. davidianus by more than 2,000 m and a greatly disjunct population (Chen 1989). The ge- ography and the geological history of this region (Yin 2010) suggest the possibility that the gap between this Qinghai record and other known localities for A. davidi- anus represents a true biogeographical break, and this population may be important for conservation purposes. From 6 June to 2 July 2012, we used a variety of methods to survey Qinghai for A. davidianus. We were unsuccessful in locating any Andrias, but here we report the environmental conditions of the historic locality and others, anecdotal reports of Andrias from locals, and sug- gestions for future efforts to locate Andrias in Qinghai. Methods Throughout our stay in Qinghai, we frequently talked to officials from the Bureau of Forestry to obtain permis- sion to search for Andrias. During this process, we also inquired about anecdotal Andrias sightings from fisher- men. This amounted to discussions with approximately Correspondence, ^twpierso® uga.edu; twpierson® gmail.com Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (1) January 2014 | Volume 8 | Number 1 | e74 Pierson et al. fifteen government officials and five local fishermen. Af- ter talking with government officials and locals of Qin- ghai, we selected sites to survey based on historical and anecdotal records. We trapped in three general regions — Bagan, Zhiduo, and Yushu (Table 1; Fig. 2) — including the exact locality of the collection of the 1966 specimen from Bagan (Trap Location 9, Figure 2). Browne et al. (2011) reviewed and evaluated survey techniques for cryptobranchid salamanders. Because of religious and cultural beliefs about the sanctity of fish, local people in Qinghai are overwhelmingly unsupport- ive of any attempts by biologists to survey aquatic or- ganisms. Due to these limitations, some methods (e.g., electroshocking) were not possible, and our field surveys were conducted primarily through trapping, which has been shown to be reasonably effective for surveying for Cryptobranchus a. alleganiensis and A. japonicus (Fos- ter et al. 2008; Briggler et al. 2013). Even so, we were restricted to trapping discretely, had several traps stolen, and were actively discouraged from actually entering the streams by both locals and governmental officials. These practical challenges significantly limited our trapping efforts. We primarily used two sizes of custom-made, mesh-net rectangular crab traps (approximately 81 x 61 X 28 cm; 61 x 46 x 20 cm) designed to catch Andrias of varying sizes. The traps were baited altematingly with sardines, fishmeal, liver, and sponges soaked in fish oil Fig. 1. The adult female Andrias captured in Qinghai, China in 1966. This specimen now resides at the Northwest Plateau Institute of Biology in Xining. held in bait containers. Traps were weighted with stones, anchored to shore, and entirely submerged in 0.3 - >5 m of water in suitable habitat. Typically, the traps were placed in still pools along rocky bluffs at the edge of the river and checked after approximately 24 hours. Addi- tionally, baited hook-and-line and manual searches of rocky habitat were used opportunistically when the habi- tat was suitable. Results and Discussion During our discussions with local people and govern- ment officials, we heard several anecdotal reports of Andrias being caught in recent years. Local Bureau of Forestry officials and one layman in Qumalai told of an adult Andrias that had been caught and thrown back by a fisherman at the same locality as the original record (Trap Locality 9, Figure 2) around 1992. The same of- ficials in Qumalai and several officials in Zhiduo told of an Andrias that had been caught in the Nieqia River at its confluence with the Tongtian River in Qumalai (34.016, 95.817) between 1996-1997. This individual was re- portedly sent to Xian and sold for food. An official from Zhiduo also reported that this fisherman’s brother had caught an Andrias in a slow part of the Tongtian River between Zhiduo and Yushu earlier in 2012. Finally, two residents of Yushu reported seeing dead Andrias in the Tongtian River after the earthquake of 2010. Only one other species of caudate (Batrachuperus tibetanus) is Table 1. Trapping effort in Qinghai. Numbers to right of the location indicate the corresponding points on Fig. 2. Date Placed Traps Location 13 June 12 Four tributaries of De Qu River near Bagan (1^) 13 June 4 De Qu River on the road to Bagan (5) 14 June 5 Bo Qu River near Bagan (6-8) 14 June 4 De Qu River at the bridge in Bagan (9) 15 June 12 Four tributaries of De Qu River near Bagan (1-4) 15 June 1 De Qu River on the road to Bagan (5) 16 June 12 Four tributaries of De Qu River near Bagan (1-4) 16 June 5 Upper De Qu River outside of Bagan 17 June 5 De Qu River at the bridge in Bagan (9) 19 June A Tribuatries of Hie Qu River near Zhiduo (10-11) 20 June 14 Tribuatries of Hie Qu River near Zhiduo (10-11) 21 June 14 Tribuatries of Hie Qu River near Zhiduo (10-11) 26 June 10 Tributaries of Tongtian River near Yushu (12) 28 June 14 Tributaries of Tongtian River near Yushu (12) TOTAL 116 Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (2) January 2014 | Volume 8 | Number 1 | e74 Andrias davidianus survey in Qinghai Province, China Fig. 2. Map of trapping localities and nearby towns in Qinghai. present in the region, but its limited distribution in Qing- hai and small size make it unlikely to have been misiden- tified as Andrias in locals’ reports. Although there is no hard evidence to substantiate the reports we heard, when taken in aggregate, they seem credible. We trapped for 106 trap-nights (Table 1) and were not able to discover any Andrias during our field survey of Qinghai. Foster et al. (2008) used a similar trapping protocol and caught Cryptobranchus a. alleganiensis at a rate of 0.01-0.10 captures/trap-night. Briggler et al. (2013) trapped for C. a. alleganiensis in deeper and more turbid water and reported an average capture rate of 0.042 captures/trap-night with net-mesh traps. We acknowl- edge that our limited number of trap-nights prevents us from making definitive conclusions about the presence or absence of A. davidianus at our trapping sites. Virtually all of the streams in which we trapped were turbid and swollen with silted water (Fig. 3), which is a major threat to Andrias conservation. While it is possible that some of this turbidity was due to seasonal snowmelt, it is more likely that anthropogenic causes are primarily responsible. Since the collection of the lone specimen in 1966, mining for gold and other valuable commodities has become prevalent throughout the Qinghai-Tibetan Plateau. Furthermore, dozens of active sand and gravel mining operations were stationed throughout the rivers we sampled (Fig. 4). Locals in Yushu reported an in- crease in mining activity in response to construction and reparation needs following the major earthquake of 2010. Additionally, some streamside microhabitats for Andrias have been degraded due to road and bridge construction (Fig. 5). Another contributor to the siltation of Qinghai streams may be grassland degradation and desertification driven by climate change that has been demonstrated in the region (Cui and Graf 2009). Conclusion Despite our inability to locate Andrias in Qinghai, anec- dotal reports suggest that relict populations may still exist throughout the former range of the species. However, the apparent dramatic declines in stream quality in the region probably threaten the persistence of these populations. Although more remote regions further west of Bagan have fewer roads and present more practical challenges to fieldwork, they hold large headwaters of the Yangtze upstream of significant mining activity and may repre- sent the most suitable remaining habitat. While we were not able to survey these regions during our expedition, they should be prioritized in future searches. Because Qinghai is at such a high elevation, suitable conditions for searching occur in a small window each year. We rec- onnnend that efforts be focused in August or September, after seasonal flooding from snowmelt has passed, but before winter has returned. In addition to the continued use of trapping, hook-and-line, and manual searches, we recommend the possibility of using environmental DNA, which has been demonstrated to be an effective tool for detecting populations of other cryptobranchids (e.g., Ol- son et al. 2012; S. Spear, pers. comm.). Because of the potential importance of this geographi- cally isolated population of Andrias in Qinghai, its redis- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (3) January 2014 | Volume 8 | Number 1 | e74 Pierson et al. Fig. 3. The locality where the first and only Andrias was collected from Bagan, Qinghai in 1966. Today, the water is turbid and ap- pears largely unsuitable for Andrias. Fig. 4. A mining operation on the banks of the Tongtian River, near Qumalai, Qinghai. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (4) January 2014 | Volume 8 | Number 1 | e74 Andrias davidianus survey in Qinghai Province, China Fig. 5. Stream bank degradation caused by road construction along the Tongtian River. covery should continue to be a top priority for Andrias conservation. Acknowledgments. — Funding was provided by the Key Program of the Chinese Academy of Sciences (XDB03030000), the Strategic Priority Research Pro- gram of the Chinese Academy of Sciences (KJZD-EW- L07), the National Geographic Society’s Young Explor- ers Grant (9084-12), the Explorers Club Youth Activity Fund, the California Academy of Sciences, and the Kun- ming Institute of Zoology. We would like to thank J. Che for her great help in planning and organizing the trip and Y. Zhang for aid in acquiring permission to conduct re- search in Qinghai. Additionally, the Chinese Academy of Sciences (including the Northwest Institute of Biology) and local forestry departments proved important for the success of the expedition. Literature Cited Brigger JT, McKeage BE, Girondo NM, Pitts PR. 2013. Evaluation of traps to capture Eastern hellbenders {Cryptobranchus alleganiensis alleganiensis) in deep water habitat. Herpetological Review 44(3): 423^28. Browne RK, Hong L, McGinnity D, Okada S, Zhenguan W, Bodinof CM, Irwin KJ, McMillan A, Briggler JT. 2011. Survey techniques for giant salamanders and other aquatic Caudata. Amphibian & Reptile Conser- vation (4): 1-16 (e34). Chen X. 1989. Amphibia and Reptilia. Pp. 173-227 In: Economic Fauna of Qinghai. Editor, Xining D Li. Northwest Plateau Institute of Biology Qinghai, Chi- na. [In Chinese]. Cui X, Graf H. 2009. Recent land cover changes on the Tibetan Plateau: A review. Climatic Change 94: 47- 61. Dai X, Wang YZ, Liang G. 2009. Conservation status of Chinese giant salamander {Andrias davidianus). Report of the Chengdu Institute of Biology, Chinese Academy of Sciences. Available: http://www.cepf. net/Documents/final_CIBCAS_giantsalamander_chi- na.pdf [Accessed: 03 December 2013]. Foster RL, McMihan AM, Breisch AR, Roblee KJ, Schranz D. 2008. Analysis and comparison of three capture methods for the Eastern hellbender {Crypto- branchus alleganiensis alleganiensis). Herpetologi- cal Review 39(2): 181-186. Murphy RW, Fu J, Upton DE, de Lama T, Zhao EM. 2000. Genetic variability among endangered Chinese giant salamanders, Andrias davidianus. Molecular Ecology 9(10): 1539-1547. Olson ZH, Briggler JT, Williams RN. 2012. An eDNA approach to detect eastern hellbenders {Cryptobran- chus a. alleganiensis) using samples of water. Wildlife Research 39: 629-636. Tao FY, Wang XM, Zheng HX. 2006. Analysis of com- plete cytochrome b sequences and genetic relation- ship among Chinese giant salamanders {Andrias da- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (5) January 2014 | Volume 8 | Number 1 | e74 Pierson et al. vidianus) from different areas. Acta Hydrobiologica Sinica (2006) 36: 8-11. [In Chinese]. Tao FY, Wang XM, Zheng HX, Fang SG. 2005. Genetic structure and geographic subdivision of four popula- tions of the Chinese giant salamander (Andrias davidi- anus). Zoological Research 6: 162-167. [In Chinese]. Yang LP, Meng ZN, Liu XC, Zhang Y, Huang JH, Huang J, Lin HR. 2011. AFLP analysis of five natural popu- lations of Andrias davidianus. Acta Scientarum Natu- ralium Universitatis Sunyatseni 50(2): 99-104. [In Chinese]. Yin A. 2010. Cenozoic tectonic evolution of Asia: A pre- liminary synthesis. Tectonophysics 488: 293-325. Zhang KJ, Wang XM, Wu W, Wang ZH, Huang S. 2002. Advances in conservation biology of Chinese giant salamander. Biodiversity Science 10(3): 291-297. [In Chinese]. Received: 03 December 2013 Accepted: 17 January 2014 Published: 22 January 2014 Todd Pierson’s research focuses on amphibian and reptiles ecology, evolution, and conservation. He grad- uated with a B.S. Ecology from the Odum School of Ecology at the University of Georgia in 2013. He currently works in the EHS DNA Lab at UGA, where be develops environmental DNA assays for use in detecting aquatic amphibians. Yan Fang is mainly interested in tbe pbylogeography and conservation genetics of amphibians. She gradu- ated with a Ph.D. from the Kunming Institute of Zoology (KIZ), Chinese Academy of Sciences in 2013. Now she works on the conservation genetics of Chinese giant salamander at KIZ. Wang Yunyu is staff at the Southern China DNA Barcoding Center (SCDBC), Kunming Institute of Zool- ogy^ Chinese Academy of Science. She currently works on DNA barcoding of amphibians and reptiles. Theodore Papenfuss is a Research Scientist at the Museum of Vertebrate Zoology. His current research activities involve field studies of amphibians and reptiles in Asia and Central America. He is also collabo- rating with conservation agencies that are conducting surveys of tropical forests in Guatemala in order to select areas for permanent habitat protection. Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (6) January 2014 | Volume 8 | Number 1 | e74 Copyright: © 2014 Michaels et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License, which permits unrestricted use for non-commercial and education purposes only provided the original author and source are credited. The official publication credit source: Amphibian & Reptile Conservation at: amphibian-reptile-conservation.org Amphibian & Reptiie Conservation [General Section] 8(1) :7-23. REVIEW The importance of enrichment for advancing amphibian welfare and conservation goals: A review of a neglected topic ^Christopher J. Michaels , Roger Downie, and ^Roisin Campbell-Palmer ^•‘^Preziosi Group, Faculty of Life Sciences, Michael Smith Building, University of Manchester, Manchester, ENGLAND ^School of Life Sciences, Graham Kerr Building, University of Glasgow, Glasgow, SCOTLAND ^Conservation Programmes, The Royal Zoological Society of Scotland, Edinburgh, SCOTLAND Abstract. — Enrichment, broadiy the provision of stimuii to improve the weifare of captive animais, is known to be important in husbandry practice and in the success of ex situ conservation and reintroduction programs. Practicai evidence of the importance of enrichment exists for a number of taxa, yet amphibians are pooriy represented. There is no reason to assume a priori that amphibians wouid not benefit from enrichment and, given their increasing prominence in captive programs, their requirements in captivity beyond basic husbandry shouid be the focus of more intense study. We review the existing body of research on enrichment for amphibians, as weii as that for fish and reptiies, which may be regarded as behavioraiiy and neuroiogicaiiy broadiy simiiar to amphibians. We aiso briefiy discuss mechanisms by which enrichment might affect amphibian fitness and, therefore, reintroduction success. Our review supports the contention that there may be important consequences of enrichment for both captive weifare and ex situ conservation success in amphibians and that amphibian enrichment effects may be highiy variabie taxonomicaiiy. In the face of increasing numbers of captive amphibian species and the importance of exs/fupopuiations in ensuring their species ievei persistence, enrichment for amphibians may be an increasingiy important research area. Key words. Behavior, conservation, environmental enrichment, re-introduction, welfare, ex situ, fish, reptiles Citation: Michaels CJ, Downie JR, Campbell-Palmer R. 2014. The importance of enrichment for advancing amphibian welfare and conservation goals: A review of a neglected topic. Amphibian & Reptiie Conservation 8(1) [General Section]: 7-23 (e77). Introduction A wide range of amphibian species is currently main- tained in captivity. Some species are used as models in laboratory research, including the ubiquitous Xenopus laevis and the dendrobatid frogs used to study skin pep- tides (reviewed by Daly 1998) and caecilians used in bio- mechanics research (e.g., Summers and O’Reilly 1997) and leaf frogs involved in conservation research (Ogilvy et al. 2012a, b). Several species are farmed (in addition to the many collected from the wild) for food or other products and others are maintained by private individuals as hobby or pet animals (Gascon et al. 2005). In addition, the ex situ conservation response to the on-going global amphibian extinction crisis (e.g., Gagliardo et al. 2008; Lee et al. 2006; Norris 2007) has drawn much public- ity to the growing number of amphibians maintained for conservation breeding and education in zoos and similar institutions. This increase in captive amphibians (both in actual numbers and species held) and their mounting conservation importance, has highlighted the need for a more thorough understanding of amphibian captive husbandry (Gascon et al. 2005), particularly for species that have no history in captivity and for those that are intended for release into the wild (Gagliardo et al. 2008; Gascon et al. 2005). For many other taxa, the importance of enrichment has been identified for not only the welfare, or the physi- cal and psychological wellbeing, of individual animals in captivity or those destined for release, but also for the overall/long-term success of reintroduction projects (Crane and Mathis 2010; Shepherdson et al. 1998; Young 2003). However, the implications of past work on the value of enrichment schemes for captive species cur- rently has limited scope because enrichment has neither explicitly used nor well researched in amphibians (de Azevedo et al. 2007; Burghardt 2013). The objective of this paper is to draw attention to this lack of knowledge Correspondence. Email: "^c.j.michaels44@ gmail.com (Corresponding author). Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (7) June 2014 I Volume 8 I Number 1 I ell Michaels et al. Table 1. Studies of enrichment in amphibians. Species Origin Type of enrichment investigated Findings Notes Reference Xenopus laevis Unknown Shelter provision No effect on growth rate. Frogs provided with shelter reluctant to leave it, even when provided with food. Small sample size; unknown origins and genetics (see Chum et al. 2013) Hilken et al. (1995) Xenopus laevis Laboratory bred Shelter provision Frogs use any shelter provided, but prefer plastic tubes to plants, rocks and wood. Frogs prefer tanks with shelter to tanks with no shel- ter. Frogs showed increased activity and reduced panic in tanks with shelter. — Brown and Nixon (2004) Xenopus laevis Laboratory bred Shelter provision Provision of plastic tubes reduced aggressive en- counters, wounds and/or cannibalisation events. — Toreilles and Green (2007) Xenopus laevis Laboratory bred Shelter provision No effect on growth rates. Reluctant to leave shelter. — Gouchie et al. (2008) Xenopus laevis Laboratory bred Shelter provision No effect on growth rates or body condition (fat bodies). Higher propensity to clump together without shelter. — Archard (2012) Xenopus laevis Laboratory bred tadpoles 1. Surface area size 2. Water depth 3. Aquatic partitioning/ maze 1 . Reduced surface area increased air-breathing behavior 2. Shallow water reduced growth rates and caused abnormal floating behavior (tadpoles could not surface to breath properly) 3. Tadpoles avoided narrower passages (2 cm) and preferred wider ones (4 cm) Enrichments are not ecologically relevant to this spe- cies; this work may have limited impli- cations for captive husbandry Galich and Wassersug (2012) Xenopus laevis Laboratory bred females 1 . Shelter provision 2. Conspecific provision (always with shelter) 1 . Refuge provision reduced daytime activity and animals used shelter when provided 2. Addition of conspecific further reduced daytime activity in increased refuge use. No aggression observed and refuges were shared — Archard (2013) Lithobates catesbeianus Farmed/wild-caught Environmental com- plexity (ramps, perches and caves) Improved general welfare (general aspect and condition of animals) High density laboratory condition Bang and Mack (1998) Lithobates catesbeianus Farmed/wild-caught Shelter provision Reduction in mortality and improvement in condition High density laboratory condition Hedge and Saunders (2002) Dendrobates tinctorius D. azureus D. auratus D. leucomelas Mainly reported as aggregate data across species Zoo bred 1. Feeding enrichment (control vs. insect dispenser vs. broadcast feed/aphid stem) 2. Enclosure switch 1 . Some effects on behavior (mainly activity) 2. Effect on activity levels (enclosure switch lead to higher activity levels) Very small sample sizes. Issues with experimental design, includ- ing few replicates and unexplained measures Hurme et al. (2003) Oophaga pumilio Zoo bred Feeding enrichment (feeding dish control vs. feeding dish with leaf cover to allow insects to disperse) Increased foraging duration, increased duration between prey capture events and reduced rapid feeding — Campbell-Palmer et al. (2006) Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (8) June 2014 I Volume 8 I Number 1 I e77 Enrichment for amphibians Table 1. Studies of enrichment in amphibians (continued). Species Origin Type of enrichment investigated Findings Notes Reference Mannophryne trinitatis Wild collected as tadpoles 1 . Shelter provision 2. Substrate type 1 . Strong, positive effect on growth rates. No effect on behavior (weak effect on time spent jumping) 2. Preferred shallow water Substrate prefer- ence predicted by habitat Walsh and Downie (2005) Physalaemus pustulosus Wild collected as spawn 1 . No/weak effect on growth or behavior 2. Preferred dig-able (sand or gravel) substrate Leptodactylus fuscus Agalychnis callidryas Laboratory bred juveniles and adults Shelter provision Frogs prefer planted to non-planted enclosures. This preference increases when animals are deprived of plants before choice test. Froglets reared with plants grow faster and are in better condition than those reared without. Frogs reared with plants have more diverse and more abundant cutaneous bacterial communities. — Michaels et al. (2014b) Cryptobranchus alleganiensis Wild collected as eggs (head-starting program) Pre-release anti-preda- tor training Hellbenders were able to learn to exhibit a fright response to trout scent after classical conditioning; control animals showed no such improvement. — Crane and Mathis (2010) and to call for more research in order to better understand the importance of enrichment for this taxon. We will ex- plore the meaning of enrichment for amphibians, review the body of existing research (Table 1), and discuss the neglect of this field as well as how and why enrichment may be important as a focus for both amphibian conser- vation and welfare research activity. Finally, we will sug- gest a potential structure and goals for future research in this area (Table 2). Concepts of enrichment Enrichment for captive animals has been defined in vari- ous ways, but in general, is any intervention designed to improve animal welfare beyond the basic requirements for survival, usually taking the form of modifications to enclosures or husbandry protocols. Well known exam- ples include the provision of bamboo stems filled with grubs for captive Aye-aye (Daubentonia madagascarien- sis) (Quinn and Wilson 2004), running wheels for cap- tive rodents (Hutchinson et al. 2005) and the spraying of unfamiliar scents on parts of the enclosures for big cats; e.g., Szokalski et al. 2012 in tigers (Panthera tigris). Enrichment is often sub-divided into environmental, behavioral, and social categories. Shepherdson (1998) defined environmental enrichment as any intervention that provides “the environmental stimuli necessary for optimal psychological and physiological well-being.” This is distinct from behavioral enrichment, which is de- signed to elicit or allow the expression of specific behav- Amphib. Reptile Conserv. | amphibian-reptile-conservation.org (9) iors or behavioral repertoires (Shepherdson 1994). Social enrichment, the provision of access to other individuals (usually, but not always, conspecifics) to cater for social interaction needs (including both environmental and be- havioral components), has also been identified as impor- tant for a number of taxa (Berejikian et al. 2001; Lan- termann 1993; Miranda de la Lama and Mattiello 2010; Polverino et al. 2012; Saxby et al. 2010; Sloman et al. 2011; reviewed by Hayes et al. 1998 and Young 2003; see below). Enrichment can influence behavioral repertoires and stress levels beyond addressing stereotypical behavior and physical health problems (reviewed by Young 2003) and can affect physical brain structure in species as di- verse as mice {Mus musculus) and crickets {Acheta do- mestica) (Lomassese et al. 2000; van Praag et al. 2000). These findings have led to a current view of enrichment, which recognizes the importance of all three categories for the psychological as well as the physical welfare of captive animals (Dawkins 2006; Young 2003). The three forms of enrichment can be used to improve conservation success by training animals with the aim of improving survivorship upon release; e.g., anti-predator training in the black footed ferret {Mustela nigripes; Dobson and Lyles 2000). Although some forms of train- ing may be beneficial, the use of enrichment may result in conflict between maximizing individual welfare in captivity and equipping animals destined for release with the most appropriate survival skills (Caro and Sherman 2013; Harrington et al. 2013), and both objectives should June 2014 I Volume 8 I Number 1 I ell Michaels et al. Table 2. Key areas of species biology knowledge required for effective enrichment research, potential tools for assessing enrich- ment needs and effects and areas of amphibian captive husbandry for which enrichment may be important. Key areas of amphibian bioiogy, to be integrated into enrichment research Potential measures of welfare and fitness Potential areas of captive husbandry for enrichment research focus Cognition Learned and hard- 1 wired behavioral components | Catalogue existing issues in 1 captive amphibians and their husbandry | Enclosure design • Size • Complexity - Permanent (furniture and decor) - Temporal (novel objects, timed misting) • Refuges • Lighting - Wavelength - Photoperiod - Intensity Perception of environment 1 Behavior and behavioral assays 1 Behavior 1 Natural behav- ioral repertoires and i activity levels of 1 species 1 Foraging success 1 Environmental parameters • Gradients • Fluctuation (seasonal and diel) Foraging strategies ' and dietary compo- = sition 1 > 1 O 1 15 Growth and development 1 ^ 1 Reproductive ^ behavior g, • Breeding "g strategies ^ • Mate choice . o. • Competition for | -g mates/breeding S sites 1 Q. s-2- 1 ns